PREFACE

SRM University launched its first satellite, SRMSAT in collaboration with Indian Space
Research Organisation on 12 October 2011. SRMSAT has completed 23000 orbits, being operational,
on-orbit for more than 5 years. Since the design and development of SRMSAT, the work on space
technology development has continued under the Department of Research. Under the administration of
Director (Research), Prof. D. Narayana Rao and project architect Mr. M. Loganathan, the project was
conceived in its preliminary formulation in late 2014.

Assessing possibilities and prospects of a challenging small-satellite mission, Moon was chosen
as the target. Moon, the nearest neighbour and scientifically crucial part of solar system exploration
presented with an excellent mission possibility, with an orbiter type configuration. SRMSAT-2, a lunar
orbiter mission was preliminary developed with priority on technology demonstration of extending
small-satellites’ applications beyond Earth. Priority was set on unconventional trajectory design by
choosing a “Ballistic Capture” complemented “Low Energy WSB Transfer Trajectory” which would
decrease propellant requirement and also increase the flexibility of lunar transfer with compensation of
longer transfer time. With baseline set, the design of SRMSAT-2 began with a major focus on multi-
body trajectory dynamics for low energy transfer, a propulsion system for orbital maneuvers and
communication system for reliable deep space communication. Constraints and possibilities were
analysed, and primary feasibility analysis was done, and the generic mission profile was given a “go”
for preliminary design. With the conclusion of the preliminary design phase, astrodynamic feasibility
has been analysed, payload options for lunar on-orbit science have been chosen, and a comprehensive
preliminary systems design has been completed.

This Design Report, “SRMSAT-2: THE MISSION TO THE MOON” consolidates, the project
and mission overview with objectives. The mission profile and spacecraft specifications in current
design are comprehensively covered followed by Mission requirements on system and sub-system level.
Currently, science payload options and their accommodation and implementation on-board the
spacecraft has been comprehensively laid down in the “Payload” section featuring a Microwave
Radiometer, Radiation Dose Monitor and Narrow Angle Camera among the options aimed at acquiring
low to medium resolution geological/photo-geological and morphological data of the Moon which
forms a crucial part of the requirements of the global lunar science community.

Aspects of synchronised mission analysis and systems engineering for the current design are
described in “Design: Drivers and Methodology” and complete highlights, facts, figures and
specifications of elements of current design are described in concluding sections. The document then
complements the design reports of work under each subsystem of Spaceflight Dynamics, Astrionics,
and Mechanical systems divisions.

A total of 50 undergraduate students, from various disciplines of Engineering and Technology,

contributed to the project over the span of 2 years, with a current work force of 25 students with positive
support from the University administration.

1
 1. BACKGROUND
1.1 SMALL SATELLITES
Much of the interest in small satellites stems from a desire to "do more with less" with least possible
cost owing to large-scale budget realities of space missions. Small satellites (upto 500 kg) enable
demonstration of low-cost space application. This benefit is derived from the relative simplicity of the
small missions (or the pre-existence of mission elements) and from the size of the satellite. Small
missions generally have less stringent requirements as measured by performance, calibration, or
longevity. Less complex missions require shorter development time, which goes to the heart of lower
costs. Because simpler missions may also be less capable, the science and operational needs must be
carefully evaluated to ensure that they are adequately addressed. One of the emerging benefits of small
satellite missions is a reduction in the "time to science." Large, complicated missions take many years
to conceptualise, develop and implement during which time they lose on both scientific understanding
and evolved technology. In addition, an increasingly cost-constrained environment makes large
missions especially vulnerable to budget instabilities which often hampers the scientific and
technological benefits associated with the mission. In this case, small satellites come as unique
architecture to support application specific mission scenarios.

In terms of science aboard a small satellite platform, acquiring large amounts of data albeit
lower quality can provide a more robust estimate and the scientific observation requirements can be
reasonably met with low-to-medium resolution data. Further, Small satellites have become increasingly
important in the context of capacity building in space technology for developing nations. Coupled with
realistic and focused goals with the available off-the-shelf technologies, such satellites make it possible
to flexibly adapt their development, launching, and operation, providing faster and cheaper access to
space.

The main aim for small satellites is to set up specialised observational units or serve as a platform for
testing the viability and reliability of the new and advanced technology. Newly developed science can
be easily employed to check it’s outcomes in Space. Thus, Space Exploration becomes a more
approachable concern for all kinds of institutions owing to the ease of developing small satellites.
Moreover, it is seen that the failure to cost ratio for small satellites is way less than that for large
satellites. This provides a larger reassurance on the scope of possible eminence of Small satellite
missions

Emphasising its importance in a developing country like India is not only necessary but also
elementary at all levels of development. Indian Space Research Organisation with its established state
of the art space technologies has been at the forefront of global space science, technology, and
exploration scenario and is one of the most reliable space systems developer and launch provider. It is
necessary that in a country like India with a history of student-designed satellite successes, that student
developed and led space missions and applications can advent beyond conventional space where the
ideal candidate for development are small satellites.

2
 1.2 SRMSAT
Introduction
SRMSAT is a nanosatellite project with a primary
objective of monitoring the greenhouse gases and water
vapor levels in the atmosphere using near infrared (nIR)
spectroscopy. The project was a student-led mission and
was designed and undertaken by a team of nearly 40
undergraduate students from the Engineering and
Technology Dept. of SRM University, Chennai. The
satellite was launched into space by ISRO in PSLV C-18
as a piggyback and obtained a stable, nearly circular low
earth orbit, of altitude of 867 km in 2011. It is a 20°
inclination LEO (Megha-Tropiques) .The payload used to
carry out the scientific objective was Argus 1000
spectrometer. The mission objective was to provide
hands-on experience to the students and the faculty of
SRM University.

The satellite was designed with a mission life of 12

months but is still in orbit, having completed 23000 Figure 1.1 : SRMSAT
orbits. A ground station that contains two Yagi –Uda
antennae along with LNAs and transceivers was established on the terrace of the University Building
for independent communication with the satellite. The communication system employs an amateur V-
U band. The satellite like most small satellite uses COTS components. The primary structure of the
satellite is completely made of Al – 6061.

The satellite consisted of 7 subsystems namely, Structure, Thermal, ADCS, Telemetry, Power, OBC
and Ground Station. The ADCS (Attitude Determination and Control System) uses a magnetometer as
the primary attitude sensor and SCDM as secondary sensors. It uses magnetorquers for minor
adjustments made in the spacecraft attitude. The On-board computer (OBC) consists of a
microcontroller which uses 32- bit ARM- 9 architecture. The structure of the satellite was designed
such that the PCBs were removable. The satellite is powered by three solar panels mounted on three of
the four sidewalls with a LI-ion battery for storage and backup during eclipse periods. An ambient
temperature of 0-40°C was maintained throughout the operation times for thermal control of the satellite
and its components. The structural was approved during the first trial of testing with a vibrational
frequency of 115 Hz, satisfying the requirement of a mere 45 Hz for quasi-static launch loads.

The structural bus of the SRM Nanosatellite has a payload carrying capacity of 20 kg and 5 kg on the
bottom plate and the top plate respectively. It can mount up to 7 PCB trays in the split structure. The
net volume is around 11 liters. A monopole helix antenna is mounted on the bus. It uses a GPS (Global
Positioning System) for tracking. The system overview has been provide in the table given below:

3
 Table 1.1: SRMSAT Specifications

Orbit LEO 200 inclination (Megha-Tropiques)

Time Period 102 minutes
Passes 14 orbits/ day with 6-7 visible passes
Altitude 867 km
Positioning GPS
Power Generation 10 W (average)
Dimensions 286 x 286 x 288 mm3
Mass 10.08 kg
Stabilization 3 axis
Design Life 12 months (minimum)
OBC 32 – bit processor with ARM 9 architecture
EEPROM, Flash and NVS RAM
ADCS Magnetometer HMR 2300
3 Magnetorquer
SCDM and GPS
Thermal Control Passive thermal control using MLI, Conformal
coating, and Black Paints

4
 2 THE MISSION TO THE MOON

Moon, the nearest celestial body, has been a center of observation for centuries. It presents a window
to the origins of the solar system and bears undisturbed evidence of processes and phenomena that are
crucial to our understanding of the Moon, the inner planets and the whole solar system in general. Also,
being the nearest target bound in the sphere of influence of Earth, it makes the best technology proving
ground for deep space exploration. A large part of the global roadmap for exploration relies on
comprehensive robotic- orbiting and landing, and human exploration, and Moon, the ideal choice for a
combined Space Science and Technology demonstration. Lunar orbiters can be crucial specifically, in
characterising landing sites, evaluating and mapping Resources, and observing phenomena on the
Moon. With survey and background study of past and present missions and the lunar exploration
roadmap, ‘technology demonstration of a lunar transfer’ and ‘lunar science aboard a small satellite’
were chosen as the design driving baselines. Proving mettle of in-orbit science by small spacecraft could
revolutionize Space Exploration, just how small satellite applications have revolutionised Earth
sciences and observation.

Apart from crucial possibilities of lunar science, Moon, in the sphere of influence of Earth provides an
additional advantage for a small satellite. Chaotic regions of critical gravitational interactions between
Earth and Moon along with Sun, i.e. Lagrange points of Earth-Moon and Earth-Sun offer various
dynamic regimes of motion. This can be exploited to save propellant required and thus the mass as a
whole.

While evaluating various transfer types like conventional Hohmann transfer, Bi-elliptic Transfer, 3-
body resonance transfer and WSB transfer, it was found that most optimally used Hohmann transfer
would be costly in terms of propellant as well as too fast, especially with respect operations depending
on small-satellite components with reliability levels below normal. Other transfers were eventually
classified infeasible and discarded, and the WSB transfer type trajectory profile was chosen. First,
practically applied in JAXA’s HITEN mission, WSB trajectory, considered for the mission, saves up to
15-20% of the transfer ∆V. Also, the dynamical flexibility of the problem governing the motion offers
various possibilities including inherent inclination change, saving on huge plane change maneuvers.
Keeping in considerations, all the constraints of small satellites, a trajectory design as shown in the
mission profile are fixed.

5
Project Objectives
 To design a university-class microsatellite, completely by undergraduate students of SRM
University, for a lunar orbiter mission.
 To provide first-hand experience to students in space mission design and analysis.
 To induce student research in planetary/interplanetary small-satellite technologies
 To prove capabilities of small satellite systems beyond Earth.
 To demonstrate the value of small-satellite science platform for lunar/planetary exploration.
 To execute efficient administration-student-industry interactions for optimal mission output
 To Foster awareness and developments of space science and technology.
 To collaborate with, as well aid the global space science community with acquired lunar science
data.
 Efficiently execute a multidisciplinary project with students from various STEM backgrounds
 To effectively coordinate and utilize knowledge base at individual, subsystem and system level
in the team
 Prepare students to become valuable resources to the Aerospace community
 To develop spacecraft and mission design infrastructure at SRM University.
 To develop a generic space exploration architecture and design platform for missions beyond
Earth by small-satellites
 To lay down a path for future small-satellite missions in lunar/planetary exploration
 To update current university tracking station to S-band capabilities for communication in Earth
Phasing Orbits

Indian Space Research Organisation’s launch support along with IDSN support for Telemetry,
Tracking and Command is aspired and elementary to the mission.

For the mission operations, support with respect to Space Link Extension (SLE) for interfacing
Flight Dynamics facility, Data processing center and mission operation centre established at SRM
University, with ISTRAC and IDSN stations, is aspired.

A primary data center (SRM Lunar Science Data Center) will be established to collaborate with
DSN and University tracking station with capabilities to support handling of downlinked science data
and, maintaining and regulating data archives. It will be responsible for interfacing with the scientific
community and students of SRM University in providing acquired science data.

NOTE: Considerations in the design report have been made for launch similar to Chandrayaan aboard
PSLV and ISTRAC and IDSN ground systems for TT&C.

6
 2.1 SPACECRAFT SUBSYSTEMS
For, preliminary design, the following classification was made for the formulation of mission and
spacecraft systems.

Figure 2.1: Spacecraft Subsystems

2.2 SPACEFLIGHT DYNAMICS DIVISION

Trajectory Design and Mission Analysis:
This subsystem, though not physically associated with the spacecraft, is associated critically with
spacecraft design. The subsystem’s preliminary work is to choose a transfer profile to achieve final
mission orbit and lay out total propulsion requirements. The Mission analysis segment is responsible
for checking the astrodynamic feasibility of each solution in a dynamic envelope of possibilities in
terms of analyses of error sources, risks involved, injection constraints and sensitivity.

GNC

Guidance Navigation and Control is responsible orbit determination, prediction, and control. The
primary task of the subsystem would be to give preliminary and precise OD to support all phases of
trajectory within bounds. The subsystem will be responsible for the pre-flight design of optimal

7
deterministic maneuvers and in-flight design of statistical maneuvers for trajectory/orbit maintenance
and control within capture corridors for the transfer trajectory. GNC will comprise of a Ground Flight
dynamics segment with personnel and infrastructure to support the mission during its lifetime to suffice
successful transfer, Lunar Orbit Injection, and Stationkeeping.

ADCS

The ADCS subsystem is responsible for determining and controlling the spacecraft attitude in space.
The main functions of the ADCS subsystem can be divided into Determination, Prediction, and Control.
Attitude determination is the process of computing the spacecraft orientation, within a pre-defined
accuracy, using the data obtained from the on-board sensors. Attitude prediction is the process of
forecasting the future attitude using dynamic models, and attitude control is the process of orienting the
satellite in a specified direction using actuators and momentum exchange devices.

Spacecraft Mechanical Systems Division

Thermal

The thermal subsystem mainly aims at regulating the thermal conditions of the internal components
satellite. It controls the internal conditions of the satellite and analyses the effects of the space
environment, to prevent large temperature changes and radiation damage to electronics. The system
design mainly consists of passive thermal control with the addition of heaters if needed for active
control.

Propulsion

The propulsion subsystem basically provides thrust to the satellite for controlled movement by
providing acceleration. It consists of a primary and a secondary system, used to maneuver in space and
change the attitude of the satellite body. It is essential for satellite missions which are not earth bound.

Structure

The structural subsystem of a space mission design is responsible for the mechanical design of all
components and the Bus configuration to accommodate all components. This subsystem is also
responsible for ensuring the structural stability of the satellite against quasi-static loading factors and
excitations of the Launch Vehicle.

Astrionics Division
OBC

The on-board computer subsystem is the brain of the satellite which is responsible for platform data
processing, payload data handling, and interfacing with components of other spacecraft subsystems.
The software architecture provides a functional overview of the entire on-board software system. The
software is developed around a microcontroller that plays a crucial role in computation, scheduling and
executing various tasks, organizing and collecting telemetry data, receiving and responding to
telecommands.

8
Power

The Electrical Power System generates, stores, conditions, controls, and distributes power within the
specified voltage band to all bus and payload equipment. The Power subsystem is required to supply a
continuous source of electrical power to spacecraft loads during the mission life, distribute power,
provide switching interface for OBC and protect components from electrical faults.

TTC

The Telemetry, Tracking, and Command (TT&C) or communication subsystem provides the interface
between the spacecraft and ground systems. The data collected by payload equipment is transmitted to
ground station through this subsystem along with housekeeping data. TT&C performs functions like
Carrier tracking, Command reception and detection, telemetry modulation and transmission and
ranging.

9
 3. THE MOON

The Moon, our nearest celestial neighbour, is of crucial scientific importance and is the base
target for space exploration because it represents a body frozen in time that in itself, has records of some
of the earliest processes in solar system evolution (e.g., planetary differentiation, impact flux, and
processes). The oldest material excavated from the surface of the Moon by the lunar missions is
estimated to be around the same age as that of the solar system; 5 billion years. The Moon represents a
celestial body preserving the unique history of planetary formation, space environment, and cosmic
radiations for years, which has not been erased by weathering or any other evolutionary processes as in
the case of the Earth. There is no atmosphere on Moon and a subsurface abundance of igneous (fire
formed) rocks. Seeing through the early history of chemical segregation, bombardment and volcanism
to the Moon’s primordial origins is one of the prime objectives of space science. While most of the
lunar terrain is dry and dusty, the most primary evidence of lunar segregation is the existence of two
distinct classes of terrain: the Highlands and the Maria.

Figure 3.1:

During its early evolution, the severe bombardment of meteorites and other small bodies formed the
highlands, which are intensely cratered regions of greater heights and sharp brightness. Highlands
comprises of ‘Anorthosite’ (slowly cooled igneous rock) and ‘Breccias’ or fragments of different rocks
compacted and welded together by impacts. Unlike mountains on the Earth, mountainous highlands did
not result from plate tectonic type crustal dynamics. Maria are basaltic regions, smooth and plain in
nature, visible as dark spots on the Moon, due to less reflectivity, formed by historic volcanic activity.
These constitute 16 % of the Moon’s surface. These are also associated with significantly noticeable
gravitational anomalies called ‘Mascons’ These mares more in number towards the near side than the
far side. These regions contain higher iron and Ilmenite abundances

The stratigraphy of the Moon comprises of the top ‘rolling’ layer of powdery soil and fragments of
moon rocks called the ‘Regolith.' The source of regolith formation is linked to the meteoritic impacts
which formed the craters. The Moon has either a very small core rich in iron ore or no core at all. It is
predicted that the lower mantle is rich in Ortho-pyroxene (opx), Olivine and extends up to 1000kms.
The compounds are Fe-Mg-rich silicate mineral, Cliono-pyroxene (cpx) – Ca-rich silicate mineral with

10
Fe and Mg Plagioclase –an Al-rich mineral. Quakes inside Moon occur in the band around 800 -
1000km in depth, a level that is interpreted as the base of the lunar lithosphere. Seismic S waves
apparently do not traverse the region below the zone of Moonquakes.

Figure 3.2:

Moon Earth

Mass 7.353*1022 kg 5.976*1022 kg

Radius(spherical) 1738 km 6371 km
4 3
Surface area 37.9*10 km 510.1*104 K\km3
(land = 149.8*104 Km2)
Flattering 0.0005 0.0034
Mean density 3.34 g/ cm3 5.517 g/ cm3
Gravity at equator 1.62 m/sec2 9.81 m/sec2
Escape velocity at equator 2.38 Km/sec 11.2 km/sec
Sidreal rotation time 27.322 days 23.9345 hr
Inclination of equator/orbit 6041’ 23028’
Mean surface temperature 1070C day; -1530C night 220C
Temperature extremes -2330C(?) to 1230C -890C to 580C
Atmosphere ̴104 molecules/cm3 day 2.5 * 1019 molecules/cm3(STP)
2*105 molecules/cm3 night
Moment of inertia(1/MR2) 0.395 0.3315
Heat flow (average) ̴29 mW/m2 63 mW/m2
Seismic energy 2*1010 (or 1014?) J/yr 1017-1018 J/yr
Magnetic field 0 (small paleofield) 24-56 A/m

Table 3.1:

11
 3.1 ORIGIN OF THE MOON

The theory of the origin of the Moon has been a consistent element of speculation for a long time with
a number of hypothesized theories discarded since its inception as a study. The earlier missions hoped
to get complete information to formulate a consistent theory of the origin. However, it was realised that
it was only possible if the Moon were still a primitive, undifferentiated homogeneous body. However,
differentiation of Moon into zones of chemical distribution imposed over active geological past has
obscured the consistencies required for the origin theory. It was clear that the celestial data (motion),
orbital properties (gravitational influences) and dynamical constraints (shape, gravity, a moment of
inertia, etc.) would not suffice in the formulation of the Origin of the moon.

Studies using chemical, mineralogical, isotopic and chronological data led to the postulation of five
major theories on the origin of the moon.

 THE FISSION THEORY:

It states that the Moon separated from the Earth at some point in the distant past. It is supported
with an assumption of irregular geometry of the Earth and thus based its explanation on the
imbalance that caused it to split in two.

 THE CAPTURE THEORY

Considers Moon as a separate entity which formed somewhere in the solar system, later captured
by the gravitational field of the Earth.

 THE CO-ACCRETION THEORY:

The Earth and the Moon may have been formed at the same time from the solar nebula by co-
accretion.

 THE COLLIDING PLANETESIMAL THEORY:

The Moon is a product of condensation of debris from the interaction of Earth-orbiting and Sun-
orbiting planetesimals.

 THE GIANT IMPACT THEORY:

A celestial object (Mars-sized) impacted Earth, early in its history, ejecting large volumes of matter
from the evolving Earth, with aggregated and formed the Moon.

Although every new reformed/combined hypothesis goes a step beyond towards reconciling the strong
points and solving the dilemmas in dynamics, chemistry, and geophysics, there is still a need for further
data acquisition to answer the question of the origin thoroughly. This one issue was taken into
consideration while developing the science mission objectives. Also, in explaining the events and
developing several models, major assumptions were required to be made. Thus, a study structure was
formulated to focus on identifying important geological features and processes on the Moon, its
importance in the evolution history and inconsistencies with developing models. This lays requirements

12
for the science data, which would be relevant in forming a steady explanation of the current state and
the historic evolution linked to it.

3.1 GEOLOGICAL PROCESSES ON THE MOON

Meteorite impact is a major transport process on the Moon, because it redistributes materials both
laterally and vertically, mostly by sedimentation. Depending on the size of the event, impacts can
pulverize surface boulders, shatter subsurface bedrock, or even dislodge crustal sections tens of
kilometers deep. The impact is also the most important metamorphic process on the Moon, causing
characteristic shock deformation, heating, melting, vaporization, and even ionization of the pre-existing
minerals and rocks. Impact processes alter the textures of rocks and generate new ones, such as
pyroclasts, impact melts, and fragmental rocks (breccias) containing either a single rock type
(monomict) or multiple rock types (polymict). The ejection of material from impact structures called
lunar ejecta, and the subsurface effects beneath them, combine to produce a lunar crust and surface
composed of lithologically diverse impact deposits. Some ejecta deposits from individual young impact
basins are distinctive enough to be mapped as discrete geological formations and stratigraphic units;
they have distinct surface characteristics and well-defined structural/temporal relations.

Processes And The Regolith

Regolith is the top layer in the lunar stratigraphy. One of the minor geological processes linked to this
on the Moon is lunar rock erosion. This erosion is due to ‘abrasive’ effect of small impacts of a
micrometer to centimeter size particles, leading to gradual erosion and fragmentation of lunar rocks.
This is slow but nonetheless a significant geological process typically linking erosion of rocks and the
small-scale evolution of the lunar surface and the ‘gardening’ layer of the Regolith. Gardening process
gradually over the time develops a fine-grained powdery layer which is continuously ruffled up or
‘gardened’ to form homogeneously mixed top layer of the regolith. Below this layer of regolith is the
layer of actual bedrock slabs. When developing regolith models, it is seen that although gradual
geological evolution and gardening sustain the homogeneity, a crucial factor influencing stratigraphic
inconsistencies in their history i.e. layers, only minor depth apart can reveal different historic and
depositional history. This is due to infrequent large-scale regolith evolution by centimetre to meter big
particles. Thus, a sampling of regolith is necessary, if done, be at a depth of few meters containing
consistent deposition history directly linked with the gradual process on the underlying bed-rock.

Large-Scale Bombardment Of The Highlands

The epoch of intense bombardment that occurred in early lunar history has little or no experimental or
observational constraints, and major assumptions must be made in their development. There are also
uncertainties about the shapes and sizes of the transient crater cavities and the volumes of material
excavated from them. These give rise to divergence in the model and thus, inconsistencies in the
explanation.

13
Lateral Mass Transport
Models for impact craters, generally, have ejecta spread centro-symmetrically around the crater. The
models are based on photogeologic studies of the “continuous” deposits around lunar craters.
Unfortunately, the ejection velocities for the more distant “discontinuous” ejecta are not sufficiently
known to be meaningfully incorporated into these models. This omission is a significant flaw in these
models, for the discontinuous deposits clearly consist of high-speed ejecta, and the amount of such
material ultimately determines the transport efficiency of the impact event at large distances from the
crater. Distant ejecta is also geologically significant because it may distribute “exotic” components
across the Moon (or even farther away). At the global scale, our views on the homogeneity or
heterogeneity of the present lunar crust also depend on the efficiency of lateral mass transport.

Thus, science data is still needed on various aspects, of properties of lunar surface/ sub-surface to assist
the accurate formulation of stratigraphy, depositional and compositional history and evolution.

Figure 3.3:

Basaltic Volcanism
The presence of volcanic activity and lavas on the surface of a planet reflects the planet’s thermal state
and its evolution. Unless the planet somehow generates high enough internal temperatures to melt rocks,
volcanic activity may never begin or, having begun, may not last long. On the Earth, melts form in the
mantle, the deep layer that underlies the Earth’s outer crust; the necessary heat is provided by the decay

14
of natural radioactive elements. The mechanism of melting in the Earth’s mantle can be by the
movement of large crustal plates that drag down (subduct) crustal rocks to lower depths where they can
melt. Subduction does not occur on the Moon, where all lunar volcanism apparently resulted by melting
of mantle rocks mixed with little or no recycled crust. ‘Partial Melts’ are formed by heating of mantle
materials. These do not have much correlation with the composition of the original mantle. These partial
melts, called basalts have a lower level of SiO2. With high MgO and FeO content. For the necessary
melting and basaltic volcanism to occur on the Moon, temperatures of >1100°C at depths of 150–200
km are required. The mere existence of basaltic lavas at the lunar surface, therefore, provides important
data about past temperatures within the Moon. The duration of lunar volcanic activity is related to the
thermal state of the mantle and thus to the thermal evolution of the Moon. Current data indicates that
major lunar volcanic activity ceased at approximately 3 b.y. ago, implying that mantle temperatures had
then dropped below those necessary to sustain partial melting.

All of the smooth, dark regions visible on the Moon’s nearside consist of lavas that partly or completely
fill the multi-ring basins. Studies of lunar basalts are important for determining the composition of the
Moon’s mantle. Because they are “partial” melts, the lunar basalts are chemically complementary to the
residual unmelted mantle materials left behind. Using the thermodynamic principles involved in
equilibrium partial melting, the complementary residue may be characterized, and the original bulk
composition of the Moon’s mantle may then be calculated.

Not all products of basaltic volcanism are necessarily lava flows. “Fire is fountaining,” driven by gases
emanating from the magma, occurs during basaltic lava eruptions, dispersing the melt as fine droplets.
These droplets form pyroclasts or volcanic ash, that may be spread over large areas. Pyroclastic
materials can be identified by remote-sensing techniques, especially by infrared reflectance spectra that

Figure 3.4:

15
reveal their glass-rich nature. Pyroclastic deposits on the Moon are widely dispersed and are readily
worked into the regolith by impact gardening; identification of these deposits in surface photographs is
possible only at a few well-preserved locations. Filling of the Maria Basaltic volcanism and large-scale
impact basins are spatially related to the Moon. Basin formation, followed immediately by the collapse
of the transient basin cavity, leads to large-scale fracturing of the crust and thus provides both the
structural framework and the conduits along which basaltic magmas generated at depth may ascend to
the surface. Most such conduits seem to be located just within the edges of the mare basins, where large-
scale slump zones should be prominent. Deep melting can be produced by several mechanisms
associated with the formation of large impact basins. The lower thermal conductivity of thick basin
filling ejecta blankets, which is related to their high porosity, can produce significant upwarping of
mantle isotherms, thus promoting localized melting below a basin. Other factors include the actual
physical uplift of mantle materials during basin formation, the decrease in hydrostatic pressure caused
by the sudden removal of overburden, and the resulting decrease in melting temperature with decreasing
confining pressure. In addition, massive, unfractured sheets of fresh impact melt produced during basin
formation, possibly a few kilometers deep, may act as an impermeable plug that forces the ascending
lavas from the center of the basin to its edges. Significant time may elapse between the formation of a
large mare basin by impact and its subsequent filling with basaltic lava flows to form the dark lunar
maria. Current information fixes the formation of the Imbrium Basin at about 3.9 b.y. ago. However,
despite violent and intense early lunar history, geological activity has remained relatively dormant. The
heavily cratered, more active Moon that existed before >3.8 b.y. ago is more difficult to understand.
Impact cratering clearly dominated these early surface processes.

Improved understanding of the lunar craters and surface is needed not only for lunar stratigraphy, but
the data would also help us understand the intense collisional environment that must have affected all
the planets in the early inner solar system, including Earth.

Other science prospects considered while developing the lunar science objective was scanning/mapping
Olivine and Pyroxene. The first important event in the geologic evolution of the Moon was
the crystallization of the near global magma ocean. It is not known with certainty what its depth was,
but several studies imply a depth of about 500 km or greater. The first minerals to form in this ocean
were the iron and magnesium silicates, olivine and pyroxene. Because these minerals were denser than
the molten material around them, they sank.

The silicate minerals, especially pyroxene and olivine, are the most abundant minerals in rocks of the
lunar crust and mantle. Meteoroid impacts over time have broken up and pulverized the lunar bedrock
to produce a blanket of powdery regolith several meters thick, which forms the interface between the
Moon and its space environment.

Pyroxene

Pyroxenes are the most chemically complex of the major silicate phases in lunar rocks. They are also
informative recorders of the conditions of formation and the evolutionary history of these rocks.
Pyroxenes are compositionally variable solid solutions, and they contain most of the major chemical
elements present in the host rocks.

The pyroxene crystal structure, which basically consists of chains built up from linked silicon-oxygen
tetrahedra, combined with metal-oxygen octahedra. Oxygen atoms define the corners of all of the
polyhedral sites, and the cations (Si and other metals) are located inside the polyhedra. The structure is
composed of octahedral layers containing infinite chains of edge-sharing bands of six-cornered
(octahedral) polyhedra (called the M1 sites); the chains run parallel to the crystallographic c-axis. These

16
chains are cross-linked by distorted six-cornered octahedra or larger eight-cornered polyhedra
(collectively called the M2 sites). This M1- M2 layers are in turn separated from each other by layers
composed of infinite chains of silicon-oxygen tetrahedra that also run parallel to the crystallographic c-
axis.

One of the first-order differences between terrestrial and lunar pyroxenes is the abundance of Fe3+ in
the former and its absence in the latter; this difference reflects the more reducing conditions (lower
oxygen partial pressure) in gases on the Moon than on Earth. Some compositional differences between
lunar and terrestrial pyroxenes.

Lunar pyroxenes, including those from both mare basalts and highland rocks, have great chemical
variability. They also differ in several respects from terrestrial pyroxenes. Pyroxenes from terrestrial
ocean floor basalts, which are the most abundant basalts on Earth, are more magnesian than those from
lunar mare basalts, and the lunar mare basalt pyroxenes are Fe2+ rich. this difference implies that the
source region in the Earth’s mantle is significantly enriched in Mg relative to the source regions for
lunar mare basalts.

Olivine

The crystal structure of olivine, (Mg,Fe)2SiO4, consists of serrated chains formed of edge-sharing
octahedra. These chains run parallel to the crystallographic c-axis (Fig. 5.10). The octahedral chains are
cross-linked by isolated SiO4 tetrahedra. The major cations in the octahedral sites, Fe2+, and Mg are
distributed with a high degree of disorder (randomness) over both the M1 and M2 octahedral sites.
However, the small amounts of Ca that may occur in olivine occupy only the M2 site.

Two main differences between terrestrial and lunar olivines are present. First, although significant
Mg:Fe variation is present in both olivine groups, lunar mare olivines include more iron-rich
compositions. Second, the most magnesian terrestrial olivines are more magnesian than the most Mg-
rich lunar olivines. As with the lunar pyroxenes, this difference almost certainly reflects more
magnesian composition of the Earth’s mantle relative to that of the Moon

Helium And Ilmenites

The Moon has abundant resources of oxygen, hydrogen and other solar wind gases trapped in its
regolith. Also, there is a relative abundance of Helium-3 isotope on the Moon compared to that of
Earth. 3He can be used as a fusion element and is thus considered as one of the most important fuels for
power generation in the future. Since 3He has high diffusivity, it normally gets lost from the lunar grains.
However, the mineral Ilmenite (FeTiO3) is abundant on the Moon and has high retentivity for 3He. The
distribution of 3He associated with Fe and Ti can be determined by geochemical mapping since it would
have the same distribution as (Fe+Ti). Over the four billion year history of the Moon, several hundred
million tonnes of 3He have impacted the surface of the Moon from the solar wind. The analyses of
Apollo and Luna samples showed that over 1 million tonnes of 3He still remain embedded in the surface
of the Moon. Even a small fraction of this could provide the world’s electricity for centuries to come.
A large number of studies are being carried out to determine the technical feasibility of utilising this
resource in-situ and provide baseline architecture for human exploration platforms on the moon. The
abundance of 3He / Ilmenites can be inferred from thermal and dielectric properties combined with
microwave brightness temperature readings from the sub-surface levels. The microwave radiometry

17
payload currently decided presents an excellent scope for relevant science, which will aid the lunar
science community with respect to the Ilmenite abundance mapping from acquired data.

Radiation Environment Around Moon

The space around the Moon contains many types of ionizing radiation: large fluxes of low-energy solar
wind articles, smaller fluxes of high-energy galactic cosmic rays, and rare but occasionally intense
particle fluxes emitted by solar flares. The radiation fluxes and energies typically span at least eight
orders of magnitude. The lunar radiation environment also changes with time, usually reflecting the
level of solar activity, such as the modulation of galactic cosmic rays and the irregular emission of
energetic particles from the sun. Three major types of radiation at the Moon are discussed here: the
solar wind, solar-flare-associated particles (also called solar energetic particles or solar cosmic rays),
and galactic cosmic rays. The radiation consists mainly of protons and electrons with some heavier
nuclei. These particles interact with the Moon in different ways, depending on their energy and
composition, resulting in penetration depths that vary from micrometers to meters. The results of
interactions with exposed lunar rocks and soils also vary considerably; the effects include solar-wind
implantation, heavy-nuclei tracks, spallation reactions, and the generation of secondary neutrons and
gamma rays.

3.2 PAST MISSIONS

The Japanese mission, HITEN (formerly called MUSES-A) was successfully launched in 1990, to
perform a sophisticated Earth-Moon circumnavigation, the first practical application of WSB transfer.
In 1994 Clementine mission, equipped with a laser image detection and ranging system (LIDAR) and
high-resolution cameras (HIRES) that photographed nearly the whole lunar surface in UV, visible, NIR
and LWIR bands from a 425km lunar orbit. It provided the first global data sets for lunar gravity,
topography, and multi spectral imaging with about 200m resolution. Lunar Prospector launched in
January 1998 carried remote sensing instruments such as a gamma-ray spectrometer, neutron
spectrometer, alpha particle spectrometer, etc. This mission designed to provide answers to
longstanding questions about the Moon, its resources, structure and the origin and provided valuable
scientific data on the distribution of Thorium, Potassium, other radioactive and stable elements eg., Fe,
Ti, etc. Radar reflectivity experiments performed by Clementine hinted at the possibility of the existence
of a large amount of water frozen on the permanently shadowed polar regions of the Moon. Lunar
Prospector’s neutron spectrometer detected bursts of slow neutrons over the Moon’s poles, suggesting
the presence of hydrogen atoms and hence the possible presence of water/ice. Chang’e orbiter and
landers provided invaluable information to the scientific community with unmatched resolution data
sets.

Chandrayaan-1, the first Indian mission to Moon, played a major role in reinvigorating research in
fundamental science, helped in upgrading technological capabilities for future space systems, and
became a meaningful contender in the international space arena for exploration and commercial
exploitation of resources in the immediate neighbourhood of our planet. Chandrayaan-1 mission was
instrumental in passing on the baton to sophisticated future lunar and planetary missions with possible
landing and sample return capability. This needs to be further supported by an infrastructure of small
satellite missions paving application-specific and cheaper access to the moon. SRMSAT-2 was

18
conceived with a vision to kick start an era of such missions with emphasis on role educational
institutions and academia to support in different ways, the Indian space, and exploration program
With a preliminary study on the Moon and scope of Lunar Science, with respect to the global roadmap
for future Lunar Exploration strategies. A converging analysis was done on the various basis of lunar
science possibilities and requirements.
In the broadest tier of requirement analysis, any mission to moon would need to satisfy one or more of
the following to be relevant :

 Prospects of human Exploration and life sustenance (Bio-astronautics)

 Provide advancement in exploration (Geodesy)
 Polar Resource assessments and evaluation (Volatiles and prospects of ISRU)
 Landing site selection and safety (Characterization)
 Assist scientific data of astronomic importance (Origin of Moon/SS)

A general exploration architecture to establish partial duplication of operations on a small satellite level
can be carried out with scientifically relevant data acquisition on the basis of:

 Geodetic global topography –

 Medium spatial resolution Hydrogen mapping
 Temperature data in polar shadowed regions
 Imaging of surface in permanently shadowed regions
 Assessment of meter and smaller scale features for landing sites
 Characterization of polar regions
 Definitive identification of the composition, abundance, and distribution of volatiles in lunar
PSRs.
 Characterize sub-surface thermal and dielectric properties
 Polar region resources assessment (and associated landing site safety)
 High spatial resolution global resource assessment
o Elemental composition
o Mineralogy
o Regolith characteristics

Detailed measurement sets important to lunar exploration community: (Mission goals to be selected
from scaled down versions of the following)

I. Characterize deep space radiation environment and impact on the physiology of living systems
II. a. Global topography with 10-m vertical accuracy at 3-km equatorial cross-track and 30 m
along-track sampling
b. Global orbit determination to 100 m along- and cross-track, 10 m radial (only nearside
achievable without sub-satellite).

III a. Characterize surface morphology in regions of permanent shadow at ~50 m spatial resolution.

III a. Characterize abundance of hydrogen within the upper 1 m to 20% accuracy and ~5 km
resolution, with 100 ppm detection limit.

19
III a. Characterize the temperature, from 35-200 K, of the polar cold traps to 1 km spatial resolution
and 5 K precision.

III b. Identify putative deposits of appreciable near-surface water ice in polar cold traps at ~100 m
spatial resolution.

III c. Measure illumination conditions, within 5 deg of the poles, to ~ 100 m spatial resolution and 5 hr
temporal resolution.

IV a. Characterize surface morphology through visible imaging at 1 m feature identification resolution

over targeted areas of 10x10 km

On-orbit remote sensing can be an important asset characterize the distribution of surface volatiles,
provide some insight into buried volatiles and aid understanding the cycling of polar volatiles

 Directly or indirectly detect surface frost and subsurface ice can support understanding the
contemporary migration of volatiles
 Low light imaging within PSRs to reveal subsurface ice deposits by detection of anomalous
fields of bright ejecta of small craters associated with other volatile indicators
 Active spectroscopy using orbital reflectors or multispectral lasers in the 3um region to directly
detect surface frost and understand the lunar water cycle
 Low light imaging spectroscopy exploiting indirect lighting such as innovative application of
an M3-like instrument may also provide maps of surface frost in PSRs

20
 3.3 REGIONS OF INTEREST

Polar Regions
Precursors to surface exploration (lander/impactor/rover), orbiters need to observe regions of interest
to aid the planning of surface mission operations. Hydrogen distribution and concentration is currently
one of the prime factors in site selection studies. Enhanced hydrogen is widespread across the polar
regions and is sometimes concentrated in permanently shadowed regions (PSRs). Data show that
average annual surface temperatures below 110K are also widespread, including both PSRs and areas
sometimes illuminated. This characteristic allows preservation of shallow buried ice for geologic time
Arguments derived from lunar surface processes suggest volatiles will be distributed irregularly and the
high water abundance observed by LCROSS was not consistent with the regional H abundance
indicating sampling of a local concentration

Optimizing a number of volatile relevant and environmental parameters: H abundance, temperature,

slope, proximity to PSR and Earth visibility for direct communications is required. An investigation on
identifying regions based on H abundance and average annual temperature supplemented by examining
and visibility from Earth for a line of sight communication total lighting over a lunar diurnal cycle.

At both the poles, there are regions that need orbital reconnaissance to

1. Cabeus vicinity (South Pole)

2. Shoemaker/Nobile vicinities (South Pole)

3. Peary vicinity (North Pole)

These areas are volatile-rich (H > 150 ppm) and can maintain subsurface ice (average annual surface
temperature < 110K). Topography in this region comprises of modest slopes (10 degrees). When
including lighting explicitly, the north polar Peary vicinity is slightly favored owing to somewhat more
persistent lighting, with the Cabeus vicinity showing the least persistent lighting.

Lunar science profiles were laid down with respect to the specific sites of interest, exploration
requirements, mineral composition. A few examples are laid down and the basis of importance for the
respective profile.

South-Pole Aitken (SPA):

 Still a very important high priority for constraining the impact flux of the inner solar system
and is part of the last decadal survey that has not yet been addressed.
 Age of potentially the oldest giant impact on the Moon is a vital scientific goal not only for the
impact history of the Moon but also for understanding the evolution and impact history of the
inner solar system.
 The SPA impact excavated the crust deeply, maybe even exposing the upper mantle. Returning
samples could shed light on the nature of the lower lunar crust and possibly the upper mantle.
 Mare patches and cryptomare within SPA allow the examination of the diversity of mantle
sources from the lunar far side.
 Determination of the ages of the large craters within SPA will potentially give a definitive test
to the cataclysm hypothesis.

21
  Volatiles are present in permanently shadowed craters around the South Pole (i.e., the southern
extent of the basin).

Lunar Poles
 It is important to explore the lunar poles (specifically the volatile deposits) before the lunar
environment is disturbed by future human landings, as the volatiles emitted from such landings
may compromise the pristine nature of the volatile deposits.
 Peary Crater in the North Polar region is a flat-bottomed crater that would facilitate landing
and as part of it is permanently shadowed it may contain volatile deposits.
 The topography of the lunar South Pole is not ideally suited for long distance roving. The
South-Pole Aitken basin contains rough terrain from a traversing perspective. Lunar North Pole
topography appears gentler (for traverses) and is closer to Mare sites. The location allows a
variety of science questions to be answered (mare basalt diversity, cratering history, polar
volatiles, resource potential, etc.), to be explored, and lunar outpost suitability evaluated.
 A natural follow-on to LADEE would be a rover to polar areas (permanently shadowed
regions).
 The lunar poles should be explored because this is where the Hydrogen is concentrated and to
answer the following questions:
o What other elements are concentrated with Hydrogen?
o What are the origins of these deposits? ⇒ How are these deposits preserved within the
regolith?
o How do these volatiles get transported around airless bodies?

Nectaris Basin
 Nectaris Basin is another old impact basin (older than Imbrium) so returning impact melt from
this basin will also help constrain the impact flux of the inner solar system.
 The advantage is that it is on the lunar nearside as compared to SPA.
 Nectaris also has well-defined ejecta features and impact melts within its interior to tie in with
ejecta blanket.
 Age dating of Nectaris will help constrain the end of planetary accretion and whether or not
there was a terminal cataclysm.

Aristarchus Plateau
 Pyroclastic deposits from early lunar explosive volcanism may have been more common
during the early history of the Moon (i.e., pre—3.8 Ga).
 The pyroclastic deposits probably contain internal volatiles from deep within the moon –
raising the possibility that the Moon never fully outgassed, which at a fundamental level is
difficult to reconcile with the Giant Impact hypothesis.
 Where are the volatiles? They could be trapped on grain surfaces, cracks, pores, and crevices.
 Lunar pyroclastic deposits are not well explored – Aristarchus permits such exploration.
 Gaining a better understanding of pyroclastic eruptions on the Moon has direct implications
for similar eruptions on other places of the Moon, as well as, for example, Mercury and Io.
 The possibility of fossil regolith trapped between lava flows/pyroclastic eruptions – this could
give a snap shot in time of the Suns activity through trapped solar wind particles.

22
Others
 Lunar Dust Toxicity: Studies of lunar dust toxicity need to be conducted in situ because the
returned Apollo lunar samples may have been compromised (adsorption of water and other
volatiles).
 Exploration of Lunar Magnetic Anomalies (e.g., Reiner Gamma):

o Understanding the plasma environment within and outside the lunar magnetic
anomalies allows the nature of such anomalies to be evaluated (i.e., remnant ancient dynamo
or transient impact event?).

o Direct applicability for human exploration (protection against solar flares during
extended stays at an outpost)

23
 4. PAYLOAD
Characteristics of a generic lunar orbiter mission for relevant science mission has a common
attribute of a polar mission orbit. Also, lunar orbits for respective data sets and their acquisition
requirements can have allowable altitude as low as 50km. However, higher altitude is preferred due to
higher magnitudes of extremely fluctuating thermal stresses are induced on the spacecraft at lower
altitudes. Thus, a 250 km near-polar near-frozen orbit is chosen.

While assessing payload selection, a general background of missions and their payloads was done with
respect to scientific relevance of the data obtained. Current trends on large/medium spacecrafts carrying
relevant science instruments are as follows:

4.1 CONVENTIONAL BASELINE PAYLOAD SPECS

(I) Life science payload ~ 10 kg; low/mod data rate; 20 W

(II) Laser altimeter with good orbit determination (15 kg; 20 W; mod/high rate)
(III) SAR (~50 kg; 60 W; high rate)
(IV) Neutron imaging (20 kg; 5 W; low rate)
(V) Radiometer (5 kg; 5 W; low rate)
(VI) Narrow-angle camera (10 kg; 5 W; high rate)
(VII) Wide-field camera (2 kg; 5 W; low rate)

Out of the above, practicality of on-orbit remote sensing, complimented by advancements in

miniaturisation (size, mass, and power) of the components available commercially, positively presented
the basis of selection for relevant payloads. Simplicity and independence of operations (without
imposing a requirement for additional science instruments), available power and mass paved the way
towards the selection. As described in the earlier sections, prospects of lunar exploration intercepted
with small satellite missions were put into iterations. Also, the primary consideration of characterizing
and proving the capability of small satellite science missions was taken, while keeping minimal
duplication of science data acquired by existing or planned missions.

SRMSAT-2, focusing on radiometric- geoscience, morphology, and weathering, in its current design
features 3 payloads. The mission and spacecraft bus are defined for a primary, a secondary and an
outreach payload.

A. PRIMARY PAYLOAD 1. PMS- Passive Microwave Sensor

(Microwave Radiometer)

B2. SECONDARY PAYLOAD RDM- Radiation Dose Monitor

C3. OUTREACH PAYLOAD NAC – Narrow Angle Camera

Table 4.1: Payload and priority

Along with configuring primary scientific mission, it was realised to complement payload subsystem
with two more payloads. The selection of these was driven by one of the objectives of Lunar exploration
laid out in the previous sections. Selection of the secondary payload was driven by the exquisite and

24
non-conventional nature of the trajectory combined with the use of COTS components configured in
the mission design. Extended Earth-bound phasing and, cis and trans-lunar cruise periods directed the
attention towards the importance of measuring fatal radiation dose. This would not only lay out a
required shielding for any similar mission profile but also characterize radiation in near deep-space
(trans-lunar) which is currently, the prime choice for simulating and technology proving of future life-
science robotic or human missions.

Public outreach started with Voyager’s Golden Disks and is an important constituent of any distinct
mission. Outreach photographs from ISRO’s Mars Orbiter Mission and JAXA’s SELENE mission have
proved immensely impactful on the public. Through the years, in space exploration, public outreach has
reserved a spot in the missions with continuously increasing importance. It thus seemed important to
emphasize on educational and inspirational impacts of lunar exploration by small-satellite.

Some of the operation would be during Earth-departure in the final cruise orbit, Lunar approach, same
frame Moon/Earth rise and set, and during orbit stabilisation to final mission orbit.Camera, other than
outreach purpose may also be used in characterizing shadowed regions and illumination variation. Also,
targets, especially potential ISRU, human exploration, landing and historic Apollo sites will be focused
upon if can be incorporated in Event and Orbit Schedules of operations.

 Science objectives may be extended to incorporate collaborations with national/international

scientific community for the instrument acquisition, payload operations, and payload data
utilisation.
 A primary data center (SRM Lunar Science Data Center) will be established to collaborate with the
DSN and University tracking station with capabilities to support handling of downlinked science
data and, maintaining and regulating data archives. It will be responsible to interface with the
scientific community and the students of SRM University for providing acquired science and
outreach data. The Datacenter will make lunar science data available to the students to induce and
assist research thus furthering developments in Lunar and Planetary Sciences at SRM University.

25
 4.2 PASSIVE MICROWAVE SENSOR
(Primary Science Payload)

PMS is the currently chosen as the primary scientific payload onboard. The objective is to
obtain radiometric regolith measurements to infer surface/sub-surface thermal and dielectric properties
by remote on-orbit sensing, mainly over lunar poles. Final mission orbit was formulated to general
attributes of a lunar orbiter mission with importance on polar orbit and science over lunar poles.

For a long time, visible light and infrared have been the preferred wavelengths for lunar observation,
though microwave band was also used occasionally. Because of the synchronization of the Moon’s
rotation and revolution, visible and near infrared light cannot ‘see’ the dark side of the Moon, whereas
microwaves can. Moreover, it can penetrate a certain depth under the surface of the dielectric material.
For these reasons and for its relatively low weight and simple design, it is among the proposed payloads
for the mission.Since, SRMSAT-2 will be in a frozen near-polar orbit, observing through a full lunar
day/night cycle, PMS will achieve full diurnal coverage in 6 months of continuous observation after
commissioning. This is also set as the Science Mission lifetime. However, keeping in accordance with
other mission operations, PMS would not be able to observe Moon at all times. Thus, duty cycle along
the total mission will be optimally configured to cover total or maximum surface observation in this
time period.

The primary data measurement obtained will be Brightness Temperature (TB). PMS is a two frequency
microwave radiometer operating at 3 GHz and 37 GHz. Brightness temperature measured at 37 GHz
provide surface temperature information ( ~5cm) while the Brightness temperature measured at 3GHz
gives subsurface temperature information( ~5m).

Data in terms of Brightness Temperature (TB) obtained can directly be used to derive/infer :

i) Microwave emissivity of the Regolith

ii) Regolith thickness: Inversion of Brightness temperature at low-frequency channel
iii) Sub-surface thermal and dielectric properties: (Brightness Temperature measurement)
iv) Diurnal temperature variation
v) *Regolith Anomalies: By complementing microwave sounding
vi) *He3abundance : Relating microwave brightness temperature to lunar soil properties.

Observation
Instrument pointing and observation is currently set to push broom lunar center pointing type
observation with two operating frequencies 3GHz and 37 GHz. Patch antennae will be used for
radiometric sensing including calibration. The observation strategy will be laid out by specifying Orbit
Schedules (OSC) and Event Schedules(ESC) . The Orbit Schedule is a periodic schedule that repeats
itself within a time period. Each orbit within a schedule is allocated an operation/function. Event
Schedules are occurrences of events within an orbit that include events related to mode changes within
LCOM, LTARP, LNORM and LSTK, and execution of tasks under the modes . Observation execution
is specified when OSCs are mapped to ESCs. Since OSCs and ESCs will be the drivers for on-orbit
observation operations, they would need to be critically designed, assessed and mapped to define a
reliable observation strategy.

26
 Figure 4.1: Lunar Regolith

Moon’s layers as shown in Fig. 1, consist of a series of layers of material ejected from individual
meteorite craters over a period of time and embedded lava flows and pyroclastic deposits. Regolith is
the crucial, top most layer, up to 5m deep in mares and up to 15 m deep in the older highlands. It is a
result of meteorite impacts. The regolith ideally consists of two zones: (1) a near-surface ‘reworked’ or
‘gardened’ zone, typically from few centimetres to tens of centimetres thick, in which all layers have
been homogenized or mixed together, and (2) an underlying slab or sequence of slabs in which the
original layering is still undisturbed. As the distance from the lunar surface increases, some intrusion
of rocks can be found.

The Radiometer (Hardware And Operations)

The selection of the radiometer can be based either on exploiting total power which offers simple
realisation but poor calibration or the pulsed noise injection which offers relatively easy self-calibration
but complex implementation. Total power and pulsed-noise injection configurations can be represented
by a two-load radiometer (TLR) configuration presented in Fig. 5. Briefly, this type of radiometer
consists of an antenna, a switch, and a receiver. The switch connects, at different instants, the antenna
and the one or two thermal reference loads to the receiver. This type of radiometer combines a simple
design together with a good performance in terms of radiometric precision, by using both the chopping
post processing filtering technique and an optimal choice of the switching time of the SP3T.

Total Power Radiometer (TPR)

In the schematic (Fig 2) showing the working, G is the simplified amplifier and B is a band-pass filter.
The microwave power has to be detected to find some measure of its mean. Two straightforward
detector types can be made, using microwave semiconductor diodes: the linear detector and the square-
law detector. Then the output voltage will be proportional to the input power and hence the input
temperature. Finally, we indicate where the integration takes place: The signal from the detector is
smoothed by the integrator to reduce fluctuations in the output, and the longer the integration time, the
more smoothing there is.

𝑉𝑂𝑈𝑇 = 𝑐. (𝑇𝐴 + 𝑇𝑁 ). 𝐺

27
 (4.1)

VOUT is the voltage obtained, TA in the antenna noise temperature, TN is the Noise Temperature and G
is the Gain.

The sensitivity of the total power radiometer is given by:

𝑇𝐴 + 𝑇𝑁
∆𝑇 =
√𝐵 . 𝒯
(4.2)

Where 𝒯 is the integration time constant.

Figure 4.2: Total Power Radiometer

Noise-Injection Radiometer (NIR)

The noise-injection radiometer represents the final step towards stability; that is, the output is
independent of gain and noise temperature fluctuations.The output from the Dicke Radiometer is
independent of Gain ‘G’ and noise temperature ‘TN’.

Figure 4.3: Noise Injection Radiometer

The noise-injection radiometer is a specialization of a Dicke radiometer. If the reference temperature

and the antenna temperature are equal, the output from a Dicke radiometer is zero. A servo loop adjusts
TI as the noise injection to maintain the zero output condition, or rather the near zero output condition:
The loop gain can be made large but not infinite.

𝑉𝑂𝑈𝑇 = 𝑐. (𝑇𝐴′ + 𝑇𝑁 ). 𝐺

(4.3)

VOUT is the voltage obtained, TA in the antenna noise temperature, TN is the Noise Temperature and G
is the Gain.

And as,

𝑇𝐴′ = 𝑇𝐴 + 𝑇𝐼 and 𝑇𝐴 = 𝑇𝑅 − 𝑇𝐼

28
 (4.4)

TI is the Injected Noise Temperature.

The sensitivity of the noise-injection radiometer is given by,

𝑇𝐴′ + 𝑇𝑁
∆T = 2 .
√𝐵.𝒯

(4.5)

But, as TA’ is equal to TR, we find;

𝑇𝑅 + 𝑇𝑁
∆𝑇 = 2.
√𝐵 . 𝒯
(4.6)

Calibration

Two Load Radiometer (TLR) Configuration

PMS will perform routine in-flight radiometric calibration to calculate signal gains and offset values to
correct for thermal and electronic drifts. Calibration sequence will prioritize space as zero references
for long-term calibrations, and short in-flight communication could rely totally on Internal hot/cold
loads. The current design will encompass a two load radiometer or TLR, which is a simple total power
radiometer that uses two reference loads/references of stable brightness temperature to calibrate the
radiometer. These loads are called the hot and the cold load w.r.t the magnitude of the reference
brightness temperature. These are stable calibration loads (e.g. Noise diodes and COLFETs) in addition
to which references such as deep space observation and internal reference black body can also be used.
The two-load calibration procedure used is based on the assumption that a linear relationship exists
between the radiometer output and the brightness temperature. Hence, measurements of two different
stable loads/references are sufficient to establish the linear relationship necessary for instrument
calibration. Advances in technology allow us to implement the radiometer operations within the
software instead of the hardware implementations used in early radiometer designs. For example,
sampling and averaging data from the radiometer antenna and a calibration load and then subtracting
these two measurements was historically performed by synchronous detection circuitry. These same
operations are now performed using digital means (i.e., digitization of analog voltages followed by
software manipulation of the digitized data). An SP3T switch changes calibration reference being
observed/measured. The switch calibration switch is assumed to have an identical loss in all three states
and no thermal gradients. The calibration loads have well-characterized brightness temperatures T1 and
T2 that are stable over long time periods. Other terms are shown in Fig. 1 is the antenna temperature,
Ta, combined loss of antenna and antenna feed, Lo , receiver gain, G , receiver detection bandwidth, B,
and receiver noise temperature, Tr.

29
The relation between the output voltage of a receiver, V and the input of a target radiation T M is
expressed as (Fig),

TA = aV + b

(4.7)

The microwave radiometer is calibrated by

connecting to the observation antenna, cold space
antenna, and internal matched load, respectively.
During each calibration cycle, the receiver would
receive the temperatures from warm load TH and
cold space TC, with outputs of VH and VC,
respectively. The coefficients of calibration are then
given by the equations,
𝑇𝐻 − 𝑇𝑐 𝑇𝑐 𝑉𝐻 − 𝑇𝐻 𝑉𝑐
𝑎= , 𝑏=
𝑉𝐻 − 𝑉𝑐 𝑉𝐻 − 𝑉𝑐

(4.8)
Figure 4.4: Two point calibration of PMS

Figure 4.5: Two load Radiometer

The radiometer’s Noise equivalent Delta Temperature (NEDT), ∆T A, will depend on greatly upon the
errors of both the antenna measurements and the calibration measurements. Increasing antenna dwell
time with a push broom observation than others like whisk-broom observation indeed, reduce the errors
relatively but only to a certain extent. Most error reduction will depend on dwelling longer on the
calibration loads. Also, selection of references of higher and lower brightness temperature is the best
option to reduce NEDT. A second desired feature will be Dicke-switching the radiometer source
between antenna and calibration loads has the important byproduct of high-pass filtering the radiometer
noise power measurements. This so-called “chopper filtering” is used to remove low-frequency flicker
noise from the measurements that would otherwise reduce TA accuracy.

A possible drawback of a Two Load Radiometer can be represented by the short amount of time lost
for the observation of the Moon emission. Nevertheless, this does not seem to be a dramatic problem
since a 250 km mission orbit is chosen.

Microwave Brightness Temperature (Lunar Measurement Case)

Surface Microwave Brightness Temperature Cycles (Day/Night)

30
 The Moon’s daytime thermal environment is controlled by the solar radiation flux. Because of
the low thermal conductivity of the lunar soil, and the length of the lunar day, the temperatures of
illuminated surfaces on the Moon are always close to radiative equilibrium.The low thermal
conductivity of the uppermost layers of the lunar regolith inhibits the flow of heat from the subsurface
to the surface, and as a consequence, lunar nighttime temperatures hover near 100 K for the duration of
the 14-day lunar night. Also, the depth of penetration of the diurnal temperature wave on the Moon is
∼30 cm. Thus, an analysis of brightness temperatures to be detected, to aid the selection and preliminary
parameterization of the on-orbit radiometer along with setting a baseline for Models, needs to be done.A
multi-layer model of lunar regolith as is shown in Fig.1, the surface is treated by a vacuum-soil interface,
which has no atmosphere and no rough-surface-scattering. The background Brightness temperature, TB
is 2.7K. Below the surface, all regolith layers are divided at a given interval considering their
temperatures and attenuation variations. Due to the large fluctuation of the top layers’ temperature, the
TBemitted from the lunar surface are continuously changing with time in a lunar day cycle.

Fig 4.6 : Lunar Surface Temperature Variation during a lunar Fig. 4.7 : Lunar Surface Temperature Variation during a
day at (Lat,Long) = (0 , 0) lunar day at (Lat,Long) = (0 , 0) at various depths

Fig 4.8: Lunar Surface Temperature Variation Fig 4.9: Temperature Profile at various depths at Lunar
during a lunar day at North Pole Poles

Fig.2 shows the equator surface temperature changes in a lunar day (29.5 Earth days), a cycle from
sunset to dawn. We find that the maximum temperature is reached at about 392.82K at noon and the
minimum of 133.77K at dawn. Fig 3 shows the temperature profile at noon in the first graph and at

31
dawn in the second graph. We also find there is a great change in the top layers, especially at the top
20cm. Fig. 3, the third figure gives the average of the whole day and the fourth gives the differences
between the two times of the day.

For contrast, Fig. 4 and Fig. 5 show the surface temperatures and profiles at the polar area.The
anticipated reason being the length of the day and nights being nearly 12 times that at the equator. Thus,
in this region, the maximal and minimal temperature went up to only 157.93K and 64.97K, respectively.

Surface Microwave Brightness Temperature (Frequency Variation)

Fig. 7 shows a whole day TB changes at the lunar equator and the latitude of 50°N while the regolith
depth is 1m. The left and the right figure are very similar. But in Fig. 8, the TB at depth 0.5m and 5m
at the lunar north polar are very different because of the penetrated depths of each frequency are
different and the temperatures at the surface are larger fluctuated while at the bottom the temperatures
are basically the same all the time.

Fig 4.10 -12: Detected brightness Temperature changes

during a lunar day at equator (for various Microwave
frequencies at 1m, 0.5m and 5m depths)

Sub-Surface Microwave Brightness Temperature (Depth Variation)

32
To see the property of TB of different frequencies change with regolith depths, Fig. 8 shows Tb change
with depths at four microwave frequencies at the equator. Because of fluctuation of temperature
profiles, the trends of TBs at day and night are very different. Fig13a is plotted for T b at local lunar
noon, where the four lines have no intersection and the line shapes are all different from each other
from 0.5n to 5m regolith depth, and Tb increases with increasing frequencies and depths. While the right
four lines in Fig. 13 intersect each other, the values are very close when the depths are smaller than
50cm. At local lunar dawn, the Tb at 3.0GHz and maybe reach the maximum of the four channels’ with
depths bigger than 2.5m.

Fig. 4.13a: Brightness Temperature variation with depth at Lunar Equator( at various Frequencies)

Fig. 4.14: Brightness Temperature variation with depth at Lunar North Pole (at various Frequencies)

Based on the heat conduction equations (Coarse Idealisation) lunar temperature profiles have been
analysed. Microwave radiometry at 37GHz gives unique brightness temperature values just below the

33
surface; 3GHz gives information at few meters below the lunar surface. 3GHz gives Tb variation at
greater depth (up to 5m). However, as observed in Fig. 9, the trend of brightness temperature repeats
itself (left) at lower depths at 3GHzshowing the same Tb for two different depths. 37 GHz gives a unique
Tb curve for lower depths and less variation over the depths. Thus, readings can be compared of 3GHz
with that of the 37GHz detection to obtain a certain value. Therefore, a minimum of two frequencies is
required for accurate microwave lunar remote sensing

Observation Model
Current coarse design includes a simplified observation model. The Temperature Profile mapping can
be explained with a two-layer structure for lunar regolith divided into vertical-uniform di-electric
permittivity and homogenous temperature profile layers.Further, conclusions about the vertical
temperature of the lunar regolith and the lunar bedrock layer as functions of zenith angle(θ), azimuth
angle(ϕ), and the selenographic latitude(ξ) as T2(θ,ϕ, ξ) and T3(θ,ϕ, ξ) and effective
emissivitiese2(θ,ϕ, ξ) and e3(θ,ϕ, ξ) respectively with data over one scan of IFOV, or one PMS shot.

For lunar craters, the surface is assumed flat and therefore, the emissivity can be used to represent the
effective emissivity of a crater. The cold and warm areas due to effects of shadow and surface tilts
exist inside a crater when the microwave radiometer intercepts the discontinuity of the crater. The
percentage of cold and warm areas varies in daytime TB in an oscillatory fashion.
* Simplified Crater Model:
The depth and floor diameter as a function of crater diameter employs a simple statistical model
involving images from the spacecraft. The crater parameters (depth and floor diameter) as given as:

yp= g ・Dk
(4.9)
where g and k are constants and D(km) is the diameter of the crater.

The physical temperature of the lunar regolith at night time is influenced by the temperature at
daytime. The physical temperature of the lunar regolith within a few cms has high conductivity and
inner sources, hence stays warm.

∬𝑃𝑀𝑆 𝑆𝐻𝑂𝑇 (𝑒2(𝜃, 𝜙, 𝜉)𝑇2(𝜃, 𝜙, 𝜉) + 𝑒3(𝜃, 𝜙, 𝜉)𝑇3(𝜃, 𝜙, 𝜉)) 𝐹𝑛(𝜃, 𝜙) 𝑑Ω

𝑇𝑏 (𝜉) =
∬𝑃𝑀𝑆 𝑆𝐻𝑂𝑇 𝐹𝑛(𝜃, 𝜙) 𝑑Ω

(4.10)

34
Science Data Product Plan

PMS Data Product Level Description

Pre-Flight Calibration Data Pre-flight calibration data (Spectral Response, Blackbody
Response)
Stage 0 Time-sequenced raw science and housekeeping data (on-orbit)
Stage 1 Calibrated radiances and housekeeping data merged with timing
information (on-orbit)
Stage 2 State coupled (Lat, Lon, Local time) global brightness temperature
(on-orbit)
Stage 2* State coupled (Lat, Lon, Local time) global brightness temperature
and monthly max, min and average surface temperature.
(Processed Data-On Ground)
Stage 2** Derived quantity: Regolith thickness density, Emissivity, and
Dielectric constant (Processed Data- On ground)
Stage 2*** Temperature observation of permanently shadowed regions,
localized temperature observations of derived surface and
subsurface characteristics (Processed Data- On ground)
* Certainty of the stage to be decided based on level of science mission and data priority set according
to the instrument and operation profile in the critical design

Conclusive Comments
 Dielectric Characterisation: PMS’s radiometric data along inferred regolith depth, will be
needed to be compared with the results of available models to infer bulk surface and sub-surface
(depth-wise) thermal and dielectric properties. While PMS’s resolution is currently assumed
not to be enough to resolve significant details, over a considerable cycle portion of a diurnal
cycle , PMS’s data would be enough to indicate coarsely/map surface distribution, microwave
emissivity at locations receiving reasonable coverage.
 He3 abundance on the Moon: The investigations to quantise the existence of He3 embedded in
Ilmenites (TiO2+Fe-oxides) in the lunar sub-surface layers is highlighted as one of the prime
focus areas in Global Lunar Exploration Roadmap since it provides excellent aspects for lunar
exploration.These compounds fill in for a crucial in-situ power generation resource, furthering
advances in human exploration prospects. This analysis is carried out by determining the
relations between microwave brightness temperature and lunar soil properties. The microwave
brightness data provides the precursor, which along with derived data for soil properties can be
processed to quantise He3.
 The anticipated payload configuration will not be operationally relevant with slews. Thus, an
automatic data halt will be called for the instrument in case of predicted slew in payload
operating mode or unpredicted slews, while in orbit. Also, in case the operation of primary
while simultaneously, requiring attitude change for other operation (like outreach payload
pointing ), the instrument will be called in for a data halt. Data log will be restarted with pre-
operation calibration and lunar pointing Payload Operating Mode attitude.
 Although there are no payloads to complement, in order to investigate Permanently Shadowed
Regions or Polar Cold Traps, that it could infer volatile presence, the data on anticipated
locations of PSRs is intended as raw microwave brightness temperature data.

35
 4.3 RADIATION DOSE MONITOR
(Secondary Payload)

Objective
The primary objective of the Radiation Dosimeter is to characterize the space radiation environment by
measuring the total radiation dosage acquired along the trajectory and into the lunar orbit. The radiations
are mainly particulate in nature, in the form of plasma radiations but may be of varied energies and
origins. Data obtained will be used for the evaluation of radiation environment and radiation shielding
requirements for future manned and robotic lunar missions.

Payload Priority
The radiation dosimeter is the secondary payload in accordance with the mission. It signifies that it is a
measurement followed by an analysis being done en route to the lunar orbit and not the primary cause
in driving the mission design itself. The payload will be commissioned at the first payload
commissioning stage, just after the completion of initial orbit sequence in the Normal Mode. RDM will
be nominally ON in all the normal modes after commissioning

Instrument
Radiation dosimeter measures the total radiation dose, which is defined three principle ways –
Absorbed, equivalent and effective dose. Absorbed dose is the amount of energy cumulatively absorbed
per unit mass by a body due to exposure to radiations. As it is a measure of energy being imparted to a
body, it can be used to measure both particulate and electromagnetic wave radiations. The latter two
ways of definition are concerned more with the medical field to measure the effect of exposure of tissues
to nuclear radiations.

A radiation dosimeter is used to study radiations in space. The radiation environment in space is
classified as the particulate radiations which need not be electromagnetic in nature, and hence do not
account for net heating instantaneously. These radiations consist of plasma or highly energetic particles
whose energies have a wide range of few keV to MeV. The particles originate either as a result of
coronal discharges from the sun or are of cosmic origin. They are charged particles and hence can get
deflected by the magnetic field encountered in the path. The magnetic field of the Earth is considerable
to cause significant deflections and form a meteoric structure. Hence, trapped radiation belts are formed.
The most significant dosage in a deep space mission is in the two Van Allen Radiation belts that extend
up to 60,000 km. Thus, the monitoring of radiation activity is hugely important from both, the mission
point of view and for terrestrial beings.

The cosmic waves of galactic nature originate outside the solar system and are modulated by the
heliosphere. They have colossal energies that can range up to few GeV and consist of mostly protons
(~90%) with a small amount of alpha particles (~8%) and electrons (~2%) with traces of oxygen and
fluorine. The cosmic waves are highly influenced by solar activity and have very small fluxes at high
energies.

The Van radiation belts of the earth extend from 1000 to 3700 km and 13000 to 60000 km respectively.
The inner belt consists of mainly electrons and proton at energies up to 10MeV and 100MeV

36
respectively. The energies of the outer radiation belt are highly variable and need more monitoring as
it can have large changes with small incoming flares, due to a large area in pace and lower energy
particles. The radiation in these regions thus, must be monitored with precision. The plasma radiation
arising due to solar activity have energy ranges of 100MeV/amu and have highly variable particulate
flux densities. They can classified into solar winds and flares. The former is a more continuous and
uniform radiation whereas the latter is a relatively less predictable burst of energy due to solar activity
in the corona.

The magnetic field of the moon is not as significant as the Earth’s magnetic field, and thus, doesn’t have
any known significant trapped region of energetic particles. The cosmic rays and radiations of the solar
origin, however, are equally significant. The particulate fluxes can be measured and studied using a
Radiation dosimeter. It is an instrument that can measure the absorbed doses in Grays (J/kg) indirectly.
A more conventional, non-SI unit used for measurement is rad and krad, which is defined as radiation
acquired dose. It has a numerical value equal to 100 ergs/gram. (erg/g).

Operation
The device primarily contains one semiconductor detector, one charge sensitive pre-amplifier, a fast
12-channel analogue-to-digital converter (ADC), a discriminator, a real-time clock, two or more
microcontrollers and a flash memory. More custom components modifications and additions can be
made with respect to the mission like the use of a GPS or a temperature sensor.

The unit is managed by the microcontrollers, and serial links provide it with the transmission of the data
stored on the memory towards a telemetry system of the spacecraft. Different power sources are used
with various instruments which include 3.6 or 7.2 V rechargeable or primary batteries, 28 or 43 V DC
aircraft and satellite power.

The detector measures the amplitude of pulses of the incoming radiations, which is proportional to a
factor of 240 mV/MeV to the energy loss in the detector and, respectively, to the dose and linear energy
transfer (LET). These amplitude signals are then converted to digital signals and organized in 256-
channel spectrum using the former 8-bits of the analog-to-digital converter. The absorbed dose is
calculated using the formula:
256

𝐷 = 𝐾 ∑ 𝑖𝑘𝑖 𝐴𝑖 /𝑚𝐷
𝑖=1

(4.11)

wheremD is the mass of the detector in kilograms,kithe number of pulses in the ‘ith’ channel, Ai the
amplitude in volts of pulses in channel ‘i’, Kiki.Ai the deposited energy (energy loss) in Joules in channel
‘i’ and K a coefficient. All 256 deposited dose values, depending on the deposited energy for one
exposure time, form the deposited energy spectrum.

37
SSTL Radiation Monitor
The radiation monitor by Survey Satellite Technology Ltd. was chosen based on the success of past
missions, low mass and power requirements, and availability. The functions of the dosimeter include:

 Particle detection and measurement of dose rate and total acquired dose
 Measurement of high-energy proton fluxes and heavy ion Linear Energy Transfer (LET spectra
 Measurement of resulting ionization current

It consists of an adjustable LET range and an automated threshold trigger and working life of more than
7 years. It has been used up to six missions in the past, including KitSat-1 (1992, LEO), PoSat-1 (1993,
LEO), Tiungsat-1 (2000, LEO), Amsat-Oscar 40 (2000, HEO), GIOVE-A (2005, MEO),
TechDemoSat-1 (2014, LEO) and DMC3 Constellation (Launching 2015). The given specifications are
as follows11

PHYSICAL PROPERTIES
Dimensions ~10 × 7 × 6.5 cm3
Mass ~700 g
Power 1.8 W, at 28 V unregulated
Lifetime ~7.5 years for LEO
Temperature -20°C to 50°C (Operational range)
-30°C to 60°C (Survival Range)
Radiation dose 5 krad

PERFORMANCE CHARACTERISTICS11
Sensors 1 internal RadFET pair.
3 external RadFET pairs.
1 UV photodiode.
1 large area PIN diode detector.
Total Dose Typical measurement in Krad up to 10 Krads.
Dose Rate Typical measurement in mrad/s up to 50 mrads/s.
Particle Monitoring Nominal LET range 0.04 - > 6.4 MeV*cm2/mg
in63 channels
Integration Time 75 s steps in 5 minute periods (Programmable
inorbit)
Data Rate ~ 220 Kbytes/ day
TM/TC interface CAN-SU or Optional (RS422 and RS485)

38
Liulin-Type Spectrometry-Dosimetry Instrument
The Liulin-Type Spectrometry-dosimetry Instruments (LSDIs) are used for measurement of amplitudes
pulses incoming particle and wave radiation using a silicon detector. The measurement made is
proportional to the total energy absorbed, and is measured in Gray (Gy). Until now, more than 90 LSDIs
have been developed, built and used on the Earth’s surface, at aircraft-cruising altitudes, at LEOs, which
is used both inside and outside of the Earth’s magnetosphere and on the Moon’s orbit. They can be
custom made with the smallest instalment of 98g for Chenadrayaan-1 in 2008-2009 to the largest
modification of 450g.

The instrument can be custom made for space, and thus has favourable masses and power requirements,
as well as volume, was taken.

PROPERTY VALUE
Dimensions 10 × 5 × 2 cm3
Weight 100 g
Detector 0.3 mm thick Si-PIN Semiconductor Type, 2 cm2
area

39
 4.4 NARROW ANGLE CAMERA
(Outreach Payload)

Through the years, in space exploration, public outreach has reserved a spot in the missions with
continuously increasing importance. It thus seemed important to emphasize on educational and
inspirational impacts of lunar exploration by small-satellite. Camera is an ideal choice for an outreach
instrument for a small satellite.

These are classified broadly over their angle of view. A

larger angle of view gives more coverage area but a lower
resolution whereas a narrow angle gives more depth and
resolution over a specific object or location of interest. A
wide angle camera encounters distortions and less
resolution for same lens size. With an interest of imaging
of specific sites of interest finally, a narrow-angle camera
was chosen. The instrument currently selected for the
narrow-angle camera is Hyperion Technologies’ IM200.

The IM200 is a versatile imager with a 4 Mega pixel

camera lens(2048x1944). It has a sampling frequency of Figure 4.15: Selected Narrow Angle Camera
5Hz, and the image buffer can store up to 25 images.
The chosen lens configuration is 50mm F2. It gives a field of view of approximately 19° in both
horizontal and vertical axes. Given lunar orbit altitude of 250 Km, it gives a sweep area of around 80 x
80 sq. Km. IM200 is tested for space environments and has a radiation tolerance of 12.5 kRad and a
low power consumption of less than 600 mW.
The data product type is classified into miscellaneous and includes image acquisitions as :

1. Images of the Indian subcontinent (if commissioned) in the phasing orbits

2. Images of the Earth in departure phase (after TLCM)
3. Images of Earth and Moon in the Cruise phase including one from the apogee (~ 1.5 x 106 km)
4. Images of the Moon in the lunar approach phase

5. Images of specific lunar sites after final commissioning into the mission orbit (250 km altitude)
around the moon
NAC Data Product Level Description

Pre-Flight Calibration Data Pre-flight calibration data

Stage 0 On-orbit test images (untargeted)

Stage 1 Calibrated images of targets (on-orbit)

Images captured by the narrow-angle camera would be used to study the various features of the lunar
surface in final data product stage. These captured images will be sent to the ground station using the
HGA.

40
41
 5. MISSION OVERVIEW

5.1 SPACECRAFT
SRMSAT-2 is a micro-satellite class spacecraft with a total mass of nearly 86 kgs. The
dimensions of the spacecraft bus are 50cm x 50cm x 60cm with two deployable and one side mounted
solar panels for power generation. The base plate features the conventional IBL-298 launch interface.
A high gain deployable antenna is mounted on the top plate, and two low gain (QFH) antennae are
mounted diametrically opposite for omnidirectional communication. The bus provides a budget of
maximum allowable 10 kgs of mass and 15 W of power consumption for the payload. The total
propellant mass is 42 kg which is about 48% of the total spacecraft mass.

5.2 PAYLOAD
The current bus allows a total mass of 10 kg, dedicated for housing payloads with maximum allowable
power consumption of 15 W in Payload Operation Mode. Currently, the SRMSAT-2 spacecraft is being
designed with consideration of three payload options. PMS (Passive Microwave Sensor) is the primary
scientific payload with the objective of studying the Moon by microwave remote sensing of the Moon’s
surface and subsurface. Radiation Dosimeter is selected as the secondary payload for monitoring
radiations doses along the trajectory and a Narrow-Angle Camera as the tertiary payload.

5.3 TRAJECTORY
A trajectory assuming PSLV launch and GTO orbit profile of Chandrayaan with multi-burn phasing
strategy, WSB type transfer trajectory, Weak lunar capture and 250 km circular polar lunar science orbit
are the baselines. A total Delta-V of 1.718 km/s (nominal case) including delta-V for lowering to lunar
science orbit with pre-capture delta-V of 0.801km/s.

PHASING ORBITS: The current spacecraft trajectory design involves 2 phasing orbits from launch into
initial Geostationary Transfer Orbit. This provides the benefit of lower magnitude burns, flexibility in
injection corridor targeting and excess time for acquiring statistical correlation coefficients to improve
OD.

WSB TRAJECTORY: the Main feature of current trajectory is associated with lower propellant
requirements offered by utilising sun’s perturbative angular momentum transfer in Weak Stability
Boundary of Earth. While weak capture offered because of the Weak Stability Boundary of the Moon.
The spacecraft will traverse to the boundary of Earth’s sphere of influence and while returning towards
the Earth will get captured weakly around the moon. The final Trans-libration Cruise maneuver at the
phasing orbit periapsis a low magnitude patch point maneuver at the WSB to target negative C3 at the
moon.

WEAK CAPTURE AND FINAL LUNAR ORBIT: On targeting a negative C3 about the moon, spacecraft
will get captured in a highly eccentric unstable orbit. Efficient burns will be executed to bring
eccentricity to 1 with comparatively stable circular orbit. From there, an efficient multi periapse-point
burns will be executed to transcend in final lunar science orbit at 250 km altitude.

42
 5.4 PROPULSION
Dual Propulsion system comprising of PPS (Primary Propulsion System) for Orbit, Maintenance and
Control and RCS (Reaction Control System) for attitude maintenance and Control. The PPS is
configured, as a bipropellant (N2H4/ N2O4) system with helium pressurizer while, RCS uses
monopropellant hydrazine as fuel. The on-board system houses one 22N thruster as the Primary
propulsion system and four 1N thrusters as RCS.

5.5 COMMUNICATION
Communications system onboard SRMSAT-2 is designed for S-band deep space communication and
tracking with request for IDSN and ISTRAC support. The system is made compliant to necessary
standards and has been designed according to the parameters of ISTRAC and IDSN stations. Updation
of current VHF/UHF university ground station will be updated to facilitate S-Band communications
will form a part in integrated ground segment as SRM GS. This will provide support during the Earth
phasing orbits. On-board communications system comprises of one HGA and two near-omni directional
QFH-LGA. Also, to support deep space tracking and communication attributed to cruise profile a Deep
Space Transponder is configured in the design will be used to communicate to the IDSN.

5.6 GUIDANCE, NAVIGATION, AND CONTROL

Guidance Navigation and Control is responsible orbit determination, prediction, and control. Navigation
will be done by Navigation by a single station, two-way S-band ranging and Doppler tracking With
stringent constraints are placed on Orbit Determination. the subsystem will quantify OD requirements
for the mission and design optimal schedules of preliminary and precise OD to support all phases of
trajectory within bounds. Earth phasing orbits and long cruise duration will be used to improve OD
solutions for statistical un-modelled force coefficient determinations. The design of optimal pre-flight
deterministic maneuvers and in-flight design of statistical maneuvers will be done for trajectory/orbit
maintenance and control within flight envelopes. Dynamic determination of optimal low magnitude
will be done of the WSB maneuver at more than 1.3 x 106 km from Earth to target weak capture. On-
ground propagation and maneuver planning will be executed by ground segment.

5.7 ATTITUDE DETERMINATION AND CONTROL

ADCS will provide necessary fulfilment of orientation requirements for operation modes configured in
the Flight and operations plan along with providing required pointing for acquisition of relevant Payload
data in the respective Event and Orbit schedules. SRMSAT-2 features Active Attitude Control with
Reaction Wheels and RCS thrusters. Attitude Determination on-board system comprises of a miniature
star tracker , 2 IMUs and 8 PSD-based Sun sensors.

43
 5.8 STRUCTURE
The SRMSAT – 2 bus, is a cuboid enveloping a volume of 50 x 50 x 60cm3. It is a microsatellite with a
net mass of about 85kg. The structure uses honeycomb sandwich panels for the skeleton of the Bus. All
major components are made of Al-6061, a grade of Aluminium alloy. It provides sufficient room for
accommodating a propulsion module, a set of 4 reaction wheels, 4 Reaction Control Thrusters, a
parabolic antenna, two QFH antennas and other COTS components.

5.9 THERMAL
The thermal subsystem consists of a three-axis stabilized thermal design which considers survival range
of temperatures to survive the harsh space environmental load differential in the final lunar orbit. The
satellite uses MLI (Multi-Layered Insulation), conformal coatings and Black paints for passive thermal
control. Patch and Cartridge heaters will be used for active control according to the requirements.

5.10 ON-BOARD COMPUTER

The on-board computer system consists of dual micro-controller architecture. One microcontroller
works as the master computer while the other as flight dynamics computer. This configuration enables
full redundancy in case of hardware failure.

5.11 GROUND SYSTEMS

Ground segment systems would support various operational requirements of the mission. The flight
dynamics facility (SRM FDF) , SRM Lunar Science Data Centre (SRM LSDC) , S-band University
Ground station (SRM GS) , Mission Data Center (SRM MDC) and Mission Operation Center (SRM
MOC) together comprise of an integrated ground segment interfacing with ISTRAC/IDSN for telemetry
and telecommand relay through a SLE. The segment systems would be supported by SLE and would
be CCSDS compliant. SRM Lunar Science Data Centre will handle and maintain downlinked payload
data.

44
45
 6. MISSION OBJECTIVES

6.1 SCIENCE
 To design and demonstrate the effective operation of On-board Science Instrument (OSI)
subsystem.
 To acquire lunar surface and subsurface geology and morphology data of the Moon.
(For current payload options):
 To acquire and make available low to medium resolution data sets of measurements
 To acquire local brightness temperature data over moon’s surface.
 To investigate regolith characteristics by microwave radiometry.
 To characterize radiation environment in the chosen mission profile.
 To acquire photos of the Earth and Moon during departure and approach respectively
 To image specific lunar sites of interests. (Impact Craters, Permanently Shadowed Regions,
basins and landing sites)
 To qualitatively and quantitatively assess the importance of on-orbit science data acquired from
COTS payload instruments aboard small-satellite platforms.

6.2 TECHNOLOGY DEMONSTRATION

 To design and develop a lunar orbiter spacecraft within micro-satellite mass constraint.
 To design and demonstrate an unconventional Low Energy Transfer trajectory to the Moon.
 To transfer the spacecraft from GTO parking orbit to the Lunar Mission Orbit with lowest
imposed ∆V
 To implement an on-board propulsion system to execute effective orbit and attitude control.
 To design and implement a reliable on-board communication architecture to support tracking
(orbit determination) and telemetry optimally, at cruise apogee, the complete transfer trajectory
and lunar complying with design performance in respective phases.
 To demonstrate the capability of COTS component in lunar orbiter type spacecraft systems in
harsh earth-bound and deep space radiation environments.
 To design Spacecraft Thermal Control system to effectively counter thermal loads, especially
high fluctuation in lunar mission phase and contain internal temperatures within respective
tolerance ranges.

46
 7. SUBSYSTEM REQUIREMENTS
7.1 TRAJECTORY
 To design transfer trajectory from GTO to the final lunar mission orbit with the minimum ∆V
requirement to achieve the desired final orbit.
 To design optimal manuevers for orbit raising, plane/phase change during the early stages of
the trajectory to avoid redundant plane change manuevers in the later phases.
 To design optimal transfer with minimum waiting time after orbit injection.
 To assess the dependency of trajectory and waiting period for minimum energy transfer
opportunity in case of launch delays (from minutes to days scale).
 To design optimal WSB maneuver in the WSB region of Earth (1.5 million to 2 million
kilometres) to attain the final inclination and eccentricity.
 To quantify flight envelops and injection corridor to layout the requirements of orbit
determination.
 To assess contingent trajectory options in case of injection errors and missed transfer
opportunities.

7.2 COMMUNICATION
 Provide effective high data rate communication uplink and downlink from the satellite to
ground station for payload and engineering data transmission using high gain antenna when
necessary
 Provide low data rate receive only communication link throughout the duration of the mission
using omni directional antenna
 Modulate downlink data for transmission suitability and demodulate uplink data for data
interpretation by the onboard computer
 Provide an in orbit reconfigurable communication system
 Facilitate satellite tracking using doppler ranging
 Achieve stable communication link within the power, size, and mass constraints

7.3 ADCS
 To 3-axis stabilise the satellite against any external disturbances
 To provide adequate pointing and determination accuracy as required by mission
 To identify requirements and select component accordingly

7.4 ON-BOARD COMPUTER

 To provide effective interfacing with all the subsystems on board the satellite.
 To select a microcontroller that is suitable for the mission based on core architecture, instruction
sets, processing speed, interface support, power consumption, operating temperature range,
development environment.
 To formulate a Software Architecture from the kernel layer to the subsystem application layer
as a layered hierarchy with Fault Detection, Isolation, and Recovery.

47
 To collect and organize telemetry data and encapsulate them in the form of packets and to
extract telecommand packets in compliance with CCSDS Space Packet Protocol.
 To use error detection and correction schemes to prevent errors that might occur as a result of
exposure to harsh environments.

7.5 POWER

 Supply required power to all subsystems throughout the duration of the mission
 Store energy and supply uninterrupted power during eclipse and peak power demand.
 Provide an optimum bus voltage that would best fit the power system architecture.
 Regulate the power supply to maintain the electrical conditions at satellite components’
requirements
 Provide Electrical Fault Detection and Protection mechanisms.
 Minimize single event effects and maintain reliable power supply during single event transients.

7.6 STRUCTURE
 To design a bus of capability to envelope all the required components in a volume available in
the Launch Vehicle
 To configure a structure with minimum mass while maintaining the required structural
efficiency.
 To design a bus adequate strength to withstand the Quasi-Static loads and avoid the dynamic
coupling during the launch.
 To provide sufficient Margin of Safety to sustain all 2𝜎 and 3𝜎 loads for the spacecraft structure
 To select appropriate material to meet the structural and thermal design requirements.

7.7 THERMAL

 To control the effects of the external environment to maintain the temperatures of the internal
components in their operational and survival ranges during operation and otherwise
respectively.
 To protect the internal components, especially the thermally delicate components like COTS
electronics from the large thermal stresses developed in the lunar orbit phase.
 To protect the electronic components from plasma radiation of cosmic, coronal and trapped
radiation belt origins.
 To maintain an ambient temperature within the limits of the weakest component throughout the
mission.

48
 8. CURRENT DESIGN SUMMARY

8.1. TRANSFER DESIGN

Lunar Orbit Parameter

Final Lunar Orbital Parameters
Periapsis Altitude 250km
Eccentricity 0
Inclination 90°
RAAN 60°
Argument of Periapsis --- (270o)
w.r.t injection
trajectory

TRANSFER DESIGN CONSIDERATIONS

 Launch: Similar to Chandrayaan into GTO (180 inclination)

 Two Earth Phasing orbits after launch injection to GTO
 Trans-Libration Cruise Maneuver (TLCM) design to inject into cruise trajectory
 Cruise to about 1.5 x 106 km (Simulated Nominal trajectory – 1.467 x 106 km.
 WSB patch point maneuver to target lunar orbit injection (ideally 0 m/s ) executed at near
apogee of cruise trajectory
 Injection into weakly captured polar (900) lunar orbit
 Stabilisation to 250 km circular lunar orbit (Final science mission orbit.

Delta-V Analysis
(Sample Nominal Transfer)

Maneuver ΔV (km/sec)

Phasing Maneuver (GTO to EPO1) 0.3899

Phasing Maneuver (EPO1 to EPO2) Vx Vy
0.1134 0.1192
TLI Maneuver 0.2436
WSB Patching Maneuver 0.05 - 0.08
Lunar Capture 0.1 – 0.2
Descent to Final Science Orbit 0.7 – 0.8
Additional Orbit transfers to Final Science --- (Statistical)
Mission Orbit
Total 1.7605

49
 8.2. POWER BUDGET
(Lunar Phase Design )

Operating
Power Duty Orbit Margin Total
Subsystem Component
Requirem Cycle Average % Power
ent
Sun Sensors 400mW 100% 400mW 10% 440mW
Star Sensor 700mW 100% 700mW 10% 770mW
Reaction
ADCS 6W 90% 5.4W 10% 5.94W
Wheels
RCS Thrusters 20W 15% 3W 5% 3.15W
IMU 45mW 100% 45mW 10% 49.5mW
RF Network 1W 100% 1W 10% 1.1W
Power
75W 0% 0W 10% 0W
Amplifier
LNA 3.6W 100% 3.6W 10% 3.96W
Communications
Transponder –
30W 52% 15.6W 10% 17.16W
Rx & Tx
Transponder -
5W 48% 2.4W 10% 2.64W
Rx
On-board data On-board
4W 100% 4W 10% 4.4W
handling computer
Primary
Propulsion 40W 0% 0W 10% 0W
Thruster
Payload Payload 15W 18% 2.7W 10% 2.97W
Thermal Battery Heater 10W 100% 10W 10% 11W
Average
Average Power
48.85W 53.58W
Power with
Margin

50
 8.3. MASS BUDGET

Components Mass
(kg)
Antenna Mount Back Face 0.081
Base plate Honeycomb 1.96
Battery Module 1.95
He Tank 1.14
Hinge × 4 0.81
Hydrazine Tank 4.02
Narrow-Angle Camera 0.03
NTO Tank 5.48
Parabolic Antenna Dish 1.57
PCB Module 0.62
Power Amplifier 0.38
PPS Thruster 0.61
QFH Antenna × 2 0.3
Reaction Wheels x 4 3.84
Radom 0.1
RCS Mount × 4 0.13
RCS Thruster × 4 0.87
Side Panel Hcp 1_3 × 2 4.63
Side Panel Hcp 2_4 × 2 4.47
Solar Panel 9.92
Star Tracker 0.4
Sun Sensor Top × 2 0.16
Sun Sensor Module × 6 0.44
Top Deck Honeycomb 1.97
Transponder 2.6
Hydrazine 13.72
NTO 27.78
Pressurizing tank 0.29
Total Mass 86.43

SUBSYSTEM MASS (kg) PERCENTAGE

STRUCTURE 25.60 28.68 %
TTC 4.85 5.43 %
PROPULSION 42.10 47.17 %
ADCS 4.84 5.42 %
POWER 11.87 13.29 %

51
 9. OPERATIONAL MODES
1. INITIALISATION
2. SYSTEM COMMISSIONING – 0
3. DETUMBLING
4. ATTITUDE HOLD MODE
5. SYSTEM COMMISSIONING
6. SUN POINTING MODE
7. COMMUNICATION MODE
8. SAFE MODE
9. MANEUVER MODE

1. INITIALISATION
After the spacecraft detaches from the retainer lug of the IBL interface, initialization mode
commences. The OBC boots up and checks all the peripherals, sensors, and other hardware units. The
battery status is checked to make sure enough power is available to assist detumbling and solar panel
deployment. Otherwise, OBC moves into low power, and the batteries are allowed to charge. FDIR
(Fault Detection Isolation and Recovery) is initiated if a fault is detected, logged and transmitted to
the Ground Station.

2. SYSTEM COMMISSIONING – 0
For safety reasons, the fuel, oxidizer and pressurizer tanks are closed using NC pyrotechnic valves. As
RCS Thrusters are required for detumbling, it is necessary to commission the propulsion system. The
functioning of all the valves are checked, and any residues in the pipes are vented out. After which, the
pyrotechnic valves connected to the propellant and oxidizer tanks are signalled to open, thus priming
the field lines with oxidizer and propellant. Once priming is complete, the pressure regulator is set to
the desired pressure and the pyrotechnic valve connected to pressurising tank (Helium) is opened. Any
faults detected during this process are logged.

3. DETUMBLING MODE
After being deployed into an orbit, the spacecraft experiences high angular rates about its orthogonal
axis. The ADCS calculates the angular rates using various deterministic methods while the control
algorithm computes the required torque, and thereafter signals the RCS Thrusters to execute the
manuever. The Reaction Wheels remain inactive during this mode as it might get saturated during the
process.

4. SYSTEM COMMISSIONING – 1
In this operation, the solar panels are deployed, and the IMUs are calibrated with respect to the stabilised
body axis.

52
5. SUN POINTING MODE
This mode corresponds to the normal operation mode of the Spacecraft.
For maximum power generation, the Spacecraft is orientated such that the solar panels face the Sun
directly. As the spacecraft has omnidirectional communication, it will be ready to receive any
commands from the Ground Station.

6. COMMUNICATION MODE
The omnidirectional QFH (Quadrifilar Helical) Antenna provides the capability of receiving signal
throughout the mission without reorienting the antennas. On reception of a signal from the Ground
Station, the spacecraft switches to Communication Mode. Normally, QFH Antennas are used for
telemetry and telecommand. On occasions when high data transfer rates are required, such as to transmit
payload data, the satellite can be commanded to operate through the high gain parabolic antenna. Once
commanded, the Spacecraft reorients itself to point the parabolic antenna towards Earth. To differentiate
the communication between the two, the QFH uses LHCP (Left Hand Circular Polarization) while the
parabolic antennas employ RHCP (Right Hand Circular Polarization).

7. ATTITUDE HOLD MODE

This mode is used for mission-specific tasks such as to assist payload operation, calibration, and any
other targeting operations. The spacecraft reorients and maintains the required attitude until the
completion of the operation.

8. MANEUVER MODE
During orbital manoeuvers, the spacecraft thrust vector needs to be properly aligned to the desired
attitude with strict regard for spatial accuracy. The spacecraft will be three-axis stabilised using RCS
Thrusters alone. For sufficient delivery of power to the Propulsion system, the Communication Systems
are shut down.

9. LOW POWER MODE

The Spacecraft switches to this mode on command from the Ground Station during the prolonged cruise
phase. During this mode, all spacecraft operations are minimal, and it moves into a state of hibernation
for a stipulated time period as commanded by the ground station. The Spacecraft will be Sun pointing
and the frequency of operation of ADCS will be brought down to a bare minimum. Albeit this, the
Radiation Dosimeter will be functional and logging the data.

10. SAFE MODE

This mode is activated by the FDIR algorithm in situations like low power, prolonged eclipse time, etc.
All actuators and sensors are switched off, and OBC operations are minimized or idled or put to sleep.

53
 10. FUNCTIONAL MODES
The functional modes are mission specific modes, which governs the spacecraft's operations during
mission phases.

Sr FUNCTIONAL
DESCRIPTION
No MODES
System Check Mode
1 SYSCHECK Checks all peripherals, hardware for anomalies, logs data, statuses, executes
FDIR
Normal Mode
This is the normal functional mode of the spacecraft during the earth phase. The
satellite will be able to receive commands using the omnidirectional antenna.
2 NORM
Commissioning of all systems excluding that of the primary payload will be
performed. The spacecraft will be three-axis stabilised and sun pointing. All
systems will be checked, analysed and any problems will be rectified.
Earth Pointing Communication Mode
This mode corresponds to the Communication Mode of the spacecraft during
the earth phase. It employs omnidirectional QFH and high gain parabolic
antenna for its operation. Incase of a parabolic antenna, the spacecraft will be
reoriented so as to point it towards the earth. This functional mode is specific to
the spacecraft alone and doesn’t pertain to any ground segments.

All spacecraft tracking and orbit determination fall under this. As

tracking is crucial to determine the course of the spacecraft, it is
3 EPCOM 0/1
0 necessary that the spacecraft always responds to Doppler ranging
signals during the mission lifetime. The QFH antenna will be used to
track the spacecraft as we have necessary margins for tracking even
from the farthest point.

Generally, the QFH antenna will be associated with uplink and

1 downlink of data packets. This includes transmission of housekeeping
and other logged data as well as reception of tele commands.

Navigation-Communication Mode (Ground Segment)

4 NAVCOM 0/1
(Mode is predominantly defined for operations of Ground Segment)
Preliminary Orbit Determination
Used primarily after launcher separation and after apogee/orbit raising
maneuvers. The Ground segment operates to determine coarse orbit
0
parameters to obtain acquisition information for POD by tracking from
other stations. The tracking is supported on-board mainly by the LGA.

Precise Orbit Determination (POD)

After signal is acquired at the respective ground station, the ground
1 segment prepares for precise orbit determination with active operations
of FDF with ISTRAC/IDSN station. Data is processed in batch arcs
from AOS to LOS and orbit solutions are generated.

54
 Guidance Communication Mode (Ground Segment)
5 GCCOM 0/1 (Mode is predominantly defined for operations of Ground Segment)

Test burn
Pre-designed test burn strategy mainly associated to the RCS and thrust
0 vector calibrations in the second phasing orbit. Test burns before the
second phasing orbit are included in system check. The test burn schedule
and guidance data telecommands are uplinked.
Maneuvers
After system calibration and decision making from navigation and
calibration test data, system readiness is validated. The FDF reviews the
1
transfer and targets the closest and optimal transfer window with guidance
plan. Approval is followed by uplink of maneuver information
telecommands.
Apogee Raising Maneuver Mode
All trajectory and orbital correction within earth orbit come under this mode.
6 ARM
The spacecraft is 3-axis stabilized in the required thrust vector and all other
hardware like communication systems are switched off.
Deep Space Operations Mode
The operation of the spacecraft under the final phasing orbit and until the
7 DSO 0/1
spacecraft reaches the lunar orbit is governed by this mode.

Normal mode
0 In normal mode, the spacecraft is 3-axis stabilised and sun pointing. The
spacecraft is ready to receive at any point using the Omni-antenna.
Hibernation
In Hibernation mode, the spacecraft is made to which under low-power
1
mode. The attitude corrections will be sampled down. The radiation
monitoring device will log the data.

8 DSM TC/1 Deep-Space Maneuver Mode

Mid-course trajectory correction

(Deterministic and Statistical)
After setting the maneuver time, communications system is turned off.
TC
Attitude is set to 3 axis stabilised with required thruster pointing. It is
followed by Sun acquisition and transfer maneuver execution
information downlink by LGA.
Patch point maneuver
(Deterministic)
1
Same as DSM_TC. It would be operated once at the WSB patch point
for injection targeting.

9 LIM Lunar Insertion Maneuver

55
 Lunar Orbit Stabilization Mode
The mode is initiated to bring the spacecraft to the final science orbit. The
communication system is switched off. The spacecraft is three-axis stabilised,
and orbital correction manuevers are executed.
10 LOSM

Primary Payload Commissioning

In payload commissioning mode, the primary payload is switched on, and
11 PCOMS
necessary calibration is done.

Lunar Target Pointing Mode

This mode is for payload operations. The primary payload will be made to
12 LTARP
point at the required lunar surface, and the spacecraft will be 3-axis stabilised.

Lunar Operations - Normal Mode

This is the normal mode of spacecraft in lunar orbit. The spacecraft will be
13 LNorM sun-pointing and will be ready to receive command using omni-
communication

Lunar Operations – Communication Mode

This is the lunar communication mode in which, the omni-antenna will be
14 LCom used for transferring logged data. When payload data has to be transferred, the
high gain parabolic antenna will be used.

Lunar Operations - Station Keeping Mode

In this final science, the orbit will be periodically corrected to account for
15 LSTK
perturbations. The communication system will be switched off.

56
 11. GROUND SEGMENT

11.1. INTRODUCTION
The Ground Segment is composed of ground operations, organizations and ground systems that together
form the ground infrastructure required to support the mission from its preparatory stage to its
completion. Spacecraft operations mainly consist of monitoring of telemetry data received at the ground
station, handling housekeeping data, planning mission executing and take decisions, and configuration
of the on-board equipment to execute the mission. This comprises not only the design, implementation,
and testing of the ground segment but also planning of the operations in the earlier stages of the mission
development.

The Ground Segment comprises of:

 THE GROUND STATION

o University S-band Tracking Station (SRM SGS)
 THE FLIGHT DYNAMICS SYSTEMS
o SRMSAT Flight Dynamics Facility (SRM FDF)
 GROUND DATA SYSTEMS
o SRM Mission Data Centre (SRM MDC)
o SRM Lunar Science Data Centre (SRM LSDC)
 FLIGHT OPERATIONS SYSTEMS
o SRMSAT Mission Operations Centre (SRM MOC)

This infrastructure would be established with an interface with ISRO’s IDSN/ISTRAC by an SLE after
consultation.

11.2. THE GROUND STATION

The Ground Station provides the communication interface with the spacecraft. The interface is provided
by antennas present at the ground station. The demodulation of telemetry data (Data Processing) and
transmitting command to the spacecraft are systems within the ground station which play an equally
important role.

1. Antenna

The characteristics of the mission which has to be supported determine the type of antenna that is
required. The ground station will be upgraded from its current VHF and UHF capabilities to S-band
capability to communicate with the spacecraft during its Earth bound phase in the phasing orbits.
However, because of the inevitable need for high fidelity communication architecture to suffice the
communication requirements, optimally, during the cruise phase, IDSN support is elementary and thus
requested.

2. Data Processing

57
After reception of the signal by the antenna, it must be processed further before being ready to be sent
to the mission control center. The frequency of the signal then decreased by a down-converter, from RF
carrier level (S-Band) to an intermediate frequency level. This down-converted frequency is then
processed by the Baseband Unit, which is the central component of the ground station. The baseband
unit checks the data for errors or transmission losses by performing Cyclic Redundancy Checks, bit
synchronizations, etc. The result of this data processing is a data stream containing telemetry data which
is sent to the mission control center for further processing. As spacecraft and ground segment comply
with CCSDS recommendations, the data from the ground station will be sent to Mission Control Center
using Space Link Extension.

The data processing is similar for data to be transmitted for commanding the spacecraft. The baseband
unit receives command bit stream form the Monitoring and Control System (MCS, see section ) which
is modulated and converted from a digital to an analog signal. This processing is done at an intermediate
frequency which is then increased using a High Power Amplifier (HPA). This final amplified signal is
radiated by the antenna.

3. In-Orbit Tests

In-Orbit Tests (IOT) of the spacecraft are performed to verify the compliance of the on-board
instruments with their specifications and to ensure that their operations have not been degraded during
launch. IOT or system check schedules also allow the measurement of the characteristics of these
instruments and are often repeated throughout the mission to assess performance variation and suitable.
Owing to COTS component systems, these IOTs or system check schedules will form an elementary
part of an operation during Earth Phasing Orbits. The current design of IOTs is bifurcated into routine
system checks and System calibration schedule. Routine system checks would be an elementary part of
post-maneuver operations and performed routinely after each major operation event like a maneuver.
System calibration schedule will be the main focus of operations in the second phasing orbit where
instruments will be calibrated, and sufficient data will be acquired to correct for any errors pertaining
to the operations. The total on-board and ground system readiness to carry on the mission with
compliance of test results with the pass criteria in the flight rules from the critical design will be
evaluated. It’d be an exhaustive schedule of pre-defined tests and calibrations, a major part of which
would be with respect to the PPS and RCS, with test burn strategy to identify and fit un-modelled thrusts
in the flight dynamics systems. The sensor readings, attitude and orbit data from the previous orbit
(Phasing orbit 1 ) would be used to process and identify errors to enhance system performance for
critical phases of the mission requiring stringent operation constraints.

11.3. FLIGHT DYNAMICS

Flight dynamics team will overlook the systems in the SRM FDF and will deal with all aspects of the
mission, related to the spacecraft orbit and attitude monitoring, changes, and execution during the
mission. While before the mission, in the preparatory phase, they will mostly be involved in the mission
formulation and analysis. Based on this analysis, they supply other systems (ground station, flight
operations, etc.) with predicted orbital information (Operational modes, Orbit Schedules, Event
Schedules, etc.) to support their preparation. During the mission, flight dynamics experts perform orbit
and attitude determination and control, executed in real time and deliver updates of the flight profile
and changes to the other operation systems.

58
1. Mission Analysis and Operational Products

Flight Dynamics experts are involved in the following processes which outline the main outputs
required by the ground segment:

 Identification and selection the ground station or network of ground stations that satisfy the
mission requirements in terms of ground coverage.
 Orbit/Trajectory maneuver planning to reach the final orbit
 Launch and transfer window and opportunity selection, taking into account the requirements in
terms of final orbit, maneuver planning, and launch availability.

The strategies for final orbit acquisition and station keeping also come under mission analysis and its
main outputs. The orbit acquisition strategy can only be finalized after completion of the orbit
determination after the last maneuver. Apart from these main outputs, the specialized operational
products are:

 A list of orbit related events: This is the basis for mission planning. It defines events in
chronological order. Eg. Periapsis and apoapsis crossings, eclipse entry/exit, etc.
 Ground Station Events: This is similar to the orbit-related events but focused on the ground
segment. Eg. Acquisition and loss of signal.
 Ground Station Ephemerides: It provides information (azimuth and elevation) to point the
antenna during a spacecraft pass or during communication/tracking mode.

All these operational products are specified in the Interface Control Documents (ICD).

Another category of operational products to be specified are the command and telemetry files that are
exchanged between flight dynamics and flight operations during mission execution.

2. Orbit and Attitude Determination

The first opportunity for state update comes during the launch phase or immediately after separation.
Based on actual launcher performance, flight dynamics experts have to decide whether or not, to update
the ephemeris of the first acquisition in the ground station. It is a critical decision that has to be taken
possibly before the acquisition of signal (AOS) or during the first contact. This update will determine
whether the spacecraft can be successfully tracked during first contact. Usually, tracking using the
orbital elements provided by the launcher becomes increasingly difficult with time. In such cases, two
line elements (TLE) provided by the US Strategic Command can be used. Though, the availability of
these TLEs may vary from a few hours to several days after launch.

Immediately after first successful contact, the flight dynamics experts can begin with their major task
of Orbit Determination. Measurements are required over at least one complete orbit revolution. Orbit
determination is required after each orbit maneuver. The comparison of orbital elements prior to and
after the maneuver allows the determination of the ΔV actually achieved during the burn. The difference
between this and the planned ΔV is used to determine the calibration factor (usually applied to thrust
level) that can be taken into account in the planning of the next maneuver. The thrust contribution to

59
ΔV due to attitude maneuvers performed using thrusters needs to be separated by performing
measurements just before thruster firings.

Attitude Determination is also the responsibility of the Flight Dynamics during the mission.
Measurements, which are carried out autonomously, depend on the sensors and actuators implemented
on board the spacecraft’s attitude control system. The result of the attitude determination can be used
in the preparation of attitude maneuvers, in the same way, orbit determination can be used for orbit
manoeuver planning.

11.4. THE GROUND DATA SYSTEM

The Ground Data system personnel are responsible for the ground segment infrastructure required to
support ground operations. The exchange of data within the control center or with external connections
is done via communication links. Spacecraft specific implementation and testing of the whole ground
segment is necessary to ensure its proper functioning.

1. The Control Center and Network Infrastructure

The Control Center hosts all personnel involved in the mission. Many computers with redundant critical
hardware are accommodated together and combined into a network. The redundancy switching is
autonomous in order to optimize system availability. All the data in the control center is backed up
periodically to avoid any loss. For precise timing applications, all computers are synchronized to
Coordinated Universal Time (UTC), which is distributed over the whole center network. The public
power supply is often a single point of failure, and it is necessary to install Uninterrupted Power Supply
(UPS) based on a separate source to bridge the possible outages. The switching to the secondary power
supply is controlled autonomously by software. Documentation related to control center design is
essential, especially for long-term missions when changes in personnel occur periodically.

2. Communication Links

The most obvious need for a communication link is between the ground station and the control center,
through which telemetry data is forwarded from the antenna to the Mission Control Center (MCC).
Command data obviously moves in the opposite sense, to be radiated to the spacecraft. All these data
are usually carried by leased IDSN lines or through the internet. Attention must be paid to the data
formats - so called ‘gateways’ are often required for protocol conversion.

Communication links are used for delivery of operational products to the different systems involved in
the operations. Apart from all data exchanged between machines, a voice conference system with
different channels called ‘voice loops’ is often implemented for efficient communication between all
personnel involved in the mission.

All these communication links must be documented in the ground segment description, but more
importantly in interface control documents.

11.5. FLIGHT OPERATIONS SYSTEM

The flight operations team is in charge of conduction the operations, which consist mainly of
commanding the spacecraft and monitoring its telemetry data. Both these tasks are performed with the
help of monitoring and control system, which processes telemetry and command data in near real time
and archives them.

60
1. The Monitoring and Control System

The Monitoring and Control System is the heart of operations. On the telemetry side, it processes the
bit stream generated at the ground station and transmitted to the control center. It also presents these
processed data to the flight operations in a meaningful way. On the command side, it translates the
orders from the operators into a bit stream that the spacecraft can interpret. The command process is
performed directly from the command queue either manually by sending them one by one, batched
together or automatically.

There are two additional important functions in the command process. The first is the Pre-Telemetry
Verification (PTV) – this ensures that the values of a list of telemetry parameters must be within a
specified range before the command can be sent. The second is the Command Execution Verification
(CEV) – this checks that a list of telemetry parameters are within a specified range to indicate that the
command execution has been successful.

All the information required by the MCS is stored in a database, which is delivered to the flight
operations team with proper documentation.

2. Archiving and Off-line Processing

Telemetry data processing is usually performed near real time, that is to say that there are only a few
seconds delay between the data generation on-board and its processing on the ground. This is generally
the case for geostationary satellites. Near real-time processing, however, is not always possible for deep
space missions. In such cases, data is usually stored on-board and dumped together with current real-
time data, during a contact with a ground station.

All data down-linked from or up-linked to the spacecraft must be permanently archived, including off-
line processed data that has to be merged with the real-time data. The temporal ordering of the data in
such an archive is usually done using the on-board time, especially if off-line data is present. Attention
must be paid when a time reset occurs. Eg. After a reboot of the on-board computer.

3. Flight Procedures and Validation

Due to the high cost and complexity of a spacecraft, there is no place in operations for improvisation.
Many constraints, either at sub-system level (equipment temperatures, warm up times, fields of view,
etc.) or at the system level (power balance, available memory, angular momentum, etc.), need to be
taken into account when commanding the spacecraft. All the relevant information must consequently
be summarized in Flight Operation Procedures that are executed during the mission.

A flight operations procedure can be seen as a list of steps in chronological order, itemizing the actions
to be performed. Flight procedures need to cover not only all nominal activities expected during the
entire mission but also a set of well-chosen contingency cases.

During the validation process, the definition of a common set of display pages is agreed between the
flight operations team and the spacecraft fabrication team, which is important for the mission execution.
Since the spacecraft fabricator is often asked to support operations, it is important that experts from
both teams interact easily. The validation status of flight procedures is presented at the Operational
Readiness Review (ORR)

4. Mission Planning

61
The Mission Planning System (MPS) builds the interface between the science team, the space segment,
and the ground segment. It aims at using space segment resources optimally to supply the science team
with products (images, measurement data, etc.).

The timeline produced for the low lunar orbit phase by the mission planning process is a detailed
sequence of all the events foreseen during this phase of the mission, timed accurately to the second. It
is based on operational products delivered by the flight dynamics and augmented by the activities from
flight operations and ground data systems. The sequence of events is the guideline for all personnel
involved in the mission, and consequently takes all aspects of the mission into account.

5. Mission Execution

The success and overall quality of the mission execution is dependent upon the thoroughness of
activities undertaken in the preparatory phase. Great efforts must be made to get all personnel involved
in the mission execution to a level of proficiency where no question remains unanswered, no uncertainty
remains about the roles and tasks to be performed.

The role of each person involved in the mission execution must be defined clearly. Personnel from the
flight dynamics team, the ground data team, the flight operation team and the representatives of the
science team are involved and ultimately decide what is to be done in the case of non-nominal
operations. All these people form the combined flight operations team, which is lead by a flight director.
This organisational structure must be documented in the mission operations plan. Decision authorities
are clearly identified. The list of flight operational procedures and the mission plan combine to complete
the flight operations plan. From this, a training and simulation plan may be developed to give the
required proficiency.

62
 1. TRAJECTORY DESIGN
BALLET
(BALlistic Low Energy Transfer)

1.1 INTRODUCTION AND BACKGROUND

University class small-spacecraft missions to Moon as apprehended, rely on piggy-backing with
a pre-scheduled launch, restrict any control over launch opportunities. Also, non-conventional small-
spacecraft platforms incorporating propulsion and other systems, increase the uncertainty of the overall
mission and impose a greater ΔV requirement. Thus, establishing a mission design in typically preferred
fast transfers (like Hohmann transfer) is not chosen. Low Energy Transfers, like the chosen WSB (Weak
Stability Boundary) transfer increase flexibility, relieve transfer constraints and offer contingent capture
options at the cost of increased transfer time. Apart from these factors, most importantly, it reduces the
total transfer ΔV, by an ideal maneuver-free ballistic capture. However, chaotic nature of low energy
transfer will exert reliability requirement on trajectory estimation, demanding for more robust
navigation. Thus, orbit uncertainty quantification and propagation are crucial factors and will be the
drivers for planning navigation strategies. Thus, an elaborate study on dynamics of the transfer problem,
its constraints and benefits is done along with the preliminary trajectory.

When assessing several possibilities including gravity assist to escape the Earth-Moon system,
circumnavigation and flat angle re-entry et cetera to save transfer energies, in the early days of space
exploration, three body consideration served as an invaluable mathematical formulation. The three
Body Problem (3BP) in its various forms (PCR3BP / R3BP) and its energy integral gave solutions to
most of the problem statements numerically or otherwise with an exception of automatic/ballistic
capture of a projectile launched from Earth. V. A. Egorov explicitly ruled out the possibility of energy
free lunar capture in a restricted 3 Body Problem formulation involving the Earth, the Moon and the
Spacecraft.

When it first came to light from the work of Belbruno and Miller to salvage Muses-A, it was eminent,
that role of Sun’s influence on the trajectory is crucial. Thus, a Four Body Problem was defined as the
base of these low energy transfer trajectories. Unlike the three body problem, a complete analytic
solution is not available. More importantly, an aiding relation like the Jacobi Energy integral based
static surfaces or singular points, is not available for the problem. However, sufficiently aiding Weak
Stability Theory and Dynamical systems theory to assist trajectory design by numerical methods using
synchronised simplification or articulation into two three body problems patched together by
constrained parameter targeting or dynamical systems theory is applicable.

1.2 BALLISTIC CAPTURE

Spacecraft propellant consumption is related to the velocity change ∆V. The only spacecraft propulsive
velocity change required for lunar capture is a small 30m/s maneuver that is performed near the weak
stability boundary region and a small maneuver less than 90 m/s to achieve high eccentricity stable
orbit. By contrast, the Hohmann transfer trajectory requires 200m/s insertion ∆V at the Moon. Low
Energy Transfers through WSB result in a ‘weak capture’ at the Moon where the Two body energy of
spacecraft with respect to Moon is negative. This means that motion around the Moon is unstable and
will escape the Moon after a short period of time. C3 remains negative for finite time. A minimal ∆V
is required to achieve relatively stable orbit.

63
 Out of this weak capture, when propagated forward or backward, the trajectory escapes Moon
and transitions into an elliptical orbit around Earth resonant with orbit of the Moon. In other perspective,
a weak capture is transition between two resonance orbits around Earth and the idea of these transfers
can be conceived as a resonance orbit construction that arrives near Moon WSB with negative C3. Koon
et al found numerically that dynamical channels in form of invariant tori were associated to the change
brought about in transitioning between resonance orbit.

Classical Method of Earth Moon Transfer and comparison

The travel between Earth and Moon involves the solution of a two-point boundary value problem. This
can be realised by solving WSB or Dynamic system equations for Four Body problem.
The problem of travel to the Moon maybe conveniently separated into three separate phases; being the
launch into the Earth orbit, WSB maneuver and injection into the stable lunar orbit.

Hohmann transfer trajectories require an ellipse with the periapsis at the point of tangency with a
circular Earth orbit and the apoapsis at the Moon’s orbit. Large thrusting maneuvers, Moon at lunar
nodes and a circular orbit constrain the transfer. The accessibility of these orbits from Hohmann transfer
trajectories is limited by the direction of the hyperbolic excess velocity vector.

For convenience, the orbit about the Earth is described in an inertial coordinate frame of reference, with
the x axis pointing to the Moon, the z axis normal to the lunar orbital plane and the y axis forming the
right handed rectangular coordinate system.

The elements Ωe and ie relate to the direction of the Earth departure velocity vector. The element Ωe is
associated with the time of launch and ie is associated with the launch azimuth. Constraints on periapsis
radius or altitude of the Earth injection point, the element rae, defines the launch energy or propellant
required to burn out near Earth orbit. The inclination of the orbit is free to be specified and is selected
to place the orbit insertion point on the Earth side of the Moon resulting in a prograde orbit. The angles
Ωm and ωm define the orientation of the orbit selected to be the same as the elements of the approach
hyperbola and the orbit insertion maneuver performed at the periapsis. The comparison of two classes
of transfers is given below as a coarse analysis to insert spacecraft from 300 km

Earth Injection Moon Injection Total ∆V Duration

∆V (km/s) ∆V (km/s) (km/s) (Days)
Hohmann 3.103 ~0.709 3.812 ~5
Transfer (Ideal-
Moon at nodes)
Low Energy 3.171 0.496 ~3.667 ~84-120
Transfer
(Ideal)
Table 1.1: Comparison between conventional Hohmann Transfer and Low Energy Lunar Transfer

NOTE: Bi-elliptic transfer of similar kind would require 250 m/s correction maneuver at the boundary
of escape trajectory from moon to patch it to the ellipse outbound from the Earth.

Phasing Orbits
Orbit Phasing is the adjustment of the time-position of the object along its orbit, which can be describes
as changing the object’s true anomaly. In general, a phasing maneuver could be considered a two-
impulse Hohmann transfer which takes the object away from, then back into, its original orbit. To

64
optimise the burn maneuvers and for operational relevance two phasing orbits with 90,000 km and
1,30,000 km are selected. The ΔV required for the first and second maneuver is 0.44km/sec and
0.20km/sec respectively. Phasing increases the flexibility and provides with supplementary time in case
of emergency in the initial critical operation. Analysis was conducted on phasing orbits with plane
change maneuver. The initial inclination was taken to 18° similar to Chandrayan’s inclination, and the
ΔV was found to be 0.389km/sec and 0.1134km/sec respectively with VY as 0.12km/sec to achieve
inclination of 28° at 1,30,000km at the apoapsis to enable WSB. The plane change is combined with
the final phasing at the apoapsis to maximise efficiency.

Figure1. 1:Phasing Orbits with apoapsis of 90,000km and 1,30,000km.

Figure1. 2:Plane change maneuver combined with phasing maneuver at the apoapsis of 1,30,000km

1.3 WSB

WSB is elucidated as the region near the Moon where sensitive chaotic dynamics occur. It can
be stated as a region in position-velocity space, the generalisation of the Lagrange points. WSB region
was mapped out by Belbruno in 1986. A spacecraft reaching this chaotic region can be ballistically
captured at the Moon. In 1991 Japanese spacecraft Hiten was the first to demonstrate this ballistic lunar
capture transfer applying weak stability boundary (WSB) theory. Due to efficient and optimal fuel
saving attribute, this type of transfer has evolved since and known to be called as WSB lunar transfer.
The WSB lunar transfer saves up to 15% in the ΔV required for Lunar capture and reduces the ΔV
required for orbit maintenance. This transfer increases the flight time of the spacecraft and the requires
the spacecraft to attain an Earth apoapsis of 1.5 million kilometres.

65
The dynamic nature of the region is chaotic and trajectory in this region is very sensitive and unstable.
It is possible to approximately locate the region and thus confine to the location of the region rather
than its complex mathematical formulation.

Keplerian Ellipse

Low Energy Trajectory

Figure 1.3: Geometry for WSB lunar transfer

Restricted Three and Four Body Models:

This section provides with detailed description of the restricted three and four body models used to
understand and study the WSB region. Two masses m1 and m2 are considered as the primary bodies of
motion. They move around their common centre of mass and satisfy the condition m1 > m2. A third
body of negligible mass m moves in the gravity field produced by the two primaries with the condition
m << m2 < m1. The rotating reference frame used to study the motion of the third body is defined with
the origin located at the centre of the mass of the primary bodies, x-axis being along the motion of the
primaries, z-axis being normal to the plane of the motion of the primaries and y-axis forms a right hand
reference frame. Circular Restricted Three Body Problem (CRTBP) is stated as the model in which two
primary bodies move in a circular orbit around their centre of mass. If the primaries move in an elliptical
orbit the model is called Elliptical Restricted Three Body Problem (ERTBP). When in the case of
CRTBP the presence of Sun is taken into consideration with Sun rotating around the Earth-Moon system
with a constant radius ρ and constant relative angular rate ωs, the model is known to be Bi-circular
Restricted Four Body Problem (BRFBP). This model is a fair approximation for the Earth-Moon-Sun
system. The eccentricity of the Earth’s orbit around the Sun (eEARTH = 0.016) and that of the Moon’s
orbit around Earth (eMOON = 0.054) are very small, the inclination of lunar orbital plane with respect to
the ecliptic plane is only 5°.

Equations of Motion:
The equations of motion for the mass m for CRTBP model can be expressed as a function of potential
Ω3(x, y, z) as:
1 1−𝜇 𝜇 1
Ω3 (𝑥, 𝑦, 𝑧) = (𝑥 2 + 𝑦2) + + + (1 − 𝜇)
2 𝑟1 𝑟2 2

Where (1.1)
𝑚2
𝜇= 𝑚1 +𝑚2
is the mass parameter of the system,

66
𝑟1 = √(𝑥 + 𝜇)2 + 𝑦 2 + 𝑧 2 and 𝑟2 = √(𝑥 + 𝜇 − 1)2 + 𝑦 2 + 𝑧 2 being the distances of the third
body from m1 and m2 respectively.
1
2
𝜇(1 − 𝜇) is defined as a potential constant.

Obtaining the second order derivatives from 1, the equations of motion can be stated as:
(1−𝜇)(𝑥+𝜇) 𝜇(𝑥+𝜇−1)
𝑥̈ − 2𝑦̇ − 𝑥 = − 3 −
𝑟1 𝑟23

1−𝜇 𝜇
𝑦̈ + 2𝑥̇ − 𝑦 = − ( 𝑟3 + 𝑟3 ) 𝑦
1 2

1−𝜇 𝜇
𝑧̈ = − ( − )𝑧
𝑟13 𝑟23

(1.2)

For ERTBP model the instantaneous distance between m1 and m2 and the angular rate of the reference
frame will vary with time due to motion in elliptical orbits. Thus the equations of motion for this model
can be given as:
1−𝜇 𝜇
𝑥̈ − 2𝜔𝑦̇ = 𝜔̇ 𝑦 − 𝜔2 𝑥 − 𝑟13
(𝑥 − 𝑥1 ) − 𝑟3 (𝑥 − 𝑥2 )
2

1−𝜇 𝜇
𝑦̈ + 2𝜔𝑥̇ = −𝜔̇ 𝑥 + 𝜔2 𝑦 − ( − 𝑟3 ) 𝑦
𝑟13 2

1−𝜇 𝜇
𝑧̈ = − ( 𝑟13
+ 𝑟3 ) 𝑧
2

(1.3)

Where ω is the actual angular rate of the reference frame, x1 and x2 are the coordinates along the x axis
of the primary bodies and r1 and r2 are the instantaneous distances between the spacecraft and the Earth
and the spacecraft and the Moon respectively, defined as

𝑟12 = (𝑥 − 𝑥1 )2 + 𝑦 2 + 𝑧 2 and 𝑟22 = (𝑥 − 𝑥2 )2 + 𝑦 2 + 𝑧 2

(1.4)

The following dimensionless entities are considered for the motion of the three body:

1. The unit of distance is equal to the mean distance between the primary bodies.
2. The unit of mass is equal to the sum of the masses of the primaries.
3. The unit of time is taken as such the period of m2 about m1 is 2π.

For BRFBP model the equations of motion will be as a function of potential, given as:
𝑚𝑠 𝑚𝑠
Ω4 (𝑥, 𝑦, 𝑧) = Ω3 (𝑥, 𝑦, 𝑧) + 𝑟3
− 𝜌2
(𝑥 𝑐𝑜𝑠𝜑(𝑡) + 𝑦 𝑠𝑖𝑛𝜑 (𝑡))

(1.5)

Where ms is the mass of the Sun,

𝑟3 = √(𝑥 − 𝜌𝑐𝑜𝑠𝜑)2 + (𝑦 − 𝜌𝑠𝑖𝑛𝜑)2 + 𝑧 2 is the distance between the spacecraft and the Sun.

𝜑=𝜑
̅ (t) is the phase angle between the Sun and the Earth-Moon direction.

67
Hence the equations of motion for the spacecraft can be written as:
𝜕Ω4
𝑥̈ − 2𝑦̇ = 𝜕𝑥

𝜕Ω4
𝑦̈ + 2𝑥̇ =
𝜕𝑦

𝜕Ω4
𝑧̈ =
𝜕𝑧

𝜈̇ = 𝜔𝑠

(1.6)

The dimensionless characteristic values used in this model are:

1. The relative Sun angular rate is ωs = -0.9251

2. The distance between the Sun and the Earth-Moon centre of mass is set equal to ρ = 3.8881x1012
3. The mass of Sun is set to be ms = 3.289x105

Weak Stability Boundaries (2D):

The definition of the weak stability boundaries based on Belbruno’s algorithm is defined in the synodic
frame of reference. Consider a radial line l(v) from m2, defining an angle v with the x-axis of the
reference frame. Following criterion must be set in order to define WSB:

 The initial velocity vector (v) of the mass m must be normal to the radial line in a direct or
retrograde direction with respect to m2.
 The Keplerian energy of mass m must be negative with respect to the small primary body
m2.
 The eccentricity of the initial osculating orbit of the spacecraft is constant at each variation
of the initial point on the radial line.

The motion of the body will be stable if the following conditions are met:

1. The spacecraft must perform a complete revolution around m2 and must cross the radial line at
a point b belonging to l(v), being different from the starting point.
2. The Keplerian energy must be zero or a negative value with respect to m2 without going about
m1.

The motion will be deemed as unstable if the motion of the spacecraft undergoes the following
conditions:

1. After leaving the radial line, the spacecraft moves away from m2 towards m1 and makes a cycle
about m1 without crossing the radial direction l(v) or colliding with m1.
2. The Keplerian energy of the spacecraft is positive with respect to m2.

The transition of the motion of the spacecraft from stable to unstable can be used to define a set of
values for the radius r and anomaly v if the distance between the spacecraft and m2 along the radial line
is varied with a fixed eccentricity value. Keeping r* as the first transition radius value for each anomaly
v, with the help of Belbruno’s definition it can be stated that the orbit is deemed to be stable if r < r*
and unstable if r ≥ r*.

68
 Figure 1.4: WSB in Planer Restricted Circular Three Body Problem according to the definition given by Belbruno (with
osculating orbit eccentricity as zero).

Figure 1.5: WSB geometry depending on the initial eccentricity.

It can be said that slight variations in the initial conditions such as global or local variations of the
escape geometry, strongly affect the motion of the third body and its stability in the WSB region. A
small change of about 100km in the radius can cause the motion to be unstable. As illustrated in Figure
5, on increasing the eccentricity the stable zones reduce and become closer to Moon conditioning
capture only at low periselene altitudes. The low lunar transfer involves being close to the Moon with
high eccentricity and low periselene altitudes of the capture orbit. Both direct and retrograde initial
velocities of the spacecraft in WSB have the same effect on the size of the WSB when the eccentricities
are increased in both the cases. As stated above the size of the WSB region reduces and becomes close
to Moon with low periselene capture orbit altitude.

The stable points are the so called capture zone, corresponding to the conditions for the mass m that
lead to a stable orbit around the Moon, basically the requirement of negative Keplerian while

69
performing at least one complete revolution around Moon. The WSB arms extend perpendicularly to
the Earth-Moon direction and can be related to the Earth gravity-gradient. Let the spacecraft have a
direct direction of the initial velocity around the Moon, where the gravity gradient opposes the motion
of the third body subtracting energy from it, there are a greater number of stable points that can be used
for capture. This further implies that in the m1 - m2 direction where the gravity-gradient moves the
spacecraft away from the Moon, there are no stable points beyond the lunar sphere of influence.

Figure 1.6: Jacobi constant for stable points (e = 0).

Stable points exist also for Jacobi constant values smaller than that corresponding to the libration points
(CL1 and CL2), that is when the zero velocity curves are fully opened and escapes are possible through
L1 and L2.

This analytic approximation thus gives the description of the WSB built considering only the first
stable/unstable transition.

70
WSB (3D):

Figure 1.7: Planes ρ(γ) and ρ(ν) for three – dimensional definition of WSB.

To study the three-dimensional definition of the WSB the following case is considered. The initial orbit
plane of the third body m does not lie in the plane of motion of the two primaries, but on a plane ρ(γ)
rotated by an angle γ around the x-axis of the synodic reference frame (illustrated in Fig. 6). Here due
to the out of plane component of motion, for t > t0 the mass m on leaving the plane γ will not intersect
the radial line, thus the motion will always be classified as unstable. To avoid this the radial line l(v) is
replaced by a semi-plane ρ(v), normal to the plane ρ(γ) and including m1 and m2.

For this study, as similar criterion needs to be specified as done in the previous segment. The motion of
the spacecraft m will be said to be stable if after leaving the periselenium of the oscillating orbit, it
crosses the ρ(v) again with zero or negative Keplerian energy with respect to m2 without going around
m1.

The motion will be unstable if the following conditions are met, namely:

1. After a complete revolution around m2, it crosses the plane ρ(v) with a positive Keplerian energy
with respect to m2.
2. After leaving the ρ(v) plane, the spacecraft moves away from m2 towards m1, making a cycle
about m1 without crossing the plane ρ(v) or colliding with m1.

The figure 8 illustrates the configuration of the WSB for different values of the angle γ in the framework
of CRTBP with initial eccentricity of the oscillating orbit to be zero. For increasing values of the out of
the plane angle γ, the WSB structure reduces, thereby decreasing the extension of capture zone. If the
spacecraft has an eccentricity of a value 0.92 which is typical for a trans-lunar low energy transfer, the
WSB structure changes as per Figure 9.

71
 Figure 1.8: WSB structure for increasing values of γ with e = 0 in the framework of the CRTBP.

Figure1.9: WSB structure for increasing values of γ and e = 0.92 in the framework of CRTBP.

72
Thus it can be concluded that with high eccentricity the stable zone is reduced and for increasing the
value of the angle γ the number of stable points decrease drastically. For initial polar conditions that is
at γ = 90° the WSB is reduced to a very small region close to Moon.

Figure 1.10: WSB structure for increasing values of γ and e = 0 in the BRFBP.

For BRFBP, particularly the presence of Sun slightly extends the region of WSB, especially with high
out of plane angle γ. It is vital to state here that from the view point of the WSB definition or the view
point of identification of the stable/unstable transition conditions, the influence of the eccentricity of
the Moon around the Earth and of the Sun gravitational effect is approximated to be negligible. The
perturbations though have a major role in the dynamics of the low-energy lunar transfer, they are almost
negligible for WSB study. Same cases are considered for BRFBP and simulations are illustrated in
figures 10 and 11.

73
Figure 1.11: WSB structure for increasing values of γ and e = 0.92 in the BRFBP.

Figure1.12: Rotation of the plane ρ(γ) around the z axis.

74
Assuming δ being in the range 0°-90° and γ being 90°, that is at polar initial conditions for the
spacecraft, the ρ(γ) plane will pass from the (z-x) reference plane to being parallel to the (z-y) reference
plane. With high eccentricity of 0.92 it is observed that the extension of capture region increases with
δ varying from 0° to 60° and then decreases on further increasing δ to 90°.

Figure 1.13: WSB structure for increasing values of the angle δ with e = 0.92 and γ = 90°.

Trajectory evolution after Capture:

The identification of stable/unstable points is vital for successful trajectory design but the minimum
periselene altitude is the deciding factor as it is required to design safe capture orbits without colliding
with the Moon. Thus categorization of certain (r, v) values becomes of great importance for a safe
flyover the lunar surface. As from previous segments it can be marked that for increasing value of the
out of the plane angle γ, the number of stable points decreases. For each value of γ, points that avoid
collision with the lunar surface exists along with conditions that lead to collision with the lunar surface.
For polar initial conditions, it is more difficult to obtain stable orbits that in equatorial case, but the
possible orbits are more probably safe.

WSB and Lunar Low-energy Transfers:

The study performed so far is of vital use for designing lunar transfer with stable captures. The Lunar
low-energy transfer use the Earth-Sun WSB in order to obtain the first ∆V to raise the perigee to the
lunar height, and the Earth-Moon WSB to reach capture with the minimum amount of energy. These

75
transfers require fine synchronization between the position of the Moon and that of the Sun at the
beginning of the transfer. The value of Jacobi constant C is not constant due to the presence of the Sun
since it is a Four – body problem. As the perturbation of the Sun acts, the Jacobi constant in the Earth
– Moon CRTBP changes in magnitude. It can be said that the transfers with equatorial capture have a
greater value of C, hence require lower amount of energy as compared to polar orbits. Requiring high
value of C, equatorial captures occur in the vicinity of L2 libration point, with a small opening of the
zero velocity curve (Figure 14).

Figure 1.14: Polar and equatorial captures for WSB lunar transfers.

Once the WSB geometry is known, it is possible to determine if the capture orbit is stable or unstable.
To execute circulization maneuver with low- thrust propulsion systems, a stable capture in WSB is
required. As illustrated in the Figure 15, the selected point is out of the stable zone and thus the
circulization maneuver needs to be executed at the periselinuim passage. This devoids the spacecraft
the advantage of the stable motion. This implies that if the motion of the spacecraft is integrated forward
in time from these conditions out if the stable region, the trajectory will be unstable and thus escape to
L1 point. If the line of apsides of the capture orbit is rotated by 30°, the initially unstable point enters
the stable region and thus the motion of spacecraft will be stable and it will complete at least one
revolution of the Moon

76
 Figure 1.15: WSB structure for equatorial orbits and eccentricity capture e = 0.5.

Figure 1.16: WSB lunar transfer with stable capture orbit.

Thus it is very important and crucial to distinguish the dynamics of the motion of the spacecraft before
and after capture. Figure 16 shows the improved stable trajectory when the line of apsides is rotated by
30°, although the eccentricity of the capture orbit changes from 0.5 to 0.48, changes in geometry of the
WSB are negligible and the spacecraft is captured with stable conditions.

77
 1.4 METHODOLOGY
Trajectory Design: Dynamical Method
 To compute Sun-Earth manifolds emanating from SEL1 or SEL2 periodic orbits, that come in
the Earth’s vicinity and intersect with the final phasing orbit.
o A cut section with origin at Earth can be used to compute Poincare section. Planar case
of CRTBP (has DOF=3) computes to a 2-dimensional hyperplane (subspace of one
dimension less than the original phase space and disconnects it into two discreet parts)
in the energy surface. Since stable/unstable manifolds are in form of tube-like
structures, it casts a distorted ellipse as the Poincare section on the hyperplane.
o Any point within the ellipse of the Poincare section with same Jacobi energy constant
as the manifold’s, must come from exterior region outside the Earth’s orbit i.e. outside
Earth’s sphere of influence bounded by the zero velocity curve in the Sun-Earth system.
This cannot be used to compute the manifold that comes from the Earth’s vicinity.
However, any point outside the cast ellipse with the same Jacobi energy constant as
manifold’s would come from inside the Earth’s SOI.
o Also, points near the manifolds tend to stay near the manifolds. Thus an appropriate
point selection outside the Poincare section, the ellipse, would ensure that it lingers
around the manifold for a long time.
o A point on the Poincare section of an unstable manifold when propagated backwards,
will wind around corresponding periodic orbit at L2. Thus, a point near the Poincare
section, not entirely on it, will wind around L2 periodic orbit but not indefinitely and
will, after some number of orbits leave the periodic along the stable manifold.
o Same transfer, in forward propagation, will require to traverse the stable manifold from
near the earth and then towards the moon in an unstable manifold. All of these
manifolds are described in the SE3BP.
o The unstable manifold will lead the spacecraft away from the SEL points and towards
EML2 points. Once inserted in the manifolds, by theory, must be transported into the
lunar region between EML1 and EML2.
o Thus an overlay diagram of Poincare section of unstable manifolds of Sun-Earth and
Earth-Moon systems, gives an intersection which can be used to transport the
spacecraft into final leg of Lunar approach. Essentially providing the patch point of the
two legs.
o Considering the reference frame changes, additional 4-body perturbations, energy
difference of manifolds will give an essential patch point burn required to send the
spacecraft towards the moon. These burns in common offer patching in around 30 – 70
m/s.
o Complete synchronisation of energies can be mapped in rare occurrences and a totally
free transfer is possible.
o An end-to-end trajectory can be designed by an approach analogous to conventional
patched conic trajectory design. A transfer ellipse is designated after the final trans-
libration maneuver moves spacecraft along the stable manifolds of Lyapunov orbit of
the Earth-Sun 3-body system. The outbound trajectory is energetic enough to follow
the stable manifold of the Lyapunov orbit but it doesn’t have enough energy to insert
itself into the Lyapunov orbit. Thus it starts falling back to the earth. This is when it
should encounter stable manifolds of the Earth-Moon system. Jacobi Energy difference

78
 in this regime is used to calculate patch maneuver or what, in mission is named as WSB
maneuver as it generally occurs at the Weak Stability Boundary region of the Earth.
o An infinite set of trajectories to serve different purposes with design contingent
trajectory options can be used choose efficient and desired trajectory options with
features like eliminating plane change maneuver requirements. Different dynamical
regimes, chaotic in nature governed by Lagrange points are alternative of Porkchop
plots used for patched conic problems, for these types of trajectories.
o The role of dynamical systems theory, specifically the invariant manifold theory of
periodic orbits about the Lagrange points, is crucial in the solution of this problem.
Thus, dynamical systems analysis would be an elementary part of the detailed mission
design. The focus will be on the trajectory arcs on the computed stable and unstable
manifolds as initial guesses for the desired end-to-end trajectory.

Once the spacecraft enters the green tube (Figure

17) of the exterior stable manifold of the Earth-
Moon L2 Lyapunov orbit, it must go into the lunar
region between the Earth-Moon L1 and L2. In order
for this orbit to then leave the lunar region, it must
leave via the interior stable manifolds of the L1
Lyapunov orbit of the exact same Jacobi constant.
 A surface (or "space") of section, also called
a Poincaré section is a way of presenting a trajectory
of n-dimensional phase space in an (n-1) -
dimensional configuration space.

Figure 1.17: The stable (green) and unstable (red)

Manifolds of Lyapunov Orbit.

Figure 1.18:(a) The Sun-Earth Unstable Manifold (Red) (b) Its Poincare Section.

79
 Figure 1.39: Using Poincare Section to Find the Transfer Trajectory.

1.5 PARAMETER SEARCH AND OPTIMISATION

Forward Targeting
Baseline Algorithm

A point x is taken with respect to Earth at a radial distance of RE and the motion is numerically
integrated. The orbital parameters at x are taken in spherical coordinate system as RE, longitude (αE),
latitude (δE), velocity magnitude (VE), flight path angle (γE) and flight path azimuth (σE) described in
Earth centred inertial spherical reference frame. 2X2 targeting algorithm comprises of varying VE and
γE to target RM and iM (desired radial distance and inclination with respect to Moon respectively).
Symbolically, it is denoted as

VE, γE → RM, iM

Initial VE is taken such as to account the trajectory on the Keplerian ellipse of Earth apoapsis of more
than 1.3 million kilometres as seen by the preliminary evaluation of change in angular momentum
required in the graph (Fig.15). Initial value of γE is chosen to be near zero. The initial epoch is chosen
such that it takes into account the desired Sun-Earth-Moon angle. A second order Newton’s differential
corrector is used as the targeting algorithm to converge RM and iM, resulting in lunar eccentricity less
than one. Thus the spacecraft attains the elliptic state in the lunar WSB when it reaches 1.4 to 1.5 million
kilometres in the Earth apoapsis after about 45 days of Trans-Lunar injection, approaching the Moon
with negative C3 energy.

Visualization of the WSB

Only the Earth and Moon are modelled and it is assumed that they move about their common centre of
mass in a uniform circular motion. It is also assumed that the motion of the spacecraft lies in the same
plane as the Earth and Moon and the spacecraft is assumed to be have zero mass. For the modelling of
the planar restricted three-body problem, a rotating coordinate system is taken as such that the Moon

80
lies at its centre and the Earth is located on the x-axis to the right of the Moon. The average distance
between the Earth and the Moon is normalised to 1 and the mass-ratio µ is taken to be 0.0123.

Figure 1.20: Lunar WSB Surface (eM, θ, E), θ = π.

The Earth is taken to be at angle θ = 0° in the polar coordinate system with respect to Moon. The exterior
of the Lagrange point in the anti-Earth direction θ = π. The Jacobi energy, denoted by E has a special
value of -3.184 when evaluated at θ = π. The capture at Moon can occur when E>E1.

eM, θ, E are used to parameterize WSB and gives the distance rM of the spacecraft between capture and
escape, and where the spacecraft is captured so that eM <1, resulting in unstable capture. The distance
from the Moon where the WSB is located is given by

rM = rM (eM, θ, E)

(1.6)

This equation can also be written as

E = E (eM, θ, rM)

(1.7)

Assuming the distance to be at the periapsis, the inertial velocity vM in the rotating coordinate system
can be formulated as

−1/2
𝑣𝑀 = (√((1 + 𝑒𝑀 )µ) ) 𝑟𝑀 − 𝑟𝑀

(1.8)

81
The selection of θ = π for WSB as the capture occurs at the lunar far-side.

The Nominal Trajectory (300Km Circular Orbit)

The nominal trajectory is generated using a modified form of the forward targeting algorithm described
above. These transfers can be designed for every month. The baseline lunar parameters can be taken as
RM = 3000 kilometres and iM = 90 degrees.

The sequence designed for this analysis commenced by insertion into a 300 km orbit. Orbit coast The
spacecraft is propagated in the coast mode in the orbit and propagation stopping conditions are selected
as a control parameter. This is when the impulsive Trans – Lunar Injection (TLI) maneuver is executed.
Through the times of transfer, capture at the Moon and in the lunar orbit the TLI is propagated using
numerical integration.

Determining the Launch Date:

Once the month of launch has been selected the first step is to
determine the launch date as per the Sun Earth Moon
geometry suitable for the transfer. (Figure 21) For this
analysis II quadrant transfer having the apogee towards the
Sun is selected. The Sun Earth angle λ is set to 130° for the II
quadrant transfer based λ is targeted to 130° to converge using
Astrogator’s differential corrector to find the date in
November 2018. The date found was 26th November 2018.

Calculating the Trans – Lunar Injection (TLI) ΔV:

Now, to allow the spacecraft enter the transfer trajectory the Figure 1.24: Quadrant II WSB transfer in Sun-Earth
necessary TLI ΔV is required so that the spacecraft has the Rotating Coordinates.
sufficient energy to leave the Parking Orbit. Considering it to
be a Keplerian 2 – body problem, the semi major axis (a) is calculated for an apogee radius and perigee
radius. Using the equation

𝑣2 𝜇 𝜇
− =−
2 𝑟 2𝑎
(1.8)

Thus TLI ΔV can be stated as

ΔVTLI = vf - vi = 3.17084 km/second

Since there might be uncertainties regarding the orbit parameters to be on the safe side, it is convenient
to calculate the orbital energy along with velocities. C3 is defined as

C3= -µ/a

(1.9)

The C3 was calculated to be -0.59km2/second2.

Calculating the Launch and Coast Times:

It is vital to design a trajectory that encounters Moon with suitable arrival conditions. It is required to
assign Right Ascension (α) and Declination (δ) of the line – of – apsides in the Sun – Earth rotating

82
system as the defining parameters since the trajectory required is not hyperbolic. Thus δ can be stated
as the measure of the relationship of the trajectory with the ecliptic plane. Since the Moon’s orbit varies
within 5° of the ecliptic plane, δ is an important parameter to control. δ is chosen to be zero to avoid
complications. Right Ascension of the Ascending Node (RAAN) is determined by the launch time
whereas the coast time embarks the Argument of Perigee (ω). The range of α is between 25° to 40° as
the desired trajectory is of the II quadrant transfer. The control parameters are modified in Astrogator
to converge at the target value of δ and α. The ΔVTLI directly corresponds to VE as it is the difference
between V and the velocity in GTO. The coast duration is directly related to the control parameter γ, as
it is the angle between the velocity vector V and the local horizon (which is perpendicular to the position
vector R). Coast duration controls the position of R being the time during which the spacecraft
transverse from the GTO till the TLI.

Targeting the Lunar Capture:

When the targeting so far is achieved, the trajectory obtained attains the lunar periselene of radius
30,000 kilometres to 250,000 kilometres. The spacecraft will encounter three periselene ranging from
1.4 million kilometres to 200,000 kilometres, thus Astrogator is set to stop the propagation at the
periselene with the radius of less than 50,000 kilometres. A different approach can be applied to avoid
stopping on different periselene during targeting, by targeting rM to 10,000 kilometres thus refining
ΔVTLI. On Achieving this target, Astrogator can be further used to target r M to 3,000 kilometres and iM
to 90° for easy convergence of the differential corrector. The spacecraft can be further controlled in
being captured at the Moon by advancing the targeting from 2X2 to 3X3 by targeting the Lunar C3 as
well. Astrogator’s differential corrector is set to target the Lunar C3 of -0.01 to further stabilize the
capturing process.

Trajectory Construction using STK Astrogator:

STK’s Astrogator advanced module is used for trajectory construction and analysis by above mentioned
algorithm along with Astrogator’s MCS (Mission Control Sequence) containing various elements of
the mission. Earth Moon and Sun were included as celestial bodies. A differential corrector is used for
targeting. It calculates the numerical sensitivity matrix describing the changes in constrains as a function
of the changes in the controls. The perturbation size can be controlled to calculate the numerical partial
derivatives along with the speed of the convergence. Astrogator’s GUI automates the above processes.
Our targeting sequence uses Newton’s Differential Corrector (Central Difference) to target the equality
constraints with respect to control parameters. Root finding algorithm taken is Newton-Raphson
method. STK/Astrogator has several user-
configured orbit propagation algorithms which
consist of variable step Runge-Kutta numerical
integrators. There are several force models
provided such as the central body of integration; the
spherical harmonic model along with order and
degree; the atmospheric model; 3rd body
perturbations and SRP. For this analyzation Jet
Propulsion Laboratory’s DE-405 Ephemerides
were used to calculate the position of the planets.

Figure 1.22: Quadrant II WSB transfer in Sun-Earth

Rotating Coordinates.

83
BALLETTRANSFER2019.sc

Three planets were added as objects for analysis namely ‘Earth’, ‘Moon’ and ‘Sun’ to create reference
frames of analyses. The spacecraft SRMSAT-2 was added with orbit propagator set as ‘Astrogator’.
Various components and reference frames needed to be configure into the Astrogator Component
Browser and analysis workbench which would be necessary in the mission sequence design and
analysis.

Reference Frames and Vector Configuration (Analysis Workbench)

SUN

 ‘Earth-Sun’ displacement vector from origin point ‘Earth Centre’ to ‘Sun Centre’ is created for
reference to Sun-Earth-Moon Angle.
 A new set of ‘Trajectory’ axes ‘ES_BBR’- Earth-Sun Body-Body rotating axes is created.
‘Trajectory’ type axes are aligned and constrained to the position and velocity vectors along the
trajectory of the point relative to reference system. Trajectory Point is the Sun Centre and
Reference system is Earth Inertial. This is the Earth-Sun Co-rotating at Sun Centre.
 A new coordinate system is created of ‘Libration’ type, ‘SE_L1’ is created with ‘L1’ point of
primary body sun and secondary body Earth. This system is central bodies aligned axes of
selected libration point between one primary and multiple secondary central bodies with X-axis
along position of libration point relative to primary centre of mass and Z-axis along orbit normal
of libration point relative to primary centre of mass.

EARTH

 An ‘Assembled’ Coordinate system, ‘EC_SE-1’ with origin at Earth centre and correlated
reference axes as SE_L1 created above in Sun components.
 A similar assembled coordinate system, ‘EC ES_BBR’ with origin at Earth centre and reference
axes as ‘Sun ES_BBR’. This is the counterpart of Body-Body rotating coordinate system
centred at Earth.
 A ‘Fixed in axes’ reference vector, ‘MJ2000_X’ is also created which is fixed along X of
MJ2000 axes.

MOON

 A lunar reference frame, ‘EME’ is configured as ‘Aligned and Constrained’ type. Moon’s
angular velocity vector and Earth ‘MJ2000_X’ created in previous mentioned steps are chosen
as aligned and constrained vector respectively.
 An Assembled reference ‘EME2000’ frame from the above defined axes is created.

COMPONENTS CONFIGURATION:

Two propagator models are added for the analysis.

 ‘Earth-3body’ Propagator consisting of primary gravitational force model JGM2 gravity field,
Sun and Moon’s influence as third bodies.
 ‘Selenocentric’ Propagator comprising of two body gravitational force and Sun and Earth’s
third body effects.

Orbital Elements in Selenocentric ‘EME’ frame

84
  All the Keplerian elements are duplicated with Moon centred origin and EME frame configured
above
 The position magnitude component is duplicated to configure ‘Lunar R Mag’ as a new
component.

Target Vectors:

 ‘Lunar C3 Energy’ component is added in the form of osculating element with respect to Moon
centre.

Two other target vectors are used as a substitute to flight path angle in the targeting method
mentioned in the previous section of trajectory design.

 ‘SR Outgoing Asymptote Dec’ – The Declination of the outgoing asymptote or apsides which
is calculated in above configured sun rotating ‘EC_SE L1’ point.
 ‘SR Outgoing Asymptote RA’ – The Right Ascension of the outgoing asymptote or apsides
which is calculated in above configured sun rotating ‘EC_SE L1’ point.

A sample base line initial (scenario) start epoch of 1 Jul 2019 00:00:00.000 was taken for the analysis.
The target sequence defined in the MCS contains the following operations that are needed to be
performed in the given order to obtain the desired trajectory, they are namely:

1. Launch Epoch
2. Coast Propagation
3. TLI Maneuver
4. Propagation to Apoapsis
5. Propagation to Targeted Rmag
6. Propagation for Capture
7. Propagation to Periselene

Each segment with its characteristics has been explained below:

Spacecraft Configuration:
Drag Area: 0.72 m^2
1. Launch: The launch epoch in the starting of analysis of 1 SRP Area: 0.72 m^2
July 2019. It is selected as a control parameter for targeting, Dry Mass: 45 kg
Fuel Mass: 36 kg
primarily, the Sun-Earth-Moon (SEM) angle as well as Total Mass: 81 kg
Right Ascension and Declination of outgoing asymptotes.
Tank Pressure: 150000 Pa
The latitude and longitude of the launch facility and the Fuel Density: 1020 kg/m^3
Cr: 2.000000
burnout parameters were kept default and the ascent type Cd: 2.000000
as Elliptical quartic motion. The equality constraint SEM Rad Press Area: 0.72 m^2
Rad Press Coeff: 1.000000
angle created as a new component in Astrogator’s target
vectors is selected from results of this segment.

2. PO Coast: This segment is propagation in the 300 km orbit until the TLI maneuver is executed.
The ‘Earth-3body’ Propagator configured in the Astrogator’s propagator components is used.
Duration of 3000 seconds is set as the stopping condition and the trip time is selected as the
control parameter for targeting various parameters such as SEM angle, C3 and Periselene.

3. TLI Maneuver: Impulsive maneuver segment is inserted next with thrust vector selected for
attitude control. Velocity-Normal-Co-normal (VNC) frame of reference centred on the
spacecraft is used. The control parameter, the velocity of the X-component of executed
maneuver is set to an initial value of 3.17080 km/s as a reasonable first guess for departure orbit

85
 anticipated. The equality constraints for this segment, Earth_C3 energy and asymptotes are
selected from the results.

4. Propagation to Apoapsis: In the next segment, the propagator is conditioned to stop at the
apoapsis with Earth as the central body. The propagator model selected is Earth-3Body model.
We idealise the WSB patch point maneuver to occur at apoapsis. However, an optimal WSB
maneuver can be anywhere on the return leg. Thus, the end of this propagate segment marks
the WSB maneuver.

5. WSB Maneuver: At the apoapsis, lunar targeting WSB maneuver segment is inserted. Although
this is not exactly where the WSB burn is executed to target the moon, it is a reasonable way to
quantify changes in energy requirements if the trajectory deviates from nominal. This Impulsive
maneuver segment is inserted with the thrust vector selected for attitude control in the Velocity-
Normal-Co-normal (VNC) frame of reference centred on the spacecraft is used. The maneuver
ΔV components are set to zero and all three components are set as target parameters.

CONTROL PARAMETERS EQUALITY CONSTRAINTS

Launch Epoch (Nominal Launch) SEM_Angle (Nominal Launch)
StoppingConditions.Duration.TripValue C3_Energy (TLI)
(Coast)
ImpulsiveMnvr.Cartesian.X(TLI) SR_Outgoing_Asymptote (TLI)
ImpulsiveMnvr.Cartesian.X(WSB) SR_Outgoing_Asymptote (TLI)
ImpulsiveMnvr.Cartesian.Y(WSB) BDotR (ToMoon)
ImpulsiveMnvr.Cartesian.Z(WSB) BDotT (ToMoon)
BTheta (ToMoon)
Lunar C3_Energy (ToMoon)
LunarInclination (ToMoon)
LunarEccentricity (ToMoon)
R_Mag (ToMoon)
Table 1.2: A list of control parameters and equality constrains

NOTE: In the nominal trajectory analysis, the WSB maneuver magnitudes will be kept to zero to show
an idealised Low Energy Transfer. In the later analyses, this will be varied with respect to the fixed
launch date and the magnitude of X, Y and Z component will cover for the launch epoch deviation from
the nominal launch epoch.
6. Propagation to Targeted R_mag: This segment has propagator set to Earth-3body propagator.
and it specifies the stopping condition to be the targeted - Rmag = 300000km with Moon origin
as the reference point.

7. Propagation to capture: This propagator has ‘UserSelect’ stopping condition that is, it stops at
a value specified as per user specified calculation object. Here, the user specified calculation
object is defined as the lunar eccentricity with value ‘1’ and the central body is set to Moon.

NOTE: Several periapsis with respect to the moon occur while traversing on the
return/approach leg. The 6 and 7 segment are used remove any ambiguity of targeting higher
periapsis that occur at distances like 1.2 x 106 km (first perigee).

86
 8. Propagation to Periselene: This is the last segment of the target sequence in the MCS. Moon
is set as the central body. The periapsis of the lunar capture orbit is set as the stopping condition;
thus it propagates the spacecraft to the periapsis of the capture orbit. The propagator model is
Earth Moon model. The results to be evaluated in this segment are lunar C3 energy, inclination,
eccentricity and Rmag of the lunar capture orbit.

Targeting Sequence Overview

Once the Mission sequence operations are set, it becomes necessary to prepare the differential correctors
for targeting the parameters. A total of six differential correctors with varying parametric tolerance and
perturbation levels are set up.

1. Target SEM: This differential corrector is conditioned to

target the SEM angle to 130 degrees to provide with suitable
starting launch date as per the Sun-Earth-Moon geometry to
converge on the capture. The control parameters required for
the target to converge are launch epoch, coast stopping
condition and velocity of the x-component of the TLI
maneuvers (Vx).

2. Target Earth C3 Energy: Here the target is C3 energy of the

spacecraft with Earth as the governing body. The starting
value C3 value is taken as -0.590 km2/sec2 and the control Figure 1.23: Targeting Sequence.
parameter for targeting is TLI Maneuver ΔVx. This is
converged upon after iterating the SEM angle.

3. Target Asymptotes: This differential corrector, in this segment, targets RA and Dec of
outgoing asymptotes in solar rotating frame. After converging on the C3 value this target
sequence is set to Iterate. The starting values of Dec as 0 degrees to avoid any complications
and keeping the nominal trajectory planar while ascension asymptote is set to 164 degrees.
The targeting of this differential corrector becomes crucial as its convergence ensures
encountering the Moon with suitable arrival conditions when iterated with C3. Launch epoch,
coast duration and TLI Maneuver ΔVx serve as the control parameters for this target.

4. Target Lunar Capture: To initialize and ensure capture, first the C3 energy of the spacecraft
is targeted to -0.01 km3/sec2. Coast duration and Vx are set as the controlling parameters.

5. Target Periselene: When the targeting so far is achieved, the trajectory obtained attains the
lunar periselene of radius 30,000 kilometres to 250,000 kilometres and the initialization of the
capture process is done. Therefore, now it is necessary to define the parameters capture orbital
parameters. The desired targets here are the inclination and Rmag of the lunar capture orbit.
Inclination is set to 91 degrees while Rmag is 5000km. The control parameters are coast
duration and Vx.

6. Target Capture: The last differential corrector is set to stabilize the capture and achieve the
desired capture orbit. The declination asymptote is set to 0 degrees while the C 3 energy is

87
 targeted to -0.09 km3/sec2 so as to assist the differential corrector to converge easily. The
control parameters set are coast duration and Vx.

ANALYSIS:
We initiate by targeting the Sun-Earth- Moon Angle to converge at 130deg. After that converged Earth
C3 Energy of the orbit is targeted at -0.59 km2 /s2 with TLI Maneuver as control parameter. This gives
the Keplerian state after maneuver, an appropriate eccentricity that lingers in the WSB long enough to
change perigee on the return leg. After this, Coast Stopping Condition is used as a control parameter to

Figure 1.24: Unconverged trajectory approaching Moon with Figure 1.25: Trajectory approaching Moon with converged SEM
targeted inclination and Lunar c3 and converged SEM and and Asymptote RA and Dec Values at lower ΔV =
Asymptote RA and Dec Values ΔV = 3.17023km/sec. 3.17020km/sec.

converge on Outgoing Asymptote Right Ascension

and Declination. The RA is set to converge on 164° while to construct a planar transfer, Declination is
set to converge on 0°. With above trials done, we get trajectory returning from the WSB approaching
the moon in a correct way. Some trials are shown

A numerical approach, divergent in nature, was chosen to target lunar weak capture. SEM was relieved
as a target parameter and now with Launch Epoch, stopping conditions of Coast segment and TLI
maneuver X, a negative lunar C3 was targeted with 0° asymptote declination. Every situation was seen
in two cases lower ΔV ranges from 3.16090 to 3.17020 and higher ΔV ranges from 3.17090 to 3.17120.
Some of these results in the initial stages were inferred from to converge upon nominal trajectory and
understanding the working of differential corrector. Numerous experimentation was done based on
changing the initial control and constraint values, varying tolerances of control parameters and resetting
to get a baseline geometry. Some of the results from which a trend was observed are given below. Also,
as given by the theory various motions in dynamical regime of the Earth-Moon system were seen:

 On targeting constraints, some trajectories converged to conditions after more than one orbit
around Earth. Some of them also encountered Moon on the second pass.
 Some trajectories within same ΔV ranges escaped the Earth-Moon systems when different
conditions were imposed.
 Some trajectories encountering moon with near zero C3 energy degraded to a lower orbit
around Earth.

88
Figure 1.26: Trajectory approaching Moon with convergence Figure 1.27: Trajectory approaching Moon with convergence of
of Earth Orbit C3, RA and Dec with higher. Earth Orbit C3, RA and Dec with higher.

Inferences:

A transfer trajectory could be made in both high ΔV ranges (3.17090 to 3.17120) as well as in lower
Delta-V ranges (3.1690 to 3.17025). It was evident from the experimentation that lower ΔV trajectories
would encounter the Moon behind the Earth with respect to the final apogee. While, higher delta ΔV
values meet the moon at the same side of the Earth w.r.t to the apogee. Any of these transfers can be
targeted. Having inferred that, a baseline lower delta-v of 3.17023 was taken and targeting was initiated
in the normal sequence starting from SEM angle.

Figure 1.28: Trajectory approaching Moon with Figure 1.29: Trajectory approaching Moon with
ΔV=3.17010 with near zero lunar C attaining negative C3 ΔV=3.17094 with positive lunar C3 energy uncoverged for
about Moon as shown by the pink segment. Moon encounter.

89
Critical parameter deciding the weak capture is the Lunar C3 Energy. A negative C3 energy about moon
would mean a weak capture is attained and for ‘some time’ major part of the motion in governed under
the Sphere of Influence of the Moon. Targeting negative C3 was nearly impossible in a one shot.
Reiterative setting and converging equality constraints with control parameters in hit and trial method
was taken up. Near zero lunar C3 energies were targeted initially. The subsequent motions were
investigated to adjust tolerances and control parameter resetting along with convergence perturbations.

C3 was finally converged upon from 0.01 km2/s 2 to -0.002 km2/s 2 to achieve negative C3. A desired
value of -0.18 km2/s 2 was aimed at for targeting. However, optimal weak capture was finally converged
upon by targeting -0.090 km2/s 2. The orbit although not meeting inclination, got weakly captured with
eccentricity of 0.707625. Figure 30 shows the weak capture segment of the trajectory construction

Figure 1.30: Weak Capture Segment of the Nominal trajectory.

90
Below is the graph of C3 variation over the trajectory.

Figure 1.31: C3 Variation over the trajectory.

It is clearly seen in the left side zoomed view that Lunar C3 energy goes negative which is clear
indication of a weak capture. An optimal stabilisation maneuver during this capture would be required
to enter into a more stable high eccentricity orbit. After which, the multi-burn periapsis lowering
sequence is initiated to inject spacecraft into desired lunar mission orbit.

An interesting feature of the nominal trajectory found without performing a stabilisation maneuver
when propagated was that the spacecraft, after sometime having left the moon, encounters the moon
through resonant set of Earth orbits as shown below and gets captured in weaker state again by the
moon. The resonant earth orbits are shown by magenta while captured state is shown by yellow.

Figure 1.32: Trajectory profile without lunar orbit Figure 1.33: Trajectory profile without lunar orbit
stabilisation maneuver. Second capture shown in yellow. stabilisation maneuver. Second capture shown in yellow.
(Earth Inertial Frame)

91
 Figure 1.34: Trajectory profile without lunar Figure 1.35: Zoomed view of the second capture, in yellow
orbit stabilisation maneuver leading to second (Selenocentric Frame).
capture. (Selenocentric Frame).

The tables below show the converged values constraints as well as control parameters for weak capture
at the moon through a WSB transfer in the ideal situation of a planar transfer without WSB maneuver
requirement.

CONTROL PARAMETERS
Name Final Value
Launch Epoch 21 Jun 2019 21:04.13.205 UTCG
StoppingConditions.Duration.TripValue 7175.16 sec
ImpulsiveMnvr.Cartesian.X(TLI) 3.1692 km/sec
ImpulsiveMnvr.Cartesian.X(WSB) 0 km/sec
ImpulsiveMnvr.Cartesian.Y(WSB) 0 km/sec
ImpulsiveMnvr.Cartesian.Z(WSB) 0 km/sec
Table 1.3: Final Values of the Control Parameters

EQUALITY CONSTRAINTS
Name Desired value Final Value
SEM_Angle 129° 128.472°
2 2
C3_Energy -0.6 km /sec -0.627384 km2/sec2
SR_Outgoing_Asymptote 0° -4.95939e-011°
SR_Outgoing_Asymptote 164° 188.499°
BDotR -40000 km 4340.88 km
BDotT 0 km 37249.8 km
BTheta 88° 6.64694°
2 2
C3_Energy -0.09 km /sec -0.0923756 km2/sec2
Lunar Inclination 90° 26.5799°
LunarEccentricity 0.8 0.707625
R_Mag 3000 km 15517.7 km

Table 1.4: Desired and Final Values of the Equality Constraints

NOTE: The simulated transfer trajectory mission summary generated from STK Analysis Bench is
given in Appendix A.
92
 Figure 1.36: SRMSAT-2 Preliminary transfer trajectory design (Earth Inertial).

Figure 1.37: Transfer trajectory (Earth – Moon Co- Rotating frame) Figure 1.38: Transfer trajectory with weak capture
(Selenocentric Frame)

93
Launch Window During a Single Day
The Launch Window for the desired trajectory can be stated as the time during a given day when the
spacecraft can be launched and returned to its nominal orbit around the Moon. Taking into consideration
that the coast duration and the ΔVTLI will hardly be altered when the launch time is changed by an order
of a few minutes, it was concluded that these parameters will not be used to retarget the trajectory. Thus
a new maneuver named Mid-Course Correction Maneuver (MCCM), needs to be executed at the
apogee. This maneuver is established in the Astrogator’s VNC (velocity-normal-co-normal) local
coordinate system. The x-axis is aligned to the velocity vector V, the y-axis is aligned to the orbital
angular moment, N = RxV and the z-axis (co-normal) is given by C = VxN. When the launch time was
varied by 10 minutes, each trajectory was targeted with Astrogator. At first the co-normal component
of MCCM was controlled such that nominal distance from the Earth was around 555,610 kilometres
when the spacecraft was within 300,000 kilometres within the vicinity of the Moon. Next r M and iM
were targeted to 3000 kilometres and 90° respectively to control the normal and the co-normal
component of the MCCM together. The final constraint to be targeted was the Lunar C3 to -0.12
kilometres2/second2 to control the velocity component of the MCCM. The result data concluded that a
ΔV of less than 50metres/second could be achieved if the launch window of 20 minutes is considered.

Launch Period Over Several Days

The Launch Period for the designed trajectory can be stated as the days of the month for which the
spacecraft can be launched and returned to its nominal trajectory. Here as stated above the ΔVTLI will
not vary but the coast duration may change depending upon the launch vehicle, thus is defined as a
control parameter. Returning to the nominal orbit will be a challenge in this case since Moon moves
approximately 13° a day. Energy Correction Maneuver or the ECM will be applied along the inertial
velocity vector using the VNC axes. ECM may be direct or retrograde and the best suited region to
apply this maneuver is the perigee where the TLI takes place. Along with ECM, MCCM will also be
applied here with six control parameters namely:

1. Launch epoch
2. Coast duration of the GTO
3. ECM velocity component
4. MCCM velocity component
5. MCCM normal component
6. MCCM co-normal component

The result obtained is a unique trajectory due to the presence of six control parameters. With
experimentation it was concluded that it will be easier for the differential corrector to target at 300,000
kilometres from the Moon, instead of at Periselene.

Next the nominal state vector in the Moon-centred inertial coordinate system were targeted, namely:

1. Epoch
2. Right Ascension
3. Declination
4. Lunar C3
5. Horizontal flight path angle
6. Inertial velocity flight path azimuth.

94
6X6 targeting has been proved to be rapid in converging for ‘nominal plus 1 day’ to ‘nominal minus 3
day’ and the final result show that a 5-day launch period including the nominal could be achieved for a
possible ΔV of 50 metres/second.

1.6 FROZEN ORBIT

An unstable spacecraft orbit around moon will have more fuel requirements for orbit maintenance
without which it will crash into moon’s surface. Hence lunar orbits need to be designed in such a way
that they not only meet ideal observation conditions but also have less fuel requirements for orbit
maintenance. It is well known that the orbit of a lunar spacecraft is mainly influenced by the
gravitational field of Moon, Earth and Sun. The Primary perturbation on a low-altitude (100 km to 300
km) lunar parking orbit is mainly due to Moon’s non-spherical gravity field which also has Mascons.
The analysis of the factors affecting the stability and lifetime of lunar orbits thus plays a vital role.
Frozen Orbits exist due to the three body permutation influence at higher altitudes and non - spherical
gravity effects . When Earth elliptical orbit with critical inclination is pertubed by Moon, there exists a
region of stable libration motion in the e - ω phase space. Argument of perapsis (ω) needs to be 90 or
270 degrees as a prerequisite for the location of apoapsis and coverage by payload at the south pole or
north pole respectively. This region, though being surrounded by large region of chaos, provides with
possiblities of focused coverage at the poles due to the stable libration motion. Non spherical nature of
the Moon’s gravity controls the lifetime of a lunar orbit. It can be seen that gravity effect of Sun causes
very less variation on the periapsis variation and hence on the orbital lifetime. The perigee variation
shows a very high lifetime if the gravity effect of Sun and Earth are not considered. Hence for the
lifetime estimation of lunar orbits, the gravity effect of Earth is essential. Detailed simulations have
been carried out to analyse the effect of altitude on lifetime.
For proper and sufficient scope for the payload, certain orbital parameters are constrained. To enhance
payload features and to incorparate the need of producing continuous coverage, a low lunar orbit stands
as the basic prerequisite. The argument of periapsis is set to 270 degrees to allow South Pole
observations. Keeping the above constrains in mind, a detailed examination of variations in each
parameter with respect to one another, has been done. The effect of inclination and right ascension of
the ascending node on the lifetime variation due to the altitude is also studied. For each analysis, the
point mass effect of three bodies has been neglected.

Propagation Model Design:

STK’s Astrogator software has been used for the simulations. The gravity model chosen is LP150Q
(50X50) and spherical solar radiation pressure has been included. Moon Inertial Coordinate system has
been used and the epoch date of 13th October 2019 has been chosen for the analysis. The stability of the
orbit was judged upon the least gradual change in the altitude of both periapsis and apoapsis for an
elapsed period of 200 days. The variation of each orbital parameter is reported below:

Inclination:
When the inclination was varied from 0° to 150° (Figure 39 and 40), least amount of variation in
altitude of the perigee and the apogee was observed in orbits with inclination degrees in the range 80°
to 100°. It was percieved that the most stable orbits existed in the range of 86° to 92°. A detailed study
for orbits with inclination between 86° and 92° demonstrated (Figure 41 and 42) that the least variation
was least in the altitude was in orbits 90° to 92° of inclination. The study was concluded with the
deduction that 91° provided with the most stable frozen orbit with sufficient coverage swaths over the
poles for the payload.

95
Figure1.39: Effect of various degrees of inclination on the apogee altitude of the circular 250 km frozen orbit.

Figure 1.40: Effect of various degrees of inclination on the perigee altitude of circular 250 km frozen orbit.

Figure 1.41: Effect of 86° to 92° of inclination on the perigee altitude of circular 250 km frozen orbit.

96
 Figure 1.42: Effect of 86° to 92° of inclination on the perigee altitude of circular 250 km frozen orbit.

Altitude:
Setting the inclination to 91°, an analysis was conducted for different altitudes for circular lunar orbit.
The altitude was varied from 150 km to 950 km accounting the need of low lunar orbit as per the
payload. The orbits below 150 km were not considered due to extremely fluctuating thermal loads. The
variation in inclination was noted to be of 0.4° (Figure 43). Maximum descripency seen in eccentricity
was 0.008 (Figure 44) when altitude was varied from 250 km to 950 km, thus it was concluded that
different altitudes have almost negligible effect on the eccentricity and inclination of the orbit. It was
inferred that 250 km was the best suited altitude for the mission, payload features being the most
highlighted criteria of selection.

Figure 1.43: Effect on inclination when altitude of the circular frozen orbit is varied.

Figure 1.44: Variation in the altitude of the apogee when RAAN is changed from 0° to 150°.

97
RAAN:
Varying RAAN from 0° to 150°, the graph (Figure 45 and 46) showed that the least deviation in altitude
was at RAAN of 60°. Thus a further detailed analysis was conducted varying RAAN from 45° to 75°
(Figure 47 and 48). For different values of RAAN (45° to 75°) inclination of the frozen orbit was also
looked upon (Figure 49). The data conveyed hardly any deviation for each RAAN, thus the RAAN was
decided upon as 60°.

Figure 1.45: Effect on eccentricity when altitude of the circular frozen orbit is varied.

Figure 1.46: Variation in the altitude of the perigee when RAAN is changed from 0° to 150°.

Figure 1.47: Variation in the altitude of the perigee when RAAN is changed from 45° to 75°.

98
 Figure 1.48: Variation in the altitude of the apogee when RAAN is changed from 45° to 75°.

Figure 1.49: Variation in the inclination of the frozen orbit when RAAN is changed from 45° to 75°.

With the results of the above studies the final orbital parameters of the frozen orbit are as follows:

Final Orbital Parameters

Periapsis Altitude 250km
Eccentricity 0
Inclination 91°
RAAN 60°
Argument of Periapsis 90°
Table 1.5: Final Frozen Orbit Parameters

Maneuver ΔV (km/sec)

Phasing Maneuver (GTO to EPO1) 0.3899

Phasing Maneuver (EPO1 to EPO2) Vx Vy
0.1134 0.1192
TLI Maneuver 0.2436
WSB Patching Maneuver 0.05 - 0.08
Lunar Capture 0.1 – 0.2
Descent to Final Science Orbit 0.7 – 0.8
Additional Orbit transfers to Final Science --- (Statistical)
Mission Orbit
Total 1.7605
Table 1.6: ΔV Budget Table

NOTE: The ΔV for additional orbit transfers to final science mission orbit has been
considered within the margins. By possible Lunar gravity assist ΔV can be further reduced.
99
 2. GUIDANCE NAVIGATION AND CONTROL

2.1 INTRODUCTION
SRMSAT-2 will employ a GNC system encompassing various ground and space segment
elements. SRMSAT-2 will use a radiometric ground based tracking with support of ISTRAC and IDSN
for tracking requirements in respective phases of the mission. Spacecraft is configured, within the
TT&C subsystem, to house an S-band Deep Space Transponder facilitating 2-way Doppler and ranging
based tracking measurements for Orbit Determination (OD) which will explicitly require the support of
IDSN, during OD schedules of Trans-libration or WSB Cruise leg and Lunar Approach leg, before the
TCM and the LOI maneuvers. The SRMSAT-2 trajectory design utilizes multi-burn or phasing
strategies to divide the total ∆V requirement magnitude into multiple small maneuvers. The trans-
libration maneuver required to send the spacecraft to an apogee of about 1.4 - 1.5 x 106 km is divided
into two more orbits phased after the initial insertion into the GTO by launch vehicle and before the
final TLCM, of apogee about 90000 km and 130000 km. These were included in the design to
accommodate for thrust limitations of micro-satellite propulsion systems, and to slow the transfer down
to obtain operations benefit.

Phasing orbits form an important factor in increasing mission/transfer success, due to

comparatively lower reliability levels incurred from hardware/software/operations on-board a small
satellite. The phasing orbits are dedicated to, apart from low magnitude burns, to remove injection
bias/launch separation errors, in-flight instrument tests and calibration, consistent OD solution
generation and model calibrations. The final phasing orbit is meant to serve mainly as a coasting orbit
to target the transfer. Although, it is mentioned as a static element in the trajectory design, would be
variable depending on coast times in lower orbits, transfer opportunity, transfer windows, OD
constraints and readiness of the spacecraft and ground segment to execute the transfer. This orbit will
be the main element of critical design in both Trajectory Design and mission analysis. A series of year
round transfer opportunities with varied launch scenarios and inducing various variables anticipated
during the actual mission will be simulated to make an optimal choice of the final phasing orbit to
configure nominal and contingent transfer trajectories sufficing injection requirement at the moon. A
similar multi-burn deceleration strategy would be used to commission spacecraft into its final science
mission orbit from the initial weak capture orbit.

The flexibility offered in the WSB transfer trajectories used in the mission design with respect
to launch constraint, final inclination and a wide range of transfer opportunities is counter balanced by
the stringent requirement placed on navigation due to highly sensitive nature of the dynamics and
corresponding divergences within it. Very low tolerances are expected to be imposed on the trajectory
for injection at the moon. Thus, before the spacecraft is inserted into the cruise trajectory, exhaustive
OD schedules if required, would need to be included to improve on prediction errors. A consistent OD
solution with calibrations in phasing orbits for adverse translational thrusts, SRP coefficients, and other
un-modelled forces should be known with accuracy levels necessary for guidance and control within
the effectiveinjection corridor. These accuracy levels would be set by sensitivity analysis for the
transfers. The transfer timing, however, relieves the constraint on navigation information turn-around
and decision –making.

100
 A set of deterministic Trajectory Correction Maneuvers would be planned subsequently to
counter theoretical navigation propagation errors. Apart from deterministic maneuver, the design would
include a set of mid-course statistical maneuvers usually smaller in magnitude than deterministic,
utilised to counter in-flight OD and execution errors of previous maneuvers. The subsystem’s chief task
would be to devise a robust navigation strategy to account for requirements within operational
constraints.

Guidance, Navigation and Control subsystem will be a part of space as well as ground segments. The
tasks of GNC as a scope of future work after the mission geometry is considerably fixed and constrained
will be:

PRE-LAUNCH
 Selection from available ground stations (ISTRAC) to support all the phases of mission.
 To analyse dynamic sensitivities of the transfer, quantify capture corridor and lay down flight
path margins
 Lay down Orbit Determination requirements (radial, cross-track and out of the plane) to
maintain injection corridor in the trans-lunar trajectory
 To supplement astrodynamics feasibility with frequency and accuracy of orbit determination
observations to validate mission possibility
 Lay down Orbit Determination schedules for required orbit knowledge and specify coast
periods in phasing orbits before each maneuver
 Plan Deterministic TCMs in the transfer trajectory
 Evaluate dynamic injection targeting by creating various mission scenarios and lay down
contingency levels and counter measures
 Devise test burn strategy in second phasing orbit to model adverse translational thrusts due to
Angular momentum desaturation maneuvers
 Develop dynamic and measurement models
 Quantify, with statistical margins, TCM, and lunar orbit station – keeping ∆V budget
 Phase specific use of high gain parabolic and low gain QFH antennae for navigation
 Lay down nominal and contingency levels and respective flight rules of operations
 Establish Orbit Determination, Prediction and Control infrastructure within the ground segment
to interface with ISTRAC/IDSN/University Ground station to optimally fulfil all the mission
requirements

101
LAUNCH AND POST LAUNCH
 Evaluate separation states and execute preliminary orbit determination for acquisition at ground
stations for POD sequence
 Fit un-modelled effects and calibrate developed models
 Execute test burn strategy for examining adverse translational thrusts
 Provide unambiguous and consistent OD solutions after each maneuver
 Improve Orbit prediction errors by subsequent maneuvers and injection parameter data
 Develop and update small force data to improve orbit knowledge and maneuver planning
 Select injection opportunity and window from the catalog of nominal and contingency transfers
 Fit outgassing and SRP model inconsistencies
 Evaluate status of the mission and select from nominal/contingency levels of operation laid in
the flight rules for executing the transfer
 Execute multi-burn phased deceleration from capture to final mission orbit
 Station – keeping and orbit maintenance for payload operation for 6 months observation period
 Generate ephemeris for attitude and other data for statistical investigation on mission execution,
anomalies, and others
 Provide ephemeris data to tag payload data and for data re-processing for scientific
investigation.

102
103
 2.1 ORBIT DETERMINATION
Baseline Principles
The case of orbit determination falls into generic category of collaborative orbit determination. The two
essential elements of an orbit determination problem are orbits and observations. Orbits are solutions
of an EOM:
𝑑𝑦
= 𝑓(𝑦, 𝑡, 𝜇)
𝑑𝑡
(2.1)

which is an ordinary differential equation; y ∈Rp is the state vector, μ ∈Rp’ are the dynamical parameters,
such as the geo-potential coefficients, t ∈ R is the time. In the transfer case, the equation of motion is
the 4-body problem, the spacecraft orbit being perturbed by the gravitational attraction of the Sun, Earth,
and Moon. The orbit in the vicinity of a pla being mostly perturbed by the asymmetric part of the
geopotential, but also by non-gravitational perturbations. The initial conditions are the value of the state
vector at an epoch t0:

𝑦(𝑡0 ) = 𝑦0 ∈ R𝑝

(2.2)

In the two simple cases cited above we have p = 6, i.e., the vector of the initial condition is just formed
by the position and velocity of the small body in some inertial reference system. The orbits are specific
solutions, for a given value of y0 and μ, of the equation of motion (initial condition problem). All the
orbits together form the general solution

𝑦 = 𝑦(𝑡, 𝑦0 , 𝜇)

(2.3)

also known as integral flow when considered as a mapping from the initial conditions (and dynamical
parameters) to the current state at time t:

𝑦(𝑡) = Ф𝑡𝑡0 (𝑦0 , 𝜇)

(2.4)

For the second element, we introduce an observation function

𝑅(𝑦, 𝑡, 𝜈)

(2.5)

depending on the current state, directly upon time, and also upon a number of kinematical parameters
ν ∈Rp’’. The function R is assumed to be differentiable. The composition of the general solution with
the observation function is the prediction function

𝑟(𝑡) = 𝑅(𝑦(𝑡), 𝑡, 𝜈)

(2.6)

which is used to predict the outcome of a specific observation at some time ti, with i = 1,...,m. However,
the observation result ri is generically not equal to the prediction, the difference being the residual

104
 ξ𝑖 = 𝑟𝑖 − 𝑅(𝑦(𝑡𝑖 ), 𝑡𝑖 , 𝜈), 𝑖 = 1,. . . , 𝑚

(2.7)

The observation function can depend also upon the index i, the most common case being the use of a
two-dimensional observation function like (right ascension, declination) or (range, range-rate), in which
case R has two different analytical expressions, one for i even, the other for i odd. All the residuals can
be assembled forming a vector in Rm

ξ = (ξ𝑖 )𝑖=1,…,𝑚

(2.8)

which is in principle a function of all the p + p’ + p’’ variables (y0, μ, ν).

The above equations define a fully deterministic model: each residual is a single valued function of the
p+p’ +p’’ parameters. This function is obtained from the observation function, for which we assume an
explicit analytical expression, by using the general solution, which is not known as an analytical
expression but is uniquely defined by the differential equations; both functions are assumed to be
differentiable.

The random element is introduced by the assumption that every observation contains an error. Even
assuming we know with perfect accuracy all the true values (y0∗, μ∗, ν∗) of the parameters, that our
model is perfectly complete (both for the equation of motion and for the observations), and that our
explicit computations are perfectly accurate (they are computed in “exact arithmetic”, not with a
realistic computer), nevertheless the residuals

ξ∗𝑖 = 𝑟𝑖 − 𝑅(𝑦(𝑦0 ∗ , 𝑡𝑖 , 𝜇∗ , 𝑖) = 𝜀𝑖

(2.10)

would not be zero but random variables. The joint distribution of 𝜀 = (𝜀𝑖 )i=1..., m needs to be modeled,
that is we need some assumption, in the form of a probability density function.

The basic tool of the classical theory of orbit determination is the target function Q(ξ) depending on the
vector of residuals ξ. The target function cannot be chosen arbitrarily, but needs to satisfy suitable
conditions of regularity and convexity. For e.g. a target function with sum of squares of residuals:
𝑚
1 𝑇 1
𝑄(ξ) = ξ = ∑ ξ2𝑖
𝑚 𝑚
𝑖=1

(2.11)

Since each residual is a function of all the parameters,

ξ𝑖 = ξ𝑖 (𝑦0 , 𝜇, 𝜈)

(2.12)

the target function is also a function of (y0, μ, ν). The next step is to select the parameters to be fit to the
data: let x ∈ RN be a subvector of (y0, μ, ν) ∈Rp+p’ +p’’, that is x = (xi), i = 1, N, with each xi either a
component of the initial conditions, or a dynamical parameter, or a kinematical parameter.

Then we consider the target function

105
 𝑄(𝑥) = 𝑄(ξ(x))

(2.13)

as a function of x only, leaving the vector of the consider parameters k ∈Rp+p’ +p’’−N (all the parameters
not included in x) fixed at the assumed value.

The problem of orbit determination based on the transfer function is then based on minimum principle,
which selects a nominal solution at the point x∗∈ RN where the target function Q(x) has its minimum
value Q∗. The principle of least squares is the minimum principle with as target function the sum of
squares Q(ξ) = ξT ξ/m, or some other quadratic form.

Nominal Trajectory

In an inertial geocentric reference frame (ECI), the

motion of the spacecraft is governed by the following set
of differential equations:

𝑟̇ = 𝑣
𝐸 𝑟𝑆𝑠𝑐 𝑟𝑆𝐸 𝑟𝑀𝑠𝑐 𝑟𝑀𝐸
𝑣̇ = − 3
𝑟 − 𝑠 ( 3 − 3 ) − 𝑀 ( 3 − 3 )
𝑟 𝑟𝑆𝑠𝑐 𝑟𝑆𝐸 𝑟𝑀𝑠𝑐 𝑟𝑀𝐸

(2.14)

These are EOM, where r and v are respectively the

position and velocity vectors of the spacecraft with
respect to the Earth-in the J2000 inertial reference frame, rSsc and rSE are the Sun-spacecraft and Sun-
Earth vectors, rMsc and rME are the Moon-spacecraft and Moon-Earth vectors, µE, µS and µM are the
planetary constants of Earth, Sun, and Moon respectively. The position of Sun and Moon with respect
to the Earth and the spacecraft are calculated using planetary ephemeris, accounting for secular
variations in the orbital elements of both the Earth and the Moon. Secular variation of the angles
between the Earth equatorial and the Moon-equatorial reference frame need to be considered. Above
equations describe the perturbed motion of the spacecraft where the second and third terms in the
variation of the velocity vector are the gravity perturbation of the Sun and the Moon respectively. If
compared to a more traditional Hofmann transfer, this kind of trajectory increases the transfer time by
approximately 3 months. The figure shows a typical WSB transfer for SRMSAT-2. In principle, 3
maneuvres would suffice to perform a complete transfer – one maneuver at the Earth, followed by a
single WSB or patch point maneuver, and finally a very small maneuver to inject into orbit at the Moon.

Dynamic Models
SRMSAT’s trajectory will be influenced by solar radiation pressure, adverse thrust due to ADMs, small
and short duration outgassing events due to the evaporation of residual substances and parasitic gravity
effects. These contribute to forces/accelerations not accounted/modelled for in the EOM for Orbit
Determination and need to be modelled by statistical fit of systematic errors in the measurement process.

106
Thrust Calibration
To characterize the RCS adverse-thrust behaviour, in addition to the passive trending analysis, active
calibrations will be conducted during the test and calibration schedules in the second phasing orbit. A
series of test burns would need to be strategized to remove un-modelled thrust effects on orbit
knowledge and prediction. Plane-wise adverse thrust components would be determined in various RCS
thrusting scenarios during the test sequence. This calibration involves slewing the spacecraft to three
mutually orthogonal attitudes and estimating perturbing delta v.

Gravitational Model
The gravitational model consists of Newtonian point masses for 4 body analysis of Sun, Earth, Moon,
and Spacecraft. In reasonable approximation, the problem is bifurcated into two CRTBPs, and the EOM
are integrated/propagated in specific regimes with Sun-Earth-Spacecraft and Earth-Moon-Spacecraft
(circular restricted three body problem) in the preliminary analyses. In detailed design phases, analyses
would be conducted with planetary ephemerides and constants, will be selected and used for all mission
phases.

Solar Radiation Pressure (SRP)

The solar radiation pressure (SRP) model consists of a parabolic antenna, six flat plates to represent the
spacecraft bus and two flat plates to represent the solar power arrays. Each component has a diffuse and
specular reflectivity coefficient. Diffuse reflectivity imparts acceleration normal to the reflecting
surface while the specular term imparts accelerations both normal to and in the plane of the reflecting
surface. Trend analyses in the detailed design will be done by simulating mission operating modes in
mission timeline to evaluate effective solar radiation incidence for each attitude which will be integrated
for respective time periods. Since the geometric shape of the SRMSAT-2 bus is simple, the scope of
SRP modelling both prior and post the launch are anticipated to be easy to accommodate and update in
the OD models.

Outgassing
Outgassing events occur due to the evaporation of residual volatile substances.The pre-launch analysis
will be carried to estimate the maximum possible outgassing acceleration which is most likely to occur
soon after launch. They are likely to decay exponentially. The magnitude of acceleration is, however,
quiet small and will not affect the overall trajectory considerably.

The Capture Corridor

In order to derive the accuracy requirement for orbit determination throughout the WSB transfer,
leading to lunar insertion, the required level of accuracy at lunar insertion needs to be defined. An error
in the determination of the exact lunar injection maneuver will lead to the spacecraft entering in
undesired orbit increasing mission orbit acquisition ∆V simultaneously increasing risk factor of overall
mission success. Therefore, prior analyses focused on investigating the influence of insertion errors (i.e.
sensitivity) in the capture lunar orbital elements would be required. This early analysis will permit the
definition of a capture corridor that guarantees the correct lunar insertion.

Orbit Determination
The Orbit Determination, OD, is performed by processing raw measurements coming from the one or
more of ISTRAC ground station for Earth-orbiting phase and 18m IDSN antenna, located in Byalalu.

107
The set of measurements includes range and range rate  and ̇ , from the ground station, plus the
pointing angles Az, El (respectively azimuth and elevation). Since the actual position of the spacecraft
is given in the Earth Centred Inertial (ECI) reference frame, it is necessary to write the state as it was
seen in the local South East Zenith (SEZ) or any such local reference frame of the ground station. The
range in the ECI reference frame is given by the difference in position of the spacecraft and ground
station location.

ρ𝐸𝐶𝐼 = 𝑟𝐸𝐶𝐼 − 𝑟𝑠𝑖𝑡𝑒−𝐸𝐶𝐼

(2.15)

Where rsite-ECI is the ECI position of the ground station. The range and velocity vectors in the SEZ frame
are given by the following transformations:

ρ𝑆𝐸𝑍 = 𝐴𝑆𝐸𝑍−𝐸𝐶𝐸𝐹 𝐴𝑇𝐸𝐶𝐼−𝐸𝐶𝐸𝐹 ρ𝐸𝐶𝐼

ρ̇ 𝑆𝐸𝑍 = 𝐴𝑆𝐸𝑍−𝐸𝐶𝐸𝐹 𝐴𝑇𝐸𝐶𝐼−𝐸𝐶𝐸𝐹 𝑣𝐸𝐶𝐼

(2.16)

where ASEZ-ECEF and AECI-ECEF are respectively the transformation matrix from the Earth Centred Earth
fixed (ECEF) reference frame to SEZ and the transformation matrix from ECEF to ECI:
cos  − sin  0
𝐴𝐸𝐶𝐼−𝐸𝐶𝐸𝐹 = ( sin  cos  0 )
0 0 1
cos  sin  − sin  sin  − cos 
𝐴𝑆𝐸𝑍−𝐸𝐶𝐸𝐹 = ( −sin  cos  0 )
cos  cos  cos  sin  1
(2.17)

.λ and υ are the latitude and longitude of the ground station;  280.4606 t is the rotation angle
between the ECI and ECEF reference frame about the z-axis; Ω is the Earth’s angular speed expressed
in deg/day and t is the time expressed in MJ2000. Finally, the set of simulated measurements is obtained
from the SEZ position and velocity:

ρ = |ρ𝑆𝐸𝑍 |
ρ𝑍
𝐸 = 𝑠𝑖𝑛−1
ρ
ρ𝐸
𝐴 = 𝑠𝑖𝑛−1
√(ρ2𝑠 + ρ2𝐸 )

ρ̇ 𝑆𝐸𝑍 ∗ ρ𝑆𝐸𝑍
ρ̇ =
ρ

(2.18)

ρS, ρE, ρZ, being the components of the range in the SEZ frame. The actual measurements were simulated
by perturbing the nominal ones defined in above equations with a random noise with normal
distribution. Ionospheric and tropospheric refractions were not taken into account since their effects are
expected to be corrected at ground station level before the information is provided to flight dynamics

108
for orbit determination. The orbit state vector is obtained by means of a suitable filtering technique able
to deal with nonlinearities in the system dynamics.

Filtering Techniques
The dynamics and measurement model used in the filtering takes the following form :
v 0
̇ = [ ṙ ] = [
x(t)
v̇ f(x(t), t)] + [aw + aSolar ]

yk = h(xk ) + vk

(2.19)

where f(x(t),t) is the set of nonlinear continuous-time equations in the second in the system of EOM, aw
is the random noise acceleration produced by the attitude system, asolaris the solar pressure modelled as
follows:
𝐴 𝑟𝑆𝑠𝑐
𝑎𝑆𝑜𝑙𝑎𝑟 = −𝐶𝑅 𝐹𝑆𝑜𝑙𝑎𝑟 (𝑟𝑆𝑠𝑐 )
𝑀(𝑡) 𝑟𝑆𝑠𝑐

(2.20)

where A, M are the satellite cross section area and mass respectively, CR the reflectivity coefficient and
FSolar is the solar flux. In above equations the measurement model h(xk) corresponds to the set of four
equations, giving the values of range, elevation, azimuth and range-rate, and yk is the set of actual
measurements corresponding to the state vector xk at the measurement time tk,. The term νk is the
measurements noise defined as νk = R σnormal, where σnormalN (0,1) is a vector of random numbers taken
from a normal distribution. R is the observation error covariance matrix whose components along the
diagonal are the squared values of the ground station errors associated with the measurements. Since it
is assumed that all the measurements are uncorrelated and independent, extra diagonal elements in R
are null and the matrix is diagonal.

In the following three sequential filtering techniques were analysed to establish which approach would
be most suitable to navigate a spacecraft to the Moon along a WSB transfer. The three filters are: an
Extended Kalman Filter (EKF), an Unscented Kalman Filter (UKF) and a Kalman filter based on high
order Taylor expansions.

Extended Kalman Filter

The EKF is a well-known dynamic optimal filter which was first employed in the Apollo program
(Battin and Levine, 1970). The EKF linearizes the filtering problem about the estimated state and
covariance matrix. It is composed of two conceptually distinct phases: the time update and the
measurements update. The time update phase consists of the propagation of the latest estimate 𝑥̂𝑘+ to
− −
obtain an a-priori estimate at current epoch 𝑥̂𝑘+1 with the corresponding covariance matrix 𝑃𝑘+1 :
−
𝑥̂𝑘+1 = 𝑥(𝑡𝑘+1 ; 𝑥(𝑡𝑘 ) = 𝑥̂𝑘+ )
−
𝑃𝑘+1 = Φ𝑘+1 𝑃𝑘+ Φ𝑇𝑘+1

(2.21)

109
 −
where the predicted state 𝑥̂𝑘+1 is obtained by integrating forward the above equations starting from the
latest estimate state 𝑥̂𝑘+ ; Φ𝑘+1 is the state transition matrix (STM) coming from the linearization of the
dynamic equations about the updated state:
𝑡𝑘+1
Φ𝑘+1 = Φ𝑘+1 (t 𝑘+1 , t 𝑘 ) = Φ𝑘+1 (t 𝑘 , t 𝑘 ) + ∫ 𝐹(𝑥, 𝑡) Φ𝑘+1 (t , t 𝑘 ) 𝑑𝑡 ≈ 𝐼 + 𝐹𝑘 ∆𝑡
𝑡𝑘

f( 𝑥𝑘+ )
𝐹𝑘 =
𝑥𝑘+

(2.22)

Being Φ𝑘+1 (tk, tk )equal to the identity matrix. The measurement update phase consists in the
− +
computation of the Kalman gain Kk1 and the state estimate 𝑥̂𝑘+1 and covariance matrix, 𝑃𝑘+1 updates:
𝑇
𝐾𝑘+1 = 𝐻𝑘+1 [ 𝐻𝑘+1 𝑃𝑘− 𝐻𝑘𝑇 + 𝑅]−1
+ + − ))
𝑥̂𝑘+1 = 𝑥̂𝑘+1 + 𝐾𝑘+1 (𝑦𝑘+1 − ℎ(𝑥̂𝑘+1
h( 𝑥̂𝑘+1
−
)
𝐻𝑘+1 =
𝑥̂𝑘+1
+

+ −
𝑃𝑘+1 = (𝐼 − 𝐾𝑘+1 𝐻𝑘+1 )𝑃𝑘+1

(2.23)

Where Hk1 is the Jacobian matrix of the measurement function. The Kalman gain as reported in
equations minimizes the a posteriori error covariance matrix. It represents a function of the relative
certainty of the measurements and current state. As the gain increases the measurements are trusted
more and the estimates rely less on the prediction model. On the contrary as the gain decreases,
measurements tend to be ignored and the estimate rely more heavily on the prediction model.

Unscented Kalman Filter

The unscented Kalman filter, first introduced by Julier et al. in 1995, works on the premises that by
using a limited set of sample, optimally chosen, it should be easier to approximate a Gaussian
distribution than to approximate a nonlinear function. The UKF was shown to be preferable to the EKF
in the case of nonlinear systems as the expected error in terms of mean and covariance matrix is lower,
and it does not require the derivation of the Jacobian matrix. The Kalman filter updates are represented
as follows:

𝑥̂𝑘+ = 𝑥̂𝑘− + 𝐾𝑘 𝑣𝑘

𝑃𝑘+ = 𝑃𝑘− − 𝐾𝑘 𝑃𝑘𝑣𝑣 𝐾𝑘𝑇

(2.24)

Where 𝑣𝑘 and the Kalman gain𝐾𝑘 are:

υ𝑘 ≡ υ𝑘 − 𝑥̂𝑘− = 𝑦̃𝑘 − h(𝑥̂𝑘− , 𝑘)

𝑥𝑦
𝐾𝑘 = 𝑃𝑘 (𝑃𝑘𝑣𝑣 )−1

(2.25)

110
𝑣𝑘 is called innovations process and represents the difference between the current measurements and
the predicted ones, 𝑃𝑘𝑣𝑣 is the covariance matrix of the innovations process at the sample time t k . The
matrix xy Pk is the cross-correlation between 𝑥̂𝑘− and 𝑦̂𝑘− . The approach used in the filter design requires
augmenting the covariance matrix with:

P𝑘+ 𝑃𝑘𝑥𝜔 𝑃𝑘𝑥𝑣

𝑃𝑘𝑎 = [ (𝑃𝑘𝑥𝜔 )𝑇 𝑄𝑘 𝑃𝑘𝜔𝑣 ]
(𝑃𝑘𝑥𝑣 )𝑇 (𝑃𝑘𝜔𝑣 )𝑇 𝑅𝑘

(2.26)

Where, 𝑃𝑘𝑥𝜔 is the correlation between the state error and process noise, 𝑃𝑘𝑥𝑣 is the correlation between
the state error and measurement noise, and 𝑃𝑘𝑥𝑣 is the correlation between the process noise and
measurement noise, which is zero in this case. If L is the number of elements per column of the
augmented covariance matrix 𝑃𝑘𝑎 , then a set of 2L samples a χ𝑎𝑘 (i) , with i=1,..,2L, called sigma points,
are generated such that:

χ𝑎𝑘 (i) = σ𝑘 (𝑖) + 𝑥̂𝑘𝑎

σ𝑘 = ±Ƞ√𝑃𝑘𝑎

χ𝑎𝑘 (0) = 𝑥̂𝑘𝑎

(2.27)

where η is a suitable scaling factor and 𝑥̂𝑘𝑎 is the augmented state :

𝑥̂𝑘
𝑥̂𝑘𝑎 = [0𝑞∗1 ]
0𝑧∗1

(2.28)

q is the dimension of w, and z is the dimension of yk. The sampled sigma points are then:

𝑥̂𝑘𝑥 (𝑖)
χ𝑎𝑘 (i) = [𝑥̂𝑘𝑤 (𝑖)]
𝑥̂𝑘𝑣 (𝑖)

(2.29)

χ𝑘𝑥 is the vector of the first n (size of xk) elements of a χ𝑎𝑘 , χ𝑤

𝑘
is a vector of the next q elements of a χ𝑎𝑘
and χ𝑣𝑘 is the vector of the last z components of a χ𝑎𝑘 . The sigma points are transformed or propagated
through the system dynamics equation by the so called unscented transformation (UT)

γ 𝑘+1 (𝑖) = 𝑓( χ𝑘𝑥 (𝑖), χ𝜔

𝑘
(𝑖), 𝑢𝑘 , 𝑡𝑘 )

𝑥 (𝑖),
μ 𝑘+1 (𝑖) = ℎ( χ𝑘+1 𝑢𝑘+1 , χ𝑣𝑘+1 (𝑖), 𝑡𝑘+1 )

(2.30)

111
The predicted mean of the state vector, the covariance matrix and the mean observation can be
approximated using the weighted mean and covariance of the transformed vectors:
2𝐿
−
𝑥̂𝑘+1 = ∑ 𝑊𝑖𝑚 χ 𝑘+1
𝑥
(𝑖)
𝑖=0

2𝐿
−
𝑃𝑘+1 = ∑ 𝑊𝑖𝑐 [χ 𝑘+1
𝑥 (𝑖) − ][χ 𝑥 (𝑖)
− 𝑥̂𝑘+1 𝑘+1
−
− 𝑥̂𝑘+1 ]𝑇
𝑖=0

2𝐿
−
𝑦̂𝑘+1 = ∑ 𝑊𝑖𝑚 μ 𝑘+1 (𝑖)
𝑖=0

(2.31)

where m 𝑊𝑖𝑚 , 𝑊𝑖𝑐 are suitable weighting factors. Finally the updated covariance and the cross
correlation matrix are:
2𝐿
𝑦𝑦
𝑃𝑘+1 = 𝑣𝑣
𝑃𝑘+1 = ∑ 𝑊𝑖𝐶𝑂𝑉 [γ 𝑘+1 (𝑖) − 𝑦̂𝑘+1
−
][γ 𝑘+1 (𝑖) − 𝑦
−
̂𝑘+1 ]𝑇
𝑖=0

2𝐿
𝑥𝑦
𝑃𝑘+1 = ∑ 𝑊𝑖𝐶𝑂𝑉 [χ 𝑘+1
𝑥 (𝑖) − ][ 𝑥 (𝑖)
− 𝑥̂𝑘+1 −
γ𝑘+1 − 𝑦̂𝑘+1 ]𝑇
𝑖=0

(2.32)

In this way it is possible to update the filter to the next observation and prediction at time tk+1.

Conclusion:
The role of Guidance, Navigation and Control in the mission profile selected for SRMSAT-2 has been
laid out. The on-ground GNC elements will be a part of Mission Operations and Flight Dynamics
facility of Team SRMSAT ground segment. The relevance and criticality of GNC operations directly
translates to the mission success. A robust navigation strategy will need to be configured in the detailed
design with available ground stations after consultation with ISRO for access to ISTRAC and IDSN.
With available support, optimal strategy to utilise ISTRAC, IDSN, University S-band Ground station,
interfaced with Flight Dynamics and Mission Operations facility at SRM University will be designed
to overcome constraints of the mission/project. The navigation and guidance requirements for the
mission will be need to be reviewed and validated by experts at ISRO to initialise detailed design of the
sub-system

112
 3. ATTITUDE DETERMINATION AND CONTROL SYSTEM

3.1 INTRODUCTION
Attitude Determination and Control System (ADCS) is a completely autonomous subsystem
responsible for determining, predicting and maintaining the spacecraft’s attitude throughout the mission
life of the spacecraft. Primary functions of ADCS includes achieving the pointing accuracy defined by
other subsystems such as TTC, Power, Payload and maintain the required attitude while overcoming
external disturbances due to space environmental factors such as atmospheric drag, solar radiation
pressure, and gravity gradients.
The main functions of the ADCS subsystem can further be divided into Attitude Determination and
Attitude Control. Attitude determination is the process of computing the spacecraft orientation, within
a pre-defined accuracy, using data obtained from the on-board sensors. Attitude control is the process
of orienting the satellite in a specified direction using actuators such as reaction wheels and RCS
thrusters. In order to execute attitude control, formulation of a spacecraft specific dynamic model is
necessary; Attitude Dynamics is responsible for formulating and designing dynamical models which
describe in detail the evolution of the spacecraft attitude due to controlling torques and external
disturbances.
SRMSAT-2 is a micro-class lunar orbiter following a low energy transfer trajectory using weak stability
boundary conditions. ADCS system design for SRMSAT takes into account reliability, efficiency,
redundancy and is able to provide the required mission performance accuracies. A robust and adequate
system design is, therefore, inevitable to meet the above stated stringent requirements.

3.2 SUBSYSTEM OBJECTIVES

 Stabilization and reorientation of the satellite in the desired attitude based on mission operating
modes while overcoming external torsional disturbances.
 To achieve the pointing accuracy required by the payload(s) and other system operational
modes
 Designing a completely autonomous ADCS system using COTS components and
incorporating them into a micro class satellite through optimum hardware (sensors and
actuators) configurations to provide redundancy, while also providing sufficient efficiency so
as to meet mission requirements.

113
 3.3 DESIGN OVERVIEW
The ADCS hardware consists of sensors and actuators which are designed and configured based on the
reliability of the components. The design also accommodates for redundant sensors and actuators. The
selected components and their configuration are shown in Table 3.1.

Component Quantity Configuration

The star tracker is placed on the top panel of the satellite, adjacent to
one of the QFH antennas. The 15° FOV of the star tracker is not
Star tracker 1
affected by the adjacent components unless they are too near (2.6cm
for 20cm height)

SENSORS 4 out of 8 sensors are placed on the side panel 3, perpendicular to it,
Sun Sensor 8 at 45° to each edge of the panel. The other two face along the solar
panels on face 1 for redundancy and one facing along the face 3.

IMU 2 The IMUs are integrated into the PCB with other electronics.

The RCS is placed on the bottom face of the spacecraft, in-plane and
at 45° to each edge of the face. The thruster configuration is designed
RCS 4
so as to provide maximum control in all the directions with a minimum
number of thrusters possible.
ACTUATORS
The reaction wheels are placed in a tetrahedral configuration with one
Reaction of them mounted on a face and the other three inclined with respect to
4
wheel the body axis. This configuration provides control in all the three-axis
as well as redundancy in case of failure of one of the wheels.
Table 3.1: ADCS Components overview

3.4 OPERATIONAL MODES

Operational modes are used to define a set of possible operating scenarios for a spacecraft. Some
operations are common to every spacecraft such as initialization, de-tumbling, etc., whereas others
could be specifically designed based on mission requirements and mission profile.

Initialization
On spacecraft separation from the retainer lug of the IBL interface, the initialization mode commences
the OBC boots up and performs a system check on all peripherals and hardware units. FDIR confirms
the status of the system check and proceeds to the next mode of operation.

System Commissioning Mode– 0

In this mode, the propulsion system commissioning is executed. Venting, priming and pressurizing are
key operations performed in this mode. The propulsion systems are made ready to perform the required
burns for maneuvers.

114
Detumbling
After being deployed into an orbit, the spacecraft has high angular rates about its orthogonal axis. ADCS
measures the angular rates in various deterministic loops, the control algorithm then computes the
required torques and generates the signals required by the actuators in order to de-tumble the satellite.
As IMUs cannot be used before calibration and star tracker has low accuracies and sampling rates, sun
sensors will be used in the preliminary stage to estimate angular rates so as to reduce the high angular
rates firsthand for coarse stability. Once the rates are reduced, star tracker can be used optimally.

Attitude Hold Mode

The spacecraft is 3-axis stabilized with a reference attitude and the attitude is maintained until it enters
another operation mode. The IMU's are calibrated with respect to stabilized body axis using star tracker
and sun sensor.

System Commissioning Mode– 1

In this operation mode, the solar panels are deployed.

Sun Pointing Mode

In order to generate maximum power for charging the batteries, the spacecraft needs to be oriented such
that the solar panels are directly facing the sun. This mode of operation is called the sun pointing mode.
In this mode, the ADCS carries the required attitude maneuver and orient the solar panels towards the
sun.

Communication Mode
The satellite is generally in the ‘ready to receive mode’ due to the omnidirectional QFH antenna. When
high data rates are required for large amounts of data, the QFH is used to command ADCS to reorient
the spacecraft such that the parabolic antenna points towards the earth in order to communicate with
the ground station.

Maneuver Mode
During orbital maneuvers, the thrust vector of the spacecraft needs to be aligned along the desired
attitude. Instead of spin stabilization during a maneuver, continuous 3-axis stabilization with high
pointing accuracy is maintained throughout the burn. Three-axis stabilization was chosen over spin
stabilization due to its long durations of finite burns.

Low Profile Mode

This mode will be operational during the cruise phase of the mission. The frequency of ADCS
operations will be minimum and only sun pointing is required (at lower pointing accuracy) from ADCS
during this mode. This mode is activated when a state of emergency is declared by the FDIR algorithm.
The ADCS operations are minimal during this mode. The actuators are turned off, and the sensors
receive low or no power, depending on the battery status.

Payload Operation Mode

The primary payload of SRMSAT (Radiometer) requires 3-axis stabilization with a position accuracy
of 0.1°. The spacecraft is oriented towards the target, defined by the mission profile. Under this mode,
the payload operation is given a higher priority over sun pointing. This means that the satellite is
reoriented to point the solar panels towards the sun only if it does not disturb the pointing requirements
of the payload operations.

115
 3.5 ADCS ARCHITECTURE
The ADCS architecture is shown in Fig 3.1. The attitude information obtained from the sensors is of
different formats and measured at different sampling rates. The attitude data type converter combines
all this information to provide the sensed attitude data in the required format and frequency. The
Attitude determination algorithm provides the best estimate of the attitude with respect to an inertial
frame.
The operating modes are identified from the on-board trajectory model. These operating modes have a
pre-defined attitude towards which the spacecraft needs to be oriented for the duration of an operation.
The error, which is the difference between the predefined attitude and the determined attitude is
computed and fed into the control algorithms. The attitude propagator estimates the attitude change
using dynamic models, and the predicted attitude information is also used by the control algorithms (to
determine any non-optimal performance, initiating FDIR). The control algorithms calculate the required
torques required to perform attitude maneuvers. This command torque is distributed among the
actuators according to their operation limits.

116
Figure 3.5: ADCS ARCHITECTURE

117
 3.6 ATTITUDE DETERMINATION
Introduction
The attitude determination system is responsible for detecting the attitude of the spacecraft with respect
to a reference frame. The reference frames can be multiple, based on operation modes defined. The
determination system is required to achieve at least threshold deterministic accuracy requirements so as
to increase ADCS pointing accuracy for the payload operation. In spite of the various requirements of
other subsystems, the determination system needs to be consistent and reliable for the control system to
perform the required attitude control and maneuvers. It has been aimed to design a system with
maximum attitude determination accuracy and redundancy for the spacecraft to achieve the various
attitude constraints.

Star tracker
A reliable attitude determination system is critical to the success of the spacecraft. As high accuracy
and high reliability devices, star trackers play an important role in attitude determination of the
spacecraft. The star trackers determine the attitude of the satellite by capturing star images and
processing them to locate the position of the stars. Based on the star images sampled by the star tracker,
the locations of the star image in the star system coordinate database are identified with identification
algorithms. With the help of the coordinates of the star images in the field of view (FOV), a three-axis
attitude of the star tracker in the body reference coordinate is calculated and then the attitude of the
satellite is determined by coordinate transformation. The accuracy of star trackers lies within the range
of few arc-seconds which further aids to its high reliability over the attitude determination system. All
the final calibration of other determination sensors like IMU rely upon the working of the star trackers.
Hence, the success of the attitude determination system, on the whole, is dependent majorly on the
accuracy and consistency of the star trackers.

Figure 3.6: Basic principle of star tracker

This makes the reliability of the star trackers a crucial part of the mission success. Star trackers are
generally large size, heavy, expensive, and power consuming due to large computational requirements.
This makes this hardware difficult to miniaturize for the small satellites without compromising the
deterministic accuracy. However, there are some star trackers like ST-400, which are specially designed
for the small satellite application.

118
Configuration
The star trackers give the relation between the coordinate axis of the satellite and the star system in the
form of quaternions as the output. The general configuration of star trackers is decided based on the
placement of other hardware units that could hinder in the functioning of the star trackers. Also, it needs
to be pointing to a region where there is the minimum possibility of getting an eclipse. Considering all
the above-mentioned parameters, the most suitable and beneficial position for the star trackers is
perpendicular to the direction of the solar panels, payload pointing, and away from the position of the
thrusters. The configuration decided is on the side panel 2, behind the deployable solar panels. This
position provides sun shielding to the star tracker which in turn increases the life of the hardware. It
also assures that it points towards a clear sky without any hindrance from the sun or the moon and also
avoid the blocking of the field of view of the sensor due to the emission from the thrusters.
Hardware Selection
Star trackers are basically a camera with enhanced electronics and a microcontroller to process the
images, detect the stars, and calculate the quaternion data. The primary criteria for selecting the right
star tracker for the mission are size, mass, accuracy, and cost.

Criteria ST-200 ST-400 VST-41M RIGEL-L Procyon

Mass 50 gm 210 gm 0.7-0.9 kg 2.6 kg 1.75 kg
Size 30 x 30 x 38.1 53.8 x 53.8 x 80 mm x 100 mm
mm 90.5 mm x 180 mm
Accuracy X/Y <30 arcsec X/Y <15 arcsec X/Y <18 arcsec X/Y <5arcsec X/Y <5arcsec
Z <200 arcsec Z <120 arcsec Z <122 arcsec Z <25 arcsec Z <30 arcsec
Power consumption 3.65V, 600mW 5V, <700mW 12V, 2.5W 28V, 6.5W 28V, 6.5W
Radiation tolerance 12.5 kRad >25 kRad >20 kRad <5 kRad 5 kRad
Cost $24,000 $45,000 $430,000 $330,000
Update Rate 5 Hz 5 Hz 4Hz 1 to 16 Hz 1 to 4 Hz
Table 3.2: A brief comparison between the various star trackers considered

From the above table of comparison, the best option of a star tracker is Hyperion Technologies’ ST-
400.It is an improved version of the ST-200, which is aimed at applications in pico- and nanosatellite
platforms, with increased performance and increased radiation tolerance. It has nominal power
consumption and size without compromising with the accuracy level. With 700mW of maximum power
consumption at 5V, it does not consume much power even with a duty cycle of 100%. Being a major
part of the attitude determination system, it is essential that it consumes the least power which can be
provided by the solar panel. Apart from consuming less power, the size of the sensor is also a major
consideration factor as it has just a size of 53.8 x 53.8 x 90.5 mm.

119
 Figure 3.3: The basic mechanical characteristics of ST-400

Hyperion Technologies is a Berlin-based company. The architecture of ST-400, on the whole, consists
of an optical lens with a field of view of around 15°, and a microcontroller to perform the processing
and to calculate the attitude values. The information collected from the star image is processed, and the
final quaternions and Euler angles are generated by the sensor unit itself. The transformation matrix is
edited so that the final quaternions are generated after the transformation with respect to the body axis
of the spacecraft.

120
 Figure 3.4: Hyperion Technologies’ ST-400

The ST-400 has passed all the space grade qualification tests and is commercially available for use since
October 2013.
The specifications of ST-400 are listed in the table below. Apart from the general specifications of ST-
400, the major advantage of it is having its own processor pre-installed and calibrated by the
manufacturer. Thus, the chances of getting the deterministic error due to calibration or interfacing of
the hardware with the software are reduced to a minimum.

PROPERTY VALUE UNIT

Attitude determination accuracy (yaw, 15 arc seconds
pitch)
Attitude determination accuracy (roll) 120 arc seconds
Maximum slew rate (tip/tilt) 0.5 °/sec
Maximum slew rate (roll) 1 °/sec
Interfacing RS422/RS485 Bidirectional interfacing
Operating temperature -20 - +40 °C
Supply voltage 3.6 – 15 V
Power consumption 700 mW
Table 3.3: Basic properties of ST-400

Sun Sensor
Sun sensors provide the sun position vector with respect to the body axes of the satellite. Sun sensors
are one of the most frequently used attitude sensors in the determination system. Unlike the star trackers,
they don’t have any slew rate restriction for measuring the values. Unlike the Earth, the angular radius
of the Sun is nearly orbit independent and sufficiently small (0.267° at 1 AU). For most applications a
point-source approximation is valid. This simplifies both, the sensor design and attitude determination
algorithms. The Sun is sufficiently bright to permit the use of simple, reliable equipment without
discriminating among sources. Sun sensors are broadly categorized as analog and digital sensors.
The digital sun sensors are more accurate than the analog sensors. The general accuracy of the analog
sensors is 0.1° whereas for digital sensors it can go up to 0.01°. However, the sensor made is analog
which is converted to digital, for better accuracy and less white noise created due to analog signals.

121
Configuration
The simplicity of sun sensors makes them a very reliable unit in attitude determination system. They
don’t have any stringent criteria under which they are bound to function. This reduces any errors that
can result in due to any other source of light. The primary criteria for selecting an adequate configuration
for sun sensors is to cover all the directions under its field of view. Depending upon the structural design
of the satellite bus, the placement of these sensors vary. This makes the placement criteria to be
completely mission and structure dependent. The designed configuration for the sun sensors for
SRMSAT can be shown in the figures below:

Figure 3.5: Sun Senor Configuration

The final configuration consists of 8 sensors. Four of them are placed on the longitudinal edges at 45°
to each side with two of them in the direction of the solar panels. The other four are placed parallel to
the surface: one to the solar panel, one on the top, and one at the bottom. It has been aimed to use the
minimum number of sun sensor units as well as provide redundancy for the failure of any unit. It has
been aimed to use the minimum number of sun sensor units with redundant units for the solar panels

122
(Sun Pointing Mode). The one near the thruster is not used during maneuvers as it will result in distorted
value due to the fumes.
Sensor Selection
The sun sensor being used in SRMSAT is a PSD based analog sensor, designed in-house. A sun sensor
primarily converts incident light information into specific electric signals that are used to calculate the
angle of the sun with respect to the face of the sensor. For this generation of electric signals based on
the incident light rays, PSD is used in the design. PSDs or position sensitive devices are optoelectronic
devices that measure the position of the incident light and provides outputs in the form of current. In
this case, a two-dimensional tetra-lateral PSD is used to get a two axis output. The incident light beam
is limited through a pinhole placed above the active face of the PSD. Smaller the pinhole more is the
precision. However, a much smaller hole would also result in diffraction. The advantage of PSDs over
the CCDs is that PSDs can track very small positional changes over the photosensitive area. On the
contrary, CCDs resolve distances limited by their pixel size and thus can lose microscopic spot
information.
The chosen PSD is the Hamamatsu S5990-01.It is a tetra-lateral pin cushion type two-dimensional PSD.
It has four current outputs whose values are decided by the position of the incident beam on the active
area.

Figure 3.6: Tetra lateral configuration of S990-01

The angular rates can be given in the terms of elevation (θ), and azimuth (𝛷) which are the attitude
angles to be determined. The positions x and y are calculated by the formulae given by the Hamamatsu
itself:
2𝑥 (𝐼2 + 𝐼3 ) − (𝐼1 + 𝐼4 )
=
𝐿 𝐼1 + 𝐼2 + 𝐼3 + 𝐼4
2𝑦 (𝐼2 + 𝐼4 ) − (𝐼1 + 𝐼3 )
=
𝐿 𝐼1 + 𝐼2 + 𝐼3 + 𝐼4
(3.1)
Using these x and y we can calculate the required elevation and azimuth value.
𝑦
𝜑 = 𝑎𝑟𝑐𝑡𝑎𝑛
𝑥

123
 √𝑥 2 + 𝑦 2
𝜃=
ℎ
(3.2)
Where h is the height of the pinhole from the PSD. The first equation for 𝛷 however, has a singularity
𝜋
when x=0. This happens for 𝛷=±2 , where the sign is decided by the sign of y. The equation for θ does
not have this problem since the height of the pinhole, h, is fixed and is never zero. If both x and y are
zero, the equation for 𝛷 has no solution. This happens when θ=0. In this situation, 𝛷is not defined. This
problem can be solved if we calculate θ before and calculate 𝛷 only if 𝜃 ≠ 0.
These electric currents generated by the PSD are still in an implicit form and can’t be used directly for
the position calculation. In the various electronic stages, the first stage involves converting these current
values to voltages and amplifying them. The saturation current for the S990-01 is 500𝝁A. Hence, the
average value of currents generated would be less than 50𝝁A. The variable amplifier converts these low
level current values to voltages and amplifies them to the required value. After the amplification of the
voltage levels they are passed through a low pass filter to filter out the noise generated by the PSD.
These voltages are converted into digital signals and finally detected by a microcontroller for final
attitude calculations. The variable amplification of the voltages, A/D conversion, and communication
of the data with the control system are all managed by the microcontroller.
The Hamamatsu PSD has an active area of (4x4) mm with a positional resolution of 0.7𝝁m. With an
operating temperature of -20 to +60°C, it is ideal for space use. It has a maximum detection error of
±70𝝁m which makes the maximum angular detection error to be 0.01°. The size of the pinhole is
400𝝁m. The FOV of the resultant system is a rectangular pyramid with 120° between opposite sides
and 135° at the diagonals. The peak spectral response of the PSD is 960nm, which is optimal for the
usage as a sun sensor. The final data collected with the microcontroller about all the four voltages are
processed to calculate the angular position of sun through the sun sensor.

Inertial Measurement Units (IMU)

IMUs are the inertial sensors being used in the attitude control system of SRMSAT. These sensors
measure the rate of change of angular velocity of the body. The basic principle of MEMS IMUs is a
vibrating structure. The Coriolis Effect causes the object to exert a force on its support, and by
measuring this force, the rate of rotation is determined. IMUs generally are a combination of a
gyroscope and an accelerometer. Gyroscopes measure the angular drift of the system in the inertial
reference frame. Using the above-stated output from the IMUs, the inertial change in the state of the
satellite is measured. The gyro data is also used in the dynamic modeling of the spacecraft for attitude
prediction.
Configuration
Inertial measurement units (IMUs) measure the current orientation, and rate of turning along with the
linear acceleration of the body. IMUs are placed in the PCB units along with other units and placed on
one of the side panels of the spacecraft.
Hardware Selection
The IMU sensor used for SRMSAT is the xsens’ MTi-1 series IMU. The MTi-1 series is a full-featured
3D Inertial Measurement Unit. With a 12.1 x 12.1 mm module and low power consumption of just
45mW at 3V, it is the smallest and most dynamic sensor in the mission. The MTi series-1 IMU features
a 3D accelerometer-gyroscope combo sensor, a magnetometer, a high accuracy crystal, and a low-power
MCU. It is a fully tested self-contained module that can output

124
 1. 3D orientation data such as Euler angles (yaw, pitch, and roll), rotation matrix and quaternions.
2. Coning and sculling compensated orientation increments and velocity increments (∆q and ∆v)
from its Attitude Engine algorithm.
3. Sensors data (acceleration, the rate of turn, magnetic field).
The sensitivity goes up to 0.001°/s/g. The low power MCU inbuilt in the system handles the data
processing for the inertial values. The MCU coordinates the synchronization and timing of the various
sensors, it applies calibration models (e.g. temperature models) and output settings and runs the sensor
fusion algorithm. The messages and the data output are fully configurable, so that the MTi 1-series
limits the load, and thus power consumption, on the application processor.

Figure 3.7: MTi-1 Series IMU

The MTi 1-series is a self-contained module, so all calculations and processes such as sampling, coning
and sculling compensation and the Xsens XKF3TM sensor fusion algorithm run on the board. The
calibration of the values, however, can be overwritten according to the mission requirements.
The XKF3 is a sensor fusion algorithm used in the MTi 1-series, based on extended Kalman filter
framework that uses 3D inertial sensor data (orientation and velocity increments).

125
 Figure 3.8: Information flow in MTi-1

MTi 1-series gyroscopes have a sensitivity variation of only 0.05% over the temperature range, enabling
them to produce efficient and accurate results every time. These sensors have the bandwidth of 180 Hz
and a natural frequency of 26 KHz.

Full-scale range ±2000 º/s (dps)

non linearity 0.1% FS
Noise density 0.01 º/s/√Hz
Bias stability 10 °/hr
Table 3.3: Gyroscope specifications

Full-scale range +-16g

Non linearity 0.5% FS
Noise density 200 μg/√Hz
Bias stability 1.1 Mg
Table 3.4: Accelerometer specifications

126
 Parameter Typical (unit)
Roll/Pitch Static 0.75 deg
Dynamic 1.0 deg
Yaw Static/dynamic, 2.0 deg
(heading) Magnetic field referenced

VRU general filter profile <1 deg after 60

(unreferenced yaw) min.
Sensitivity Minimum count that can be taken 0.001 deg/s/g

Output data rate 0-1000 Hz

Table 3.5: Orientation specifications

MTi has its own calibrated equation showing its stability under high-temperature ranges, given by:

𝑆 = 𝐾𝑇−1 (𝑢 − 𝑏𝑇 )
(3.3)
Where,
S = sensor data of the gyroscopes, accelerometers and magnetometers in rad/s, m/s2 or a.u. respectively
𝐾𝑇−1 =Gain and misalignment matrix (temperature compensated)
u = Sensor value before calibration (unsigned 16-bit integers from the sensor)
bT = Bias (temperature compensated)

Xsens calibrates IMU’s for many parameters and all calibration values are temperature dependent and
temperature calibrated. Also, the calibration values are stored in non-volatile memory in the MTi.
The operating temperature is -40 to +85 °C, which is wide enough for space applications.
The interface of the sensor with the ADCS controller is made through the I2C protocol. There are also
SPI and UART protocols available for communication. The I2C protocol was chosen as it needs fewer
connections and works on addressing of devices. It is a faster and better option for small range
communication.

Figure 3.9: I2C Interfacing of MTi-1

127
Attitude Processing
All the above determination sensors generate their own respective readings. However, they need to be
processed before feeding into the prediction algorithms. The different values from different sensors
have to be synced at the same sampling rate, and all data types converted to generate a common attitude
quaternion.

The star tracker has a frequency of 5Hz and gives out the attitude in the form of quaternion values. The
sun sensor has a resolution of 0.01°. It generates the sun vector with respect to the center of the body
axis. The IMU measures the inertial motion of the spacecraft, and its output is the rate of change of
angular velocity. The values from all the three sensors are processed by the Extended Kalman Filter to
generate the final attitude. As the sampling rate of the star tracker is comparatively less (5Hz) than the
sun sensor (100Hz) and IMU (100Hz), and is more accurate the star tracker values are used to calibrate
the sun sensor and IMU. The IMU has an in-run bias stability of 10 °/hr. This demands a stabilization
which will be provided by the sun sensor and the star tracker. The IMU values are also filtered by a
threshold drift. If the sampled values have a large deviation compared to the last recorded value, instead
of finding the error value we just skip that particular value and move to the next one.
All attitude filters use some variant of the KF, which proceeds by iteration where each iteration
involves three steps: measurement update, attitude vector reset, and propagation to the next
measurement time. The measurement update step updates the error state vector. The reset moves the
updated information from the error state to the global attitude representation and resets the components
of the error state to zero. The propagation step propagates the global variables to the time of the next
representation. The error state variables do not need to be propagated because they are identically zero
over the propagation step. These KF filters are used to incorporate the gyro information as part of the
dynamic model instead of using it as a Kalman measurement update. This is further used to generate
the dynamic model of the spacecraft used for attitude prediction and feeding into the attitude propagator
for comparison with the trajectory model to generate required commands for the control system.

128
 3.7 ATTITUDE DYNAMICS

Introduction

Trajectory dynamics governs the motion of spacecraft’s center of mass relative to some inertial frame,
whereas the motion of the rigid body about the center of mass is governed by attitude dynamics. The
knowledge of attitude dynamics is necessary for attitude prediction, interpolation, stabilization and
control of the spacecraft. Physical notations such as inertia tensor, angular momentum, angular kinetic
energy and moment about the center of mass, etc. are stated which are necessary to determine the
angular state of the rigid body. Further, the fundamental laws of angular motion based on Euler's
equations are mentioned.

Inertia Tensor
The inertia tensor or inertia matrix depicts the mass distribution of various components in the spacecraft.
It is a 3x3 matrix with moments of inertia about an orthogonal axis as its diagonal elements. Rest of the
elements are called products of inertia which are unwanted disturbances. Thus, inertia matrix has to be
transformed into a diagonal matrix. The diagonal elements of the new matrix are the principal moments
of inertia.

𝑰𝒙𝒙 −𝑰𝒙𝒚 −𝑰𝒙𝒛

I = [−𝑰𝒚𝒙 𝑰𝒚𝒚 −𝑰𝒚𝒛 ]
−𝑰𝒛𝒙 −𝑰𝒛𝒚 𝑰𝒛𝒛

(3.4)

In the case of SRMSAT-2, the principal moments of inertia as provided by the structure subsystem are
not along the body axes. It is represented by the following matrix.

𝟔. 𝟏𝟓𝟏𝟐 −𝟒. 𝟓𝟎𝟑𝟐 𝟒. 𝟒𝟐𝟗𝟕

I = [−𝟒. 𝟓𝟎𝟑𝟐 𝟒. 𝟖𝟎𝟑𝟓 𝟐. 𝟔𝟖𝟗𝟔]
𝟒. 𝟒𝟐𝟗𝟕 𝟐. 𝟔𝟖𝟗𝟔 𝟓. 𝟖𝟔𝟏𝟏

(3.5)

The above matrix is used to obtain principal moment of inertias as

𝟔. 𝟐𝟗𝟏𝟏 𝟎 𝟎
I=[ 𝟎 𝟓. 𝟗𝟏𝟔𝟐 𝟎 ]
𝟎 𝟎 𝟒. 𝟔𝟎𝟖𝟓

Note: The principal moments of inertia triad do not coincide with the body axes triad. As all the
actuators are positioned with respect to the body axes but dynamic equations are formulated with respect

129
to principal axes, all the control parameters have to be calculated with respect to principal axes.
Therefore, if any attitude maneuver is to be executed to reorient the body axes, all the parameters are
calculated for the principal axes through frame transformation and then executed. The frame
transformation matrix or direction cosine matrix is,

𝟎. 𝟗𝟒𝟖𝟐𝟕𝟏𝟏𝟒𝟓 −𝟎. 𝟑𝟎𝟑𝟕𝟒𝟓𝟑𝟑𝟔 −𝟎. 𝟎𝟗𝟐𝟑𝟎𝟕𝟏𝟑𝟏

𝑫𝑪𝑴 = [𝟎. 𝟏𝟓𝟎𝟓𝟓𝟓𝟏𝟒𝟒 𝟎. 𝟏𝟕𝟒𝟐𝟗𝟓𝟗𝟓𝟖 𝟎. 𝟗𝟕𝟑𝟏𝟏𝟓𝟔𝟓𝟎 ]
𝟎. 𝟐𝟕𝟗𝟒𝟗𝟎𝟓𝟖𝟎 𝟎. 𝟗𝟑𝟔𝟔𝟕𝟒𝟖𝟎𝟓 −𝟎. 𝟐𝟏𝟏𝟎𝟏𝟎𝟐𝟓𝟎

(3.6)

Ellipsoid of Inertia
The moment of inertia of a rigid body represents the distribution of mass within a body with respect to
an axis of rotation. The moment of inertia varies with the axis of rotation. The variation is represented
by the following equation, where ax, ay, az are the direction cosines of the arbitrary axis.

𝒂𝟐𝒙 𝒂𝟐𝒚 𝒂𝟐𝒛

𝟏= 𝑰 + 𝑰 + 𝑰 = 𝑿𝟐 𝑰𝒙 + 𝒀𝟐 𝑰𝒚 + 𝒁𝟐 𝑰𝒛
𝑰ɛ 𝒙 𝑰ɛ 𝒚 𝑰ɛ 𝒛

(3.7)

The moment of inertia about any arbitrary axis can be calculated, and the rotational motion of the
spacecraft can be analyzed using the plot of the ellipsoid of inertia. The moments of inertia of SRMSAT
about its principal axes are given as
Ix = 6.2911 kg m2
Iy= 5.9162 kg m2
Iz = 4.6085 kg m2
(3.8)
The plot of an ellipsoid of inertia of SRMSAT-2 is given in Fig.3.10.

130
 Figure 3.10: Ellipsoid of Inertia of SRMSAT

Note: Due to mass ejection during orbital and attitude maneuvers, the moments of inertia change and
does not remain constant and thus, there is a need for a dynamic plant.

Angular Momentum
The angular momentum of a rigid body is the measure of torque impulses that is required to generate
its rotational motion. If the angular velocity vector is defined on the body frame, then the corresponding
angular momentum vector can be calculated by product of inertia tensor and angular velocity vector
through rules of matrix multiplication.

𝑰𝒙𝒙 −𝑰𝒙𝒚 −𝑰𝒙𝒛 𝝎𝒙

𝒉 = [−𝑰𝒚𝒙 𝑰𝒚𝒚 −𝑰𝒚𝒛 ] [𝝎𝒚 ] = [𝑰]𝝎
−𝑰𝒛𝒙 −𝑰𝒛𝒚 𝑰𝒛𝒛 𝝎𝒛

(3.9)

Rotational Kinetic Energy

Rotational kinetic energy is analogous to linear kinetic energy and together with the knowledge of
angular momentum, defines the complete rotational state of a rigid body. The Rotational kinetic energy
of a rigid body can be expressed in terms of inertia tensor and angular velocity through the following
mathematical relation.
𝐼𝑥𝑥 −𝐼𝑥𝑦 −𝐼𝑥𝑧 𝜔𝑥
1 1
𝑇𝑟𝑜𝑡 = [𝜔𝑥 𝜔𝑦 𝜔𝑧 ] [−𝐼𝑦𝑥 𝐼𝑦𝑦 −𝐼𝑦𝑧 ] [𝜔𝑦 ] = 𝜔𝑇 [𝐼]𝜔
2 2
−𝐼𝑧𝑥 −𝐼𝑧𝑦 𝐼𝑧𝑧 𝜔𝑧

(3.10)

131
Ellipsoid of Rotational Kinetic Energy and Angular Momentum
The angular momentum along with the rotational kinetic energy completely define the rotational
dynamic state of the spacecraft in space. All the instantaneous values of moments of inertia and angular
velocities are defined by the ellipsoids of rotational kinetic energy and angular momentum with known
rotational kinetic energies and angular momentums. All the possible values of ω are along a polhode (a
curve formed by the intersection of two ellipsoids). The following two equations are the ellipsoid
equations for angular momentum and rotational kinetic energy respectively.

𝜔𝑥2 𝜔𝑦2 𝜔𝑧2

+ + =1
ℎ 2 ℎ 2 ℎ 2
(𝐼 ) (𝐼 ) (𝐼 )
𝑥 𝑦 𝑧

𝜔𝑥2 𝜔𝑦2 𝜔𝑧2

2+ 2+ 2 =1
2𝑇 2𝑇
(√ 𝐼 ) 2𝑇 (√ 𝐼 )
𝑥 (√ 𝐼 ) 𝑧
𝑦

(3.11)
Note: These equations cannot be evaluated at this stage as the kinetic energy, and angular momentum
cannot be determined.

Euler's Moment Equations

The rotational dynamics of the rigid body is governed by Euler's moment equations. It is used in
examining the rotational motion of a body caused by applied moment.

𝑀𝑥 = 𝐼𝑥 ὠ𝑥 + 𝜔𝑦 𝜔𝑧 (𝐼𝑧 − 𝐼𝑦 )

𝑀𝑦 = 𝐼𝑦 ὠ𝑦 + 𝜔𝑥 𝜔𝑧 (𝐼𝑥 − 𝐼𝑧 )

𝑀𝑧 = 𝐼𝑧 ὠ𝑧 + 𝜔𝑥 𝜔𝑦 (𝐼𝑦 − 𝐼𝑥 )

(3.12)
For stability of rotation of asymmetric bodies about principal axes with no external moments acting on
the body, the following equations are taken. ἐ is a small disturbance about the ZB axis.

𝐼𝑥 ὠ𝑥 + 𝜔𝑦 𝜔𝑧 (𝐼𝑧 − 𝐼𝑦 ) = 0

𝐼𝑦 ὠ𝑦 + 𝜔𝑥 𝜔𝑧 (𝐼𝑥 − 𝐼𝑧 ) = 0

𝐼𝑥 ἐ + 𝜔𝑥 𝜔𝑦 (𝐼𝑦 − 𝐼𝑥 ) = 0

(3.13)
The first two equations are linear (assuming ωz is constant). Taking a second order derivative of first
equation with the second equation, and using a Laplace transform yields stability conditions as
𝐼𝑧 > 𝐼𝑥 , 𝐼𝑦 𝑜𝑟 𝐼𝑧 < 𝐼𝑥 , 𝐼𝑦

(3.14)
In other words, if the body is spinning about the axis of maximum or minimum moment of inertia then
the angular motion will be stable along these axes. Therefore, any rotation about any other instantaneous
axis of rotation can cause instability leading to nutation.

132
3-Axis Stabilized Equations
The behaviour of the spacecraft in space is completely defined by the attitude dynamics equations along
each body axis. These attitude dynamics equations are obtained from Euler's Moment Equation. The
total external moment acting on a spacecraft is equal to the inertial momentum change of the system.
The equations mentioned here were derived specifically for SRMSAT, by modelling all the disturbing
torque elements and control torque parameters based on the configuration of the actuators used .These
equations are valid in earth bound phase of the mission, and critical for attitude maintenance to execute
optimal orbital maneuvers

𝑇𝑑𝑥 + 𝑇𝑐𝑥 = 𝐼𝑥𝑥 Ф″ + 4𝜔𝑜2 (𝐼𝑦𝑦 − 𝐼𝑧𝑧 )Ф + 𝜔0 (𝐼𝑦𝑦 − 𝐼𝑧𝑧 − 𝐼𝑥𝑥 )𝜓 ′ + ḣ𝑤𝑥 − 𝜔0 ℎ𝑤𝑧 − 𝜓 ′ ℎ𝑤𝑦
− Ф𝜔0 ℎ𝑤𝑦 − 𝐼𝑥𝑦 𝜃 ″ − 𝐼𝑥𝑧 𝜓 ″ − 𝐼𝑥𝑧 𝜔02 𝜓 + 2𝐼𝑦𝑧 𝜔0 𝜃 ′
1 𝑏
+ 𝜌(𝑟𝑥𝑧 𝐶𝐷𝑥𝑧 𝑣𝑦2 𝐴𝑥𝑧 + 𝑟𝑥𝑦 𝐶𝐷𝑥𝑦 𝑣𝑧2 𝐴𝑥𝑦 ) + √2𝐹𝑇
2 2

𝑇𝑑𝑦 + 𝑇𝑐𝑦 = 𝐼𝑦𝑦 𝜃 ″ + 3𝜔02 (𝐼𝑥𝑥 − 𝐼𝑧𝑧 )𝜃 + ḣ𝑤𝑦 − 𝐼𝑥𝑦 (Ф″ − 2𝜔0 𝜓 ′ − 𝜔02 Ф) + 𝐼𝑦𝑧 (−𝜓 ″ − 2𝜔0 Ф′
1 𝑏
+ 𝜔02 𝜓) + 𝜌(𝑟𝑦𝑧 𝐶𝐷𝑦𝑧 𝑣𝑥2 𝐴𝑦𝑧 + 𝑟𝑥𝑦 𝐶𝐷𝑥𝑦 𝑣𝑧2 𝐴𝑥𝑦 ) + √2𝐹𝑇
2 2

𝑇𝑑𝑧 + 𝑇𝑐𝑧 = 𝐼𝑧 𝜓 ″ + 𝜔0 (𝐼𝑧𝑧 + 𝐼𝑥𝑥 − 𝐼𝑦𝑦 )Ф′ + 𝜔02 (𝐼𝑦𝑦 − 𝐼𝑥𝑥 )𝜓 + ḣ𝑤𝑧 + 𝜔0 ℎ𝑤𝑥 + Ф′ ℎ𝑤𝑦 − 𝜓𝜔0 ℎ𝑤𝑦
1
− 𝐼𝑦𝑧 𝜃 ″ − 𝐼𝑥𝑧 Ф″ − 2𝜔0 𝐼𝑥𝑦 𝜃 ′ − 𝜔02 𝐼𝑥𝑧 Ф + 𝜌(𝑟𝑦𝑧 𝐶𝐷𝑦𝑧 𝑣𝑥2 𝐴𝑦𝑧 + 𝑟𝑥𝑧 𝐶𝐷𝑥𝑧 𝑣𝑦2 𝐴𝑥𝑧 )
2
+ √2𝐹𝑇 𝑎
(3.15)
Where, rij = CPij - CG, is the torque arm for aerodynamic drag, and ‘a’ and ‘b’ are dimensions along XB
and ZB axes.

Primary disturbing external torques experienced by spacecraft during earth bound phase of the mission
are gravity gradients, aerodynamic torque, magnetic torque and solar radiation torque. In earth bound
mission phases of SRMSAT, major effects are produced by aerodynamic torque and gravity gradients
as the trajectory design specifies perigee at low earth orbit (200-400 km). These torques must be
considered to perform attitude maneuvers and to provide optimum attitude stability maintenance
required for communication, power generation and most importantly for the apogee raising maneuvers.
Nutational Motion
External disturbing torques and torque due to thruster misalignment tend to deviate the ZB axis to
deviate from the momentum vector h causing nutation. It forces the spin axis to deviate from the
nominal desired direction. Maintenance of the nutation angle to a minimal amount is significant.
Nutational angle (θ) is given by the following equation.
ℎ𝑥𝑦 𝐼𝑥 . 𝜔𝑥𝑦
tan(θ) = =
ℎ𝑧 𝐼𝑧 . 𝜔𝑧
(3.16)
In the presence of energy dissipation, a spinning body is in stable angular motion only if the spin is
about the major axis. Energy dissipation is related to nutation by

ℎ2 𝐼𝑧
𝑇̇ = cos(θ) sin(θ) ( − 1)θ̇
𝐼𝑧 𝐼𝑥

(3.17)
133
On analyzing the above equation, the following cases were studied.

Figure 3.11: Nutational motion for body with IZ>IX

Let the spin be along z-axis, and if IZ<IX, then the body is said to have nutational instability. This is
because during energy dissipation initial nutational angle will increase until the spin is transferred from
minor axis to the major axis and if IZ>IX , then the body is said to have nutational stability as during
energy dissipation initial nutational angle will reduce to zero.

Figure 3.12: Nutational motion for body with IZ<IX

Nutational stability at Δv
The angle of nutation must be maintained as low as possible to reduce Δv loss during boosts. Thruster
misalignment is the major reason for torques that cause nutation, but there is a considerable contribution
by other disturbing torques too. The spacecraft may even miss its target completely if boosts are
performed without appropriate reduction of nutation angle. The only torque element considered is
aerodynamic torque as periapses pass through the low earth orbit. The optimum way for reducing the
angle of nutation is increasing the angular velocity about the spin axis (minor axis). The average
nutation angle is inversely proportional to the square of the angular velocity ω Z (given that ZB is the
minor axis). It is given by
4𝑇𝑑
𝜑𝑎𝑣 =
𝐼
𝜋𝜔𝑧2 𝐼𝑧 ( 𝑧 − 1)
𝐼𝑥
(3.18)

134
 Figure 3.13: Average Nutation angle vs Angular Velocity

Denutation using RCS

Denutation control is based on the fact that if any component of angular momentum is not along the
major or minor axes of the body, the body nutates and therefore, the RCS are used to eliminate and
cancel out momentum along the other axes so as to execute the process of denutation.
Considering the torques provided by thrusters as equal, and ZB axis as major axis,
𝜋 𝜋
1 4 2 4
𝑇𝑎𝑣 = 𝜋 ∫ cos(𝛼) 𝑇𝑥𝑦 𝑑𝛼 = ∫ cos(𝛼) √2𝐹∆𝑥𝑑𝛼
𝜋 𝜋 −𝜋
2 −4 4

2 4
𝑇𝑎𝑣 = √2𝐹∆𝑥𝑠𝑖𝑛(𝛼) = 𝐹∆𝑥
𝜋 𝜋
(3.19)

Tav is the average torque provided by thrusters.

Time required for denutation is given by,
ℎ𝑥𝑦 (0) ℎ𝑥𝑦 (0) 2
𝑡𝐷𝐸𝑁𝑈 = = ( )
𝑇𝑎𝑣 4
(𝜋) 𝐹∆𝑥 𝑁𝑖

(3.20)

hxy is angular momentum in the X-Y plane.

Fuel consumption is given by the following equation.
𝜋ℎ𝑥𝑦 (0)
𝑚𝐷𝐸𝑁𝑈𝑇𝑂𝑇 = 𝑚𝐷𝐸𝑁𝑈 𝑁𝑖 =
2𝑔𝐼𝑠𝑝 ∆𝑥

(3.21)

135
Spin Stabilization vs. Three Axes Stabilization during Delta V Stage
In the preliminary design of SRMSAT-2, it was conceived that given the duration of finite burns to be
executed during apogee raising maneuvers, the satellite should be three-axis stabilized rather than spin
stabilized because for a spin-stabilized satellite, execution of a finite maneuver, requires precession of
the nominal thrust vector to align continuously with VNC. Due to the fact that in a spin-stabilized satellite
angular momentum vector is fixed in inertial space, and precession is inversely proportional to nominal
angular momentum vector, adjusting the precession using RCS will be less fuel efficient and would also
require the need to accommodate for ANC (active nutation control) as parasitic torques are induced due
to misalignment between thrust vector and center of gravity. On the other hand, if the body is three-axis
stabilized it can still align with VNC with more fuel efficiency and considerable accuracy to execute a
maneuver.

Attitude Prediction
Attitude prediction is carried out using attitude propagator model so as to reduce the computational
burden during maneuvers and state change as it requires continuous monitoring and sampling of sensor
data, and hence power consumption is reduced and system becomes more efficient.
Attitude prediction requires knowledge of current state and body angular rates to propagate the attitude.
𝑑𝑞 1 ′
= [Ω ]𝑞
𝑑𝑡 2
0 𝜔𝑧 −𝜔𝑦 𝜔𝑥
−𝜔 0 𝜔𝑥 𝜔𝑦
[Ω′ ] = [ 𝜔 𝑧 −𝜔𝑥 𝜔𝑧 ]
𝑦 0
−𝜔𝑥 −𝜔𝑦 −𝜔𝑧 0
(3.22)
‘q’ here is the present spacecraft quaternion and ωx, ωy and ωz are the angular rates.
The above differential equation is an initial value problem as initial q is known; it can be solved by
numerical integration by Runge-Kutta method.

Angular Rate Estimation

When angular rates are not measured directly through IMUs, they can be estimated through the
following algorithm, which requires knowledge of present state as well as their derivatives which can
be determined through sun sensor or star tracker.

𝑞𝑠1̇
𝑝 𝑞𝑠4 𝑞𝑠3 −𝑞𝑠2 −𝑞𝑠1
𝑞̇
[𝑞] = 2 [−𝑞𝑠3 𝑞𝑠4 𝑞𝑠1 −𝑞𝑠2 ] [ 𝑠2 ]
𝑟 𝑞𝑠2 −𝑞𝑠1 𝑞𝑠4 −𝑞𝑠3 𝑞𝑠3̇
𝑞𝑠4̇
(3.23)
Where, qsis the spacecraft quaternion.

136
 3.8 ATTITUDE CONTROL SYSTEM
Introduction
Autonomous attitude control for 3-axis stabilization in SRMSAT-2 is achieved by using four RCS
thrusters and four Reaction wheels. The primary control algorithm used for attitude control is the
quaternion error command law controller which is directly implemented to derive the torque required
by the reaction wheels while a Pulse-Width Pulse-Frequency (PWPF) modulator is used to convert the
continuous torque required into a pulsed output which is used to control the RCS thrusters.

Quaternion Error Command Law

The quaternion's basic definition is a consequence of the properties of the direction cosine matrix [A].
It is shown by linear algebra that a real orthogonal 3 x 3 matrix has at least one eigenvector with an
eigenvalue of unity. This means that any attitude transformation in space by consecutive rotations about
the three orthogonal unit vectors of the coordinate system can be achieved by a single rotation about
the eigenvector with unity eigenvalue. Refer Appendix.
There exists an equivalent quaternion error vector that expresses the attitude error between the satellite's
current quaternion and the targeted quaternion.
In quaternion notation,
𝑞𝑇4 𝑞𝑇3 −𝑞𝑇2 𝑞𝑇1 −𝑞𝑠1
−𝑞𝑇3 𝑞𝑇4 𝑞𝑇1 𝑞𝑇2 −𝑞𝑠2
𝑞𝑠−1 𝑞𝑇 = 𝑞𝐸 = [ 𝑞 −𝑞 𝑞 𝑞 ] [−𝑞 ]
𝑇2 𝑇1 𝑇4 𝑇3 𝑠3
−𝑞𝑇1−𝑞𝑇2 −𝑞𝑇3 𝑞𝑇4 𝑞𝑠4
(3.24)
Where 𝑞𝐸 , 𝑞𝑇 and 𝑞𝑆 are the error, target, and spacecraft quaternions respectively.
We obtain the following attitude control laws:
𝑇𝑐𝑥 = 2𝐾𝑥 𝑞1𝐸 𝑞4𝐸 + 𝐾𝑥𝑑 𝑝,
𝑇𝑐𝑦 = 2𝐾𝑦 𝑞2𝐸 𝑞4𝐸 + 𝐾𝑦𝑑 𝑞,

𝑇𝑐𝑧 = 2𝐾𝑧 𝑞3𝐸 𝑞4𝐸 + 𝐾𝑧𝑑 𝑟.

(3.25)
Control Using RCS thrusters
RCS thrusters are controllers which are operated in pulses (on or off states). These thrusters can provide
a fixed torque value when used in attitude control. RCS thrusters are non-linear and non-continuous
attitude controllers.
The Controller or the Reaction thruster control system converts the continuous desired torque signal to
an on/off signal and is then used to control the spacecraft attitude. A simple such model using a PD
controller is given in Fig 3.14.

137
 Figure 3.14: Simple thruster control model

The commonly used on/off controllers are:

 Bang-Bang Controller
 Bang-Bang Controller with Dead zone
 Time-Optimal Bang-Bang Controller
 Schmitt Trigger
 Pulse Modulators
 Pulse-Width Modulators
 Pulse-Frequency Modulators
 Pseudo Rate Modulators
 Pulse-Width Pulse-Frequency(PWPF) Modulator

Pulse-Width Pulse-Frequency (PWPF) Modulator:

A PWPF modulator consists of a first order filter and a Schmitt trigger along with a feedback loop as
depicted in Fig3.15(a).

Figure 3.15: (a) PWPF modulator (b) Schmitt trigger.

138
The advantages that make the PWPF Modulator superior to others are:

 PWPF modulator has an approximate linear behavior, known as ‘pseudo-linear behavior’. It

can be approximated as a linear actuator as it varies both pulse width and pulse frequency with
time. Such an operation allows for a response that is analogous to that of a linear actuator such
as a momentum wheel.
 Since the controller has the option to vary the pulse size or width and its timing, the PWPF
modulator is more fuel efficient than the other controllers mentioned above. Bang-bang
controller and other controllers which are unable to produce a linear output often excite flexible
modes causing unwanted structural vibrations, which then couple back into rigid body motion
which results in a limit cycle with continuous thruster switching and therefore fuel wastage.
The PWPF controller, besides causing fewer vibrations, can tailor its response to reduce the
number of thruster firings to increase the thruster life.
 The PWPF controller offers flexibility in design and operation. PWPF modulator components
can be tuned to optimize performance for any given plant configuration, for example, when
spacecraft moments of inertia change over the course of a mission. Since the modulator
parameters are totally independent of the spacecraft parameters, no ‘prior’ knowledge of the
spacecraft dynamics is required for system analysis and performance determination, thus
providing scope for implementing adaptive control schemes.

PWPF Modulator Analysis:

The tuning of the PWPF modulator is carried out by varying the modulator parameters. The PWPF
modulator parameters are: Km, Tm, Uon, and Uoff. Km is the filter gain, and Tm is the filter time constant.
Uon is the Schmitt trigger on-value, and Uoff is the Schmitt trigger off value. Uon – Uoff = h. Kpm is the
pre-modulator gain which is not exclusively a PWPF modulator parameter and is used to amplify the
input to the required magnitude.
The system parameters can be identified in Fig4.2.
Static Behaviour analysis:
To determine the basic behavior and response pattern of a PWPF modulator, a constant input signal is
fed to the modulator, thus providing a static behaviour analysis of the modulator.
The filter output transfer function is given by:
𝐾𝑚 𝑇𝑚 𝑓(0)
𝑓(𝑠) = 𝑒(𝑠) +
(𝑇𝑚 𝑠 + 1) 𝑇𝑚 𝑠 + 1
(3.26)
Where error signal e(s) = r(s) – u(s) and r(s) = C/s and u(s) = U/s since r(s) and u(s) are both step
functions and f(0) is the initial condition of the filter.
The system equation converted in the time domain is given by:

f(t) = Km (C − U)(1 − e−t⁄Tm ) + f(0)e−t⁄Tm

(3.27)
This is used to determine the important system variables mentioned below.

139
As shown in the figure, the PWPF modulator response pattern for a constant input is simulated. The
filter output starts increasing when the step input occurs. The filter output gradually increases towards
the Schmitt trigger on-value, Uon1. When the on-value is reached, the Schmitt trigger sets its output to
the prescribed magnitude U. This immediately leads to a decrease in the filter input value due to the
negative feedback loop and the filter output then decreases until Schmitt trigger off-value, Uoff is
reached. This gives rise to the characteristic hysteresis effect of the trigger. When the Schmitt trigger
off-value is reached, the modulator output is set to zero. The period of time in which the modulator has
a non-zero output is called the thruster on-time and is denoted by Ton. The thruster off-time is denoted
by Toff and is the time when the modulator has zero output.

Figure 3.16: PWPF modulator response

The trigger-on time and off-time are given by setting f(0) = Uon for measuring Ton time and f(0) =
Uoff in eqn.2.2. On rearranging we get the following equations:
h
Ton = −Tm ln (1 − )
Uon − Km (C − U)
(3.28)
h
Toff = −Tm ln ( 1 − )
Km C − Uoff
(3.29)
The duty cycle is a measure of how the modulator responds to an input. It can be used to determine how
well the modulator output follows the input.
Ton
Duty CycleDC = Ton+Toff

(3.30)
The input signal needs to have a certain minimum magnitude to activate the Schmitt trigger. This signal
magnitude is called the internal deadzone. We know that KmC ≥ Uon, the internal deadzoneCdz can be
given by

1
Uon, Uoff and h are parameters which determine the Schmitt trigger characteristics.

140
 𝑈𝑜𝑛
𝐶𝑑𝑧 =
𝐾𝑚
(3.31)
This internal deadzone directly determines the pointing accuracy of the spacecraft, since a command
torque signal with a magnitude smaller than Cdz won’t activate the spacecraft thrusters.
When the input value gets high enough, the modulator will enter a saturation region where the thrusters
will be on all the time, which indicates that no more torque can be applied along that axis. The saturation
level Csat is
𝑈𝑜𝑓𝑓
𝐶𝑠𝑎𝑡 = 𝑈 +
𝐾𝑚
(3.32)
For a given set of system variables, the minimum pulse width for the modulator is given by,
h
Minimum Pulse Width,∆ = −Tm ln ( 1 − Km)

(3.33)
The PWPF modulator is required to operate in an optimal region. The optimal region of the modulator
is defined as the region where the modulator behaviour is as close to linear behaviour as possible, and
at the same time, the fuel consumption and the thruster activity is the lowest. Given the highly nonlinear
behaviour of the modulator, an analytic approach is difficult and impractical. Instead, system
simulations are carried out to determine the range of the modulator parameters within which the PWPF
modulator exhibits optimal performance.
All simulations are carried out using MATLAB and Simulink. A standard setup with a range of
MATLAB files is used to create the simulation plots which are used to determine the behaviour for
varying the modulator parameters.

Figure 3.17: Simulink model of PWPF Modulator

In order to maintain the functionality of the Schmitt trigger, the hysteresis value should not increase
twice the 𝑈𝑜𝑛 value, i.e. h < 2𝑈𝑜𝑛 .
The graphs shown below are created by varying different modulator parameters in a certain range.
When the value is not varied, it is set to a fixed value as described in the table below.

141
Parameter Range Fixed Value
Km 2.5-7.5 4.5
Tm 0.1-1 0.15
Uon 0.1-1 0.45
H 0.2-2 Uon 0.3 Uon
Table 3.6: Range and constant values for PWPF modulator parameters

142
(a) (b)
Figure 3.18: Duty Cycle Vs Constant input C plots for (a) Zero hysteresis, Uon = Uoff, (b) Hysteresis h = 2Uon,
Uon = -Uoff and (c) Varying Km.b

(c)

143
Optimizing with respect to pseudo-linear behavior
On evaluating linear behavior for Schmitt trigger parameters, it is discovered that the output follows
the input with a constant offset for h = 0and as h increases this offset reduces. When Uon approaches
zero, the output follows the input with almost no offset, though small Uon values give rise to bad noise
and disturbance rejection which is undesirable. The figure shows the duty cycle behavior for a range of
Uon and Uoff values.
Fig3.18. Duty Cycle Vs Constant input C plots for (a) Zero hysteresis, Uon = Uoff, (b) Hysteresis h =
2Uon, Uon = -Uoff and (c) Varying Km.
On evaluating linearity for filter parameters, it is observed that Tm does not have any impact on the
linearity of the modulator. Km, on the other hand, has a profound impact on linearity as shown in the
figure5(c). Km < 2.5 gives undesirable non-linear behavior, while Km >7.5 produces a little increase in
linear operation region. Therefore, Km should lie in the interval 2.5 < Km <7.5.

Optimizing with respect to Thruster activity

Thruster valves are subject to wear every time they are opened and closed. In order to increase thruster
life, the thruster firings should be minimized. For filter parameters, it is evident that higher Tm and
lower Km gives less thruster activity. Due to the damping effect of Km, thruster activity disappears
almost entirely for very low Km. On increasing Km, the region with thruster activity increases in area,
therefore low Km values are desired.
Fig 3.18 clearly shows that low hysteresis (h <0.2) values give very high thruster activity and are to be
avoided. Special behavior is apparent when h = 2Uon (Uon = -Uoff), the number of thruster firings is
very high, because here, the state thruster switches directly from U to –U and vice versa without a
possibility to turn itself off.

Optimizing with respect to fuel consumption

It is obvious that thruster activity and fuel consumption are closely related. Although, two parameter
configurations that give same fuel consumption does not necessarily exhibit the same thruster activity.
Fig 3.18 evaluates the fuel consumption for filter parameters. It can be observed that very low Tm
(<0.05) values result in dramatic increase in fuel consumption and should be avoided. This demand has
already been overridden in thruster activity demands.
Km should be as small as possible to reduce fuel consumption as shown in the figure 3.18. Tm should
meet a minimum requirement as very low values lead to very high fuel consumption and higher Tm
values lead to higher number of thruster firings.
Fuel consumption is maximum when h = 2Uon because the thrusters remain on as mentioned above.
As Uon and h are reduced, fuel consumed reduces, which is depicted in Fig 3.19.

144
 Figure 3.19(a): Thruster
activity for varying
modulator parameters

Figure 3.19(b): Fuel

consumption for
varying modulator
parameters.

145
Optimization Summary
The parameter limits have been given in table 3.7. The summary of each parameter impact for all
optimization criteria is given below with a conclusion range.
Filter Gain Km:
With respect to thruster firings and fuel consumption, Km should be as small as possible (especially for
small Tm). Noise and disturbance can be rejected by providing a limit on the deadzone. High Km values
(greater than 2.5) produce a linear behavior.
Filter Time Constant Tm:
With respect to fuel consumption, Tm should meet a minimum requirement, and should be as large as
possible with respect to Thruster firings. Tm should be larger than the minimum pulse width, ∆. Tm has
no impact on linear behavior.
Schmitt Trigger On-Value Uon:
With respect to linear behavior and thruster firings, Uon should be small. Uon should be large with
respect to fuel usage. The Schmitt Trigger’s characteristic requirement is that Uon value should be at
least half the hysteresis width h.
Schmitt Trigger Hysteresis h:
With respect to linear behavior and fuel consumption, h should be small and with respect to thruster
firings, h should be as large as possible.
The slower the spacecraft can afford to move, the less is the fuel consumed, and fewer thruster firings
will occur.
To conclude, the PWPF parameters should be in the following regions

Parameter Range Selected Value

Km 2.5-7.5 4.25
Tm 0.1-1 0.25
Uon 0.1-1 0.45
H 0.2-2Uon 0.3
Table 3.7:Concluded range for PWPF modulator parameters

Control Using Reaction Wheels

Inside a spacecraft, on accelerating a symmetrical body about its axis of rotation, an angular torque is
produced. The rotating body may have an initial constant momentumhw = [hwx hwy hwz ]T. Since this
momentum is internal to the spacecraft, its increase does not change the overall momentum of the
system, but instead, the momentum change is transferred (with negative sign) to the spacecraft. This is
the principle of conservation of angular momentum.
The reaction wheel can be mounted in the satellite with its rotational axis relative to the satellite's axis
frame. The momentum vector of all the momentum exchange devices inside the satellite body can be
expressed with reference to the axes of the spacecraft body frame ashw = [hwx hwy hwz ]T . For
controlling the attitude in space, at least three reaction wheels are required.

146
Model of the Reaction Wheels
The required angular torques can be achieved by accelerating the rotor of electrical motors whose axes
of rotation are aligned with the body axes X B , YB , ZB . A complete dynamic model for the electrical
motor to be used as the torque controller is shown in Fig. 3.20.

Figure 3.20: Simple model for an electric motor used as a reaction wheel

In Figure. 3.20, V is the input voltage to the electrical motor, R M is the electrical resistance of the motor
armature, and K M is the torque coefficient of the motor. The term IW denotes the overall moment of
inertia of the rotor, the task of which is to produce the desired torque, and IS is the moment of inertia of
the satellite. Finally, ωREL is the angular velocity of the rotating part of the motor relative to the stator
and (equivalently) to the satellite body, since the stator is fixed to the body of the spacecraft. The
primary task of the electrical motor is to provide the necessary angular torque to the satellite. Assume
there are no external disturbances Td and no inertial control torques acting on the satellite. With this
assumption, according to Euler's moment equation of angular motion we have

ḣs + ḣw = 0
(3.34)
This means that, in order to apply a torque on the body about some axis, a torque in the opposite
direction must be produced by the rotor of the electrical motor. Thus, ḣs = −ḣw.

Control Loop for Linear Attitude Maneuvers

The reaction wheels are operated in the torque command mode, where the required torque for each body
axis is determined by the quaternion command law and is then directly sent to the reaction wheels based
on the configuration.

147
 Figure 3.21: Reaction Wheel model used in torque control mode.

Fig. 3.21 shows the reaction wheel model in the torque command mode. The current and voltage limits
show the maximum current and voltage that can be applied to the reaction wheel. In order to achieve
an immediate torque response for a command, there is a feedback path from the motor current which is
directly proportional to the torque provided by the electrical motor.

Analysis
During analysis, the coulomb dry friction is omitted to obtain a linear transfer function of the complete
dynamic model of the reaction wheel (including the satellite dynamical model), which is given below.
𝐾
ℎ̇𝑤 𝐼𝑤 𝑤̇𝑤 𝑠( 𝑚 )
𝑅𝑚
= =
𝑉 𝑉 1 1 𝐾𝑣 𝐾𝑚
𝑠 + (𝐼 + 𝐼 ) ( 𝑅 + 𝐵)
𝑤 𝑠 𝑀

(3.35)
The equation indicates that a step in the input voltage does not produce a pure angular torque due to the
time constant in the denominator and the differentiator in the numerator. Assuming B approaches zero
and 𝐼𝑤 ≪ 𝐼𝑠 , equation becomes,
𝐼𝑤
ℎ̇𝑤 𝐾𝑉
=𝑠
𝑉 𝐼 𝑅
1 + 𝑠 𝐾𝑤 𝐾𝑀
𝑉 𝑀

(3.36)
When used in torque command mode, the transfer function becomes
𝐾
ℎ̇𝑤 𝑠𝑅𝑀
=
𝑇𝑐 𝐾 𝐾𝑉 𝐾𝑀
1 + 𝑠𝑅 (1 + 𝐾𝐼 )
𝑚 𝑤

(3.37)

Taking K such that,𝐾 ≫ 𝐾𝑉 𝐾𝑀 /𝐼𝑀 , then,

148
 ℎ̇𝑤 1
=
𝑇𝑐 𝑅
1 + 𝑠( 𝑀⁄𝐾 )

(3.38)
This is the basic equation that converts an electric motor into a reaction wheel, which along with added
electronics becomes a reaction wheel assembly.
The time constant in the denominator, which is in the order of milliseconds, can be ignored in order to
obtain a linear model of the reaction wheels which along with the satellite model is used to carry out
the analysis.
The Simulink models for the Reaction wheels used in torque command mode in a tetrahedral
configuration is given in figure 3.26.

Momentum Accumulation and Its Dumping

One drawback of a momentum exchange device is that it cannot independently remove the angular
momentum that accumulates, owing to external disturbances, in the satellite system. As we know,
according to Euler's moment equations, any external torque disturbance acting on the body augments
the angular momentum of the whole spacecraft system. With no active attitude control, the body
accumulates an angular velocity as the angular momentum of the spacecraft increases, changing the
attitude of the spacecraft. However, if reaction wheels are used to stabilize the spacecraft attitude, then
the accumulated angular momentum will be transferred to the wheels. With external harmonic
disturbances, the stored momentum will also be harmonic; as long as its level stays well inside the
momentum saturation limits of the wheels, no control problem arises. However, with constant external
disturbances, the momentum of the wheels will increase without limit, thus saturating them and
precluding their ability to provide the necessary control torques. The unwanted accumulated momentum
in the wheel must be removed from the momentum exchange device, a process called dumping of the
momentum.

149
Component selection
Actuator selection:
Actuator selection for SRMSAT is based on the following criteria:

• Near zero slew rates QUANTIFICATION OF

during primary DISTURBANCE
payload operation ENVIRONMENT
•Mass constraints
•Payload pointing •Redundancy Level
accuracy •Worst case torques due
to the follwing forces •Power budget
• Pointing accuracy of
TTC •Aerodynamic drag
•Gravity gradient
•Magnetic moment
SUBSYSTEM LEVEL •Solar radiation HARWARE SIZING
MISSION pressure AND SELECTION
REQUIREMENTS

Figure 3.22: The above-mentioned flow chart represents the criteria adopted for actuator selection.

Accounting for mission requirements:

The first step before selecting actuators is to analyze the mission profile requirements, such as slew
rates, payload pointing accuracy, TTC pointing requirements, etc. Once the mission requirements are
analyzed, the attitude control methods are decided, and the required components for such a control
system are analyzed, sized and selected.
SRMSAT is three-axis stabilized throughout all the phases of the mission. The following are the key
system level mission requirements.

Payload Pointing Accuracy TBD

TTC Pointing Requirements TBD
Slew Rates 1 degree/sec

Quantification of disturbance environment:

The worst case for disturbance torques throughout the trajectory of SRMSAT is at the periapsis which
is about 240 km above the Earth’s surface during phasing orbits. The disturbing torques are modeled
for the worst case analysis at this altitude.
At this altitude, the primary disturbance torques are due to aerodynamic drag and gravity gradients
while the other minor factors are magnetic torques and solar radiation pressure. The disturbance torques
and their respective magnitudes are summarized in table 1.2

150
 TORQUE MAGNITUDE
Aerodynamic 3.166 x 10-2 N-m
Gravity gradients 2.789 x 10-6 N-m
Solar Radiation Pressure 6.647 x 10-6 N-m
Magnetic 1.367 x 10-5 N-m

The overall worst case the torque can be estimated to be around 0.031683 N-m

Hardware
Sizing, selection and configuration:
Taking the subsystem level mission requirements and the above estimated disturbance torques into
consideration, the selected actuators should be capable of overcoming the disturbance torques while
also achieving the required accuracies.
Reaction wheels:
A reaction wheel is a flywheel attached to an electric motor and is used to provide small rotations in a
spacecraft. When the speed of rotation of the wheel is changed, the spacecraft counter rotates
proportionally through conservation of angular momentum. It is not capable of producing translational
motion.
Reaction Wheel Assembly:
A tetrahedral configuration is selected for the arrangement of the reaction wheels. This configuration
allows for one redundant reaction wheel. This configuration is provided by the following Distribution
matrix where all the reaction wheels are aligned to each other at an angle of 109.47°.

Figure 3.23: Reaction Wheel Configuration

√3 √3 √3 √3
− −
3 3 3 3
𝐿= √3 √3 √3 √3
− −
3 3 3 3
√3 √3 √3 √3
[3 − − ]
3 3 3 (3.39)

151
 PARAMETER PROCEDURE RESULTS
DISTURBANCE Worst case anticipated torque magnitude (with In SRMSAT, reaction wheels will
REJECTION gain margin) not be used for disturbance rejection
𝑇𝑅𝑊 = 𝑇𝐷 × (𝐺𝑎𝑖𝑛 𝑚𝑎𝑟𝑔𝑖𝑛) due to power and mass constraints
in wheel selection. In worst cases
TD = 3.166 x 10-2 N-m, thus, the
operation will be carried out using
RCS.

SLEW TORQUE For maximum acceleration slews, torque For obtaining 1 degree/sec slew rate
FOR required is for operation mode switching, TRW
REACTION 𝐼 = 7.32 x 10-3 N-m
𝑇𝑅𝑊 = 4𝜃 2
WHEELS 𝑡

MOMENTUM To estimate angular momentum, The estimated hmax required could

STORAGE Worst case secular torques were integrated for
not be accommodated into a micro
IN REACTION the whole orbit and cyclic torque such as gravity
reaction wheel and therefore based
WHEEL 1
on availability, maximum angular
gradients the max disturbance accumulates in 4th
momentum storage capabilities that
of the orbit.
𝑜𝑟𝑏𝑖𝑡𝑎𝑙 𝑝𝑒𝑟𝑖𝑜𝑑 fit the profile of a micro reaction
ℎ𝑚𝑎𝑥 = (𝑇𝑐𝑦𝑐𝑙𝑖𝑐 × 4
× 0.707) +
wheel, a reaction wheel was chosen,
𝑜𝑟𝑏𝑖𝑡𝑎𝑙𝑝𝑒𝑟𝑖𝑜𝑑 and configured in a tetrahedral
(𝑇𝑠𝑒𝑐𝑢𝑙𝑎𝑟 × ).
1 geometry to maximize angular
Aerodynamic torque in worst case scenario was momentum as well as to provide
integrated for orbital duration in LEO (18 mins redundancy. Selected wheel
approx) to find hmax which is about 3.41 N-ms provides 0.485 N-ms about an axis
and thus in worst case scenarios it
would require momentum dumping
every 15.32 seconds which can be
executed in 1.14 seconds using
RCS.
POWER AND POWER: 20 W for RCS and 6 W for reaction
MASS BUDGET wheels
CONSTRAINTS MASS: 4kg (total)

Table 3.8: reaction wheel sizing

Conclusion
The final selection criteria for reaction wheels were slew maneuvers and pointing accuracy. The selected
reaction wheel is the 10 SP-M manufactured by SSTL. The datasheet for the reaction wheel is given
below.

152
PERFORMANCE

Design Life 7.5 Years(LEO)

Angular Momentum 0.42 Nms
Max Speed(+/-) 5000 rpm
Speed accuracy <0.1 rpm rms
Max Torque 0.011 Nm
ENVIRONMENTAL CONDITIONS
Operating Temperature -20 to 50 ˚C
Survival Temperature -30 to 60 ˚C
Radiation 5 krad
Random Vibration (qualification) 18 grms (all axis)
First Mode >400 Hz

MECHANICAL
Mass 0.96 kg
Volume ∅109x101 mm
Moment of Inertia (wheel) 0.0008 kg m2
Static Unbalance <0.1 g cm
Dynamic Unbalance <0.2 g cm2
Lubrication Dry

ELECTRICAL AND ELECTRICAL CONTROL

Integrated Electronics Yes
Control Mode (up to 5Hz) Output Speed
Power (standby) ~0.65W
Power (5000rpm) ~2.8W
Power (maximum torque) ~10W
Supply voltage(s) 22-34V DC
Data Interface CAN bus/ RS422
Telemetry data examples Speed, Motor current, Electronics Temperature

153
Reaction control system (RCS) thrusters
The motor of the reaction wheel has a maximum limit of torque that can be provided. Therefore, they
cannot be used for tumbling, disturbance rejection in the worst case scenarios and during maneuvers.
RCS Thrusters provides thrust by the expulsion of particles with high velocities. They can be used to
provide both translational and rotational motion. SRMSAT has one primary propulsion thrusters for
translation and four reaction control thrusters for attitude stabilization, maneuvers, and attitude
maintenance.

Thrust Direction RCS Thrusters

Figure 3.24: reaction control system

RCS Thrusters Assembly

The RCS thrusters are placed on the bottom plate at 45 degrees to the edges. These thrusters are arranged
in such a way that 3-axis control can be achieved using only four RCS thrusters.
Sizing and selection

PARAMETER PROCEDURE RESULT

SIZING THRUSTERS FOR Worst case disturbance torque is Thrust required to counteract
WORST CASE estimated, based on the disturbing torques is 0.05 N
DISTURBANCE configuration required force for
REJECTION thrusters is estimated

SIZING FORCE LEVEL 𝐹 = ℎ , h=0.485 N-ms In order to execute momentum

𝑙𝑡 dumping within 1 second, with
FOR MOMENTUM
,l=0.3m,t=1.25sec
DUMPING the RCS configuration designed
for SRMSAT, Force required
from each thruster is 0.91 N

154
Selected thrusters specifications
Performance Characteristics

Characteristics 0.2lbf(1N)
Thrust Redundant Seat
Max. operating pressure, MEOP[S.I.(bar)] 400(27.6)
Proof Pressure[S.I.(bar)] 1500(103.4)
Burst Pressure[S.I.(bar)] 2650(182.8)
Operating voltage range[Vdc] 24 to 37
Maximum open response time[msec] 10
Maximum close response time[msec] 10
Power consumption[watts] 8.69 at 28 Vdc, 45F
Cycle life[cycles] 1,000,000
Weight[lbm(gram)]excluding leadwires 0.48(218)
Operating temperature range[°F(°C)] 40 to 300(4.4 to 149)
Table 3.9: Thrusters Specifications

155
 3.9 SIMULATIONS
RCS Simulations
3-axis stabilization using the PWPF modulator is simulated using MATLAB and Simulink. The model
for this simulation is shown in fig 3.25. Quaternion error command controller is employed for each axis
to derive the torque commands for each axis. The PWPF Modulator modulates this signal and sends the
pulses to the spacecraft plant. The evolution of the error quaternion and spacecraft quaternion for three
different target axes is shown in fig 3.25.

156
Figure 3.25: Simulink model for PWPF controlled 3-axis stabilized spacecraft.

157
 (a)

(b)

158
 (c)
Figure 3.25: Error quaternion and spacecraft quaternion evolution for (a)∅ = 10° 𝜃 = 7° 𝜑 = 6°, (b) ∅ = 90° 𝜃 =
−80° 𝜑 = 75° and (c) ∅ = −45° 𝜃 = 120° 𝜑 = 30°

Figure Command Response Time (s)

11(a) ∅ = 10° 𝜃 = 7° 𝜑 = 6° 9
11(b) ∅ = 90° 𝜃 = −80° 𝜑 = 75° 17
11(c) ∅ = −45° 𝜃 = 120° 𝜑 = 30° 24
Table 3.20: Response Times for various Commands

159
Simulations Reaction Wheels
Four reaction wheels in a tetrahedral configuration are modeled using MATLAB and Simulink to
simulate 3-axis stabilization using a quaternion error command law. The distribution of torque
commands among the four reaction wheels is given by a switching matrix according to the following
formula,
𝑇1 0.866 0.866 0.866 −0.866 𝑇𝑐𝑥
𝑇2 0 0 0.866 −0.866 𝑇𝑐𝑦
[ ]= [ ][ ]
𝑇3 0 0.866 0 −0.866 𝑇𝑐𝑧
𝑇4 0.866 0 0 −0.866 0
The evolution of the error quaternion and spacecraft quaternion for three different target angles
simulated using the model shown in fig 3.25.
.

160
Figure 3.26: Simulink model Reaction Wheels controlled 3-axis stabilized spacecraft

161
 (a)

(b)

162
 (c)
Fig 3.27: Error quaternion and spacecraft quaternion evolution for (a) ∅ = 10° 𝜃 = 7° 𝜑 = 6°, (b) ∅ = 90° 𝜃 = −80° 𝜑 =
75° and (c) ∅ = −45° 𝜃 = 120° 𝜑 = 30°

Figure Command Response Time (s)

15(a) ∅ = 10° 𝜃 = 7° 𝜑 = 6° 24
15(b) ∅ = 90° 𝜃 = −80° 𝜑 = 75° 41.5
15(c) ∅ = −45° 𝜃 = 120° 𝜑 = 30° 49
Table 3.11: Response Times for various Commands

163
Redundancy/Failure Analysis
Simulations are carried out to analyze the system response in case of a failure of any one of the Reaction
Wheels. Simulations are carried out for command angles: ∅ = −45° 𝜃 = 120° 𝜑 = 30°

(a)

(b)

164
 (c)

(d)
Figure 3.28: Error quaternion and spacecraft quaternion evolution for command angles: ∅ = −45° 𝜃 = 120° 𝜑 = 30° due
to failure of each Reaction Wheel.

Figure Response Time without failure Response Time (s)

16(a) 41.5 48
16(b) 41.5 53
16(c) 41.5 80
16(d) 41.5 48
Table 3.12: Response Times comparison for failure of each Reaction Wheel.

165
 1. TELEMETRY, TRACKING & COMMAND SUBSYSTEM

1.1 OVERVIEW
The Telemetry, Tracking and Command (TT&C) or communication subsystem provides the interface between
the spacecraft and ground systems. Payload mission data and spacecraft housekeeping data pass from the
spacecraft through this system to the operations center. Operator commands are also passed to the spacecraft
through this subsystem to control the spacecraft and to operate the payload. The design must be able to pass
the data reliably for all the spacecraft’s operating modes. As part of carrier tracking, most spacecraft TT&C
subsystems generate a downlink RF signal that is phase coherent to the uplink signal. This two-way coherent
mode allows measuring Doppler shift from which range rate or line of sight velocity between the spacecraft
and tracking antenna is calculated. This knowledge allows operators to scan fewer frequencies and thus,
acquire spacecraft more quickly. This Doppler ranging operation can also be used for navigation. The farthest
point for this mission’s trajectory design being 1.5 million km, the TT&C subsystem needs to be designed to
be efficient, and the ground stations must be sufficiently powerful for TT&C operations. ISRO’s state of the
art Deep Space Network (IDSN) is necessary to support this mission and the support is aspired. The IDSN,
being a critical component of the mission, forms a basis for all of the SRMSAT – 2’s TT&C subsystem
design. The system consists of one High Gain Antenna for payload and housekeeping data transmission; and
two Low Gain Antennas for omnidirectional coverage, used for command and ranging. The subsystem is S-
Band based design with CCSDS compliance for easy interface with existing technologies and ground stations.
The On-board transponder is configurable in-orbit and undergoes different changes in configuration
throughout the mission as necessary; the changes include a change in data rate and change in transmitted
power.

1.2 FUNCTIONS
 Carrier Tracking – Lock on to ground station
 Command Reception and Detection – Receive uplink signal and process it
 Telemetry Modulation and Transmission – Accept data from the On-board computer, process it, and
transmit it
 Ranging – Receive and Transmit signals to determine the spacecraft’s position
 Subsystem Operations – Provide health and status telemetry to the On-board computer, perform
mission sequence operations as per stored software sequence, Select Omni-antenna for
communications when spacecraft attitude is lost, detect faults and recover communications using
stored software sequence.

1.3 REQUIREMENTS AND CONSTRAINTS

According to CCSDS Recommended Standard 401.0-B (February 2015), any mission with a distance less
than 2 × 106 𝑘𝑚 is considered to be a Near-Earth (Category – A) mission. Therefore, The SRMSAT – 2 is
classified as a Near – Earth mission and the planning of the mission will hence be in accordance with the
recommendations for Category – A based system.

Data Rate and Volume

For the preliminary analysis, 100 kb was selected for telecommands (including coding overhead). The
maximum uplink data is highly unlikely to take place over mission duration. So a value between typical data
and maximum data is a more suitable metric for analysis.

166
 Data Value (bits) Data Value (bits)
Tele-commands 160 × 103 Payload Data 26 × 107
Software Patches 2 × 106 Engineering Data 8 × 103
Margin (@25%) 5.4 × 105 Margin (@25%) 6.5 × 107
Max Uplink Data 𝟐. 𝟕 × 𝟏𝟎𝟔 Total 𝟑𝟑. 𝟔 × 𝟏𝟎𝟕
Typical Uplink Data 𝟏. 𝟒 × 𝟏𝟎𝟔 Table 1.2: Downlink Data Volume Estimate (Including Coding
Table 1.1: Uplink Data Volume Estimate (Including overhead)
Coding overhead)

In mission critical phases, the visibility time for uplink might be

as low as 18 minutes (taken as 10 mins in the analysis to allow Data Time Rate
for margin) whereas, for lunar orbit, the visibility time would Max.
be considerably large for concern. Therefore, the uplink data 4500 bps
Uplink 10 mins
(4.5 kbps)
rate is fairly straight-forward and is set to 4 Kbps, which is Data
close to max uplink data rate value necessary. A further Typical
2333 bps
Uplink 10 mins
discussion of downlink data rate will follow after looking at (2.3 kbps)
Data
other constraints and parameters. Thus, assuming a single data Table 1.3: Uplink Data Rate
link for mission and engineering data, the downlink data rate is
tentatively fixed to 0.1 Mbps. Based on this, the time to
effectively communicate downlink data over is 56 minutes. In addition to this, there is an omnidirectional
antenna always set to receive telecommands, and its data rate is variable throughout the mission with it being
80 bps in final phasing orbit, 1 Kbps in lunar orbit and 4 Kbps in the first few phasing orbits.

Frequency and Emission Constraints

The frequency spectrum of choice for near-earth missions are S-band for uplink, S-Band for downlink with
Bandwidth up to 6 MHz, X-band for downlink with BW up to 400 MHz and Ka-band for downlink with BW >
400 MHz

Considering the low Bandwidth and low rain attenuation, the S-band is selected as the frequency band.

Power Constraints

Based on the input provided by the power subsystem, it is fixed that the total power consumption of TT&C
should not exceed 105W in the phasing orbits and 65W in the final lunar orbit.

Mass Constraint

Based on the input provided by structure subsystem, it is fixed that the total mass of TT&C should not exceed
6 Kgs.

Ground Station

The system is designed to be compatible with other regional and indigenous GS networks such as ISRO DSN.
With that in mind, the GS is required to be CCSDS compliant. An acceptable tracking accuracy for near-Earth
missions is < 5cm/sec over 1 to 10s for Doppler tracking and < 10m over 1 to 10 sec ranging accuracy. This
necessitates the GS to be furnished with the following major services: Antenna Services, Data Accounting and
Distribution Services, Command/Control and Data Processing Services.

167
 1.4 COMMUNICATION LINK

The calculated link budget sets high demands for the required ground station parameters. The 18m antenna
in the ISRO DSN is aspired to be used as the ground station for the mission. The antenna is capable of 2 kW
S-Band uplink and both X-Band and S-Band downlink. The system has G/T of 30/39.5 dB/K (45º elevation,
clear sky) for S/X-Band. The base-band system adheres to the CCSDS Standards.
The subsystem operates using three onboard antennas, two low-gain, omnidirectional antennas for Omni-
link and communication up to GTO for larger tracking coverage and lower power consumption, and a
medium-gain (20dBi), the narrow beam-width antenna is used for link post-GTO for effective, point-to-point
communication. The power fed to the antennas is 7 dBW. A power amplifier is included in case high data rate
communication is necessary from final phasing orbit and for margin purposes. A margin of 6 dB or greater is
determined to be a requirement to qualify a link as feasible.

Modulation Schemes and Link Characteristics

The CCSDS Bluebook* made the following recommendations for the space to earth link:
 Present technology makes the implementation of suppressed carrier modulation systems practical.
 A comparison of carrier signal-to-noise ratios in a conventional residual carrier phase-locked loop
with those in a suppressed carrier loop shows that the latter provides a substantial advantage over the
former, frequently exceeding 10 dB.
 Recommendation 2.4.17A defines recommended bandwidth efficient modulation formats for high
symbol rate (>2Ms/s) space-to-Earth transmissions from Category A missions in Space Research
service 2 and 8 GHz bands.
 Modulation schemes: BPSK, OQPSK or GMSK.
 Channel Coding: Convolutional Coding, Reed-Solomon Coding, Turbo Coding, Low Density Parity
Check (LDPC) Coding.

Based on these recommendations, a suppressed carrier modulation scheme is considered. BPSK and
OQPSK modulation schemes are also considered. BPSK has less bandwidth efficiency but is quite immune to
noise. OQPSK can be seen as two orthogonal BPSK systems working without interference. So, OQPSK
almost doubles the bandwidth efficiency while preserving the noise immunity of BPSK systems. The
engineering penalty for this improvement is the complexity of the system. However, modern advances in
electronics have simplified OQPSK systems dramatically. For this reason, OQPSK is selected as the
modulation scheme for this mission. Differential encoding is used for the modulation to prevent the necessity
𝐸𝑏
of synchronization pulse. The theoretical BER vs 𝑁0
values for various channel coding schemes for OQPSK
were simulated and plotted using MATLAB.

*
”Radio Frequency and Modulation Systems – Part 1 Earth Stations and Spacecraft,” CCSDS A00.0-Y-7.
Yellow Book. Issue 7. Washington, D.C.: CCSDS, November 1996. URL:
http://public.ccsds.org/publications/archive/401x0b09s.pdf [Accessed 26 January 2016]

168
 Figure 1.1: BER comparison of various channel coding schemes

The parallel concatenated coding scheme gives the best BER performance. Concatenated scheme can be seen
as two convolutional coders connected with an interleaver. Since the interleaving length is very large, the
encoder and decoders would be of enormous complexity. Using higher number of iterations in parallel
𝐸
concatenated coding, a BER of the order of 10-6 can be obtained with 𝑁𝑏 as low as 2 dB. Due to the increased
0
complexity of implementation of concatenated coding, it is not considered for analysis. The next best coding
scheme is convolutional coding. The Convolutional coding scheme with soft decision gives the best BER
performance but is again quite difficult to implement for the scale of this mission. So, the next best coding
scheme, Convolutional coding with hard decision was selected for channel coding. The code rate (r) selected
is ½ bit per symbol, Constraint Length (K) is 7 bits with Connection Vectors G1 = 1111001 (171 Octal) G2 =
1011011 (133 Octal) and Symbol inversion on output path to G2. Along with these, the frame synchronization
for convolution coding, the ASM can be obtained either in the channel symbol domain or in the domain of bits
decoded by the convolution decoder. Since convolution codes are used, the ASM will be convolutionally
encoded. The ASM bit pattern for convolution code is 1ACFFC1D. Also, the ASM shall be attached to the
transfer frame where the length of the transfer frame will be an integral number of octet not exceeding 2048
octet. The pseudo-random configuration of the data is shown in Fig 1.2.

Figure 1.2: Pseudo-randomizer Configuration

169
Antenna

The SRMSAT-2 requires an antenna with high gain for transmission of payload data and receiving of
software patches and a low gain, omnidirectional antenna set to transmit and receive throughout the mission
for telecommand and ranging functions.

High Gain Antenna

The antenna types considered for the high gain antenna are:
i. Horn Antenna
ii. Monopole and Dipole
iii. Helical Antenna
iv. Cassegrain Antenna
v. Parabolic Antenna

Horn Antenna

The horn antennas are simpler to construct, provide medium gain and have a larger bandwidth. The optimum
horn dimensions are given by
𝛿cos(𝜃/2)
L=
1−cos(𝜃/2)

(1.1)

Where the optimum path length difference is taken as 0.2 times the free space wavelength. The flare angle is
assumed to be 𝜃=25o to provide maximum directivity. Substituting the necessary parameters, an optimum
length of the horn is obtained as 1.12m, which is an impractical length for mounting on the spacecraft. Thus,
Horn antennas are not suitable for this mission.

Monopole and Dipole antennas

The maximum gain that a dipole can provide for a half wave dipole is 2.15 dBi and this gain is irrespective of
the frequency of transmission but however, the length is dependent on the wavelength. For a 5/4 wave dipole,
the gain further increases to about 5.2 dBi making it an
undesirable component to choose as medium gain
antennas are needed for this mission. If an array of
dipoles is employed, the gain can be increased but
would cross the size constraint of the mission.

Monopoles provide twice the gain of the

corresponding dipole antenna i.e. 3dB greater gain
than a corresponding dipole antenna. Even this higher
gain is not sufficient for the mission, so Monopole and
Dipole antennas are not suitable.

Helical antenna

Helical antennas are usually preferred for usage in

axial mode as they are very simple to construct and
provide sufficient gains. Setting the gain requirement
for the mission as 20 dBi, a helical antenna was Figure 1.3: Synthesized Helical Antenna
synthesized for 20 dBi gain. The dimensions of the

170
synthesized antenna are shown in Fig.1.3. The length of the helix required to obtain 20 dBi gain is as high as
997mm with 33.4 turns. A helical antenna of this length cannot be mounted on the spacecraft. Moreover, an
antenna of this height would be difficult to point in the axial mode.

Cassegrain antenna

Cassegrain antenna consists of a main parabolic dish and a convex secondary reflector. This makes the
construction extremely bulky and deployment almost impossible for this mission. The angular width that the
secondary reflector subtends is very less as compared to the angle subtended by the primary with a front fed
dish and hence the feed horn would also be long. So Cassegrain antenna can be eliminated.

Parabolic antenna

Parabolic antennas have very high directivity, provide high gain and have narrow beamwidths, which are
suitable characteristics for antenna aboard the SRMSAT - 2. The antenna structure being non-resonant, allows
the antenna to be used for a broad range of frequencies and thereby supporting a broad bandwidth. Based on a
discussion with structure subsystem, the diameter is fixed to be 0.52m, and the aperture efficiency assumed to
be 0.7 for a frequency of transmission of 2.2GHz, an approximate gain of 20 dBi is obtained. This gain is
calculated by using the following formula
𝜋𝑑
G = ( )2 𝑒𝐴
𝜆

(1.2)

Where d = diameter of aperture (in m)

𝑒𝐴 = aperture efficiency

𝜆 = wavelength (in m)

Thus, Parabolic antenna has been chosen as the High Gain Antenna aboard the SRMSAT – 2.

Omnidirectional Antenna

Omnidirectional antennas require a simple electronic and mechanical interface and clear field of view (FOV)
to qualify as an emergency backup component. They are used to acquire the data link when spacecraft
orientation may be unknown or uncontrollable. Because they are required to emit nearly isotropic radiation
and have a simple configuration, the omnidirectional antenna is best realized by using a single feed. The FOV
of a single element antenna will be limited unless the antenna can be positioned well away from the
spacecraft. However, if the distance from the spacecraft is increased considerably, effects on attitude control
system and structural problems could be adverse. The Omnidirectional Antenna needs to have the following
qualities
i. Beamwidth of a single antenna approaching 180degrees
ii. Uniform gain over the front hemisphere
iii. Minimal radiation over the back hemisphere

The other objectives of Omnidirectional Antenna design are:

 Material – Nonmagnetic Spaceflight materials with negligible outgassing
 Temperature – Survivability in 173K to 373K and operation in 208K to 350K
 Noise and Vibration – Survivability in acoustic and vibration levels required for spacecraft
components.
 Size and Weight – Spacecraft compatibility by minimizing size and weight

171
  Polarization – Uses left-handed circular polarized signals
 Gain - Provides gain greater than - 1 dB throughout the front hemisphere

Helical antenna and QFH antenna would be the best candidates for the omnidirectional antenna because
they provide circular polarization but QFH is selected considering NASA recommendations citing the small
size and mass of the QFH antenna compared to an equivalent helical antenna. A QFH antenna was
theoretically synthesized for use in this mission with these objectives.

Figure 1.4: QFH Antenna

Figure 1.5: Synthesized QFH Antenna

The dimensions of synthesized antenna are shown in Fig 1.5. The 3D radiation pattern is shown in Fig. 1.6a
and the 2D radiation pattern is shown in Fig. 1.6b. The left winding direction is to obtain left-handed
polarization.

172
 Figures 1.6(a) and 1.6(b): Radiation Pattern of the synthesized antenna

It can be observed that the gain pattern is directive at the boresight, and the beamwidth is too narrow for the
requirement. Raising the antenna above the ground plane depresses the gain at boresight and spreads the
radiation pattern along the frontal lobe. Considering this, the port height was raised in steps to 20mm, 50mm,
70mm and 100mm and analysed.

Figures 1.7(a), 1.7(b), 1.7(c) and 1.7(d): Radiation Pattern for port lengths 20mm, 50mm, 70mm and 100mm respectively

173
The 100mm port size creates a distorted wave pattern. This distortion can be attributed to the return loss at the
central frequency of transmission. A port size higher than 100mm is not desirable as it would have adverse
effects on attitude control. The 70mm port size gives the desired cardioid pattern with beamwidth approaching
180 degrees. Thus, a 70mm port size is selected for the QFH antenna. The 3D radiation pattern for the
synthesized QFH antenna with 70mm port is shown in Fig.1.8. The return loss of 100mm port is shown in Fig.
1.9.

Figure 8: Radiation pattern for 70mm port size

Figure 1.9: Return loss for the 100mm port

The antennas are mounted on the spacecraft with directions 180 degrees apart to get a full omnidirectional
pattern. These antennas can be fed with a coupled configuration or a switched configuration. Coupled
configuration drives both the antennas at once so there would be areas where the individual antenna’s beams
overlap, where interferometry pattern is observed due to the difference in signal phase. The switched
configuration adds some RF losses but further improves the radiation pattern so that the zone of

174
interferometry is reduced to image effects. Due to sufficient margins in the Omni link and due to the
importance of the integrity of this link, switched configuration is determined to be used.

Figure 1.10(a) and 1.10(b): Omnidirectional antenna configuration

1.5 MISSION SEQUENCE

In GTO and initial phasing orbits, the omnidirectional antenna with 4000 bps speed is used for
communication. This speed is cut down to 80 bps in the final phasing orbit and is increased to 1000 bps in
lunar orbit. The parabolic antenna is unlikely to be used in the phasing orbits, but provision is made for this
case. Once the lunar orbit is achieved, Omnidirectional antenna would be used to receive telecommands and
Parabolic antenna would be used for payload data transmission. In the event of attitude loss in any of these
phases, the omnidirectional antenna is automatically selected for all communications.

1.6 LINK BUDGET

Based on the ground station and link characteristics, a link budget is calculated for the mission. The link
budget can be easily calculated using the following equations:

Space Loss (Ls in dB) = 147.55 - 20 log S - 20 log f

(1.3)
Where,
S is the path length in m, and f is the frequency in Hz.

𝐸𝑏
𝑁0
= P + Lt + Gt+Laa + Ls + La + Gr + 228.6 – 10logTs – 10 logR
(1.4)

EIRP = P + Lt + Gt
(1.5)

175
 𝐶 𝐸
𝑁0
= 𝑁𝑏 + 10 log R
0
(1.6)

𝐸 𝐺𝑟
RIP = 𝑁𝑏 - 𝑇𝑠
- 228.60 + 10 log R
0
(1.7)

Where,
Eb C
,
N 0 N0
, Lt, Gt, Ls, La, Gr are in dB, P is in dBW, Ts is in K, R is in bps.

The uplink frequency is selected to be 2110 to 2120 MHz and downlink frequency to be 2200 to 2290 MHz
according to CCSDS recommendations. The data rate for the parabolic antenna is fixed to 0.1 Mbps downlink
and 4 Kbps uplink. This downlink speed was selected to allow for sufficient link margins. The
omnidirectional data rates are 100bps uplink and downlink in final phasing orbit and 1000bps uplink and
downlink in lunar orbit. The omnidirectional data rate goes up to 4 kbps in the initial phasing orbits. The link
budget calculations for the initial phasing orbits are not shown as the link is designed for much worse
conditions than the initial phasing orbits. The onboard transmitter feeds 15 dBW to the antenna with 20 dBi
gain in the unlikely case where high data rate communication is necessary for final phasing orbit. The
efficiency of the antenna is assumed to be 0.7, spacecraft system noise to be 27.88 dB-K and GS System noise
to be 19.805 dB-K. The margins for these link budgets should account for rain losses, some equipment losses
and should include a fade margin on top of that. For this reason, as mentioned before, a margin of at least 6 dB
is necessary for communication.

176
 Telemetry
Command Omni Omni
Parameter Symbol Units Source and Data
(Uplink) Uplink Downlink
(Downlink)
Input
Frequency f GHz 2.1 2.2 2.1 2.2
Parameter
Input
Transmitter Power P dBW - 15 - 7
Parameter
Transmitter Line
Lt dB Estimate - -1 - -1
Loss
Peak Transmit Input
Gpt dBi - 20.02 - 0
Antenna Gain Parameter
Transmit Antenna Input
Dt m - 0.52 - -
Diameter Parameter
Transmit Antenna
Lpt dB Estimate - -1 - 0
Pointing Loss
Transmit Antenna
Gt dBi Gpt + Lpt - 19.02 - 0
Gain (Net)
Equiv. Isotropic
EIRP dBW Eq. 1.5 70 33.02 70 6
Radiated Power
Propagation Path Input
S km 1.5 Million 1.5 Million 1.5 Million 1.5 Million
Length (Max) Parameter
Space Loss Ls dB Eq. 1.3 -222.406 -222.81 -222.406 -222.81

Polarization Loss La dB Estimate -0.3 -0.3 -0.3 -0.3

Atmospheric Loss Laa dB Estimate -1 -1 -1 -1
Receive Antenna Input
Dr m 0.52 12 - 12
Diameter Parameter
Peak Receive Input
Grp dBi 19.616 41.805 0 41.805
Antenna Gain Parameter
Receive Antenna
Lpr dB Estimate -1 -1 0 -1
Pointing Loss
Receive Antenna
Gr dBi Grp+ Lpr 18.616 40.805 0 40.805
Gain
System Noise dB -
Ts Estimate 27.88 19.805 27.88 19.805
Temperature K
Input
Data Rate R bps 4000 100000 100 100
Parameter
𝑬𝒃 𝑬𝒃
dB Eq. 1.4 29.609 8.51 27.01 11.49
𝑵𝟎 𝑵𝟎
Carrier-to-Noise 𝐶 dB-
Eq. 1.5 65.63 58.51 45.04 31.49
Density Ratio 𝑁0 Hz
Input
Bit Error Rate BER - 10-6 10-5 10-6 10-5
Parameter
𝑬𝒃 𝑬𝒃 Input
Required Req dB 8.1 7.5 8.1 7.5
𝑵𝟎 𝑵𝟎 Parameter
Implementation
- dB Estimate -1 -1 -1 -1
Loss
𝑬𝒃 𝑬𝒃
Margin - dB - Req 21.509 1.01 18.91 3.99
𝑵𝟎 𝑵𝟎
Received Isotropic
RIP dBW Eq. 1.7 -153.706 -191.09 -153.71 -218.11
Power
Table 1.4: Link Budget – Farthest Distance (Using ISRO DSN 12m Antenna)

177
It can be seen that for the farthest distance, the downlink margin for both HGA link and Omni link are
lower than the required margin, and so it is infeasible to use the ISRO 12m antenna for the final
phasing orbits.
Telemetry
Command Omni Omni
Parameter Symbol Units Source and Data
(Uplink) Uplink Downlink
(Downlink)
Input
Frequency F GHz 2.1 2.2 2.1 2.2
Parameter
Transmitter Input
P dBW - 15 - 7
Power Parameter
Transmitter Line
Lt dB Estimate - -1 - -1
Loss
Peak Transmit Input
Gpt dBi - 20.02 - 0
Antenna Gain Parameter
Transmit Antenna Input
Dt m - 0.52 - -
Diameter Parameter
Transmit Antenna
Lpt dB Estimate - -1 - 0
Pointing Loss
Transmit Antenna
Gt dBi Gpt + Lpt - 19.02 - 0
Gain (Net)
Equiv. Isotropic
EIRP dBW Eq. 1.3 79 33.02 79 6
Radiated Power
Propagation Path Input
S km 1.5 Million 1.5 Million 1.5 Million 1.5 Million
Length (Max) Parameter
Space Loss Ls dB Eq. 1.3 -222.40621 -222.81 -222.406 -222.81
Polarization Loss La dB Estimate -0.3 -0.3 -0.3 -0.3
Atmospheric Loss Laa dB Estimate -1 -1 -1 -1
Receive Antenna Input
Dr m 0.52 18 - 18
Diameter Parameter
Peak Receive Input
Grp dBi 19.616 50.805 0 50.805
Antenna Gain Parameter
Receive Antenna
Lpr dB Estimate -1 -1 - -1
Pointing Loss
Receive Antenna
Gr dBi Grp+ Lpr 18.616 49.805 0 49.805
Gain
System Noise dB -
Ts Estimate 27.88 19.805 27.88 19.805
Temperature K
Input
Data Rate R bps 4000 100000 100 100
Parameter
𝑬𝒃 𝑬𝒃
dB Eq. 1.4 38.61 17.51 36.01 20.49
𝑵𝟎 𝑵𝟎
Carrier-to-Noise 𝐶
dB-Hz Eq. 1.5 74.63 67.51 56.014 40.49
Density Ratio 𝑁0
Input
Bit Error Rate BER - 10-6 10-5 10-5 10-5
Parameter
𝑬𝒃 𝑬𝒃 Input
Required Req dB 8.1 7.5 8.1 7.5
𝑵𝟎 𝑵𝟎 Parameter
Implementation
- dB Estimate -1 -1 -1 -1
Loss
𝑬𝒃 𝑬𝒃
Margin - dB - Req 30.51 10.01 27.91 13
𝑵𝟎 𝑵𝟎
Received Isotropic
RIP dBW Eq. 1.7 -144.705 -191.09 -144.71 -218.11
Power
Table 1.5: Link Budget – Farthest Distance (Using ISRO DSN 18m Antenna)

The ISRO 18m antenna provides sufficient margin, and a link is feasible with 5W transmitted power
using the omnidirectional antenna or 15 dBW transmitted power using the HGA antenna.

178
 Telemetry
Command Omni Omni
Parameter Symbol Units Source and Data
(Uplink) Uplink Downlink
(Downlink)
Frequency f GHz Input Parameter 2.1 2.2 2.1 2.2
Transmitter
P dBW Input Parameter - 7 - 7
Power
Transmitter Line
Lt dB Estimate - -1 - -1
Loss
Peak Transmit
Gpt dBi Input Parameter - 20.02 - 0
Antenna Gain
Transmit
Antenna Dt m Input Parameter - 0.52 - -
Diameter
Transmit
Antenna Pointing Lpt dB Estimate - -1 - 0
Loss
Transmit
Antenna Gain Gt dBi Gpt + Lpt - 19.02 - 0
(Net)
Equiv. Isotropic
EIRP dBW Eq. 1.5 84 25.02 84 6
Radiated Power
Propagation Path
S km Input Parameter 1.5 Million 1.5 Million 1.5 Million 1.5 Million
Length (Max)
Space Loss Ls dB Eq. 1.3 -222.406 -222.81 -222.406 -222.81
Polarization Loss La dB Estimate -0.3 -0.3 -0.3 -0.3
Atmospheric Loss Laa dB Estimate -1 -1 -1 -1
Receive Antenna
Dr m Input Parameter 0.52 32 - 32
Diameter
Peak Receive
Grp dBi Input Parameter 19.616 58.305 0 58.305
Antenna Gain
Receive Antenna
Lpr dB Estimate -1 -1 0 -1
Pointing Loss
Receive Antenna
Gr dBi Grp+ Lpr 18.616 57.305 0 57.305
Gain
System Noise dB -
Ts Estimate 27.88 19.805 27.88 19.805
Temperature K
Data Rate R bps Input Parameter 4000 100000 100 100
𝑬𝒃 𝑬𝒃
dB Eq. 1.4 43.61 17.01 41.01 27.99
𝑵𝟎 𝑵𝟎
Carrier-to-Noise 𝐶
dB-Hz Eq. 1.5 79.63 67.01 60.01 47.99
Density Ratio 𝑁0
Bit Error Rate BER - Input Parameter 10-6 10-5 10-6 10-5
𝑬𝒃 𝑬𝒃 Input
Required Req dB 8.1 7.5 8.1 7.5
𝑵𝟎 𝑵𝟎 Parameter
Implementation
- dB Estimate -1 -1 -1 -1
Loss
𝑬𝒃 𝑬𝒃
Margin - dB - Req 35.51 9.51 32.91 20.49
𝑵𝟎 𝑵𝟎
Received
RIP dBW Eq. 1.7 -139.705 -199.09 -139.709 -218.11
Isotropic Power
Table 1.6: Link Budget – Farthest Distance (Using ISRO DSN 32m Antenna, 84dBW EIRP)

179
The ISRO 32m antenna further improves the link and eliminates the need to use a power amplifier for
the HGA link, which significantly reduces the power consumed by the TTC subsystem and reduces
the thermal dissipations. It also ensures a strong link with the omnidirectional antennas. Despite these
advantages, considering the complexity and cost of operation of the 32m antenna, it is not required for
this mission. The spacecraft stays at the distance range of 1.5 Million Km for a few hours where a
WSB patch maneuver sets the spacecraft on its course to ballistic capture. After reaching the lunar
orbit, the distance is much less, and space loss is reduced. The lunar orbit distance is approximately
410,000 km and space loss is reduced to -211.15 dB for command and -211.55 dB for Telemetry
Data. The data rate of the Omni directional link is increased to 1000 bps after reaching the lunar orbit.
The link parameters for Lunar Orbit for 15dBW transmitted power are shown in Table 1.7 and Table
1.8.

Telemetry and
Command Omni
Parameter Data Omni Uplink
(Uplink) Downlink
(Downlink)
EIRP 79 dBW 33.02 dBW 79 dBW 14 dBW
Space Loss -211.15 dB -211.55 dB -211.15 dB -211.55 dB
𝑬𝒃
49.86 dB 28.77 dB 37.27 dB 29.75 dB
𝑵𝟎
𝑪
85.880 dB-Hz 78.77 dB-Hz 67.27 dB-Hz 59.75 dB-Hz
𝑵𝟎
Margin 41.76 dB 21.27 dB 29.17 dB 22.25 dB
Received Isotropic -198.45
-133.455 dBW - 179.83 dBW -133.449 dBW
Power dBW
Table 1.7: Link characteristics in lunar orbit for P = 15 dBW for HGA (Using ISRO 18m)

Telemetry and
Command Omni
Parameter Data Omni Uplink
(Uplink) Downlink
(Downlink)
EIRP 70 dBW 33.02 dBW 70 dBW 14 dBW
Space Loss -211.15 dB -211.55 dB -211.15 dB -211.55 dB
𝑬𝒃
40.86 dB 19.77 dB 28.27 dB 20.75 dB
𝑵𝟎
𝑪
76.886 dB-Hz 69.77 dB-Hz 58.27 dB-Hz 50.75 dB-Hz
𝑵𝟎
Margin 32.76 dB 12.27 dB 20.17 dB 13.25 dB
Received Isotropic
-142.455 dBW - 179.83 dBW -142.45 dBW -198.45 dBW
Power
Table 1.8: Link characteristics in lunar orbit for P = 15 dBW (Using ISRO 12m)

It can be seen that the margin is exceeding the requirements, and there is an opportunity to
reduce the transmitted power. Communication using the ISRO 12m antenna would require the power
amplifier but the power constraint on the subsystem does not allow the TT&C subsystem to use the
power amplifier in lunar orbit. Reducing the transmitted power to 7 dBW, the calculated link budget
is shown in Table 1.9.

180
 Telemetry and
Command Omni
Parameter Data Omni Uplink
(Uplink) Downlink
(Downlink)
EIRP 70 dBW 25.02 dBW 70 dBW 6 dBW
Space Loss -211.15 dB -211.55 dB -211.15 dB -211.55 dB
𝑬𝒃
40.86 dB 11.77 dB 28.27 dB 12.75 dB
𝑵𝟎
𝑪
76.886 dB-Hz 70.77 dB-Hz 58.27 dB-Hz 42.75 dB-Hz
𝑵𝟎
Margin 32.76 dB 4.27 dB 21.17 dB 5.25 dB
Received Isotropic
-142.455 dBW -187.83 dBW -142.45 dBW -206.85 dBW
Power
Table 1.9: Link characteristics in lunar orbit for P = 7 dBW (Using ISRO 12m)

While the uplinks are possible with the 12m antenna in the lunar orbit, Downlinks have a lower
margin than the requirement, making the 12m antenna unfit for downlink transmissions.

Telemetry and
Command Omni
Parameter Data Omni Uplink
(Uplink) Downlink
(Downlink)
EIRP 79 dBW 25.02 dBW 79 dBW 6 dBW
Space Loss -211.15 dB -211.55 dB -211.15 dB -211.55 dB
𝑬𝒃
49.86 dB 20.77 dB 37.27 dB 21.75 dB
𝑵𝟎
𝑪
85.880 dB-Hz 70.77 dB-Hz 67.27 dB-Hz 51.75 dB-Hz
𝑵𝟎
Margin 41.76 dB 13.27 dB 29.17 dB 14.25 dB
Received Isotropic
-133.455 dBW -187.83 dBW -133.45 dBW -206.85 dBW
Power
Table 1.10: Link characteristics in lunar orbit for P = 7 dBW (Using ISRO 18m)

It can be seen that the ISRO 18m antenna is required for supporting the communication operations of
this mission.

1.7 Communication Hardware

The subsystem consists of an in-orbit telecommand configurable Transponder with the Rx channel
connected to an external LNA and Tx channel connected to an external switched power amplifier to
switch between operation with and without Power Amplifier. Due to the performance characteristics
of the transponder, as discussed in coming section, LNA is used for increasing the margin of the
system and not a mission-critical component. Also, as inferred from the link budget, the power
amplifier is also not a mission critical component and is just present for emergencies and margin.

181
 Figure 1.11: Block diagram of the Communication system

Transponder

As the system requires a transmit power of 7 dBW (5W) without the use of power amplifier, and
RF output of 7 dBW is the primary requirement for the onboard transponder. The transponder has to
support OQPSK Suppressed Carrier data rate of 100 kbps Tx and 4 kbps Rx. Other requirements
include ± 5 ppm initial frequency stability setting and ± 20 ppm global frequency stability. Based on
these requirements, the Integrated S-Band Transponder§ by ThalesAlenia Space is a good candidate.
The specifications of the ISBT are:

Parameter Value
Frequency S-Band (2025-2120 MHz) Rx and S-Band (2200-2300 MHz) Tx
Carrier Acquisition Threshold -128dBm
RF Output Power 7 dBW (higher with external module)
Frequency Stability ± 5 ppm initial ± 20 ppm global
Power Consumption 5W Rx, 25W Tx
Mass 2.6 Kg
Dimensions 28 mm x 170 mm x 194 mm
Ranging ESA/NASA 2 tone, ESA MPTS, NASA SNUG 450
Table 1.11: Transponder Specifications

Power Amplifier

The power amplifier is used for communication from the farthest points of the mission. As the
power that needs to be transmitted is 15dBW, and the Transponder outputs 7dBW, the power

§
For more information, see:
https://www.thalesgroup.com/sites/default/files/asset/document/Platform_Solutions_ISBT_Integrated_Transpon
ders.pdf

182
amplifier needs to have an RF output of at least 26 dBW and small size with low thermal emissions
and good efficiency. The NuPowerTM 05E05A Power Amplifier¶ by NuWaves fits the requirement
well.

Parameter Value
Frequency 2000 MHz to 2600 MHz
RF Output 15 dBW
Technology GaN
Power Efficiency 40%
Power Consumption 75W
Mass 0.25 Kgs
Dimensions 114.3 mm x 89 mm x 15.5 mm
Table 1.12: Power Amplifier Specifications

1.8 VISIBILITY
To evaluate the visibility of the spacecraft in phasing and lunar orbits, the scenario was modeled in
STK and analyzed.

Figure 1.12: Visibility analysis scenario in phasing orbits. The radiation pattern of the antenna is also shown.

¶
For more information, see:
http://nuwaves.com/wp-content/uploads/2016/03/NuPower_05E05A_Data_Sheet.pdf

183
 Figure 1.13: Visibility analysis scenario in lunar orbit.

The analysis in phasing orbits revealed that there is an average access time of 11 hours post the initial
GTO. The analysis in lunar orbit revealed an average access time of 9 hours. However,
communication is not possible for all of 9 hours due to Figure of Merit constraints. A C/No constraint
based on link budget was added to the simulation and the access time was brought down to 6.5 hours
by the constraint. This gives access for 6.5 hours and 17.5 hours of time interval between access
intervals. Based on the data size and data rate estimates, this gives the spacecraft enough time to
transmit and receive the necessary command, telemetry, and data.

1.9 FUTURE WORK

The future work of TT&C subsystem includes implementation of better channel coding schemes to
reduce the size of the system along with the GS antenna requirement, electromagnetic simulation of
the antennas aboard the spacecraft and detailed design of the RF network.

184
 2. ELECTRICAL POWER SYSTEM

2.1. OVERVIEW

The Electrical Power System (EPS) generates, stores, conditions, controls, and distributes power
within the specified voltage band to all bus and payload equipment. Power system failure essentially
results in the loss of a space mission, and it is noteworthy that many of the early satellite missions
were lost due to such failures. Hence, the power subsystem must be designed with utmost caution and
must account for abnormalities in consumption of power during worst case scenarios. The power
subsystem should also ensure that the power supply to all parts remains uninterrupted. Also, it is the
function of the power subsystem to provide a fault detection and protection mechanism for the entire
spacecraft against all electrical faults. A Power budget analysis is done to calculate the power
requirement in each phase of the mission; solar arrays and the battery pack is sized to meet the power
requirements in these phases. The duration of the eclipse is an important parameter while selecting
and sizing the secondary storage. The satellite is designed to go into a safe mode in case of low
battery and non-availability of solar power to conserve satellite function, in the periodic event of long
eclipse due to earth during the lunar orbit phase. The selection of components and parts is aimed at
accomplishing high efficiency, small size, and low cost.
It is noteworthy that the power management and distribution system of SRMSAT-2 contains a
fault tolerant MPPTU (Maximum Power Point Tracking Unit), in addition to high efficiency
switching regulators and a battery management unit. It is designed in such a way that the power
transfer switches to Direct Energy Transfer (DET) architecture in the case of MPPT failure. As the
battery voltage is matched with the array voltage, this ensures that no failures would occur.
The main satellite bus (28V) is split into sub-buses. Buck regulators are used in most bus lines
except in the 28V line, where a buck-boost regulator is used to ensure constant voltage to the primary
thrusters and RCS thrusters.
As a protection mechanism, latching current limiters and fold back current limiters are used.
Latching current limiter provides protection to components where excess current is completely
intolerable, and a foldback current limiter is used where the excess current can be tolerated to a
certain extent.

2.2 FUNCTIONS

 Supply a continuous source of electrical power to spacecraft loads during the mission life
 Control and distribute power to the spacecraft
 Support power requirements for average and peak electrical load
 Provide command and telemetry capability for EPS health and status, as well as control by
ground station
 Protect the spacecraft components against the failures within the EPS.
 Suppress transient bus voltages and protect against bus faults

185
 2.3 GENERAL LAYOUT OF EPS

The function of the power system is split up into four parts.

 Power Generation
 Power Storage
 Power Distribution
 Fault Detection and Protection

Figure 2.1: The EPS functional architecture of SRMSAT – 2

The solar array is the source of power generation in the satellite. The Power control and distribution
unit conditions the energy and distributes power to each subsystem of the satellite. The secondary
batteries store energy for use during eclipse and peak power demand. The fault detection and
protection unit detects electrical faults occurring in components, protects and troubleshoots them. The
solar cells used in the array are ultra-triple-junction GaAs cells manufactured by Spectrolab*. Each
panel is controlled by maximum power point tracking unit. The power distribution system uses high
efficiency switching regulators and bus lines are optimized to work at the high-efficiency point of
switching regulator. Li-ion batteries are considered for use as Secondary batteries. The Fault detection
and protection unit contains hardware to sense and limit current and provide switching interface. The
functional architecture of the power system which executes these functions is shown in Figure 2.1.

2.4 POWER BUDGET AND POWER CONSUMPTION ANALYSIS

Different subsystems require different amounts of power and are required to be switched on or off
during different phases of the mission. Solar panels and batteries should be sized accordingly and
switching regulators should be selected appropriately. For this reason, a dynamic power consumption
analysis is done during different phases of the mission. The peak power requirement of each
component is shown in Table 2.1. The orbital period of the spacecraft around the moon is 132
minutes. The operation time of the Battery Heater is not predictable at this point, so it is assumed to
be on throughout the orbit. Transponder Rx & Tx duty cycle is taken as 52% for worst case analysis.

*
Spectrolab Inc., “28.3% Ultra triple junction (UTJ) solar cells,” Technical Document, 2004.
[online] URL: http://www.spectrolab.com/DataSheets/cells/PV%20UTJ%20Cell%205-20-10.pdf.

186
Mission Phases

For the power consumption analysis, the trajectory is split into different mission phases with some
sub-phases and power requirement is calculated in the following phases. This analysis would
determine which phase of the mission would be the sizing case for the solar panels and battery.

Operating
Power Duty Orbit Margin Total
Subsystem Component
Requirem Cycle Average % Power
ent
Sun Sensors 400mW 100% 400mW 10% 440mW
Star Sensor 700mW 100% 700mW 10% 770mW
Reaction
ADCS 6W 90% 5.4W 10% 5.94W
Wheels
RCS Thrusters 20W 15% 3W 5% 3.15W
IMU 45mW 100% 45mW 10% 49.5mW
RF Network 1W 100% 1W 10% 1.1W
Power
75W 0% 0W 10% 0W
Amplifier
LNA 3.6W 100% 3.6W 10% 3.96W
Communications
Transponder –
30W 52% 15.6W 10% 17.16W
Rx & Tx
Transponder -
5W 48% 2.4W 10% 2.64W
Rx
On-board data On-board
4W 100% 4W 10% 4.4W
handling computer
Primary
Propulsion 40W 0% 0W 10% 0W
Thruster
Payload Payload 15W 18% 2.7W 10% 2.97W
Thermal Battery Heater 10W 100% 10W 10% 11W
Average
Average Power
48.85W 53.58W
Power with
Margin
Table 2.1: Power Budget

187
 IMU

PMS
OBC
LNA

NAC

Power
Power
Power
wheels

Radom
thruster

demand
Primary
Reaction

amplifier

(in watts)

margin 10%
Sun Sensors
Star Sensors

Transponder
Components

demand with
RCS thrusters

Battery heater
A

43.72 39.745 4 5 3.6 0.045 20 6 0.7 0.4

B

73.32 65.745 1 4 30 3.6 0.045 20 6 0.7 0.4

C
Phase 1

73.32 65.745 1 4 30 3.6 0.045 20 6 0.7 0.4

D

79.36 72.145 1 40 4 0.045 20 6 0.7 0.4

A

83.32 75.745 10 1 4 30 3.6 0.045 20 6 0.7 0.4

B

165.83 150.75 10 1 4 30 3.6 75 0.045 20 6 0.7 0.4

C
Phase 2

90.36 82.145 10 1 40 4 0.045 20 6 0.7 0.4

D

188
47.13 51.445 10 0.7 1 4 5 3.6 0.045 20 6 0.7 0.4
A

90.36 82.145 10 1 40 4 0.045 20 6 0.7 0.4

B

56.59 51.445 10 0.7 1 4 5 3.6 0.045 20 6 0.7 0.4

Phase 3
C

83.32 75.745 10 1 4 30 3.6 0.045 20 6 0.7 0.4

A

83.32 75.745 10 1 4 30 3.6 0.045 20 6 0.7 0.4

B

90.360 82.145 10 1 40 4 0.045 20 6 0.7 0.4

C
Phase 4

83.32 75.745 10 1 4 30 3.6 0.045 20 6 0.7 0.4

Table 2.2: Power requirement in different phases (All values are in Watts)
D

73.09 66.445 10 0.7 1 15 4 5 3.6 0.045 20 6 0.7 0.4

55.82 50.745 10 1 4 5 3.6 0.045 20 6 0.7 0.4

Low

mode
power

3.3 3 1 2
Safe
Mode
 A B C D
Communication
Initialization and
Phase 1 – Earth System check and mode (EPCOM
Detumbling Mode ARM mode
Orbit normal mode 0/1, GCCOM 0/1,
NAVCOM 0/1)
Communication
mode (EPCOM
0/1, GCCOM 0/1,
NAVCOM 0/1) Maneuvering
Phase 2 – Cruise Normal mode
i) With power mode (DSMTC Payload (NAC)
Orbit (DSO 0/1)
amplifier 0/1)
ii) Without power
amplifier

Phase 3 –
Maneuvering
Insertion and Communication (
mode ( LIM, Payload (NAC) -
Lunar Orbit LCOM)
LOSM)
Control
Communication
Phase 4 – Lunar Maneuvering mode (PCOM, Payload (NAC,
Normal mode
Mode mode ( LSTK) LNORM) PMS)

The satellite, as mentioned before, will go into safe mode in the event of low battery and non-
availability of solar power to conserve satellite function. This event could occur during the eclipse in
the lunar orbit due to Earth, which could last up to 5 hours and occurs about once in about 6 months.
After the battery is charged to a certain charge level, the satellite returns to normal behavior. In the
case where power amplifier operation is required, the power is supplied by solar panels and battery
together as power from one of these sources is not sufficient.

The Power Simulation in Trajectory

It is determined that either the earth orbit phase (Phase 1) or the final lunar orbit would be the sizing
case for the satellite because the cruise and insertion phases are believed to have extended, almost
continuous lighting intervals. A simulation is built to determine the sizing case. The trajectory team of
SRMSAT used Systems Took Kit (STK) for the design of trajectory. The ephemeris of the satellite
was imported from trajectory STK scenario to another scenario, and a model of the spacecraft is made
for simulation in STK.

189
Power Simulation in Earth phasing orbits

The results of the simulation in earth phasing orbits proved that it receives sunlight almost all the

Figure 2.1: Sunlight Intervals in phasing orbits

time. Initially, when the satellite is ejected from the launch vehicle; detumbling, Solar Panel
Deployment, and Sun acquisition operations are performed. During this time, no power is generated
from solar panels and the power is supplied completely by the battery. In the phasing orbits, the
eclipse duration is found to be reducing with progress in mission trajectory.

Simulation in Lunar orbit

The lunar orbit is imported into the STK scenario with earth-centered frame and is simulated for
approximately 10 months†.

Figure 2.2: Simulation in Lunar orbit

The simulation revealed a general Light duration of 90 minutes and eclipse duration of 45 minutes.
This pattern, however, can’t be extrapolated for a long time. The simulation suggested that eclipse
with time as high as 5 hrs could occur and the moon eclipse times some days before this long eclipse

†
Although Table 3. Only shows analysis for 6-month duration, in the trajectory design scenario,
analysis has been performed for extended periods and it has been observed that eclipse times can
extend upto 5 hours.

190
are shorter than usual in duration. Due to the payload and High Gain Antenna communication
operations performed in Phase 4, additional attitude maneuvres are also required along with the
increased power requirement for the payload equipment. This result suggests that the lunar orbit
should be the sizing case for the battery and solar panels.

Eclipse Start Time Eclipse Stop Time

Obstruction Duration (sec)
(UTCG) (UTCG)
12 Oct 2019 06:08:29.226 12 Oct 2019 06:46:39.017 Moon 2289.791
26 Oct 2019 07:53:25.642 26 Oct 2019 08:36:07.113 Moon 2561.472
2 Nov 2019 18:42:29.684 2 Nov 2019 19:26:25.908 Moon 2636.224
6 Nov 2019 13:13:20.900 6 Nov 2019 13:57:44.034 Moon 2663.135
11 Nov 2019 14:38:43.999 11 Nov 2019 15:23:32.557 Moon 2688.558
26 Nov 2019 16:43:01.032 26 Nov 2019 17:27:55.642 Moon 2694.61
29 Nov 2019 06:32:07.110 29 Nov 2019 07:16:53.020 Moon 2685.91
8 Dec 2019 17:56:33.452 8 Dec 2019 18:40:18.390 Moon 2624.938
20 Dec 2019 01:31:03.895 20 Dec 2019 02:12:17.763 Moon 2473.868
25 Dec 2019 22:49:51.243 25 Dec 2019 23:29:03.409 Moon 2352.165
26 Dec 2019 16:29:43.540 26 Dec 2019 17:08:37.656 Moon 2334.116
3 Jan 2020 01:08:53.333 3 Jan 2020 01:43:53.180 Moon 2099.847
9 Jan 2020 00:41:42.477 9 Jan 2020 01:11:32.442 Moon 1789.966
10 Jan 2020 12:02:01.366 10 Jan 2020 12:30:06.665 Moon 1685.3
10 Jan 2020 14:14:32.802 10 Jan 2020 14:42:31.046 Moon 1678.244
10 Jan 2020 16:27:04.272 10 Jan 2020 16:54:55.393 Moon 1671.122
10 Jan 2020 17:41:25.570 10 Jan 2020 19:05:00.643 Earth 5015.074
10 Jan 2020 18:39:35.774 10 Jan 2020 19:07:19.707 Moon 1663.933
10 Jan 2020 19:58:36.129 10 Jan 2020 21:00:25.382 Earth 3709.253
10 Jan 2020 20:52:07.311 10 Jan 2020 21:19:43.987 Moon 1656.676
10 Jan 2020 23:04:38.881 10 Jan 2020 23:32:08.232 Moon 1649.351
11 Jan 2020 01:17:10.487 11 Jan 2020 01:44:32.442 Moon 1641.955
18 Jan 2020 01:12:37.277 18 Jan 2020 01:23:45.551 Moon 668.274
19 Jan 2020 03:47:01.670 19 Jan 2020 03:48:30.078 Moon 88.408
8 Apr 2020 16:22:34.466 8 Apr 2020 17:01:04.109 Moon 2309.644
Table 2.3: Eclipse summary for the Lunar orbit. The long eclipses are highlighted

191
 Figure 2.3: Simulation result showing a long eclipse in lunar orbit

2.5 SOLAR ARRAYS

Cell Selection

The Solar cells would have to withstand the hostile space

environment and provide high efficiency because of the
size constraints. The different types of cell technologies and
their efficiencies are:
Silicon: 14%
Gallium Arsenide: 19%
Indium Phosphide: 18%
Multijunction GaAs: ~28%
Due to their high efficiency and less damage rate due to the Figure 2.4: Typical V-I characteristics of a
solar cell
space environment, it is determined to use Multijunction
GaAs cells. UTJ GaAs solar cells manufactured by
Spectrolab would fit the scenario well. They have an efficiency of 28.3% at Beginning of Life (BOL)
and 27.5% at End of Life (EOL) for this mission. The typical V-I characteristics are shown in Fig 2.5.

Degradation Factors

Degradation in solar panel power generation comes from two factors

1. Initial degradation at BOL due to manufacturing errors and tolerances.
2. Accumulated losses at EOL due to environmental effects.

Initial degradation at BOL could be caused by:

 Cell mismatch in assembly
 Cell-to-cell wiring loss
 Plasma effect; causes current leakage

Accumulated degradation could be caused by:

 Ionized particles; cause damage due to accumulated radiation dose
 UV radiation; degrades cover glass and cell’s optical performance
 Thermo-mechanical stress, causes soldered joints to crack
 Micrometeoroid and Debris; scratches cover class and reduces effective area of panel
 Bypass diode failure; loses a string on panel

192
The design of EPS should consider these losses and account for them by taking sufficient margins.
The radiation data is acquired using SPENVIS and is incorporated in the design process.

Array Design

The design lifetime of the satellite is considered as one year. The average power required during
the light time (Pd) is 53.57 W. The typical eclipse duration (T e) is 45 minutes, and the worst case
eclipse duration is 3 hours. As the long eclipse is a rare event, the array and battery are not sized for
this case. The satellite operations would be suspended, and satellite would be operated in safe mode
during the long eclipse. The power required during an eclipse (Pe) is 75 W. This is higher than the
power needed during the light time because, as it is a worst case scenario, the communication is
assumed to be taking place during the eclipse. The sunlight duration (Td) is 87 minutes.

The power required to be generated by the array during sunlight hours can be calculated by using Eq
2.1.

𝑃𝑒 𝑇𝑒 𝑃𝑑 𝑇𝑑
+
𝑋𝑒 𝑋𝑑
𝑃𝑠𝑎 =
𝑇𝑑
(2.1)

The terms Xe and Xd represent the efficiency of the paths from the solar arrays through the batteries to
the individual loads and the efficiency of the path from the arrays to the loads respectively. The
efficiency values for eclipse and daylight depend on the type of power regulation. The sources of

193
efficiency loss in these paths include the MPPTU, DC-DC converters, and I2R losses. As the MPPTU
itself is a DC-DC converter, there is an additional loss in the array path. A good approximation for
these efficiencies would be Xd = 0.7 and Xe = 0.8. This calculation gives the power generation
required as Psa =125.034W ≈ 125W.

The maximum power generated by the cell is given by the equation

𝑃𝑚𝑎𝑥 = 𝑉𝑜𝑐 × 𝐶𝐹𝐹 × 𝐽𝑠𝑐

(2.2)

Where Voc is open circuit voltage, CFF is the fill factor, and Jsc is the short circuit current density.

Using the values from Spectrolab UTJ GaAs solar cells datasheet, Pmax is calculated as 38.55
mW/cm2. The array configuration can be obtained from the following equations:

𝑎𝑟𝑟𝑎𝑦 𝑣𝑜𝑙𝑡𝑎𝑔𝑒
𝑁𝑠𝑒𝑟𝑖𝑒𝑠 𝑐𝑒𝑙𝑙𝑠 𝑝𝑒𝑟 𝑠𝑡𝑟𝑖𝑛𝑔 =
𝑉𝑚𝑝

(2.3)
𝑎𝑟𝑟𝑎𝑦 𝑐𝑢𝑟𝑟𝑒𝑛𝑡
𝑁𝑝𝑎𝑟𝑎𝑙𝑙𝑒𝑙 𝑝𝑎𝑛𝑒𝑙𝑠 =
𝐼𝑝𝑎𝑛𝑒𝑙
(2.4)
𝐼𝑝𝑎𝑛𝑒𝑙
𝑁𝑠𝑡𝑟𝑖𝑛𝑔𝑠 =
𝐼𝑚𝑝

(2.5)

The bus voltage is set to 28V due to the easy availability of components for this voltage and to avoid
boost regulators in the subsystem. The maximum voltage required by a component in the EPS is 28V.
Using the values from the data sheet, these equations result in 12 cells per string. The array current
required is 4.5A. Assuming a cell size of 5.5cm x 4.4 cm, this results in 12 strings. It is determined to
add two more strings for margin and redundancy purposes, making total strings 14 and total cells as
168. These cells are distributed between two deployed panels and a body mounted panel.

The beginning of life array power per unit area can be calculated by

𝑃𝐵𝑂𝐿 = 𝑃𝑚𝑎𝑥 𝐼𝐷 cos 𝜃

(2.6)

Where ID is the decrease in power generation due to various degradation considerations mentioned in
section 2.5 along with the panel efficiency and temperature variations. As suggested in SMAD, this
value can be set to 0.77. Cosθ is the cosine loss or sun pointing angle error, which is assumed to be 5
degrees. This results in PBOL = 295.70W/m2.

Radiation degradation reduces the solar array’s output power. The EOL power generation can be
accurately estimated by the manufacturer if they are provided with the radiation data acquired from
SPENVIS. Statistically, the degradation for GaAs solar cells is 2.75% per year, out of which radiation
causes 1.5% per year. The rest of the degradation is due to the factors discussed above. The actual
lifetime degradation can be estimated using

Ld = (1 − 𝑑𝑒𝑔𝑟𝑎𝑑𝑎𝑡𝑖𝑜𝑛/𝑦𝑟)𝑠𝑎𝑡𝑒𝑙𝑙𝑖𝑡𝑒 𝑙𝑖𝑓𝑒

194
 (2.7)

Setting satellite life as 1 year, Ld = 0.9725

The EOL power generation of the arrays is calculated by:

PEOL = PBOL Ld
(2.8)

This gives EOL power generation as 287.57 W/m2. The solar array area required to support the
spacecraft power requirement is calculated from the following equation.
𝑃𝑠𝑎
𝐴𝑠𝑎 =
𝑃𝐸𝑂𝐿
(2.9)
Asa requirement is calculated as 0.435m2. The designed cell configuration has an area greater than Asa
and is well within the allowed limits.

Cell Spectrolab UTJ GaAs solar cells

Cell Size 5.5 cm x 4.4cm
Power Generation (BOL) 146W
Power Generation (EOL) 142W
Configuration 12S 14P (Distributed)
Array Voltage 28V
Solar Array Area 0.5 m2
Table 2.4: Summary of the Solar Array

2.6 POWER CONTROL AND DISTRIBUTION UNIT

The power control and distribution unit contains the following components: (1) Three maximum
power point tracker units, (2) A battery management unit, (3) Switching regulators. The three MPPTs
are used to transfer maximum power from solar panels to the satellite bus at an unregulated voltage of
~28V, and the Li-ion batteries are connected directly to the bus without a discharge regulator through
a battery management unit which charges the battery while monitoring the state of charge.

Maximum Power Point Tracker

The Power System can have the Direct Energy Transfer (DET) or Maximum Power Point Tracker
architecture. In Direct Energy Transfer, the solar power is transferred to the bus with no series
components in between. DET can be further divided into two classes: (1) fully regulated bus and (2)
sunlight regulated bus.
The solar array generates more power at a higher voltage at the beginning of life and when at a low
temperature, coming out of eclipse. The maximum power can be extracted only if the bus voltage is
varied with years in service and with temperature. However, the load must be supplied power at the
same voltage, generally lower than the maximum power producing voltage of the solar array. A
suitable switching regulator between the solar array and the load remedies this disparity between the
maximum power producing level by the peak power tracker, and the output voltage is stepped down

195
to the constant load voltage by varying the duty ratio as required. The peak power tracking is
activated only when the battery needs charging, or the load demand exceeds the solar array output.
Otherwise, the excess power is left on the array.
Maximum Power Point Tracker is selected for use in the SRMSAT – 2 for the following reasons:

 As the maximum power point of the solar array is automatically extracted, the maximum
power can be utilized whenever needed independent of the Solar Array voltage, current and
temperature.
 Pulse width modulated DC-DC converters used for MPPT applications are largely insensitive
to the electrodynamic behavior the solar array.
 MPPT can account for changes in solar cell performances such as temperature, incidence
angle and radiation damage.
 The input voltage and current ripples of the power conditioning stage are minuscule, which
enhances the efficiency of MPPT. High ripples can affect the electronic components which in
turn can decline the available power.
 In comparison to other busses with the same cell type, MPPT significantly reduces the size of
the array.
 As concluded by Willi Hallmann‡, MPPT is the best solution for missions with strong
variations in solar intensity (interplanetary/deep space missions).

The MPPTU design for the SRMSAT – 2 is fault tolerant. The series connected boost regulator,
which is explained elsewhere§ is used for this purpose. In the case of an MPPT failure, the current
would still flow to the bus, but it will not be conditioned anymore. This will be a serious problem if
the solar arrays and battery are operated at different voltages. However, in SRMSAT – 2, the solar
arrays voltage matches with the battery voltage and no major problem would arise. This configuration
also has high efficiency. So, this configuration is selected for SRMSAT – 2.

Figure 2.5: SCBC

‡
Wilfried Ley, Klaus Wittman and Willi Hallman, " Handbook of Space Technology," John Wiley &
Sons, Ltd., 2009
§
G.R. Walker, P.C. Sernia, Cascaded DC–DC converter connection of photovoltaic modules, IEEE
Transactions on Power Electronics 19 (4) (2004) 1130–1139.

Figure 2.6: SCBC in case of MPPTU failure

196
Maximum Power Point Tracker Algorithm

There are plenty of MPPT algorithms available, and each of them has their advantages and
disadvantages. The key requirements for the success of MPPT-PV system are stability, dynamic
response, steady state error, and robustness to disturbances. Three algorithms stood out of the crowd
for SRMSAT – 2; P&O, IncCond algorithm, and Modified Adaptive Hill Climbing MPPT.

P&O (Perturb & Observe)

P&O finds the maximum power point of PV modules by means of iteratively perturbing, observing
and comparing the power generated by the PV modules. In a fixed period, the load of the PV system
is adjusted to change the terminal voltage and output power of the PV modules. The variations of the
output voltage and power before and after changes are then observed and compared to be the
reference for increasing or decreasing the load in the next step. If the perturbation in this time results
in greater output power of PV modules than that before the variation, the output voltage of PV
modules will be varied toward the same direction. Otherwise, if the output power of PV modules is
less than that before variation, it indicates that the varying direction in the next step should be
changed. The maximum output power point of a PV system can be obtained by using this iterative
perturbation, observation and comparison steps.
The advantages are a simple structure, easy implementation, and less requirement of parameters.
The shortcomings of the P&O method can be summarized: (a) The power tracked by the P&O method
will oscillate and perturb up and down near the maximum power point. (b) There is a misjudgment
phenomenon for the P&O method when weather conditions change rapidly.

IncCond (Incremental Conductance)

The theory of the incremental conductance method is to determine the variation direction of the
terminal voltage for PV modules by measuring and comparing the incremental conductance and
instantaneous conductance of PV modules. If the value of incremental conductance is equal to that of
instantaneous conductance, it represents that the maximum power point is found. When the operating
behavior of PV modules is within the constant current area, the output power is proportional to the
terminal voltage. That means the output power increases linearly with the increasing terminal voltage
of PV modules (slope of the power curve is positive, dP/dV > 0). When the operating point of PV
modules passes through the maximum power point, its operating behavior is similar to constant
voltage. Therefore, the output power decreases linearly with the increasing terminal voltage of PV
modules (slope of the power curve is negative, dP/dV < 0). When the operating point of PV modules
is exactly on the maximum power point, the slope of the power curve is zero (dP/dV = 0).
The advantage of the incremental conductance method, which is superior to those of the other two
MPPT algorithms, is that it can calculate and find the exact perturbation direction for the operating
voltage of PV modules. In theory, when the maximum power point is found by the judgment
conditions (dI/dV = －I/V and dI = 0) of the incremental conductance method, it can avoid the
perturbation phenomenon near the maximum power point which usually happens for the other two
MPPT algorithms.

Modified Adaptive Hill Climbing

The MAHC method uses automatic tuning of "a", the incremental step of the duty cycle of the
converter. When the power changes in a broad range primarily due to environmental variation, this
tuner will change the “a” to a large value to satisfy the fast response requirement during the transient
stage. When the power change is small, the controller assumes that the system enters the steady-state

197
stage and the value of "a" becomes small to keep control signal change smooth. With this tuning
mechanism, both dynamic and steady-state requirements can be considered in the controller design,
because the critical parameter is updated and adjusted adaptively. Also, MAHC uses control mode
switching, which eliminates the problem of MPOP (Maximum Power Operating Point) shift when
environment conditions change rapidly. The switching criterion is defined as ∆P/a(k-l), where ∆P =
P(k)-P(k-1), represents the change of power condition, and a(k-1) is the historic value of "a(k)". If the
value of │∆P/a(k-l) │ is larger than the threshold "e", the controller understands that the power
variation was mainly caused by the solar insolation, so the increment of the duty cycle is set to the
same direction as ∆P, the change of power condition. If the value of │∆P/a(k-1) │is small, the
controller assumes that the system control is within the steady state, or the large change in power is
caused by the large step of "a" only. In the MAHC control mechanism, there are two important tuning
parameters, "e" and "M", which make the controller flexible to deal with different situations.

The IncCond algorithm is selected for use in SRMSAT – 2 after a trade study due to its fast power
point acquisition and low magnitude of oscillations around the power point.

Bus Lines and Loads

Power is required at different voltages for various components, for this, a distribution network is
designed using DC-DC Switching regulators. The main satellite bus is split into sub-buses of 3.3V,
5V, 12V, 28V to support different components connected to the power system. The efficiency of
switching regulators varies depending on the input voltage, output voltage, and load current.
Switching regulators have an optimum range of load currents for a given input voltage to operate at
their peak efficiency. Taking this into account, the load current on a switching regulator was tried to
be kept at its optimal point during operation. Based on this and the operating voltages, a distribution
network is designed with appropriate regulators with good efficiency, if not peak efficiency at the
typical value of load current on each sub-bus. This is illustrated along with complete power
architecture (Fig. 2.16). The main bus is at a voltage of 28V, and most sub bus lines have a voltage
less than 28V. So buck regulators are used in most bus lines except the 28V line, where a buck-boost
regulator is used to ensure constant voltage to the primary propulsion system, RCS Thrusters, and
power amplifier during times of thrust, attitude control and communication respectively. The
Transponder accepts a broad range of voltages (21V – 52V), so a regulator is not required for the
transponder as the bus voltage does not swing beyond the transponder’s operating range. Thus, the
transponder is connected directly to the main bus through a switch. Vicor Power has excellent
switching regulators for this purpose and is determined to be used for the distribution unit. The
electrical characteristics of these parts are shown in Figures 16-19.

198
 Load Efficiency (approx.)
Connected Selected At At
Bus voltage line
components Typical Maximum Regulator typical maximum
load load
Vicor Cool
3.3 V, Buck line IMU,OBC 1.225 A 1.225 A Power PI3301- 82% 82%
x0-LGIZ*
Vicor Cool
Star sensors,
5 V, Buck line 1.34 A 1.34 A Power PI3301- 88% 88%
Reaction wheels
x0-LGIZ*
Sun Sensors, LNA, Vicor Cool
12V, Buck line Payload, Battery 1.116 A 2.416 A Power PI3303- 93% 95%
heater x0-LGIZ*
RCS, Primary Vicor Cool
28V, Buck boost
thruster, Power 0A 4.8205 A Power ZVS - 98%
line
amplifier PI3749-00-LGIZ+
28V, Direct line Transponder - - - - -
Table 2.5: Bus lines, Loads and Regulators

199
 Figure 2.7: Efficiency curve of 5V line
Figure 2.8: Efficiency Curve of 28V line

*Vicor Power Inc., “Cool Power PI33xx-x0,” Technical Document, 2016. [online] URL:
http://cdn.vicorpower.com/documents/datasheets/Picor/ds_pi33xx.pdf, [cited 01 May 2015].
+Vicor Power Inc., “Cool Power ZVS PI3749-00-LGIZ,” Technical Document, 2016. [online] URL:
http://www.vicorpower.com/documents/family_overviews/picor/fo_coolpower_zvsBuckBoost.pdf,
[cited 01 May 2015].

Figure 2.9: Efficiency curve of 3.3V line

Figure 2.10: Efficiency Curve of 12V line

200
 2.7 POWER STORAGE

Cell Selection and Battery Sizing

Li-ion batteries are selected for the satellite because of their high energy density, low thermal
radiation, longer charge retention and absence of memory effect. Saft MP 144350 rechargeable
lithium ion cells are considered to be appropriate for the mission. Saft MP 144350 is space grade and
has an operating temperature of – 20°C to +60°C. The required battery amperage and capacity can be
estimated using these simple equations.

𝑃𝑒 𝑇𝑒
𝐶𝑟 =
(𝐷𝑂𝐷)𝑁𝑛

Where N is, the number of batteries and n is the transmission efficiency between the battery and
loads. For this case, N=1 and n is assumed to be 0.9. Cr = 125 W-hr is obtained. Allowing 25%
margin is battery capacity, Cr = 156.25 W-hr. The nominal voltage of the selected cell is 3.75V with a
typical capacity of 2.6 A-hr (9.75 W-hr nominal energy). Therefore, the number of cells required = 16
with 8 cells per string to obtain 28V. One extra string of cells is considered for redundancy purposes.
This gives a total of 24 cells with 8S 3P configuration.

Capacity (Ah) 8.35 Ah

Capacity (Wh) 234 Wh
Configuration 8S3P
Voltage 28V
Weight 1.63 Kgs
Volume 672 cm3
Table 2.6: Battery Pack specifications

Battery Monitor

The Battery Monitor monitors the state of charge of the battery pack and reports it to the OBC. This
data is logged for housekeeping. Battery monitors also provide protection against Overcurrent in
Discharge (OCD), Short Circuit in Discharge (SCD), Over Voltage (OV) and Undervoltage (UV)
events. The battery monitor only passively monitors all battery parameters, offers protection but
doesn’t interfere with the charge voltage. However, it controls charge current to charge the battery
efficiently and discharge current to have a controlled discharge. Texas Instruments bq769x0** family
could be used for this purpose.

**
Texas Instruments, “bq769x0 3-Series to 15-Series Cell Battery Monitor Family for Li-ion Phosphate
Applications”, Technical Document [online], URL: http://www.ti.com/lit/ds/symlink/bq76940.pdf , [cited 01
May 2015].

201
 2.8 FAULT DETECTION AND PROTECTION UNIT

Electrical Fault Detection and Protection

There might be situations where a fault occurs in any component, and it starts drawing abnormal
amount of current. The power system of SRMSAT – 2 has a mechanism to detect and troubleshoot
these situations. Depending on the load, either a latching current limiter or a foldback current limiter
is connected between the switching regulator and the component. Again, COTS approach is taken,
and a current limiter with integrated switch and current sensing was selected. These components
continually monitor the current flowing to the component and feed the data to on board computer. All
the current limiters are configured to limit current to a specific value depending on the load. In the
case of over current for latching limiter, the switch is immediately turned off, and the event is reported
to the onboard computer. In the case of a foldback current limiter, the limiter enters fold back mode
where current, and voltage are reduced below operating limits. If the error has been corrected, the
component operates normally, and if it is not corrected, the component is marked as malfunctioning
and is turned off permanently, and the event is reported to ground station. Latching current limiters
are connected to every component except the primary thruster. Foldback current limiter is used for the
primary thruster as some overcurrent can be tolerated in the primary thruster. As these components
have integrated switches which can be turned on or off at command, these components are used for
switching purposes also. MAX15091/MAX15091A†† by Maxim Integrated are suitable components
for this purpose. These devices implement a foldback current limit during startup to control inrush
current lowering di/dt and keep the MOSFET operating under safe operating area (SOA) conditions.
After the startup cycle is complete, on-chip comparators provide VariableSpeed/BiLevel protection
against short-circuit and overcurrent faults, and immunity against system noise and load transients.
The load is disconnected in the event of a fault condition. The devices are factory calibrated to deliver
accurate overcurrent protection with ±10% accuracy. During a fault condition, the MAX15091 latches
off, while the MAX15091A enters autoretry mode.

Single Event Effects

Single Event Effects can be split into two types. Single Even Upset (SEU) and Single Event Latch-up
(SEL). SEUs are soft errors and are generally non-destructive. They usually appear as transient pulses
in digital circuitry. SEL results in high operating current and might damage the hardware
permanently. The space grade components required to withstand these effects are very expensive and
out of the scope of a university class mission. To overcome this problem, all components except the
battery monitor are made analog, and all intelligence to the power system is given by analog
components. Current limiters protect the components from SEL events.

††
Maxim Integrated Inc., “MAX15091/MAX15091A,” Technical Document, 2015. [online] URL:
https://datasheets.maximintegrated.com/en/ds/MAX15091-MAX15091A.pdf, [cited 01 May 2015].

202
 2.9 POWER ARCHITECTURE

The complete power architecture of the EPS of SRMSAT – 2 is shown in Fig. 2.16.

Figure 2.11: Complete Architecture of Electrical Power System

2.10 FUTURE WORK

The future work in the Electrical Power system includes circuit simulations for the complete
system, development of an integrated simulator for the entire subsystem operations, the design of
MPPT Unit and refinement of design by eliminating assumptions.

203
 3. ON-BOARD COMPUTER

3.1 OBC OVERVIEW

The On-Board Computer system of the satellite is responsible for platform data processing,
communication handling, payload data handling, propulsion maneuvering, interfacing with various
components and attitude control operations that are executed during the course of the mission. The
software architecture provides a functional overview of the entire On-board software system. The On-
board data handling and management forms a crucial part of the satellite operations which includes
interfacing with most of the components. The microcontroller plays a critical role in computation;
scheduling and carrying out various tasks; collecting and organizing telemetry data. Thus, the software
should be developed around a microcontroller that has desired characteristics and features. Efficient
communication between space and ground segments is achieved by the usage of recommended
standards like CCSDS, which increase inter-operability and efficiency. Error detection and correction
schemes are to be implemented for correcting bit flips that might occur as a result of exposure to harsh
environments.

3.2 MICROCONTROLLER SELECTION

Core architecture
In earlier satellite missions, 8-bit and 16-bit architectures were frequently favoured because of
lesser program requirements and simpler applications. However, owing to the recent advancements and
forward strides in technology and increased system requirements, many of the embedded systems now
require an extensive amount of computational power, memory storage, processing, and clock speeds.
32-bit processors were introduced whose simplicity has been increasing over the years. Nowadays, 32-
bit processors can easily execute 8-bit, 16-bit and 32-bit instructions.
64-bit systems can directly access significantly more physical memory at a single time, but on
the downside, they require more memory to access the information bank as each instruction requires an
additional opcode. Hence, utilizing a 32-bit processor for the same number of instructions, at the same
clock frequency, a boost in the processing speed can be achieved.
Hence, a trade-off study between the system requirements and system capabilities indicate the
viability of 32-bit microcontrollers over 8-bit, 16-bit and 64-bit microcontrollers. 32-bit provided
enough memory requirements to sustain real-time operations. Selection of final components revolves
around basic architecture requirements and functional features.

CISC vs. RISC architectures

Based on the following reasons, it has been concluded that the RISC (Reduced Instruction Set
Computer) based processor architecture is better than the CISC (Complex Instruction Set Computer)
based processor architecture.
● Instructions in RISC architecture take lesser number of cycles to process; therefore, real-time
events can be performed with quicker response. This is crucial for any real-time system,
particularly in space environments where quick response to changes is vital.
● The instructions are simple and powerful, which facilitates easier code development and
optimization, and also allows for effective debugging.
● A much lesser number of cycles are utilized for each instruction, operations which would
otherwise need a few finite numbers of instructions to complete, can now be done easily using
RISC by using a single instruction. This allows for simpler and prompt decoding mechanisms
as compared to a CISC machine.

204
 ● An exponential increase in the throughput of the system is achieved by an increased number of
registers. Hence, complex computations can be executed without having to write back to some
other form of memory, thereby decreasing the number of instructions for input.
● RISC processors extensively use pipelining in an attempt to execute one instruction per cycle.
Pipelining reduces the amount of time required for executing a lot of instructions and facilitates
in improving the throughput.
● RISC-based microcontrollers are capable of performing functions even at lesser clock speeds
without any major performance degradation further leading to lesser power consumption.

Power constraints
Essentially, microcontrollers should exhibit very low power consumption. Allocated power
budget should account for the presence of two microcontrollers (primary and flight dynamics)along
with other digital components of the On Board Computer such as multiplexers and external memory
while having a margin of about 30%.

Processing Power (MIPS)

Processor speed is denoted by the number of instruction cycles that can be completed within a
second. Performance is a major issue when it comes to managing all the satellite operations. The faster
the processing capability of the microcontroller, the better is the response. This is directly linked to the
clock speed and power consumption of the system.

Interface Hardware support

The on board computer should be able to interface with components from different subsystems
2
using I C, SPI, and GPIO. This assists simultaneous communication between systems and prevents
address conflicts between devices.

Analog to Digital Converter (ADC)

Devices like the current sensors generate data as analog signals to be processed by the
microcontroller. Since processors understand only digital data, an Analog to Digital Converter (ADC)
is interfaced with the microcontroller. It is preferable to have the ADC unit embedded within the
microcontroller itself.

Watchdog Timer (WDT)

In space, radiation may lead to bit-flips in memory causing software to malfunction. This may
require a hardware reset. In order to provide system resets, a watchdog timer is incremented on a regular
basis and cleared by the microcontroller upon reaching a certain value. If the watchdog timer is not
cleared, it performs a hardware reset. These implementations either happen within the microcontroller
or on an external device to ensure the continuous operation of the microcontroller.

Temperature and Radiation Handling

The microcontroller operating temperature must lie within acceptable temperature ranges. Also,
adequate radiation protection will be provided by the satellite’s radiation shield.
Development Environment
Having a conducive development environment is of paramount importance during the
development phase. A good development environment is essential for enhancing software development
and debugging processes of microcontrollers.

205
Trade Study of Microcontrollers
In the selection process of the microcontroller, the various characteristics were examined. Due
to the increased need for fast and accurate computations, and systematic flow of tasks or functions that
would take place during the mission, 32-bit microcontrollers are the most suitable because of the reasons
stated previously.
RISC (Reduced Instruction Set Computer) architecture has better characteristics over CISC
(Complex Instruction Set Computing) architecture like much faster response to events, simpler coding,
high throughput and low power consumption. We thus s microcontrollers with RISC architectures.
Microcontrollers with specifications as per given constraints were studied extensively. Criteria
for selection include our different requirements like power, processing speed, etc. ATMEL products,
specializing particularly in the field of control systems whose functionality and being already
demonstrated in space missions, showed considerable potential for application in our systems.
Amongst all the ATMEL products, a comparative analysis was conducted for the new
microcontroller models. Their specifications with respect to speed, interfaces, GPIO, etc. were
examined with the mission’s point of view; accordingly the best cases were selected while the others
eliminated.
The power consumption of the microcontroller should be as low as possible. Taking into
account the number of interfaces that are required, the microcontrollers with only one I2C or one SPI
interface, were eliminated. The number of I/O pins is also evaluated as a factor, and a greater number
of I/O pins are preferred. The operating temperature range for the chosen microcontroller should be
about -40°C and 85°C. This is true for all the microcontrollers considered.

PARAMETER AT91SAM9G25-CM AT91SAM9CN12 ATSAMA5D35-CM

Maximum operating
400 400 532
frequency(MHz)
Max I/O Pins and Ext
105 105 160
Interrupts
Maximum Power
0.89 0.89 1.144
Consumption(W)
Temperature Range −40°C to +85°C −40°C to +85°C −40°C to +85°C
2
No. of SPI and I C interfaces 2 and 3 2 and 2 6 and 3
ADC Channels 12 12 12
ADC Resolution(bits) 10 10 12
ROM (Kbytes) 64 128 160
SRAM (KBytes) 32 32 128

Table 3.1. Comparison between microcontrollers

Based on the comparison in Table 3.1, the ATSAMA5D35-CM was selected as the microcontroller to
be used for the mission.

206
Name CPU CLOCK (MHz) POWER I/O pins SPI TWI UART ADC TIMERS PWM CAMERA COMMENTS
AT91RM9200 ARM920 180 Max:- 25 mA at 1.8 V(45 mW) 122 1 1 5- 6 0 NO less no of interfaces
AT91SAM7A1 ARM7TDMI 40 Operational Mode 607 mW 49 low speed,less i/o
PHILIPS LPC2292 ARM7TDMI-S - Max Input Power:- 600 mW 112 2 1 2 8 4 6- less i2c
TEXAS MSP430F2618 - 16 Ultra Low 912.5 uW 48 2 2 2 12 to 8 2- - low speed,less i/o
MicroChip PIC24E PIC24E ~100 54 uW/ MHz 144 4 2 4 32 9+4 16 - less speed
TEXAS MSP430fr6989 - 16 171 uW/MHz 4 2 2- 5- - low speed
Reneasas H8/3048F - 25 150 uW/MHz 70 - - - - 4 4- low speed
ATMEGA 8 8 bit AVR 16 10.8 mW at 4 MHz 23 1 1 1 8 3 3 No 8 bit,less i/o
PIC 32MZ2048ECM144 200 990 mW at 3.3 V 144 6 5 6 48 9+4 9- 2 stage pipeline, slow execution
AT91SAM9XE512 ARM926 180 13.5 mW at 1.8V 96 2 2 6 4 6 0 YES less i/o
AT91SAM7X512 ARM7TDMI 55 297 mW at 3.3 V 62 2 1 3 4 3 4 NO low speed,less i/o
AT91SAM7XC512 ARM926 55 297 mW at 3.3 V 62 2 1 3 4 3 4 NO low speed,less i/o
AT91SAM9CN12 ARM926 400 234 mW at 1.8 V 105 2 2 7 12 6 4- favorable
AT91SAM9CN11 ARM926 400 234 mW at 1.8 V 105 2 2 7 12 6 4- favorable
AT91SAM9N12 ARM926 400 234 mW at 1.8 V 105 2 2 7 12 6 4- favorable
AT91SAM9RL64 ARM926 240 typical rating-875.66mw, max rating-890mw 118 1 2 5 6 3 4 NO less i2c
AT91SAM9R64 ARM926 240 typical rating-875.66mw, max rating-890mw 49 1 1 5 3 3 3- less i/o,less i2c,less spi
AT91SAM9G45 ARM926 400 typical rating-875.66mw, max rating-890mw 160 2 2 5 8 6 4 YES favorable
AT91SAM9G25 ARM926 400 typical rating-875.66mw, max rating-890mw 105 2 3 7 12 6 4 YES favorable
AT91SAM9G10 ARM926 266 Max 890mW 96 2 1 4- 3 0 NO less i/o
AT91SAM9X25 ARM926 400 Max 890mW 105 2 3 7 12 6 4 NO high speed
AT9SAM9M10 ARM926 400 Max 890mW 160 2 2 5 8 6 4 YES high speed
ATSAMA5D35 Cortex-A5 536 Max 1.144W 160 6 3 7 12 6 4 YES favorable,but high power consumption
ATSAMA5D36 Cortex-A5 536 Max 1.144W 160 6 3 7 12 6 4 YES lcd display, not necessary
ATSAMA5D44 Cortex-A5 536 Max 1.144W 152 8 4 8 5 9 4 YES lcd display, not necessary
Table 3.2: Selection of Microcontroller

Considering all the above characteristics, the AT91SAM9G25-CM, AT91SAM9CN12, and ATSAMA5D35-CMwere shortlisted.

207
Subsystem Functionality and Interfacing
Two microcontrollers are planned to be used redundantly to ensure a fail-safe system. This will
also reduce the computational load on the microcontrollers. The primary microcontroller is used ideally
used for the TTC, Thermal, Payload and Power operations and the secondary microcontroller to perform
ADCS and propulsion operations. In the unlikely event for one of the microcontrollers to fail, the other
takes over its operations. This is realized by setting up common connections and a common data pool.

Figure 3.1: Interfacing between the microcontrollers

The On-board RTC module, the clock of the RADOM, clock of the star tracker and the clock
at the Ground station are all synced before launch.
The OBC is responsible for interfacing with all the different subsystems, establishing
communication, performing computations and making decisions. The interfacing details between the
various components and the OBC have been aggregated with respect to each subsystem.

Figure 3.2: Legend for *Subsystem Interface Diagrams

* % represents data transfer between the component and the microcontroller

** #represents Status Commands from the microcontroller

208
 State command stands for a digital 1/0 command to the components. Status commands like Sun
Sensor status, MPPTU status, etc. sends request from µC to components for data retrieval or control
configurations.

ADCS Subsystem

Figure 3.3: ADCS Subsystem interface

Power Subsystem

Figure 3.4: Power Subsystem Interface

209
Propulsion Subsystem

Figure 3.5: Propulsion Subsystem Interface

Thermal Subsystem

Figure 3.6: Thermal Subsystem Interface

210
TTC Subsystem

Figure 3.7: TTC Subsystem Interface

3.2 MEMORY BUDGET

Description
Memory is one of the essential resources on a spacecraft. Not only is it going to contain the
actual program that is to be run by the satellite and provide a workspace for the ongoing processes, as
it is going to store data obtained from different subsystems as well. Any memory lapse might lead to
unintended consequences which might, in turn, be fatal to the mission. Six different types of memory
have been classified on-board the proposed satellite. Each type of memory needs to be of certain
specifications.
The microcontroller ATSAMA5D35-CM was selected. The CPU Module for the
microcontroller provides various memory options for each purpose. Analysis of different memory types
available for consideration and the requirements for the memories to be used for a certain purpose is
done below.
It is important to consider the effects of degradation of electronic components of the on-board
computer as a result of exposure to harsh conditions. Preventive measures should be taken to prevent
any damage to any of the critical information available on the system. Even if failures occur, the design
should be robust enough to avoid catastrophic results. Hence, there should always be an option to regain
the lost information, if possible.
Different types of the memory are:

211
 ● Program Memory
● Data Memory
● Bootable Memory
● Redundant Memory
● Log Memory
● Payload Memory

Requirements
The following are the requirements based on the type, size; and read and write access speeds of
the different memories discussed above†.
Type Size Speed
Program Memory Non- ~4 Mbyte Fast read
volatile
Data Memory Volatile ~3.5 Mbyte Fast read and write
Bootable Memory Non- 64 Kbyte Fast read
volatile
Redundant Non- ~16 Mbyte Fast read
Memory volatile
Log Memory Non- TBD Fast write
volatile
Payload Memory Non- TBD Fast read and write
volatile
Table 3.3: Requirements for different memory types

†
Larson, J. W., Wertz, R. J., “Space Mission Analysis and Design”, Third Edition, Microcosm Press and Kluwer
Academic Publishers, Tables. 16-13, 16-15.

212
 Size (Kwords) Complexity Size (Kwords) rev.
Functions code size data size code complexity data complexity code size data size
command processing 1 4 4 2 4 8
Communications
telemetry processing 1 2.5 4 2 4 5
rate gyro 0.8 0.5 2 2 1.6 1
Attitude sensor processing sun sensor 0.5 0.1 2 2 1 0.2
star tracker 2 15 2 2 4 30
kinematic integration 2 0.2 2 2 4 0.4
error determination 1 0.1 2 2 2 0.2
precession control 3.3 1.5 2 2 6.6 3
Attitude determination & Control thruster control 0.6 0.4 2 2 1.2 0.8
reaction wheel ctrl 1 0.3 2 2 2 0.6
ephemeris propagation 2 0.3 2 2 4 0.6
orbit propagation 13 4 4 4 52 16
Autonomy simple autonomy 2 1 2 2 4 2
power management 1.2 0.5 3 2 3.6 1
Other functions
thermal control 0.8 1.5 2 2 1.6 3
monitors 4 1 2 2 8 2
Fault detection
fault correction 2 10 4 2 8 20
Payload payload cdh 8 40 2 2 16 80
TOTAL 127.6 173.8

executive functions 3.5 2 4 2 14 4

Run-time kernel 8 4 6 4 48 16
Operating System Equipment Handlers 28 9.8 2 2 56 19.6
test diagnostics 0.7 0.4 2 2 1.4 0.8
math utilities 1.2 0.2 2 2 2.4 0.4
TOTAL 121.8 40.8

PARAMETERS
d (number of equipment handlers) 14

TOTAL (in Kwords) 249.4 214.6

TOTAL (in Kbyte) 997.6 858.4

Table 3.4: Program & Data Memory Estimate

Approach
The various memory options provided by the proposed microcontroller CPU module,
ATSAMA5D35-CM, have been analyzed for the required memory types discussed previously.

Memory Device Size

Program Memory NOR Flash - External * 16 Mbyte
Data Memory DDR2 SDRAM ** 512 Mbyte
Bootable Memory ROM - Internal *** 160 Kbyte
Redundant Memory NAND Flash (with ECC) 256 Mbyte
Log Memory - External
Payload Memory
Table 3.5: Approach towards memory requirements

* As per our estimates, the 160 Kbyte internal ROM is not sufficient to hold the program
memory. Hence, we will store the program memory in the 16 Mbyte NOR flash. The program in the
NOR flash can be executed by either using eXecuteIn Place (XIP) or by copying into RAM before
execution.
** Since the 128 Kbyte SRAM is not sufficient for the proper functioning of the program as per
estimates, the DDR2 RAM interfaced with the microcontroller will be used. The size of the data
memory is more than the required because the SAMA5D35-CM is available only in these specifications.
*** The microcontroller boots by default into this memory. This will contain the bootstrapping
code for loading and transferring the flow of control to the program memory.

213
 3.4 SOFTWARE ARCHITECTURE
An On Board Software (OBSW) of a spacecraft platform is the program which controls the OBC
and implements the following functions:

 Telecommand processing for spacecraft command from ground

 Generation of data for telemetry
 Spacecraft control comprising of:
o Attitude and orbit control
o Power control
o Thermal control
o Payload instruments operation
 System status monitoring
 Fault detection, isolation, and recovery

Figure 3.8: Software Architecture

The representation of static architecture with the functional modes tied to various OBSW
hierarchy levels provide an adequate description of the various functional modules and their interfaces.
The OBSW static architecture consists of the following main components:

 Operating system and drivers layer

 Equipment Handlers and OBSW Data Pool (OBSW DP)
 Application Layer
 Service Handlers
 Fault detection, isolation, and recovery

Operating System and Device Drivers Layer

This layer comprises of the Bootloader, the RTOS, the Kernel and the drivers. It interacts with
the hardware which is connected to it. The bootloader programmed in the boot memory and will
initialize the microcontroller to be able to boot the Real-Time Operating System (RTOS) and the OBSW
layers built on top, from the program memory. Memory layer checksum verification and corrections are
performed during the initialization. The RTOS is booted first, and the non-operating system part of the
OBSW is started after RTOS. The RTOS kernel is used to:

214
  Control software task scheduling
 Control of OBSW start sequence after boot loading with
o Initialization of all components
o Interlinking of all OBSW modules
o Startup of task operations
 Initialize threads for applications, handlers
 Log boot reports
 System log of activated/deactivated equipment

Drivers for I/O channels are interfaced between the OBSW and hardware. These include I2C,
SPI, serial, etc. drivers. These are interface drivers.

Equipment Handlers and OBSW Data Pool (OBSW-DP)

Above the RTOS, kernel and the drivers, reside the equipment handlers. They perform
command writing from the OBSW via drivers to the connected devices like actuators, switches, etc. [3].
Furthermore, the equipment handlers will perform cyclic or on request onboard telemetry acquisition.
The implementation concept for the equipment handlers will be one handler per equipment type, for
example, one handler for a set of 8 sun sensors, one for a set of 4 RCS thrusters, etc. The handlers will
be responsible for necessary compatible format conversions of the data between OBSW data pool and
the physical line. It will retrieve telemetry data from the connected equipment, use the corresponding
driver and place them into the corresponding variable slots in the OBSW DP.
The OBSW data pool (DP) is a vector containing all operational variables which are processed
inside the OBSW such as system time, sensor data, actuator data, temperatures, current and voltages,
payload data, etc. It is mandatory that the variable name in both the OBSW DP and the OBSW code
should be same. The same name should be used in telecommand and telemetry packets too. A flag
corresponds to outdated data in the DP when handlers are unable to access an updated data from the
connected device.

Application Layer
The application layer (above the OBSW DP) controls:

 AOCS
 Power subsystem
 Thermal control
 Telecommand and Telemetry
 Payload instruments
 Health monitor
The applications read input data from the OBSW-DP and place computed output back into other
variables of the OBSW DP. Each application has access to any OBSW DP variable. Some variables are
shared with all the applications like system clock on board time. Each application runs in different
threads with different update cycles which depend on the request. Each application is defined to perform
based on the corresponding operational mode.
Health monitor primarily focusses on health check-ups of OBSW. It stores the following
information as pair of name and value:

 Preselected set of OBSW variables from each domain of OBSW DP

 Memory pointer, timing parameters, RTOS flags and other OBSW internal variables
 Task scheduling parameters for selected applications
 Data bus access flags, etc.

215
 The objective of the health monitor is to:

 Check for proper functioning of the threads

 Ensure Boundary limitations on the initialized parameters
 Direct logging of the parameters

Service Handlers
The service handlers reside beside the application layer. It is dedicated to extract and process
the services mentioned in the telecommands, and service tag ID with the telemetry for sending a report
of respective services to the ground station. Services refer to on-board software commands defined
uniformly and distinctly both in the ground station and the spacecraft. Here, PUS (Packet Utilization
Standard) defined by ECSS standards are considered as the service handlers. It reserves numbers 0-127
to represent pre-defined services and numbers 128-255 are free for mission specific use. Each of the
services has a dedicated handler. For example, as per PUS, Service 1 represents telecommand
verification service. So, whenever the service 1 is called, telecommand verification is handled by the
specific handler assigned to it. Such service handlers need a proper mechanism to route
telecommand/telemetry between different applications to perform the required service. Unique
application IDs are defined for routing of services from the service handlers to other components in the
OBSW.

Fault Detection, Isolation and Recovery (FDIR)

FDIR functionality is implemented on several levels inside the OBSW. Fault detection is
performed in all applications and handlers. Based on fault detection, fault isolation and recovery
involves:

 Compliance with Electrical Power System to handle electrical faults and failures
 Triggering a reconfiguration to the redundant counterpart of the failed equipment
 Switching to Safe Mode in case of low power or during long duration of eclipse

Requirements for FDIR design include:

 A definition of clear hierarchy, identification of failure type and management on the particular
FDIR level.
 Ability of the Ground Station to perform a detailed status analysis and failure event history
analysis for unique failure identification.
 Ability of the Ground Station to alter operational limits to avoid future Safe Modes – e.g. in
cases of failures triggered by equipment degradation.
 Provision for OBSW patch and dumps
 Accountability for all the envisioned failures during spacecraft operations based on their
symptom sets

216
 Figure 3.9: FDIR Hierarchy Level

Figure 3.9 depicts the identification and propagation of faults through different levels and the
recovery process that follows. If recovery fails even after level 1, a system reconfiguration is done, or
the operational mode switches to safe mode.

3.5 ERROR DETECTION AND CORRECTION

The Error Detection and Correction schemes discussed here target secure transmission of data from
various memory devices to the Central Processing Unit of the On Board Computer. The following
methods can be implemented to transfer data with minimal bit flips that might occur as a result of Single
Event Upsets (SEU) by using Error Detection and Correction codes. Some of the EDAC techniques are
listed below.
Triple Modular Redundancy
The general approach towards EDAC involves the use of Triple Modular Redundancy
(TMR). In TMR, the redundant memory consists of two copies of the data. A voting mechanism is
carried out, and the majority is selected to be transferred. This mechanism is capable of providing
correction for multiple bit flips in data. This would not ensure security for severe errors where different
errors might occur at different positions of each copy of data, and the voters will not be able to detect
the correct (majority) data. In such cases, we might not be able to transfer the correct data. The triple
modular redundancy is expensive in terms of memory because it involves a 200 percent increase in the
memory, but considering the memory budget, sufficient memory is available to accommodate the
storage of the program code and only the important data variables needed for computations.
In TMR, three different copies of data are present which need to be checked periodically, reducing the
efficiency of the program due to the read and runtime overhead. For important data variables, this

217
overhead cannot be avoided since they are critical to the mission, and their accuracy needs to be ensured.
This overhead in program memory is minimized in this design as a check for errors in the program
memory, and important data variables are carried out only during boot up and data scrubbing. The
scrubbing interval duration should be decided in the later stages.
Before maneuvering, a system check is carried out, and output report is sent to ground station whenever
communication link is available. As this report is critical for making maneuvering decisions, a TMR
check would be performed before the information is transmitted.
Some prominent software error correction techniques are Reed-Solomon Codes, BCH codes, and
Hamming codes.
Hamming Codes
Hamming codes are capable of detecting two errors at maximum and can correct only 1 bit without
detection of uncorrected errors. Considering this reduced capability of bit correction and the
consequences of this uncorrected data being large, Hamming codes are not preferred.
Reed Solomon
The Reed-Solomon coding method is highly efficient in terms of storage, with very less overhead
(3.125%). However, the encoding and decoding take longer time making it slower than TMR. Since the
processing unit requires faster memory access in terms of important data variables, this scheme is
preferred only for Log and Payload memory.
Depending on the frequency of occurrence of these errors, the data is corrected periodically. Since
Reed-Solomon can correct up to 8-bit flips, corrected errors are updated with the help of the check bits
that are obtained while decoding. The duration of this periodic correction will be decided in later stages.
BCH codes
Though Reed-Solomon and BCH codes have similar performances, Reed-Solomon coding performs
better in burst error cases.

218
 1. STRUCTURE
The structural subsystem was involved in designing of the preliminary structure of SRMSAT – 2 with
a primary objective of reducing the structural mass of the spacecraft without compromising its structural
efficiency. All the components for the mission were modelled using Siemens NX and the design had
ample space to accommodate all the components in a micro-class satellite.

1.1 COMPONENTS
Base Plate
The base of the satellite is a 500mm × 500mm honeycomb sandwich structure with a hole at the centre
of the base plate to provide room for the PPS thruster. The facings of the honeycomb base are 1mm
thick, and the core has a thickness of 3mm. Honeycomb sandwich construction makes the base plate
structurally efficient due to their high strength – to mass ratio. The base plate has a hole of 49mm
diameter in order to accommodate the PPS thruster.

Figure 1.1: Plan of the Honeycomb Sandwich Base Plate

Side Panels
The side panels are again honeycomb plates covering an area of 500mm×600mm. These are among the
largest components of the spacecraft. They provide housing for all the COTS (Commercially off the
shelf) components as well as the propulsion module of SRMSAT – 2. A draft of the side panels is shown
below:

219
 Figure 1.2: Plan of the Side Panel

Top Deck
The top deck of the spacecraft is also a honeycomb structure covering a surface area of 500mm ×
500mm . As the top deck also needs to be structurally very efficient, honeycomb deign is chosen for
the top deck which provides the housing for the sun sensors, an omnidirectional antenna, and the
radiation dosimeter.

Figure 1.3: Plan of the Top Deck

220
Propulsion System
The propulsion system constitutes of one of the most noteworthy modules of the spacecraft. It is
responsible for providing incremental velocities to the satellite at its Aphelion and its Perihelion.
Further, it accounts for the attitude control of the spacecraft. SRMSAT - 2 uses a liquid propulsion
system in which the PPS (Primary Propulsion System) uses a bipropellant propulsion system consisting
of Hydrazine (N2H4) as the fuel and Nitrogen Tetroxide (N2O4) as the oxidizer while RCS (Reaction
Control System) uses a monopropellant propulsion system containing the same fuel as used in the PPS.
The thrust provided by the PPS thruster is 22N, and the RCS thruster is about 1N.
Tanks: The fuel and the oxidizer are in 2 cylindrical tanks with hemispherical caps pressurized by
helium supplied from a single spherical tank. In line filters are used to filter the pressurizer and the
propellants. The fuel, propellant, and oxidizer are filled into propulsion subsystem through many fill
and drain valves which allow offloading if necessary. Check valves are used to prevent vapour
migration to the pressurizer tanks. There are pressure transducers which measure the amount of pressure
in the tanks. Further, a redundant pressure regulator is present in the Helium tank to control the pressure
in the tank. There are latch valves at the entry and exit of each tank to keep the tanks isolated from other
components when not in use or malfunctioning. There are flow control valves which control the
flow/pressure of the fluid. A latch valve near the entry of the PPS thruster prevents fuel flow when the
RCS is functioning as the PPS is not required at this time.

Figure 1.4: Oxidizer Tank Figure 1. 5: Fuel Tank

Figure 1.6: Pressurising Tank

221
THRUSTERS:
Primary Propulsion System (PPS): The Primary Propulsion System (PPS) is accountable for providing
the satellite its incremental velocities (∆v) near its aphelion. The PPS thruster chosen for SRMSAT – 2
is a 22N bi-propellant thruster available for sale by AIRBUS’ Space-Propulsion unit. It is a single seat
valve 22N bipropellant thruster and weighs about 380g. The nozzle end diameter is 52mm.

Figure 1.7: PPS Thruster

Reaction Control System (RCS): The RCS thrusters are used to adjust the attitude of the satellite in the
orbit by giving minute thrusts. SRMSAT – 2 uses 4 reaction control thrusters to stabilise the spacecraft.
The RCS thruster chosen for SRMSAT – 2 is a 1N thrust redundant seat thruster from MOOG Inc.

Characteristics Value
Thrust 1N
Weight 218g
Operating Temperature Range 4.4 to 149℃

Table 1.1: RCS Thruster Specification

222
 Figure 1.8: RCS Thruster

PCB Module: The PCB module of SRMSAT – 2 is a housing which is capable of holding 4 PCBs which
delivers high adeptness and resourcefulness in performing multiple tasks. The four PCBs in SRMSAT
– 2 are for the following subsystems:

 OBC – Flight Dynamics

 OBC – Master
 Power
 Communication

Figure 1.9: PCB Module

Battery Module
The Battery Module consists of 24 cells powering the satellite with a peak power of 145W. The battery
module is positioned on the Side Panel 3 of the spacecraft as the battery module is considerably large
by volume and placing it on Side Panel 3 avoids any collision with the propulsion module.

223
 Figure 1.10: Battery Module with 24 Cells

Sun Sensor
The spacecraft is equipped with 8 sun sensors having a field of view of 120°. There are 3 sun sensors
on the top deck and 5 sun sensors on the back panel which are facing along the – ve z-axis. Each sensor
weighs about 50gm, and the working range of the sun sensors is -20 to +40℃.

Figure 1.11: Sun Sensor

Reaction Wheels Module

SRMSAT – 2 is 3–axes stabilised using 4 reaction wheels positioned in a tetrahedral geometry. The
SSTL SP-10M reaction wheels weigh about 0.96kg and can withstand a random vibration of 18 g rms.
The natural frequency of the above reaction wheels is sufficiently high (first mode frequency >
400𝐻𝑧). Its operating temperature is -20 to +50℃ and survival temperature is -30 to +60℃.

Figure 1.12: SSTL SP-10M reaction wheels

224
Solar Panels
The satellite has 3 solar panels, 2 of which are deployed, and one is mounted on the side panel – 1 of
the satellite. The deployment of the 2 solar panels uses pyrotechnic fasteners which when fired, releases
the torsion spring, and the solar panels are deployed parallel to the solar panel mounted on one of the
faces of the satellite. The solar cell configuration of SRMSAT – 2 is 12s14p with 6 strings in parallel
mounted on side panel – 1 and 4 strings each on the deployed panels.

Figure 1.13: Solar Panel Configuration

Antenna
SRMSAT – 2 has a parabolic antenna with a maximum diameter of 52cm placed on the back face of
the spacecraft and two omnidirectional helicoidal antennas placed on top and bottom faces of the
spacecraft. Placing the antenna on the back face of the spacecraft avoids any chances of occlusion of
the parabolic antenna with the omnidirectional antenna.

Figure 1.14: Parabolic Antenna

225
 Figure 1.15: QFH Antenna with Substrate Figure 1.16: QFH Antenna

Radiation Dosimeter
Radiation Dosimeter is the secondary payload of SRMSAT – 2 which will be used to monitor the plasma
particulate radiations en route to and in the lunar orbit. It calculates the amplitudes of the incoming and
accumulated radiations in radiation absorbed dose.

PROPERTY VALUE
Dimensions 10 × 5 × 2 cm3
Weight ≈ 98gm
Detector 0.3 mm thick Si-PIN Semiconductor Type, 2 cm2 area

Table 1.2: Physical Properties of Radiation Dosimeter

Figure 1.17: Radiation Dosimeter

Narrow-Angle Camera
A narrow-angle camera will be used as a tertiary payload for SRMSAT – 2. The camera to be used is
IM-200 from Hyperion Technologies. The chosen lens configuration is F50 mm with film size 16 mm.
This gives it a field of view of around 18°. The image resolution is 4 Mega Pixels (2048 × 1944). It
has a full frame frequency of 5 Hz and an image buffer storage of 25 images.

226
 Figure 1.18: Narrow-Angle Camera

Star Tracker
Star tracker is the most important deterministic sensor and is placed on the side panel 2. It tracks the
position of the stars and generates the quaternions with respect to the body axis.

Figure 1.19: ST400 Star Tracker

1.2 DESIGN AND SAFETY FACTORS

Factors of safety are used in space missions to account for uncertainties with regard to the prediction of
loads, structural analysis, the fabrication process and material properties.

227
 Design Limit Yield Loads
Loads

Flight Limit Qualification Ultimate

Loads Loads Loads

Flight
Acceptance
Loads

Figure 1.20: Mechanical Loads to be considered for Spacecraft Design

Flight Limit Load

The flight limit load for a given design condition is the maximum occurring load. The stress that is
calculated with the applied flight limit load is called the Limit Stress.

Design Limit Load

The design limit load (also known as the Qualification load) is the limit load multiplied by the factor of
safety to avoid risks during the design and test phase.

Yield Load
The yield load is the design load multiplied by the yield safety factor. It is the load that the bus can
withstand without any plastic deformation.

Proof Load
The proof load is the design limit load multiplied by the proof factor. The proof load is used to test parts
of the structure before the entire spacecraft or launch vehicle is tested.

Margin Of Safety
The margin of safety is the ratio of the allowable stress to the actual stress multiplied by the factor of
safety minus one; which implies that MOS is never negative.
σmax
MOS = −1≥0
σactual × f
(1)

σmax = Permissible Stress

σactual = Actual Stress due to design limit loads
f = Factor of safety

228
 1.3 MATERIAL SELECTION
An efficient satellite structure demands a material capable of withstanding the conditions expected with
a suitable factor of safety while having low weight and cost. The material chosen for the satellite
structure is Aluminium-6061 T6 alloy. 6061 T6 is a precipitation-hardened aluminium alloy, containing
Silicon, Magnesium and Iron as its major alloying elements.6061 T6 offers the best strength to weight
ratio for the given cost range.

Mechanical Properties Metric

Density 2.7 g/cc CTE, linear 68°F 23.6 µm/m-°C
Hardness, Brinell 95 CTE, linear 250°C 25.2 µm/m-°C
Ultimate Tensile Strength 310 MPa
Specific Heat Capacity 0.896 J/g-°C
Tensile Yield Strength 276 MPa
Thermal Conductivity 167 W/m-K
Modulus of Elasticity 68.9 GPa
Melting Point 582 - 652 °C
Poisson's Ratio 1/3
Solid 582 °C
Fatigue Strength 96.5 MPa
Shear Modulus 26 GPa Liquid 652 °C

Shear Strength 207 MPa

Table 1.4: Thermal Properties of Al-6061 T6

Table 1.3: Mechanical Properties of Al-6061 T6

229
 1.4 MASS BUDGET
Components Mass
(kg)
Antenna Mount Back Face 0.081
Base plate Honeycomb 1.96
Battery Module 1.95
He Tank 1.14
Hinge × 𝟒 0.81
Hydrazine Tank 4.02
Narrow-Angle Camera 0.03
NTO Tank 5.48
Parabolic Antenna Dish 1.57
PCB Module 0.62
Power Amplifier 0.38
PPS Thruster 0.61
QFH Antenna × 𝟐 0.3
Radom 0.1
RCS Mount × 𝟒 0.13
Reaction Wheels × 𝟒 3.86
RCS Thruster × 𝟒 0.87
Side Panel Hcp 1_3 × 𝟐 4.63
Side Panel Hcp 2_4 × 𝟐 4.47
Solar Panel 9.92
Star Tracker 0.4
Sun Sensor Top × 𝟐 0.16
Sun Sensor Module × 𝟔 0.44
Top Deck Honeycomb 1.97
Transponder 2.6
Hydrazine 13.72
NTO 27.78
Pressurizing tank 0.29
Total Mass 86.43

Table 1.5: Mass budget of the BUS of SRMSAT - 2

230
 1.5 SPACECRAFT DESIGN LOADS
Transients such as engine ignitions and shutdowns of the Launch Vehicle, wind gusts and shears and
quasi-static load factors drive the spacecraft low-frequency loads. Other environments affecting the
spacecraft low-frequency loads may be acoustics, random vibration, sine vibration and shock.

Acoustics Random Sine Vibration Shock

Vibration

☑ ☑
Lift-off

☑ ☑
Aerodynamics

☑
Separation

☑ ☑
Ignition

Table 6: Sources of Launch Vehicle Environments

Mechanical Dynamic Loads

The mechanical dynamic loads considered for the design of SRMSAT – 2 can be divided into:

 Low-frequency sinusoidal vibrations in a frequency domain of 5-100 Hz

 Random Vibrations in a frequency range of 20-2000 Hz (Will be discussed in the critical design
phase).

Sinusoidal Vibrations
Low-frequency sinusoidal vibrations occur as a result of the interaction between Launch
Vehicle and the Spacecraft occurring during Lift-off, Ignition, and POGO vibrations. POGO vibrations
are observed just before the burn up of the launch vehicle stage. PSLV user manual was considered for
the study of Launch Vehicle environmental effects and frequency requirement for spacecraft to avoid
dynamic coupling with the launch vehicle are:

 Launch Direction: f0 ≥ 90Hz

 Lateral Direction: f0 ≥ 45Hz

Influence of Natural Frequency

The Spacecraft-Launch Interface can be equated to a mathematical model of spring-mass
system with 2 DOF as shown:

𝑚1

𝑘1

𝑚2

𝑘2

Figure 21: Mathematical Equivalent of the Dynamic Spacecraft-Launcher Interfacing

Where m1 and k1 represent the spacecraft and m2 and k 2 represent the Launch Vehicle.

231
In this case, the system with the highest natural frequency is significantly higher than the natural
frequency of the total system.

1 𝑘1
𝑓1 = √
2𝜋 𝑚1

(2)

The lowest frequency of the total system is given by Dunkerley’s equation:

1 1 1
= +
𝑓𝑡𝑜𝑡𝑎𝑙 𝑓1 𝑓2
(3)
With
1 𝑘2
𝑓2 = √
2𝜋 𝑚1 + 𝑚2

(4)
f2 is the natural frequency of the total system with System 1 acting like a rigid body.

1.6 SANDWICH CONSTRUCTION

The structure of SRMSAT – 2 designed for the base plate, top deck and side panels of the satellite are
honeycomb structures made of Aluminium-6061 T6. Using a sandwich construction for the structure of
the spacecraft has the listed advantages:

 Mass savings with respect to conventional plates

 High Strength-to-Mass ratio
 Good Fatigue Properties
 Sound damping properties
 Good thermal and acoustical insulation properties

Adhesive CORE Face Sheets

Figure 22: Sectional View of Honeycomb Sandwich

232
The design aspects considered for the design of SRMSAT – 2 are:

 The facings have sufficient thickness to carry all the tensile, compressive and shear stresses
introduced in the sandwich structure.
 The core has adequate strength to carry the general shear forces and the forces introduced via
the inserts.
 The core has adequate strength and stiffness properties to prevent local and global buckling.
 The sandwich construction has optimum equivalent stiffness properties.

Transformation of Honeycomb Sandwich Plates:

Honeycomb structures are very efficient construction for high strength light materials but can
be quite tedious to study. They are therefore studied as flat plates with isotropic behaviour by
transforming the honeycomb sandwich structure into their equivalent isotropic plate by Kirchhoff-Love
Plate theory which defines the bending stiffness and tensile of a plate respectively as:
𝐸ℎ 3 𝐸ℎ
𝐷= and𝐾 =
12(1−𝜈 2 ) 1−𝜈2

(5)
Where E = Young’s Modulus
h = Thickness of the plate
ν = Poisson’s Ratio

In order to study the honeycomb structures, we define the equivalent Young’s Modulus and equivalent
thickness of the Honeycomb Plate so that:
𝐸 → 𝐸𝑒𝑞 and ℎ → ℎ𝑒𝑞

(6)
Consider the following Honeycomb:

Figure 23: General Honeycomb Plan

ℎ𝑓

ℎ𝑐 ℎ𝑒𝑞

ℎ𝑓

233
 Figure 24: Honeycomb Elevation and equivalent Plate Elevation

Where, l = size of Honeycomb Cell

t = Rib thickness
hf = Facing Thickness
hc = Core Thickness

Transformation Equations:

 Equivalent Thickness heq = √h2f + 12h2c

h2f
 Equivalent Young’s Modulus Eeq = 9√2 h ×E
c heq
2ρf hf +2ρc (H−hf )
 Equivalent Density ρ=
heq

h
Where, H = hc − f⁄2

ρf , ρc = Density of face and core respectively

 The honeycomb sandwich chosen for the design of SRMSAT – 2 had the following dimensional
specifications:
l = 30mm
t = 4mm
hf = 1mm
hc = 3mm

 The honeycomb sandwich (facing as well as core) is made of Aluminium-6061 T6 which has
the following Mechanical Properties:

E = 68.98 GPa
kg⁄
ρf = ρc = 2711
m3
ν = 1⁄3

The Honeycomb Sandwich was transformed into an equivalent plate using the above specifications:
heq = 10.44mm

Eeq = 28032 MPa

kg⁄
ρ = 1817.7
m3
The results obtained after the transformation was compared with the FEA on the untransformed
Honeycomb Model in order to validate the transformation and the variation of the deflection of the
centroid of the plate with frequency was plotted for both the models are shown below:

234
Figure 25 - Displacement variation of the Plate Centre with Frequency

Figure 26: Variation of Plate Centre with Frequency

Modal Analysis
As discussed above from the PSLV User Manual, PSLV excites vibrations into the spacecraft
during the Launch and therefore in order to avoid dynamic coupling of the satellite with the Launch
Vehicle, the minimum natural frequencies as:
Longitudinal Vibrations: f0 ≥ 90Hz
Lateral/Transverse Vibrations: f0 ≥ 45Hz
It is therefore of prime importance to carry out a Modal Analysis on the design BUS of SRMSAT – 2
in order to determine the natural frequencies of longitudinal plates and horizontal plates with sufficient
non-structural mass.

Base Plate
The base plate of the spacecraft as mentioned above must have a natural frequency ≥ 90Hz, therefore,
its finite element model using NX NASTRAN. The Finite Element information is shown below:

235
 Type of Mesh 3D Mesh

Type of Element PSOLID (CTET4)

Element Size 8 mm

Plate Thickness 5mm

Material Al-6061 T6
Table 1.7: Finite Element Model Description for Base Plate

The Finite Element Model of the Base Plate is shown below:

Figure 1.27: Finite Element Model of the Base Plate

NX NASTRAN solver was used to perform the Modal Frequency Analysis using Real Eigen Value
method after applying the following simulation constraints:

Solution Type SOL103 – Real Eigen Values

Constraints Edges with C-C-C-C condition
Non-Structural Mass 10 kg
Table 1.8: Solver Environment Information

The first 5 modes of free vibration were calculated using FEA and the results obtained are shown below:

236
 Mode Frequency (Hz)
1 284.2
2 643.63
3 656.28
4 1022.79
5 1238.29
Table 1.9: First 5 modes of free vibration

Figure 1.28: Mode 1

Figure 1.29: Mode 2

237
 Nodal Displacement vs. Frequency
1.80E+00
Nodal-Displacement (mm) 1.60E+00
1.40E+00
1.20E+00
1.00E+00
8.00E-01
6.00E-01
4.00E-01
2.00E-01
0.00E+00
0.00E+00 2.00E+02 4.00E+02 6.00E+02 8.00E+02 1.00E+03 1.20E+03 1.40E+03
Frequency (Hz)

Y1 Y2 Y3

Figure 30: Variation of displacement of nodes at different distance from centre with frequency

Linear Static Analysis

The maximum loads that the spacecraft needs to sustain are the 2σ and 3σ loads during the
Launch Period, which is induced by the Quasi-Static loading factors of the Launch Vehicle. The PSLV
user manual was considered for preliminary study, and the PSLV quasi-static loading factors are
mentioned below:

Subcase Longitudinal Lateral 1 Lateral 2

1 13.75g 7.5g -
2 13.75g - 7.5g
3 -13.75g 7.5g -
4 -13.75g - 7.5g
Table 1.10: PSLV's Quasi-static Loading Factors

The results for the above subcases are shown below:

Subcase Maximum Maximum
Displacement(mm) Stress(MPa)
1 0.0704 2.957
2 0.0703 2.974
3 0.0703 2.972
4 0.0704 2.973
Table 1.11: Linear Static Analysis Results

238
Figure 1.31: Subcase 1 Nodal Displacement

Figure 1.32: Subcase 1 Stress Distribution

239
Figure 1.33: Subcase 2 Nodal Displacement

Figure 1.34: Subcase 2 Stress Distribution

240
Modal Analysis On COTS Components
Battery Module
Modal Frequency analysis was performed on the battery module of SRMSAT – 2 whose components
are fastened using M2 tapped bolts, and the battery module is fixed to the side panel of the spacecraft
and the results obtained are shown below:

Figure 1.35: Mode 1 Displacement

Figure 1.36: Mode 2 Displacement

241
 2. THERMAL SUBSYSTEM

2.1. OVERVIEW
The thermal system is responsible to maintain operational temperatures of the individual components
and the spacecraft as a whole, within the respective survival ranges in all the mission phases. With the
mission featuring deep space trans-liberation cruise, lunar orbiting and Earth orbiting phases, the
spacecraft faces a broad spectrum of thermal and radiation scenarios. Evaluation and quantification of
its effect on the spacecraft is elementary to the spacecraft design for the mission. In the preliminary
design, the worst case thermal conditions are identified and simulated for, using analytical and
numerical solutions. The thermal control includes passive control consisting of Multi-Layered
Insulation and conformal coatings for electronics and, black paints. A heater would be included for the
active thermal control, if needed in the critical design.

2.2. INTRODUCTION AND BACKGROUND

The SRMSAT-2 mission is designed to be launched into a GTO followed by two phasing orbits of
apogee close to 90000km and 130000km. The phasing orbits might include multiple coast cycles
increasing pass cycles through the Van Allen radiation belts. The spacecraft houses a dual propulsion
system with hydrazine and NTO, and helium as the pressurizer; sensitive to the thermal variations. The
heat transfer within the bus is only due to conduction and radiation. While, the interaction with the
environment occurs through radiative heat transfer. The earth orbiting phase spacecraft suffers thermal
loads due to the Earth’s albedo and planetary radiation along with the solar radiation. The cruise phase
just suffers from the solar radiation. While, similar to the Earth orbiting phase, in the lunar orbiting
phase the spacecraft suffers from lunar albedo and the planetary radiation. The importance of thermal
analysis and subsequent need of control elevates as the spacecraft draws in closer to the moon along the
mission timeline. The priority turns then to the scientific mission phase, when the spacecraft is
commissioned into the lunar science orbit of 250 km altitude around the moon for an observation period
of 6 months. In this observation period, the moon completes on diurnal cycle. The spacecraft in this
orbit has a period of around 132 minutes. With maximum thermal loading of 1250 W/m2 and the
minimum of 5 W/m2, the thermal suffrage averages to about 320 W/m2 attributing to the non-uniform
lunar surface distribution. The spacecraft undergoes excessive thermal load fluctuations going in and
coming out of the lunar shadow. The spacecraft faces the situation of catastrophic freezing in the lunar
orbiting science mission phase. These form the baseline for thermal analyses. The heat load
approximations for the worst case were evaluated. Initial first order analysis was done analytically
followed by numerical formulation and analysis using MATLAB.

242
Objectives
Components:
 Maintain reliability index of COTS components within the system with temperature ranges
within safe limits.
 Estimate and provide essential radiation shielding measures to prevent failure of electronic
components due to excess accumulated radiation dose
Propulsion system:
 Counter excess heating in the combustion chamber and freezing in the propulsion
plumbing.
Spacecraft Body:
 Counter thermal stress fluctuation on the spacecraft structure in the lunar orbiting phase.

Thermal requirements
The Thermal system requirements of a spacecraft consist of the survival and operational temperature
ranges of each component in the spacecraft bus. For few components, the temperature ranges or limits
are identical. The following table enlists the allowable temperature ranges of all the major components
of the spacecraft:

COMPONENT SURVIVAL LIMITS OPERATIONAL LIMITS

Fuel (Hydrazine N2H4) 0 to 65°C 0 to 45°C
Oxidizer (Nitrogen tetroxide N2O4) -5 to 50°C -11 to 38°C
Propellant and Oxidizer tanks 0 to 50°C 5 to 45°C
Valves -10 to 115°C 0 to 75°C
Thruster (fluid conditions) -9 to 104°C 0 to 75°C
Reaction Wheels -20 to 60°C 0 to 50°C
All COTS components -45 to 85°C 0 to 60°C
Microcontroller -40 to 85°C -40 to 85°C
Star Tracker -30 to 50°C -20 to 40°C
Gyros/IMUs -50 to 90°C -40 to 85°C
Battery -40 to 100°C -40 to 85°C
Solar panels -200 to 130°C -150 to 110°C
Sensors -30 to 65°C -30 to 65°C
Sun Sensor -30 to 70°C -20 to 60°C
TTC subsystem -10 to 70°C -10 to 70°C
Table 2.1: Thermal Requirements

Thermal design
The thermal design process considered is two-fold. The design is selected for the spacecraft body as a
whole followed by determination and definition of a heat sink. Then, the means to maintain
temperatures for internal and external components is devised and included in the design. The most
common thermal design for the spacecraft in 3-axis stabilized modes uses simplistic thermal control
methods. The spacecraft operations in spin stabilized modes, if any (currently none), can be configured
in the thermal design, with a spinning solar array as heat sink to the internal components of the
spacecraft. The simpler shape of the spacecraft – cuboidal bus with flat plate deployments, make it

243
suitable for the use of the former mentioned design. Major focus of the thermal design is the
accommodation of propulsion system in the spacecraft, which increases complexity of the design from
generic micro-spacecraft missions.

SRMSAT-2 during all the mission operations, maintains a 3-axis stabilized orientation. Thus, thermal
requirements in the design is fulfilled by using an external coating of Multi-Layer Insulation (MLI),
over all exposed surfaces of the spacecraft bus. An element of mission design includes a probable
configuration and use of Low Power Mode in the deep space or WSB cruise after the TLCM. For this,
heaters would be required and use of which is considered to be defined in the critical design. The
subsystem wise thermal control measures are discussed in the later sections.

2.3. HEAT SOURCES

Heat source definition is a crucial step for the analysis of any system by a thermal engineer as it defines
the thermal loads on the system. The heat sources in a spacecrafts may be of external or internal origin.
The external heat sources defines the thermal space environment for the spacecraft in its mission time
and the internal sources include the various sources of heat generated mainly due to electronic
component or the propulsion subsystem in the spacecraft that can cause considerably heating in specific
region and subsequently change temperatures.

External sources
The heat sources include the direct solar radiation from the sun, albedo or reflected radiations
and the planetary radiations of celestial bodies.

Direct solar flux

The intensity of the direct solar flux varies with the distance. However, it is quite straightforward to
obtain as the power emitted by the sun is constant (3.856 x 1026 W). Considerably accurate values can
be obtained, as solar flux at particular distance is known with a tolerance of upto 0.4% and the
magnitude varies inversely with the square of distance from the sun.

Solar flux also depends on the wavelength of the radiation. The energy distribution at the receiving end
is approximately 7% ultraviolet, 46% visible, and 47% near (short-wavelength) Infrared. The total
integrated incident flux range can be found out by adding them independently. The thermo-optical
properties change along the spectral range as well. The absorptance variation for a spectral range is
shown in the plot given below:

244
 Figure 2.1: The solar output, at the solar constant has negligible variation with time

Albedo radiation

Albedo factor of a planetary body is the ratio of solar radiation reflected by the body to the incident net
solar radiation. Since it takes into the consideration only the reflective properties of the body, it will
affect the spacecraft all the time, except when eclipsed by the planetary body. Thus, during evaluation
of heat load due to albedo, we take into consideration, only the illuminated side of the planetary body.
This consideration is taken only for phases of mission in which spacecraft is in the vicinity of the body
i.e. Earth and Moon orbiting phases.

Also, the heat flux due to the albedo factor reaching a spacecraft will decrease as the spacecraft moves
along its orbit and away from the subsolar point (the point on Earth or another planet where the sun is
at the zenith, i.e., directly overhead), even if the albedo constant, as such, remains the same. This
happens because the albedo is a factor of reflectivity, not a flux. As the spacecraft moves away from
the subsolar point, it is over regions of Earth's surface where the local incident solar energy per square
meter is decreasing with the cosine of the angle from the subsolar point. The albedo heat load on the
spacecraft will therefore approach 0 near the terminator (the dividing line between the sunlit and dark
sides of a planet), even if the albedo value (reflectivity) is 1.0. The albedo factor changes for various
other reasons as well, like the variation in surface factors, change in humidity of the atmosphere above,
clouds and terrain, etc. For practical purposes an average albedo is considered and the changes in albedo
will be considered in the later phases of analysis. The intensity of the albedo radiation, Ja , and incident
on a spacecraft is a complex function of planet size and reflective characteristics, spacecraft altitude
and the angle B between the local vertical and the Sun's rays. This can be expressed in terms of a
visibility factor F as follows:

𝐽𝑎 = 𝐽𝑠 𝑎𝐹
(2.1)

The ranges of planetary albedos of some celestial bodies are shown and the averages of the planetary
bodies that are considered are mentioned below.

245
 Planetary Body Solar Radiation Intensity, Ja (as Albedo Factor, a Considered
% of solar intensity at I AU) Average
Mercury 667.0 0.06-0.10 -
Venus 191.0 0.60-0.76 -
Earth 100.0 0.31-0.39 0.34
Moon 100.0 0.07-0.12 0.10
Mars 43.1 0.15 -
Jupiter 3.69 0.41-0.52 -
Saturn 1.1 0.42-0.76 -
Uranus 0.27 0.45-0.66 -
Neptune 0.11 0.35-0.62 -
Table2.2: Solar constants and Albedo Values

For the purpose of calculating albedo radiation inputs, the Earth can be regarded as a diffuse reflecting
sphere, in which case the visibility factor varies approximately as shown below:

Figure 2.2: Visibility factors for various values of B (angle between the local vertical the incident rays
of the sun)

Planetary radiation

Each planetary body emits radiation of its own, due to its non-zero body temperature. When the net
emitted radiation emitted by the body is equated to corresponding black body radiation, the temperature
thus found is known as the equivalent blackbody temperature. The planetary radiations also vary from
position to position depending on the surface characteristics of the planetary body. The planetary
radiations usually lie in the infrared spectrum and thus, the earth’s planetary radiation if often referred
to as Earth IR. The following plot shows the variation of the Earth’s IR over time observed by a
spacecraft in a LEO.

246
 Figure 2.3: Earth IR seen over a three-hour period by a spacecraft in LEO

It can be seen from the above plot that there are large variations of the earth’s planetary radiation with
position and thus preliminary analysis, an average was considered. The orbital averages considered for
first-order analysis are:

Earth 237 W/m2

Moon 340 W/m2

The planetary radiations of the moon have large differentials due to the large and uneven distributions
of surface temperatures. This is mainly due to differential heating of the lunar surface and the tidal
locking of moon in an orbit around the Earth. The planetary radiations for the moon go up to 1250 W/m2
in the regions on the illuminated side and as low as 5 W/m2 for the regions on the dark side. These are
restricted to some regions on the lunar surface and thus are encountered for small periods. However, it
imposes considerable thermal stress that need to be accounted, on the spacecraft. This is due to
differential heating of the spacecraft in the lunar orbit, especially in a LLO.

The effects of radiations caused due to earth’s radiations at any altitude can be given by:
2
𝑅𝑒
𝐽𝑒 = 237 [ ]
𝑅𝑒𝑜𝑟𝑏𝑖𝑡

(2.2)

It’s clear that effect of Earth’s radiations in a lunar orbit is negligible and the effect of lunar radiations
in the earth bound orbits are in turn negligible as well.

Internal sources
The heat dissipation by the internal components of the spacecraft contribute to internal sources. This
usually takes place due to Joule’s heating effect in electronic components. Apart from that, heat sources
also include some other region specific sources like the combustion chamber of the primary propulsion
system. The internal heat loads are estimated according to components selected for the current

247
 spacecraft design. The following table enlists the heat dissipation loads of all major heat emitting
astrionic system components considered for the preliminary design:

COMPONENT HEAT
DISSIPATION
Communication (DSP) 0.32 W
Low noise amplifier 1.28 W
Power amplifier 36 W
Switches and regulators ≤0.1 W
Microprocessors 0.315 W
Transceiver 15 W
Battery heat dissipation 8W
Table 2.3: Internal heat loads

The usage of power amplifier is considered unlikely and is only present for the margin and emergency
communication purposes.

Thermal load calculation

The analysis is done for worst case conditions of the earth and moon bound phases. The respective
thermal loads for the above mentioned phases and the net external loads for each worst case is given in
the following table:

CONDITION DIRECT EARTH LUNAR ALBEDO NET HEAT LOAD

SOLAR PLANETARY PLANETARY
EARHT BOUND HOT CASE 1376 237 - 468 2081
COLD CASE - 237 - - 237
MOON BOUND HOT CASE 1368 - 1250 137 2755
COLD CASE - - 5 - 5
Table 2.4: External thermal load estimation

2.4 FIRST ORDER ANALYSIS

The First Order Analysis is the initial analytical approximation made for the estimation of the spacecraft
temperature as a whole. Its primary assumption is that the spacecraft is in a steady state and there is no
heat transfer internally in the spacecraft system itself. This means that the whole spacecraft, a cuboid,
have all the 6 faces in equilibrium with each other. Also, with no internal components considered, the
system will be in thermal equilibrium. The analysis is a two-fold process, the steady state simulation
followed by a transient analysis.

Steady state analysis

This is a coarse estimate of the spacecraft temperature. It is assumed that the spacecraft is in steady
state and is analysed in the worst case condition. The absorptivity and emissivity values are taken
using area averages for each outer wall as the incident surface might be covered with a combination of
solar cell array and MLI. The basic equation used is that of steady state heat balance:

248
 𝛼𝐴𝑖𝑛𝑐𝑖𝑑𝑒𝑛𝑡 × (𝑅𝑠𝑜𝑙 + 𝑅𝑎𝑙 + 𝑅𝑝𝑙 ) + 𝑄𝑖𝑛 = 𝐴𝑡𝑜𝑡𝑎𝑙 × 𝜀𝜎𝑇 4

(2.3)

The thermo-optical properties of some materials considered in the analysis are:

MATERIAL αsolar 𝜺IR

KAPTON, ALUMINIZED, 1mil [MLI] 0.36 0.67
BLACK PAINT 0.95 0.92
DOW COMING DC-007 WHITE PAINT 0.19 0.88
ALUMINUM, BUFFED 0.16 0.03
Table 2.5: Thermo-optical properties

The results of the above analysis, not taking into consideration the view factors, and assuming that the
bus attains steady state by staying stationary for infinite amount of time in space in a particular
orientation are as follows:

CONDITION TEMPERATURE
EARTH BOUND HOT CASE 389 K
COLD CASE 256 K
MOON BOUND HOT CASE 452 K
COLD CASE 112 K
Table 2.6: Results for steady state first order analysis

Inference:

It can be seen from the results, that the ambient temperatures of the spacecraft is well outside the
survival ranges of most components, and thus cannot be operated at all through the mission without
near immediate failure following attainment of such a steady state. The reason for unreasonably large
deviations in the results from the actual temperatures experienced is due to the following assumptions
made for the analyses:

 A steady analysis is done on the spacecraft. This means heat load considered on the spacecraft
is necessarily constant. This is never actually the case, as the spacecraft is continuously moving
and changing orientation with respect to external sources, it will never have as much moderated
average of the thermal heat load as compared as that of constant heat load in the worst case
condition.
 The spacecraft is of a rather complex geometry than assumed and contains quite a few
components which would be heat sources or sinks. It is nearly impractical to attain such
complete thermal equilibrium.
 The radiations incident on the spacecraft are assumed to be perpendicular to the exposed
surfaces of the spacecraft. This usually is not the case and thus, only a fraction of the incident
radiation will actually affect the body and contribute to the thermal suffrage. For this, view
factors need to be considered once the attitude schedules or a close approximation of most
common attitude mode needs to be used for the estimation of the actual heat load.

249
Transient analysis
It is the transient analysis of the spacecraft in the earth bound phase. The spacecraft structure is
assumed to be completely made of Aluminium (not considering MLI). The temperature distribution on
the outer surfaces of the spacecraft is uniform and the other spacecraft components are assumed to be
put into one large box, in the interior of the spacecraft (called as “comp”). A GEO (Geostationary Earth
Orbit) is assumed for the simulation. The temperature of the spacecraft is plotted against time. It can be
seen that the spacecraft reaches near thermal equilibrium in the orbit. The governing equations used for
creating the transient setup are as follows:
𝑘𝐴(𝑇𝑠𝑢𝑟 −𝑇𝑐𝑜𝑚𝑝 )
𝑑𝑇𝑐𝑜𝑚𝑝 𝑥
+ 𝑄𝑖𝑛
=
𝑑𝑡 𝑚𝑐𝑜𝑚𝑝 𝐶𝑝

(2.4)
𝑑𝑇𝑠𝑢𝑟
𝑑𝑡
𝑘𝐴(𝑇𝑐𝑜𝑚𝑝 −𝑇𝑠𝑢𝑟 )
𝑥
+ 1.34𝑞𝑠𝑜𝑙 𝛼𝐴𝑆𝑢𝑛𝑓 − 𝜀𝜎𝐴(𝑇𝑠𝑢𝑟𝑓 4 − 𝑇𝑒𝑎𝑟𝑡ℎ 4 )𝐹𝑒𝑎𝑟𝑡ℎ − 𝜀𝜎𝐴𝑇𝑠𝑢𝑟𝑓 4 (1 − 𝐹𝑒𝑎𝑟𝑡ℎ )
=
𝑚𝑠𝑢𝑟𝑓 𝐶𝑝

(2.5)

The transient model was made in Matlab and Simulink and the following plot of transient
temperature variation was obtained as a result of the simulation.

Figure 2.4: Transient temperature plot

250
 2.5 NUMERICAL THERMAL ANALYSIS
Numerical thermal analysis of the spacecraft is done in Siemens NX using the spacecraft system
thermal tool. The analysis was done for two specific mission phases, the earth and the lunar orbiting
phase. The following were the assumptions taken for the analysis:

 The spacing in between the solar cells were not considered. The solar panel mesh, thus has one
mesh for the solar panel and one single mesh for the solar cell array.
 The orbital heating load is given with the nadir vector pointing normally to side panel 3 and the
velocity vector, normal to side panel 2. The solar declination error angle is assumed to be
negligible.
 The trajectory was simulated for 4000 seconds. The temperatures were found for 12 positions
in the orbit.
 Only propulsion tanks, an idealized battery box part, a PCB module and a nozzle were simulated
for internal components.
 The initial ambient temperature is assumed to be 30°C.
 The lunar phase was simulated using the moon as the primary body.
 All internal module are made of Al 6061 and given the thermo-optical properties of black paint.
Internal radiation heat transfer is allowed for all internal components considered.

The orbital parameters of the mission are as follows:

Parameter Earth phase Moon phase

Apogee 42155 km 1961 km
Perigee/Eccentricity 6618 km 1961 km (e=0)
Inclination 5.21° 91°
Angle of periapsis 180° 270°
RAAN 150° 60°
Table 2.7: Orbital parameters used

The following are the results shown for the 12th time step, (the orbit is divided into 12 phases) in each
of the orbital phase.

251
 EARTH PHASE RESULTS

Figure 2.5(a) and (b): External views for earth phase results

Figure 2.5(c) and (d): Internal view for earth phase results and the temperature distribution on the solar panel

252
 LUNAR PHASE RESULTS

Figure 2.5(e) and (f): External views for lunar phase results

Figure 2.5(g) and (h): Internal views for lunar phase results and Heat Flux

253
Conclusions

 The above shown temperatures for each view in the time invariant result. They thus show the
extremums.
 All internal electronic components are within their operational ranges in both orbital
simulations.
 The maximum temperature recorded is higher in the earth phase simulation due to the longer
periods of direct heating by the sun. Moreover, the temperature gradient across the opposite
side panels is more in the Earth phase simulation.
 The temperature ranges of the propulsion tanks are 20 to 55 °C (approximately), which is
acceptable in the current situation.

2.6 THERMAL PROBLEM IDENTIFICATIONS

Complex problems arise due to a number of reasons in each of the configurations and many iterative
changes need to be made in the structure for attaining complete safety. However, some problems can
be identified using preliminary analysis that were done using numerical methods or modelling and
simulation software. The thermal problems can be classified into two broad categories, the general
problems and the mission specific problems. The mission specific thermal problems are a result of the
mission geometry. Lunar orbiting science mission phase is the major part of mission specific problems
due to the large differential heating of the spacecraft. The thermal problems in general include the
insufficient protection for the COTS components used (usually having overheating problems), and the
temperature and heat distribution in general being radically different within the spacecraft. The
propulsion subsystem is also sensitive from a thermal point of view and needs major focus of analysis
for precision control and for avoidance of failures or damage, which can directly jeopardize the mission.
Some mission specific problems were identified in the process, and are enlisted below. It must be noted
that all these problems, though, having a high priority, may not have been solved completely, or arrived
at with an optimized solution.

Orbit analysis
Large amounts of thermal stresses are developed in the LLO due large planetary radiation differential
as enlisted in the worst case conditions of the external thermal loads. The ratio of the eclipsed time to
the time in light in the proposed LLO of apogee of 100 km is about 0.719. The orbit was simulated in
STK and the following plot of eclipse times were obtained for a 24 hour simulation.

254
 Figure 2.6: Eclipse Times for LLO

Catastrophic Freezing of Fuel

The propulsion subsystem uses a dual propulsion with hydrazine as the fuel and Nitrogen Tetroxide as
the oxidizer. For the operating pressures the fuel may catastrophically freeze in the pipelines due to its
relatively high freezing point. The freezing point of hydrazine is 2°C under standard conditions. Due to
negligibly low values of external heat loads in the eclipsed times of the LLO, the surrounding
temperatures can go low. Freezing even in small amounts can completely choke the pipelines and can
cause malfunctions in the propulsion subsystem. As far as the system is concerned, NTO has a more
sustainable temperature range (fp = -11.2°C) whereas Helium is completely off the danger zone (fp = -
272.2°C, taken at standard pressure conditions).

2.7 THERMAL CONTROL

Thermal control methods are broadly classified into, active and passive thermal control. Active control
measures require extra power supply which serves as a great part of constraint for the spacecraft design
and is desirably avoided. The common thermal control methods are:

Surface finishes
Thermal coatings are used in the spacecraft thermal design for various purposes which include using
secondary surface mirrors, white paints, silver or aluminum backed Teflon which are used to minimize
the absorbed solar energy and infra-red emissions. In spite of all this, these bodies emit energy which
is almost the same as a black-body. To minimize the absorbed solar energy and infra-red emissions,
polished metals like aluminum foil and gold plating is used. If energy is to be exchanged with the other
components the vehicle is painted black internally. This goes on to show that all the types of coatings
have wavelength dependent thermal properties which become important criteria for the selection of
thermal control methods.
The thermal control finishes are affected majorly by charged particles in the orbit, UV radiations, high
vacuum and contaminant films which are invariably found on all spacecrafts irrespective of the type of
finishing on the surface of the spacecraft. This results in the increase of solar absorptivity without
affecting the infrared spectral emissivity. The degradation not only creates a problem due to the solar
load but also ccauses change in load over the whole duration of a mission. Thus, the stability of coating
properties is very important in order to limit maximum temperatures and minimise the heater power
requirements.

255
 The thermal coatings used in the thermal design of the SRMSAT-2 is basically conformal coatins and
black paints which are primarily used in the interal components of the spacecraft for the controlling in-
situ radiations among the compnents of the spacecraft. The coatings used for analysis in the software
include Black Paints,DC-007 white paints and confromal coatings wherever required.

COATING TYPE OF α ir ε ir REMARKS

MATERIAL
Conformal Gray 0.10 0.80 Used on electronic components like
PCBs
Black Paint Gray 0.95 0.92 Used on protecting Al casing walls
Kapton, aluminized, 1 mil Gray 0.36 0.67 Used on external surfaces of the
[MLI] spacecraft
FR4 Gray 0.35 0.35 PCB matrial
Titanium Gray 0.40 0.55 Propellant tanks
Table 2.8: Optical Properties of materials used in software analysis including surface finishes and
Insulations

Insulation
Multi-layer insulation (MLI) and single layer insulation are the most common passive thermal control
elements on a spacecraft. MLI is majorly used to avert, excessive heat loss from a component or
disproportionate heating via the external environment. Most spacecrafts are covered with MLI blankets,
with cut-outs provided for radiator areas to scrap internally generated waste heat. These MLI blankets
are also used to protect internal propellant tanks, propellant lines, solid rocket motors and cryogenic
dewars. Single layer insulation is used when not much thermal isolation is required as it is cheaper to
manufacture as compared to MLI.
MLI is composed of multiple low-emittance layers which are actually films with low conductivity
between layers as show in the figure. The simplest construction is a layered blanket assembled from
embossed, thin mylar sheets with a vacuum deposited aluminium finish on one side of each sheet.

Figure2.7: Composition of a typical MLI blanket

Mli analysis and future scope

Using basic thermal balance, MLI layers selection and effect of introduction of an MLI on a wall
can be calculated. Commercial spacecrafts use 12-17 MLI layers in general. MLI layer approximation

256
is important for approximating outer surface temperature of the side walls and for the estimation of the
thermal flux values incident on the side walls. The heat flux when measured in radiosities and the
fundamental heat flow equations are written below. It uses the approximations allowing negligible
conduction.

Figure 2.8: MLI layer approximation simulation

The top row, consists of conduction heat transfer from the i-1th layer and to the i+1th layer
respectively. The second layer signifies the radiosity given from the i-1th layer and the i+1th layer
respectively. The third layer is the emitted blackbody radiation which is given by the Stefan
Boltzmann’s law. The analysis considers the 0th layer to be space, a constant temperature layer of 3K.
The respective absorptivity and emissivity values considered are area averages for each layer, made of
Kapton. The perforation coefficient is assumed to be 0.5. The steady state equations of state are as
follow:

𝑄𝑐,𝑖−1 + 𝜀 ( 𝑅𝑖 + + 𝑅𝑖 − ) = 𝑄𝑐,𝑖+1 + 𝑄𝑒,𝑖−1 + 𝑄𝑒,𝑖−1

(2.6)

𝑄𝑒,𝑖−1 = 𝑄𝑒,𝑖+1 = 𝜀𝜎𝑇𝑖4

(2.7)
𝑘(𝑇𝑖+1 − 𝑇𝑖 )⁄
𝑄𝑐,𝑖+1 = 𝑥

(2.8)

Where, Q signifies net heat exchange, R signifies radiosity and T signifies absolute temperature in
Kelvin. Using the negligible conduction approximation, the equations are reduced to:

𝑅𝑖 + = 𝜏𝑅𝑖−1 + + 𝜀𝜎𝑇𝑖−1
4
+ ∅𝑅𝑖−1 −

𝑅𝑖 − = 𝜏𝑅𝑖+1 + + 𝜀𝜎𝑇𝑖+1
4
+ ∅𝑅𝑖+1 −

(2.9)

The equation are solved numerically to obtain the temperature profile and the radiosity plot at different
temperature values.

257
 Figure2.9: Plot of Rout and Temperature of MLI

Radiators
Excess heat within the spacecraft is rejected into the space by radiators. They occur in various forms
like spacecraft structure panels, flat plate radiators, or panels that are deployed after the spacecraft is
deployed into the orbit. The basic principle of a radiator is to reject heat in form of IR radiation from
their surfaces. This radiating power depends on temperature and emissivity of the surface. The radiator
needs to reject both internal and external heat sources like the spacecraft’s excess heat as well as the
sun’s heat. Thus, radiators are given surface finishes that have high IR emissivity (>0.8). This
maximises the heat rejection and lowers solar absorptivity (<0.2) to limit the heat loads from the sun.
Typical radiator finishes include quartz mirrors,slivered teflon and white paint.
The radiating power of a radiator is a function of temperature as given by the Stefan-Boltzmann
equation

𝑄 = 𝜀𝜎𝑇 4
(2.10)
The actual sizing can be determined by a thermal analysis which considers the desired operating
temperature, worst case spacecraft excess heat, environmental heating and radiative and conductive
interactions with other spacecraft surfaces. Weights allotted for radiators vary from almost nothing to
12kg/m2 depending on the structural component used as a radiator

Louvers
Louvers are active thermal control elements and have been used in different forms in spacecrafts. It is
usually placed over the external radiators and used to modulate radiant heat transfer between internal
spacecraft surfaces or transfer heat via openings in the wall from the internal surface of the sacecraft to
the outer space. A louver in its open state allows six times heat transfer than it would otherwise let out
in its closed state. Thus louvers find applications when internal power dissipation varies widely as a
result of equipment duty cycles.
Louvers generally consist of four key elements which include blades, actuators, sensing elements and
structural elements. The louver is placed over a high emittance, low absorptance radiator, to modulate

258
the flow of radiant heat from the surface. Blades, run by the actuators give variable radiation
characteristics to the radiator surface. Meanwhile, the actuators drive the blades according to the
percieved radiator temperatures. In most louver designs, a bimetallic spring drives each louver blade
autonomously, ensuring maximum reliability. This helps in pointing out a single point failure which
may be associated to a blade rather than the entire assembly. Further, a strong path between the actuator
and the radiator is sought to minimise the temperature gradient between them.

Figure 2.10: Louvers

Heaters
Heaters are active thermal control elements which generate heat in specific regions to maintain the
operational temperature range. Commercial heaters are of two main types, which are usually installed
in an interface with thermistors for autonomous thermal control.

Patch heater

The most common type of heater used on spacecrafts is the patch heater. This type consists of an
electrical-resistance element sandwiched between two sheets of flexible, electrically insulating material,
such as Kapton. The patch heater may contain either a single circuit or multiple circuits, depending on
whether or not redundancy is desired within it. The redundancy is needed when the thermal control
becomes the key element to spacecraft survival. Redundancy is provided in one patch or by using more
than one patches generally, rectangular in shape. They are required to meet the low magnetic dipole
requirements of the magnetometer sensors.

259
 Figure 2.11: patch heaters

Cartridge heater

The cartridge heater is often used to heat blocks of material or high-temperature components such as
propellants. This heater consists of a coiled resistor enclosed in a cylindrical metallic case. Typically a
hole is drilled in the component to be heated and the cartridge is potted into the hole. Cartridge heaters
are usually a quarter-inch or less in diameter and up to a few inches long and due to them being small
in size and being compatible in almost any space, they can be added anywhere inside small pipes too.

Figure 2.12: cartridge heater

Heaters would be included for active thermal control in the detailed design, if thermal situation becomes
critical to the survival of the spacecraft and there is large enough heat control requirement to be
countered by heaters or louvers. Heater sizing will be done in thermal analysis after the creation of a
thermal mathematical model (TMM) to confirm final safety requirements.

2.8 THERMAL ANALYSIS: FUTURE SCOPE

Thermal Mathematical Model

The temperature inside a spacecraft is often a complicated continuously varying function of time and
location. The estimation of these temperature fields to high precision can be a difficult task. To simplify
this process, a Thermal Mathematical Model is conceptualized by generating an approximate
representation of a spacecraft. In this model, the characterization of isothermal nodes is done by their
respective thermal capacity, heat dissipation, radiative, and conductive interfaces with the neighbouring
nodes. Sometimes there are radiative interfaces with the external environment.

260
Conductive heat exchange

Conductive heat flow is represented by

𝜆𝐴
𝑄𝑐 = 𝑙
Δ𝑇
(2.11)

Where 𝜆 is the thermal conductivity, A is the cross sectional area, l is the conductive path length and
Δ𝑇 is the temperature difference.
𝑙
Δ𝑇 = 𝑄𝑐
ℎ𝑐
(2.12)

When it is assumed that the path length is a set of discrete conductive paths connected in series, the
temperature difference is represented as
1 1 1 1
Δ𝑇 = 𝑄𝑐 (ℎ + ℎ + ℎ + ⋯ )=𝑄𝑐 ℎ
1 2 3 𝑐

(2.13)

For n isothermal nodes, the effective heat conducted from the ith to the jth node is given by

𝑄𝑐𝑖𝑗 = 𝑘𝑖𝑗 (𝑇𝑖 − 𝑇𝑗 )

(2.14)

Where hij is the effective conductance between nodes i and j and Ti and Tj are the respective temperatures
of the ith and the jth node.

Radiative heat exchange

The radiative heat exchange is characterized by three parameters which are radiative view factors,
surface temperatures, and surface properties. The amount of radiation leaving a surface I and absorbed
by a surface j for diffuse surfaces is given by

𝑄𝑇𝑖𝑗 = 𝐴𝑖 𝐹𝑖𝑗 𝜀𝑖𝑗 𝜎(𝑇𝑖4 − 𝑇𝑗4 )

(2.15)

Where Ai is the area of the surface i, Fij is the view factor of surface j as seen from surface i and 𝜀 ij is a
parameter known as effective emittance. It is assumed that the view factor remains constant over the
surface i.

261
View factor
The view factor is represented by the ratio between radiation that leaves one surface and the radiation
that is intercepted by another. The summation of the view factors from any node i, to the surrounding
equipment is unity. Hence,
𝑘

∑ 𝐹𝑖𝑗 = 1
𝑗=1

(2.16)

Here, k is the number of surrounding surfaces.

Effective emittance
The effective emittance between two surfaces largely depends upon on surface optical properties,
mutual reflections and reflections via other nearby surfaces. Complicated expressions for 𝜀 ij occur as a
result of specular surfaces and/or emittance values less than unity. When the distance between the two
parallel diffuse surfaces is relatively small as compared to their surface areas the Effective emittance is
given by:
𝜀𝑖 𝜀𝑗
𝜀𝑖𝑗 =
𝜀𝑖 + 𝜀𝑗 − 𝜀𝑖 𝜀𝑗

(2.17)

If it is assumed that the surfaces involved are diffuse and have relatively high 𝜀 values, then the errors
introduced by its use is small. In general cases the above calculation is done when the spacecrafts
interior is painted.

Calculation of nodal temperatures

Let us assume that the TMM has n number of nodes, the heat absorbed by the i th node per unit time is
given by
𝑛 𝑛

𝑄𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙,𝑖 + 𝑄1 − 𝜎𝜀𝑖 𝐴𝑠𝑝𝑎𝑐𝑒,𝑖 𝑇𝑖 − ∑ ℎ𝑖𝑗 (𝑇𝑖 − 𝑇𝑗 ) − 𝜎 ∑ 𝐴𝑖 𝐹𝑖𝑗 𝜀𝑖𝑗 (𝑇𝑖 4 − 𝑇𝑗 4 )

4

𝑗=1 𝑗=1

(2.18)

Where,

𝑄𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙,𝑖 = 𝐽𝑠 𝛼𝑖 𝐴𝑠𝑜𝑙𝑎𝑟,𝑖 + 𝐽𝑎 𝛼𝑖 𝐴𝑎𝑙𝑏𝑒𝑑𝑜,𝑖 + 𝐽𝑝 𝜀𝑖 𝐴𝑝𝑙𝑎𝑛𝑒𝑡𝑎𝑟𝑦,𝑖

(2.19)

is the external heat input, and the effective areas receiving, respectively, direct solar, albedo and
planetary radiation, is the effective area with an unobstructed view of space and Qi is the internal heat
dissipation. The heat balance equation for node i is represented by

262
 𝑛 𝑛
𝑑𝑇𝑖
𝑚𝑖 𝐶𝑖 = 𝑄𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙,𝑖 + 𝑄𝑖 − 𝜎𝜀𝑖 𝐴𝑠𝑝𝑎𝑐𝑒,𝑖 𝑇𝑖 4 − ∑ ℎ𝑖𝑗 (𝑇𝑖 − 𝑇𝑗 ) − 𝜎 ∑ 𝐴𝑖 𝐹𝑖𝑗 𝜀𝑖𝑗 (𝑇𝑖 4 − 𝑇𝑗 4 )
𝑑𝑡
𝑗=1 𝑗=1

(2.20)

The thermal behaviour of the spacecraft is characterized by a group of n simultaneous non-linear

differential equations as above, with i varying from 1 to n. To simplify the calculations in order to
formulate a solution is to assume that the spacecraft is launched at time t from an air-conditioned launch
vehicle where all the temperatures are determined initially. The calculation of the temperature with a
small increment of time is given by
𝑑𝑇𝑖,0
𝑇𝑖 = 𝑇𝑖,0 + 𝛿𝑇𝑖,0 𝛿𝑇𝑖,0 = 𝑑𝑡
𝛿𝑡

(2.21)

Several iterations have to be carried out in order to re compute the values of 𝑑𝑇/𝑑𝑡 and this proves
inefficient because of a large amount of time is required to calculate the values and this can lead to
instability when the value of 𝛿𝑡 is too large or thermal capacities are less.

A general approach involves linearizing the equations

𝑇𝑖 4 = (𝑇𝑖,0 + 𝛿𝑇𝑖,0 )4

(2.22)

Assuming 𝛿𝑇𝑖,0 is small as compared with 𝑇𝑖,0 and expanding, we can write that

𝑇𝑖 4 = 𝑇𝑖,0 4 + 4𝑇𝑖,0 3 𝛿𝑇𝑖,0

(2.23)

And hence,

𝑇𝑖 4 ≈ 𝑇𝑖 (4𝑇𝑖,0 3 ) − 3𝑇𝑖,0 4

(2.24)

Steady-state calculations

While steady state calculations are done, 𝑄𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙,𝑖 𝑎𝑛𝑑 𝑄𝑖 are constant and dT/dt is zero. This allows
us to write the equation in the linearized form as follows:
𝑛 𝑛 𝑛

[∑ ℎ𝑖𝑗 + 4𝜎𝑇𝑖,0 (𝐴𝑠𝑝𝑎𝑐𝑒,𝑖 𝜀𝑖 + 3𝜎 ∑ 𝐴𝑖 𝐹𝑖𝑗 𝜀𝑖𝑗 )] − ∑[ℎ𝑖𝑗 + 4𝜎𝑇𝑗,0 3 𝐴𝑖 𝐹𝑖𝑗 𝜀𝑖𝑗 ]

3

𝑗=1 𝑗=1 𝑗=1

𝑛
4 4 4
= 𝑄𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙,𝑖 + 𝑄𝑖 + 3𝜎𝑇𝑖,0 𝐴𝑠𝑝𝑎𝑐𝑒,𝑖 𝜀𝑖 + 3𝜎(𝑇𝑖,0 − 𝑇𝑗,0 ) ∑ 𝐴𝑖 𝐹𝑖𝑗 𝜀𝑖𝑗
𝑗=1

263
 (2.25)

This is a set of linear equations in ‘n’ unknown𝑇𝑖 , where i= 1 to n, and can be solved by standard matrix
inversion techniques. The calculated temperatures can then be substituted for the original 𝑇𝑖,0 and the
calculation is repeated until the difference between newly calculated and previously calculated
temperatures is sufficiently small for steady-state conditions to be assumed, that is, d𝑇𝑖 /dt is sensibly
zero. It should be noted that, although the criterion that 𝛿𝑇𝑖,0 should be small may not be met initially,
it will be so for the final iterations and hence the final calculated temperatures will be numerically
correct.

Transient calculations

Many types of software have been developed over the years to perform transient calculations of which
ESA-sponsored ESATAN is the most comprehensive and will be used by us. We will use the tool
THERMXL to calculate the transient calculations.

2.9 RADIATION ENVIRONMENT

The radiation environment consists of external effects on the spacecraft which need not necessarily
cause heating in the components. The radiation environment mainly consist of particulate fluxes of
electrons and protons of various origins. The various types of sources are:

Van Allen radiation belts (inner and outer)

The Van Allen belts are a collection of charged particles, gathered in place by Earth’s magnetic field.
Van Allen belts are classified into two, the inner Van Allen belt and the outer Van Allen belt. The inner
Van Allen belt stretches from 400 to 6,000 miles above the earth’s surface. It is believed that proton
energies exceeding 50 MeV in the lower belts at lower altitudes are the result of the beta decay of
neutrons created by cosmic ray collisions with nuclei of the upper atmosphere. The outer Van Allen
belt stretches from 8,400 to 36,000 miles above the earth’s surface. The outer belt consists of high
energy (0.1–10 MeV) electrons trapped by the Earth's magnetosphere. There is an empty space
separating the two belts which is explained by the newly discovered barrier.
The mission is configured for 3 Earth orbits (GTO and 2 phasing orbits) and may include multiple coast
cycles in one or all the phasing orbits. Suffrage due to these Van Allen radiations thus becomes
important to the mission and needs to be analyzed to lay down shielding requirements
One radiation-clearing strategy involves using very large radio transmitters on the ground to beam very
low frequency (VLF) waves upward. These can in principle interact with and scatter charges in the
radiation belt and drive them into the upper atmosphere.

Coronal discharge and solar winds

Corona is the luminous envelope surrounding the sun, outside the chromosphere. Its density is less than
one billionth that of the earth's atmosphere. The corona is visible only at the time of totality during a
total eclipse of the sun. The corona consists of ionized gas at a temperature of 1 million °C. Although
the visible corona extends a few solar radii above the sun, because of its high temperature it produces a
continual flow of electrically charged particles called the solar wind that move outward through the
solar system.

264
Solar wind is a stream of ionized hydrogen—protons and electrons—with an 8% component of helium
ions and trace amounts of heavier ions that radiates outward from the sun at high speeds. The continuous
expansion of the solar corona into the surrounding vacuum of space carries away from the sun about 1
million tons of gas per sec; this blows out like a wind through the solar system and is known as the solar
wind.
Many effects result from the solar wind. The characteristic that a comet tail always points away from
the sun is explained by the pressure of the wind pushing it out. The interaction of the wind with the
earth's magnetic field is responsible in part for such phenomena as auroras and geomagnetic storms.

Cosmic rays
Cosmic rays are charged particles moving at nearly the speed of light reaching the earth from outer
space. Primary cosmic rays consist mostly of protons (nuclei of hydrogen atoms), some alpha particles
(helium nuclei), and lesser amounts of nuclei of carbon, nitrogen, oxygen, and heavier atoms. These
nuclei collide with nuclei in the upper atmosphere, producing secondary cosmic rays of protons,
neutrons, mesons, electrons, and gamma rays of high energy, which in turn hit nuclei lower in the
atmosphere to produce more particles. The secondary particles shower down through the atmosphere in
diminishing intensity to the earth's surface and even penetrate it. The size of the shower indicates the
energy of the primary ray, which may be as high as 1020 electron volts (eV) or more, almost a billion
times higher than the highest energy yet produced in a man-made particle accelerator; however, cosmic
rays of lower energy predominate.
When they react with interstellar gases, the result is a gamma ray that can be traced back. Spacecraft
results indicate that many of the gamma rays appear to come from the direction of supernova remnants.
The nature of the acceleration processes by which the primary particles achieve great velocities (very
nearly the speed of light) is also still highly speculative. Modern electronic detectors called charge
coupled devices (CCDs) are effective cosmic ray detectors; a ray can strike a single pixel, making it
much brighter than the surrounding ones. Cosmic rays play a significant role in the natural mutation
and evolution of life on earth.

Radiation effects on electronics

Radiation in Space is generated by particles emitted from a variety of sources both within and beyond
our solar system. Cosmic rays, corona discharge, Van Allen radiations are the major radiation threats
in space. The composition and intensity of the radiation varies significantly with the trajectory of a
space vehicle.

Due to the earth’s asymmetric magnetic field, a region in the Atlantic near Argentina and Brazil, known
as South Atlantic Anomaly (SAA), has relatively high concentrations of electrons. The SAA is known
to cause problems such as: single event upsets (SEU) in electronics gate arrays, and SEU’s in the Space
Shuttle Orbiter’s Star Tracker’s Analog-to-Digital converter. The March 1991 solar storms significantly
increased the charged particle distributions in the Van Allen belts, also creating a third belt.

In addition to the trapped charged particles in Van Allen radiation belts (electrons and protons), the
spacecraft experience radiation threats from high energy heavy ions in space called Galactic Cosmic
rays, and secondary X-Rays generated by particles penetrating the skin of the space-craft while they
lose energy. This type of electromagnetic radiation is a significant percentage of the total component
producing total dose effects.

265
Spacecraft charging
It is defined as the buildup of charge on spacecraft surfaces or in the spacecraft interior. It causes
variations in the potential of the spacecraft surface with respect to the surrounding environment. The
space environments which contribute to the charging include thermal plasma environment, high energy
electrons, solar radiation and magnetic fields. It has many effects including electronic discharges which
cause structural and electronic damage.
There are different environments which cause the spacecraft to charge which include plasmas, high
energy electrons, solar radiations and magnetic fields.

Radiation analysis
The radiation analysis was done in SPENVIS, an open source tool provided by ESA. The analysis was
done to study the electron and proton fluxes, their differential in the mission and finally predict the
fluence and consequent spacecraft charging. The primary body for creating trajectories is restricted
only to Earth, and thus for acquiring a differential plot of fluxes in the lunar vicinity, a lunar flyby
trajectory was constructed for analysis purpose. The orbital parameters and their respective results are
as shown below:

Case 1: Geostationary Earth Orbit

Property Value
Apogee 42155 km
Perigee 6618 km
Angle of Inclination 5.21 °
RAAN 150 °
Argument of Perigee 180 °
True Anomaly 54 °
Table 2.9

266
 Figure2.13: Flux distribution over Energy Figure 2.14: Max Orbital Flux vs Orbital Time

Figure 2.15: Ion Spectrum Figure 2.16: Patch Voltage

Case 2: Earth Orbit with Most of the Orbital Time outside the Van Allen Belts

Property Value
Apogee 89999.95 km
Perigee 6759.34 km
Angle of Inclination 5.209 °
RAAN 149.841 °
Argument of Perigee 174 °
True Anomaly 54 °
Table 2.10

267
 Figure 2.17: Flux distribution over Energy Figure2.18: Max Orbital Flux vs OrbitalTime

Figure 2.19: Ion Spectrum Figure 2.20: Patch Voltage

Case 3: Lunar Flyby

Property Value
Apogee 129999.9 km
Perigee 7261.8 km
Angle of Inclination 5.2 °
RAAN 149.7 °
Argument of Perigee 169.95 °
True Anomaly 170 °
Table 2.11

268
Figure 2.21: Flux distribution over Energy Figure 2.22: Max Orbital Flux vs
Orbital Time

Figure 2.23: Ion Spectrum Figure 2.24: Patch Voltage

Conclusion
The radiation analysis can be summed up in the following points. A differential plot was made for the
orbital average differential and integral fluxes, to compare the energy distributions side by side:

Figure 2.25

269
The above plots show the differential and average fluxes for LEO, GEO and a lunar flyby distribution
over energy.

 Largest flux distribution is in the Van Allen belt region that is 400 to 36000 miles.
 The particulate fluxes have maximum flux density for low energies. (<1 MeV). Thus the highly
energized, more harmful particulate radiations are negligible.
 The fluxes of lunar region and LEO are negligible as compared to GEO region which
completely lies in the Van Allen Belts region.

270
 3. PROPULSION

3.1 OVERVIEW
The propulsion subsystem of SRMSAT - 2 consists of a primary propulsion system for orbit and
trajectory corrections and a reaction control system for attitude corrections. The primary propulsion
consists of a bi-propellant system which uses hydrazine as a fuel and NTO (Nitrogen Tetroxide) as an
oxidizer whereas the reaction control system is a monopropellant propulsion system that uses hydrazine.
This chapter discusses the design methodology of the subsystem which consists of a selection of the
type of propellant and the type of feed system. Helium was chosen as the pressurizing gas.

3.2 INTRODUCTION
Need for a Propulsion System
Most satellites that are launched into Earth-bound orbits do not require a propulsion system. One of the
major differences between the SRMSAT and the SRMSAT -2 is the presence of a propulsion system in
the latter. The former satellite was launched into LEO (Low Earth Orbit) and does not require any
orbital corrections. On the other hand, SRMSAT -2 has large delta V requirements (~1.8 𝑘𝑚/𝑠),
making the need for a propulsion subsystem a mandate. Also, an RCS system is required to control the
attitude of the spacecraft.

Types of Propulsion Systems

A propulsion system, though mandatory for deep space missions, needs to abide by the mass and power
constraints of a small satellite. A maximum of 50% mass was allotted to the propulsion subsystem.
Thus, an optimization was made after a case study of various types of propulsion systems, and various
types of propellants available for use. The different propellant types that are compatible for this
spacecraft are:
Cold Gas
A cold gas propulsion system uses an inert gas like Nitrogen as the reaction mass. It is one of simplest
thermal rocket propulsion system types in which the thermal energy is derived from the heat capacity
of the gasses. It has low efficiency due to unheated gasses but has advantages like the low cost of
propellant, ease of installation and simplicity of use. An electric resistive element can be used to heat
the gasses to increase the efficiency but results in increased power consumption.
Chemical Propulsion
They use chemical propellants like hydrazine to produce heat energy by a catalyzed chemical reaction
to generate gasses at very high pressures and temperatures. These gasses are accelerated in the nozzle
and expelled out at large velocities to provide thrust to the satellites.
Electric Propulsion
It uses electrostatic or electromagnetic energy to accelerate charged species and eject the matter at large
velocities to provide a propelling force. Ion thrusters use large potential differences to accelerate
charged particles using Lorentz force and expel them. The ions are excited by external means like the
use of microwaves to form highly energetic plasma particles. The system has a very low mass

271
requirement, making it highly efficient regarding mass but has very high power requirements (0.1 -10
kW) and low thrust outputs.

3.3 SYSTEM SELECTION

A comparative study was done on the most conventionally used propulsion systems, that is, warm gas,
cold gas, chemical, and electric propulsion. A comparison was made on the mass fractions for various
delta – v requirements. The plot of mass fractions vs. delta – vis shown below. It can be inferred from
the observation that chemical and electric propulsion systems have smaller values for mass fractions.
Although electric propulsion has the best fractions throughout the delta – v variation, it was excluded
due to large power requirements. The next best option, chemical propulsion was selected which has a
mass fraction of approximately 56 % of the whole satellite mass for the current delta –v requirement
(~1.8 𝑘𝑚/𝑠).

Figure 3.37: Mass Fraction vs Delta –v for four types

Once the chemical propulsion class has been selected, the selection of the actual chemical propellants
is done. The main types considered were solid propellants, monopropellants, and bi-propellants.

Propulsion System Configuration

Monopropellant Systems
Monopropellant Systems use a single chemical as its propellant. The main advantages of these systems
are that they do not require a separate oxidizer. The propellants release energy through exothermic
chemical decomposition in the presence of a catalyst. The molecular bond energy of the monopropellant
that is released in the form of hot gas is expelled through the nozzle to provide propulsive reaction and
the resultant thrust. These systems are simple and reliable. They can withstand space environment for
long periods, but they have the moderate specific impulse, and the life of catalysts is limited. They are
commonly used in RCS thrusters which do not require high specific impulse. Hydrazine and Hydrogen
peroxide are the most common examples of monopropellants.

272
Bipropellant Systems
Bipropellant Systems use liquid fuel and an oxidizer as propellants. They are stored in separate tanks
and are injected into the combustion chamber using pump feed or pressure feed. A spontaneous
chemical reaction takes place in the thruster to produce high-temperature products which are expelled
through the nozzle to generate the required thrust. The amount of thrust produced is controlled by
regulating the rate of propellant flow. These systems have high specific impulse and torque capabilities.
They are more complex and much costlier than the monopropellant systems.
Solid Propellant Systems
A solid propellant system uses the solid cast mixture of the fuel and oxidizer as its propellant. It consists
of a motor case, an igniter, and a nozzle. The solid cast mixture is called grain. After ignition, the grain
burns, and the combustion products are ejected through the nozzle to produce the thrust. There are two
types, namely, homogenous (double-based) and composite propellants; the latter being the predominant
one. The fuel and oxidant are mixed chemically in double-based propellants whereas they are
mechanically mixed in composite propellants.
For the given delta- v, several system configurations are possible. Even though solid propulsion gives
a large amount of thrusts, it is excluded due to its uncontrollable nature. Thus, preliminary study for
analysis of several system configurations of monopropellant and bipropellant systems was done.

PROPELLANT ISP
PPS ( 22 N )
Hydrazine 230
Hydrazine / NTO 300
MMH / NTO 294
RCS ( 1 N )
Hydrazine 210
Butane 80
Nitrogen 67
Table 3.12: Specific Impulse Values for various propellants

MONO-PROPELLANT
PPS RCS Total Mass
Hydrazine Hydrazine 57.29
Hydrazine Butane( HG ) 60.01
Hydrazine Nitrogen(CG) 61.12
Table 3.13: Mass Estimates for Monopropellant Fuel

273
 BI-PROPELLANT SYSTEM

S. PPS RCS FUEL OXIDIZER RCS Total

No. MASS MASS MASS

1 Hydrazine / Hydrazine 16.10 22.58 1.39 40.07

Nitrogen
Tetraoxide

2 Hydrazine / Butane( HG ) 16.10 22.58 3.72 42.40

Nitrogen Tetraoxide

3 Hydrazine / Nitrogen(CG 16.10 22.58 4.46 43.14

)
Nitrogen Tetraoxide

4 Hydrazine / Hydrazine / 14.96 25.16 1.02 41.14

Nitrogen
Nitrogen Tetraoxide
Tetraoxide

5 MMH/ Butane(HG) 14.27 22.98 3.96 41.21

Nitrogen Tetraoxide

6 MMH/ Nitrogen(CG 14.83 24.30 3.96 43.09

)
Nitrogen Tetraoxide

7 MMH/ MMH/ 14.83 24.30 4.67 43.80

Nitrogen Tetraoxide Nitrogen

Tetraoxide
Table 3.14: Mass Estimates for BI-propellant combinations

The above results suggest the use of a combination of bi-propellant system of Hydrazine as fuel and
NTO as an oxidizer for the primary propulsion system and a monopropellant hydrazine for reaction
control system resulting in least mass among the combinations.

3.4 TANK CONFIGURATION

Feed
For the tank configuration, firstly the type of feed is decided and then the tank type selection and sizing
is done. The major types of feed systems are pressure regulated and blow down propulsion systems.
Blowdown propulsion system
In blowdown system, the propellants and the pressurizing gas are stored in the same tank. The main
advantage of this kind of system is that there is a single tank for pressurizing gas and the propellant.
The pressure regulator under the tank controls the amount of propellant flowing to the thrusters.
Pressure regulated system

274
In pressure regulated system, there is a separate fuel, oxidizer, and pressurizing tank. The pressurizing
gas is stored at very high pressure, and there is a pressure regulator to control the pressure applied to
the fuel and oxidizer tank. The pressurizing gas (helium in this case) is stored at 200 bar.

Hydrazine NTO
Mass 15 kg 24 kg
Required Helium mass 0.1470 kg 0.2401 kg
Tank mass 6.7460 kg 10.8483 kg
Total mass 21.893 kg 35.0885 kg
Total system mass 56.9 kg
Table 3.15: Blowdown Propulsion

Total mass of propellant (Fuel + Oxidizer) 39 kg

Helium mass required 0.26 kg
Helium tank mass 1.144 kg
Hydrazine tank mass 4.1813 kg
NTO tank mass 5.804 kg
Total mass 47.4 kg
Table 3.16: Pressure regulated and fed propulsion

From the above calculations, as the mass factor saves more than 9.6 kg in the propulsion system as a
whole, it is clear that pressure regulated propulsion is suitable for the spacecraft. Also, blow down feed
mode is rarely used in case of bipropellants system due to the difficulty in holding the two propellant
tanks at the same pressure and complexities related to bipropellant engines operating over a wide
pressure range.

Tank
Bladder tanks

A bladder tank comprises a rigid vessel containing a flexible bladder and perforated axial stand-
pipe.The propellant is contained in the bladder and pressurant gas within the tank occupies the volume
between the tank wall and bladder. The pressurant gas 'squeezes' the bladder forcing propellant through
the standpipe to achieve a positive expulsion of propellant to the thruster.

Surface Tension Tanks

In a propulsion system, as the fuel gets used up, the fuel may not be available at the output orifice due
to zero gravity conditions. This would result in the non-availability of fuel when required which is
tackled by the introduction of surface tension tanks. Surface tension tanks are used to ensure proper
flow of propellant to the tank output port by utilizing the property of surface tension.
From the above study, it is decided to use bladder tanks for the propulsion system because of the
simplicity of operation.

275
Tank Shape
The major types of tanks considered are:
Spherical Tanks
Spherical tanks are the most common tanks manufactured in the industry for pressure vessel
configuration as they are the easiest to make. They have a mass-efficient pressure vessel design and
have been used since decades in spacecraft. Most spherical pressure tanks are composed of titanium.
Further, spherical tanks can be mounted in numerous ways like girth tabs and polar bosses, thus
providing many options for the propellant tank to fit into different shaped spacecraft.
Cylindrical Tanks
The need for cylindrical tanks is felt when longer missions are planned, and more fuel is needed for the
mission. Spherical tanks, though best suited, cannot be used as the radius of the spherical tanks does
not suit the structure of the spacecraft.
The shape of the propulsion tanks chosen was cylindrical shaped for the fuel and oxidizer and a spherical
tank for the pressurizer because:

 Cubical tanks provide appropriate dimensions for the satellite but are not preferred due to the
unequal pressure distribution that may increase the complexity of the system.
 Spherical tanks overcome this drawback but are inefficient in spacing as the structure has a
larger height dimension. Moreover, use of spherical tanks would require an increase in the
dimensions of the base plate and the use of two tanks of hydrazine and NTO each, will make
the schematics much more complicated and causing an increase in mass and space
requirements.
 Cylindrical tanks with hemispherical end are suitable from the design perspective.

Tank Sizing
Tank sizing is one of the critical components in overall system design that requires a rich and
comprehensive understanding of the tank gas thermodynamics and fluid flow dynamics. As propellant
feed mode chosen was pressure regulated, care needs to be taken during specification of pressurizing
gas mass to ensure that there is a supply of the propellant at all phases of the mission.

Based on the usually approached methods, volume for Hydrazine and Nitrogen Tetroxide tanks are
determined by their given mass and density at the temperature of 30oC.Helium gas mass is determined
for the loaded mass of propellants. It is supplied to the fuel and oxidizer tank through a pressure
regulator to expel propellant from their respective tanks. The gas pressure in the tank decreases
progressively as the propellant is expelled. A pressure regulator is used to lower the pressure from 200
bar to the required supply pressure.

A high-pressure gas bottle consists of the pressurizing gas at 200 bar. The expansion of the gas
causes a decrease in temperature due to depletion of the propellant. Using the thermodynamic analysis
of the pressurizing gas, mass for it is calculated. As per the current system design including fuel tanks
and oxidizer tanks, the total mass of them is calculated as shown in table 3.6.

276
 Volume of fuel (Vf) 19L
Volume of oxidizer (Vo) 14L
Volume of pressurizing gas (V0) 2L
Mass of the pressurizing gas (mp) 0.260 kg
Table 3.17: Final Tank Volume and Mass of the pressurizing gas

3.5 SCHEMATICS
The schematics of the propulsion system are designed and described below:

 It is a dual propulsion system, which contains a bipropellant propulsion system consisting of

one 22N thruster for PPS (Primary Propulsion System), and a monopropellant propulsion
system consisting of four 1N thrusters for attitude control (RCS-Reaction Control System).

 The fuel and oxidizer are contained in cylindrical tanks with hemispherical ends. The
propulsion system is pressurized by Helium (He) supplied from a single spherical tank. In-line
filters are used to filter the pressurant and propellants. The fuel, oxidizer, and pressurant are
loaded into the propulsion subsystem through several fill and drain valves. These valves also
allow offloading if necessary.

 A normally closed pyrotechnic valve is present after each propellant tank for safety reasons to
avoid stress during launch conditions.

 Check valves are used upstream of the propellant tanks to prevent vapor migration into the
pressurant tank and maintain the direction of flow.

 Pressure transducers, placed at specific locations measure the pressures at inlet and outlet of
each tank and the outlet of the pressurizing tank. They are used to monitor the pressure in the
subsystem.

 A redundant pressure regulator is used to control the pressure from the helium tank.

 Latch valves are located at the exit of each propellant tank to isolate the tank during inactivity.

 The flow control valves regulate the flow or pressure of the fluid.

 Single seat solenoidal valves are used for each of the RCS thrusters and two double seat
solenoidal valves are used for the SPS thruster.

277
Figure 3.38: Schematics of the propulsion system

278
 3.6 PROPELLANT FEEDLINE

Propellant feedlines are basically pipe connections given in the hydraulic circuit of the propulsion
system. The material chosen for high-pressure feed lines for pressurizing gas is Titanium (grade 3 Al
2.5 V) and that for the transport of low-pressure fluids, fuel and oxidizer will be copper or aluminum.
The thickness of the feedlines is calculated with hoop stress calculations of the pipe. Here, it is assumed
to be a thin cylinder and keeping it within the yield strength.
𝑃𝑑
𝜎=
2𝑡
(3.1)
2𝑡𝑠 𝜎
𝐵𝑃 =
𝑑
(3.2)
Where P is the working pressure, 𝜎is the hoop stress, d is the diameter, t is the thickness of the tank, 𝑡𝑠
is the actual thickness of the tank (considering the factor of safety) and BP is the burst pressure. The
calculations are done considering yield strength with a factor of safety. (FOS = 1.5)

Working pressure ‘P’ 300 bar

Diameter of feedline ‘d’ 7 mm
Yield strength of Ti 3Al2.5V 5000 bar
FOS 1.5
Thickness 0.315 mm
Burst Pressure 558 bar
Table 3.18: Feedline Properties

279
 3.7 THRUSTER SELECTION
Reaction Control System
The thrust requirements for RCS is low, and hence, only 1 N thrusters were compared. The following
RCS thrusters were considered for comparison:

Engine MOOG 1N Thruster Airbus DS 1 N Chemical

Monopropellant thruster

Steady State Thrust 0.22 lbf (1N) 0.22 lbf (1N)

@275 psia

Feed Pressure 70 – 400 psia 79.77 – 319 psia (5.5 – 22 bar)

(4.8 – 27.6 bar)
Valve Power 8.6 watts 12 watts

Mass 0.48 lbm (0.218 kg) with valve 0.63 lbm (0.28 kg) – with
valve

Specific Impulse 227.5 secs 220-223 secs

Table 3.19: RCS Thrusters Comparison

From the above comparison, we have chosen the Moog 1N thruster as our RCS thruster due to less
mass and power requirement.

Primary Propulsion System

22 N thrusters were considered for the primary propulsion system. The design and performance
characteristics for thruster comparison are enlisted below:

280
 Design DST-11H DST-12 DST-13 5 lbf Airbus DS 22N
Bipropellant thruster
Propellant Hydrazine/M MMH/MON MMH/MON MMH/MON Hydrazine/ N2O4
ON
Nominal 5 lbf (22N) 5 lbf (22N) 5 lbf (22N) 5 lbf (22N) 5 lbf (22N)
Steady State
Thrust
Feed 80 – 400 psia 60 – 400 80 – 400 psia 39 - 320 psia 116 – 362.5 psia
Pressure (5.5 – 27.6 psia (5.5 - 27.6 bar) (2.8 - 22.1 bar) ( 8 – 25 bar)
bar) (4.1 - 27.6
bar)
Nozzle 300:1 300:1 300:1 150:1/300:1 150:1
Expansion
Nominal 0.85 1.61 1.65 1.61/1.65 1.6-1.65
Mixture
Ratio
Valve Solenoid Latching Solenoid Latching Bipropellant torque or
Torque Torque Motor linear motor valve (single
Motor or Solenoid and dual seat valve)
Valve Power 41 watts max 6 watts max 41 watts max 6 watts max 25 Watts including the
(2 coils wired (Latch) (2 coils wired (Latch) valve operation @ 28V
in series) 7 watts max in series) 7 watts max
(primary) (primary)
9 watts max 9 watts max
(secondary) (secondary)
(Torque Motor)
15.6 watts max
(solenoid)
Mass 1.7 lbm (0.77 1.4 lbm 1.5 lbm (0.68 1.4 – 2.0 lbm 0.837 lbm (0.38 kg) –
kg) (0.64 kg) kg) (0.64 – 0.91 Single seat valve
kg) 1.499 lb (0.68 kg) – Dual
seat valve
Chamber Platinum/Rho Platinum/Rh Platinum/Rhodi C-103 Platinum
Material dium Alloy odium Alloy um Alloy
Specific 307 secs (min) 297 secs 294 secs (min) 284 secs (min) 300 secs (At nominal point)
Impulse (min)

From the above comparison, the Airbus 22 N thruster is the best option and suits the mission
needs.

281
 3.8 FINAL SUBSYSTEM CONFIGURATION

Property Storage Pressure Mass

Fuel Hydrazine 5 bar 17.49 kg

Oxidizer Nitrogen Tetroxide 5 bar 22.58 kg

Pressurizer Helium 200 bar 0.26 kg

Thrusters

Type Thrust Quantity

Primary Propulsion
22 N 1
System (PPS)

Reaction Control
1N 4
System (RCS)

Propulsion Tanks

Purpose Type Volume Mass

Fuel Cylindrical 13L 4.18 kg

Oxidizer Cylindrical 19L 5.80 kg

Pressurizer Spherical 2L 1.14 kg

Table 3.9: Final Subsystem Configuration

3.10 PROPULSION SYSTEM COMMISSIONING PROCEDURE

System commissioning consists of venting, priming and pressurizing. It is done in the initialization
mode of the mission. The feedlines in the propulsion system are initially filled with compressed gasses
which are vented out through the thrusters in the form of cold gas thrust or through the vent valves. The
pyro valves of the fuel and oxidizer tank are opened, and the feedlines are then filled with propellant as
part of the priming process. The pressure from the Helium tank is then released to the feedlines by
opening the pyro valves to pressurize the system. The propulsion system then can be used for normal
operation.

282