Guidelines For Monitoring Stand-Alone Photovoltaic Power Systems

Guidelines for Monitoring
Stand-Alone Photovoltaic
Power Systems
Methodology and Equipment
IEA PVPS T3-13:2003

IEA PVPS Task 3 – Use of Photovoltaic Systems in Stand-Alone and Island Applications
IEA PVPS
International Energy Agency
Implementing Agreement on Photovoltaic Power Systems
Task 3
Use of Photovoltaic Power Systems in Stand-Alone and Island
Applications
Report IEA PVPS T3-13: 2003
<Guidelines for monitoring

stand-alone photovoltaic systems>
Methodology and equipment
October 2003
IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 2

Foreword
The International Energy Agency (IEA), founded in November 1974, is an autonomous body
within the framework of the Organisation for Economic Co-operation and Development
(OECD) which carries out a comprehensive program of energy co-operation among its 20
member countries. The European Commission also participates in the work of the Agency.
The IEA Photovoltaic Power Systems (PVPS) Program is one of the collaborative R&D
agreements established within the IEA and, since 1993, its Participants have been conducting
a variety of joint projects in the applications of photovoltaic conversion of solar energy into
electricity.
PVPS is headed by an Executive Committee comprising one representative from each
participating country, while the management of individual research projects (Tasks) is the
responsibility of Operating Agents. Currently nine tasks have been established. The twenty-
one members of the PVPS Program are:
Australia (AUS), Austria (AUT), Canada (CAN), Denmark (DNK), European Commission,
Finland (FIN), France (FRA), Germany (DEU), Israel (ISR), Italy (ITA), Japan (JPN), Korea
(KOR), Mexico (MEX), Netherlands (NLD), Norway (NOR), Portugal (PRT), Spain (ESP),
Sweden (SWE), Switzerland (CHE), United Kingdom (GBR), United States (USA). More
details about participating countries can be found on the PVPS program website at:
www.iea.pvps.org.
This International Technical Report has been prepared under the supervision of PVPS Task 3
by Dave Turcotte (Canada) and Farah Sheriff (Canada) in co-operation with experts of the
following countries: Australia, France, Germany, Japan, Netherlands, Norway, Portugal,
Spain, Sweden and Switzerland.
The report expresses, as nearly as possible, a consensus of opinion of the Task 3 experts on
the subjects dealt with.

Disclaimer
This document is designed solely as a guideline for monitoring in accordance with the goals fixed by the
Task 3 of the International Energy Agency for its work on Photovoltaic Systems for Stand-Alone and
Island Applications. It is not intended to be used as a standard and it does not claim to supplement,
amend or replace any of the standards referenced hereinafter nor any other existing national or
international standard.
Perspective
The Guidelines for monitoring stand-alone photovoltaic systems : Methodology and
Equipment is one document in a series of related documents prepared by IEA PVPS, Task 3.
The other papers are complimentary to this work and any person contemplating investing in a
stand alone PV power supply would be well advised to read the other papers in this series.
These are all available on the IEA/PVPS web page www.iea.pvps.org.
Report Code
[1] Guidelines for monitoring stand-alone photovoltaic IEA-PVPS
Systems- Methodology and Equipment T3-13:2003
[2] Guidelines for selecting stand-alone photovoltaic systems. Under

preparation
[3] Lead-acid battery guide for stand-alone photovoltaic IEA-PVPS
systems T3-05:1999
[4] Use of appliances in stand-alone photovoltaic systems: IEA-PVPS
problems and solutions T3-09: 2002
[5] Recommended practices for managing the quality of stand- IEA-PVPS
alone photovoltaic systems T3-15: 2003
[6] Survey of National and International Quality Assurance IEA-PVPS
Procedures and Standards for Stand Alone PV systems T3-07:2000

Table of Contents
1 Introduction.................................................................................................... 6
1.1 Objective.................................................................................................. 6
1.2 Background ............................................................................................. 6
1.3 Need for Guidelines................................................................................. 7
2 Monitoring Methodology................................................................................ 7
2.1 Reporting procedure................................................................................ 7
2.1.1 Content............................................................................................. 7
2.1.2 Aggregate results ............................................................................. 7
2.1.3 Format .............................................................................................. 8
2.2 Data Requirements ................................................................................. 8
2.2.1 General Information (Q1) ................................................................. 9
2.2.2 System Details (Q2) ......................................................................... 9
2.2.3 Quality Assurance Parameters (Q3)................................................ 9
2.2.4 Commercial Evaluation Parameters (Q4) ........................................ 9
2.2.5 Scientific Evaluation of System Performance (Q5)........................ 10
2.2.6 Battery Performance Parameters (Q6) .......................................... 10
2.2.7 Appliance Performance Parameters (Q7)...................................... 11
2.2.8 User satisfaction and adaptation to technology (Q8)..................... 11
3 Monitoring Equipment ................................................................................. 12
3.1 Level of autonomy ................................................................................. 12
3.1.1 Power supply.................................................................................. 12
3.1.2 Number of data points.................................................................... 12
3.1.3 Reliability ........................................................................................ 12
3.2 Choosing the logger .............................................................................. 14
3.3 Choosing the sensors............................................................................ 18
3.3.1 Temperature................................................................................... 18
3.3.2 DC Voltage..................................................................................... 19
3.3.3 DC Current ..................................................................................... 19
3.3.4 AC Measurements ......................................................................... 20
3.3.5 Solar Radiation Measurement ....................................................... 21
3.3.6 Wind Speed and Direction ............................................................. 22
3.3.7 Other sensors................................................................................. 22
3.4 Choosing the Signal Conditioners......................................................... 23
3.5 Wiring Techniques................................................................................. 23
3.5.1 Choosing the Correct Wires........................................................... 23
3.5.2 Wiring the System Together .......................................................... 24
3.6 Tips for successful monitoring............................................................... 25
4 Conclusion................................................................................................... 25
Appendix A - Questionnaire Templates............................................................... 27

1 Introduction
1.1 Objective
The objective of the document Guidelines for Monitoring Stand Alone Photovoltaic Power
Systems – Methodology and Equipment is to:
Describe a monitoring procedure that if followed will reassure investors, project
managers, performance auditors, equipment manufacturers and servicing firms, that the
performance data collected are functionally robust, equitable and representative.
1.2 Background
Investors, project managers, performance auditors, equipment manufacturers and servicing
firms frequently require advice about renewable energy technology options and capabilities.
More often than not, different system architectures, monitoring procedures and applications,
so complicate comparisons between systems that available performance data are rendered
next to useless. Adding to the confusion, third parties interested in “sustainable energy”
solutions may experience claims and counterclaims about product capabilities and be plied
with an assortment of performance data, some derived from the maker’s specification, some
measured in the field, but nearly all collected on a piecemeal basis without any form of quality
management.
In accordance with its current work plan, the IEA PVPS Task 3 needs to be able to analyse
the performance of case studies, to determine what comprises a successful, or conversely an
unsuccessful, installation. This embodies an assessment process conducive to equitable
comparison of different systems at different locations. Having regard to the situation, it was
decided to prepare guidelines that establish a functional and consistent approach to
performance monitoring for a range of “stand alone” PV system.
Performance assessment involves:
data collection, which is a straightforward process of measuring parameters;
evaluation of that data in a manner that provides useful information, which is a complex
exercise; and,
dissemination of useful information to interested parties.
The Guidelines supplement existing documentation dealing with performance monitoring of

renewable energy systems and their various components. Task 2 of the IEA PVPS group has
prepared a report titled Operational Performance and Design of Photovoltaic Power Systems
and Subsystems1. This document has its focus on Outcomes, in particular data
management and analysis. As such, it complements these Guidelines that have more of a
focus on the parameters that need to be monitored.
A set of monitoring Standards has been produced by the IEC, titled Standard for Photovoltaic
system performance monitoring2. The focus of the IEC standard is on the electrical
performance of PV systems, and it does not address hybrids or prescribe a method for
ensuring that performance assessments are equitable. However, units used for the Task 3
Guidelines are consistent with those used for the IEC Standard.
Otherwise, various Standards [6] are being developed around the world for hybrid system
design and performance, but none seem to address the issue of the performance comparison
of different systems on an equitable basis.
The Guidelines prescribes monitoring requirements that are believed to facilitate the
implementation of monitoring and common comparison basis for systems. It is acknowledged
that data integrity can best be served by live monitoring and automatic analysis. This is
1
“Operational Performance and Design of Photovoltaic Power Systems and Subsystems”, IEA-PVPS T2-03:2002,
Munich, 2001, 64 pages.
2
“Photovoltaic system performance monitoring – Guidelines for measurement, data exchange and analysis”, IEC
standard 61724, Geneva, 1998, 37 pages.

technically feasible, and it is recommended as a prime aim of any global data management
system established to provide a performance baseline for stand-alone power systems.
However, it is also recognised that the capability requires further development to become
universally effective, cost effective and convenient under field conditions.
1.3 Need for Guidelines

The Guidelines provide a reference manual in which the information is organised so that a
user is able to focus on a specific performance aspect, avoiding irrelevant data collection.
The Guidelines starts with a common technical approach but is adapted to the different point
of views.
To achieve this, the performance parameters have been organised into customised data sets
designed to cover the spectrum of likely monitoring needs. There is a set of general
parameters that provide the site information that is universal for all case studies. Thereafter,
purpose specific sets include:
Quality assurance parameters;

Commercial evaluation parameters;
Scientific evaluation of system performance;
Battery performance parameters;
Appliance performance parameters;
User-satisfaction and adaptation to technology.
A person proposing to monitor the performance of a system need only peruse the index to
identify the set that best suits their purpose, then turn to the relevant section of the Guidelines
to view the parameters recommended for measurement.
2 Monitoring Methodology
2.1 Reporting procedure
2.1.1 Content
Monitoring photovoltaic systems can provide useful information about their operation and
what should be done to improve performance, but if the data are not reported properly, the
effort is wasted. To be helpful, a monitoring report must provide information on the relevant
aspects of the operation in terms that are easily understood by a third party.
Appropriate performance parameters need to be selected, and their values consistently
updated with each new issue of the report. In some cases it may be beneficial to monitor the
performance of individual components in order to refine and improve system performance, or
be alerted to loss of performance in time for preventative action. For example, monitoring
battery charge/discharge profiles will signal when replacement is due before downtime from
system failure is experienced.
2.1.2 Aggregate results

In the typical monitoring situation, output will be sensed every minute, averaged every hour
and the result stored. Even with hourly averaging a huge amount of data is generated. To be
useful, the raw data must be checked for integrity then summarised.
At the minimum, data needs to be collected for at least twelve months to reflect seasonal
variations. Monthly averages are a convenient interval for summarising energy production,
consumption and efficiency. When adding or averaging results, always make sure to mention
the number of records in the data set. Results must be presented in tabular form (in
Appendix of the report) while bar graphs (in the report itself) may be added to give a
convenient view of system performance.

Power values can be measured or calculated from V*I values. In AC, the power factor
( cos(θ ) where θ is the angle between the voltage and current) must be considered to obtain
the real power. If power is calculated from the voltage and current values on a system that
performs averages, the result of the V*I product must be averaged and stored since it cannot
be calculated from the averaged voltage and current. In fact, the product of averages is
different from the average of products. The following equations demonstrate this limitation for
the calculation of average power ( P ) using n voltage ( Vi ) and current ( I i ) values.
∑V
1
Average voltage = V = i
n i =0
n
∑I
1
Average current = I = i
n i =0
Power = Pi = Vi ⋅ I i
n n
∑ P = n ∑V ⋅ I
1 1
Average power = P = i i i
n i =0 i =0
?
Is P = V ⋅I
n n n n n
∑ ∑ ∑ ∑∑V ⋅ I
1 ? 1 1 1
Vi ⋅ I i = Vi ⋅ Ii = i j
n i =0
n i =0
n i =0 n2 i =0 j =0
n n n
∑ ∑∑V ⋅ I
1 1
Vi ⋅ I i ≠ i j unless either V or I is constant
n i =0 n2 i =0 j =0
2.1.3 Format
Avoid lengthy text with numbers. Use a table format whenever possible.
If you intent to report on a regular basis on the same system, try to be as consistent as
possible on the format of presentation;
3-D bar graphs are nice but sometimes difficult to read. Use 3D wisely when reporting
results.
Choose colours and shapes wisely. Remember that a black and white copy of your report
will be made at some point.
Use SI units whenever possible.
Use the yyyy-mm-dd date format to avoid confusion when exchanging document with
people that speak different languages.
2.2 Data Requirements

Equitable assessment of the performance of "Stand Alone Systems" is complicated by system
configuration options, application possibilities and cultural variables. The Guidelines
recommend individual sets of performance parameters that reflect the use for which the data
might be intended. If adhered to, the Guidelines will improve computed data quality, simplify
performance analyses and raise the level of confidence in the conclusions inferred from the
data, especially for third parties.
In this section, the different measurement points will be explored and classified based on the
type of results expected to be achieved through monitoring. Detailed questionnaire templates
are available in Appendix A of this document. Questionnaire templates are numbered Q1 to
Q8 for easy reference.
Depending on needs and available information, case studies for the IEA database will include
General Information as described in section 2.2.1 plus any other data that may be appropriate
under the circumstances as detailed in sections 2.2.2 to 2.2.8. However, it is accepted that

there may be specific circumstances where the data sets included in the Guidelines need
customization.
2.2.1 General Information (Q1)

The General Information data set establishes the where, when, what, why and who aspects of
the project. It is included as a minimum standard for any system performance monitoring. In
the process it enables the system to be classified in accordance with the Classification
System developed by Task 3 as detailed in the IEA publication Guidelines for Selecting Stand
Alone Photovoltaic Power Systems [2]. The Classification System employs a simple key to
establish which category of stand alone system the case study best fits, in the process,
facilitating equitable performance comparison with other compatible case studies.
2.2.2 System Details (Q2)

To classify the system, detailed information about it must be gathered. The System Details
data set touches questions related to the configuration of the system, and the sizing of its key
components such as photovoltaics and battery bank.
2.2.3 Quality Assurance Parameters (Q3)

Capital outlaid to purchase a system covers hardware, any associated software and the
warranties that underpin the performance claims of the supplier. In some cases it will also
cover system installation and provide system support for a specified period.
Unfortunately, many systems are reported as performing below the expectation of the
purchaser, earning the technology (and/or manufacturer) a bad name. All to frequently
contractual responsibility is unclear and the purchaser is unable to claim the full warranty that
was paid for.
This suggests that too little attention is being given to this aspect of the investment. The
period of warranty may be inappropriate for a particular application, or the ability to legally
enforce it has been negated by an inability to prove where responsibility for a problem lies.
Was it the fault of the supplier, the installer or the user?
Questionnaire Q3 covers warranty considerations that should be addressed, both when a
system is purchased, and throughout the warranty period. Please consult the accompanying
documents for guidelines to improve and maintain quality of systems [5] and information
about international quality assurance standards [6].
2.2.4 Commercial Evaluation Parameters (Q4)

Detailed, analytical performance assessment requires costly data acquisition equipment and
involves time consuming processing and analysis. While this may be appropriate for pilot or
demonstration projects where a considerable research effort can be justified, it is rarely the
case once the technology matures commercially.
Project developers using mature technologies are interested to know how the system
performs in a commercial sense. Is the system providing the services as planned? What is its
availability? Is there a need for maintenance or repair? What is the return on investment? etc.
The commercial performance parameters provided in Questionnaire Q4 are intended to
answer these questions. By selecting only those parameters of commercial relevance, and
conducting local data-reduction (integration, averaging before storage), monitoring costs can
be kept low relative to the cost of a comprehensive scientific monitoring exercise.
Questionnaire Q4 provides a suite of parameters that might be monitored for a hybrid system.
Simpler systems such as solar home systems will require even less complexity. In a solar
home system, one need only measure the input to the charge regulator and the output from
the battery to load to gain an appreciation of how the system is performing commercially.
Energy flows can be measured with kWh meters, while outages and demand variations will
require a recording instrument such as a data logger.

From the renewable energy sources, only the “useful” energy output will be measured. This
may be less than the total energy generation if electricity demand is less than electricity
generation while the storage is full. Parasitic losses or energy dissipated in a dump load are
also excluded as useful energy.
Power Factor is included as a relevant parameter for commercial evaluation of performance
as it will impact on inverter efficiency.
Outages should be logged as soon as noticed. Whenever substantial unexplained differences
occur between measurements and expected values, the recording frequency used for the
monitoring program should be increased.
Irradiation and wind characteristics are better measured on site. In both cases it is difficult to
derive a reliable value using off-site data. Irradiation comprises global, direct and diffuse
components. Direct is very difficult to derive from global data, but the need to measure it is
limited to instances where solar concentrator systems are to be deployed.
2.2.5 Scientific Evaluation of System Performance (Q5)

The need for technical parameters to develop modelling and analysis tools for stand-alone PV
power systems is important because the optimal design of such systems is closely related to
how they are operated, and vice versa. The complexity of the modelling will range from a
very simple, straightforward exercise in the small scale PV/battery solar home systems case,
to a quite involved and complex program in the case of a PV-hybrid (PV/battery/diesel).
Task 3 case studies will, on occasions, involve monitoring of prototype systems where the
cost of detailed performance monitoring is justified, and Questionnaire Q5 provides a
checklist of the relevant parameters to be monitored in such a case.
2.2.6 Battery Performance Parameters (Q6)

The application of electrochemical storage to photovoltaic power supply systems is less
understood, than the more conventional battery applications. The requirement for regular
“deep cycling” under all possible climatic conditions seems to bring out discrepancies in the
way seemingly identical batteries perform, and this underpins debate about the effectiveness
of manufacturers management prescriptions.
Assessing battery performance is a complex exercise. Even quite simple quasi-static battery
models used in dynamic simulation programs (e.g. TRNSYS) require the following:
Current (discharging is negative and charging is positive)
Voltage (terminal voltage)
State of charge (SOC, the fraction of remaining capacity divided by total battery capacity
on an ampere-hourly basis)
Nominal battery capacity (Ah)
Temperature (°C)
In order to determine the battery SOC it is important that the people responsible for operating
the systems (or the automatic system), regularly (every 1–2 months) calibrate the total battery
capacity. One possible approach is outlined out below:
1) Charge the battery completely.
2) Allow the battery to rest (i.e. no current in or out of the battery) until an equilibrium
voltage can be observed (approx. 2–3 hours).
3) Discharge the battery completely.
4) Let the battery rest (i.e. no current in or out of the battery) until an equilibrium voltage
can be observed (approx. 4–5 hours)
If the above procedure has been followed, and current (A) and (V) has been monitored, it is
possible to calibrate the battery model using the lower and upper equilibrium voltages. An

exact battery model is absolutely necessary if an analysis on how to improve the operation of
a system with a fixed design is to be performed.
A comprehensive set of battery performance monitoring parameters is shown in
Questionnaire Q6.
2.2.7 Appliance Performance Parameters (Q7)

Investment in renewable energy generation capacity is burdened by up-front capital costs,
while fossil fuelled plant is characterized by incremental, but high, operational costs. Despite
lifetime costs generally favouring renewable energy plant, this situation tends to establish
market place bias in favour of fossil fuel plant. To reduce the bias or minimize the need for
capital, the renewable energy industry strives to fine tune system load to match system
capacity.
A site audit is prerequisite to achieving a functional balance. The audit will include auxiliary
equipment such as pumps, compressors, fans and controllers, electrical appliances such as
refrigerators, TVs, computers and tools) or lighting demands. Information collected should be
in sufficient detail to evaluate efficiency issues as well as obtain a feel for how load may vary
over time. The audit will also collect characteristic data (e.g. start-up and shut-off
characteristics) for key electrical components. Consequently, it will be possible to
recommend best practice for various pieces of electrical equipment.
Alternatively, if no information about the total or fractional electrical loads is available, one
could try to estimate these using some key pieces of information such as:
Number of electrical components from each category (auxiliary, appliances, and lighting).
Number of people living/using the building on a daily or weekly basis
Information about the type of thermal demands covered by the system should also be
collected. This may be cooling (e.g. shading devices, fans, or air-conditioning equipment),
space heating (e.g. passive solar heating, wood-burning stoves, electrical heaters, or heat
pumps), hot water heating (e.g. gas, electrical, wood, or solar collectors), or high temperature
loads (e.g. electrical, gas-burning, or wood-burning ovens). The total thermal demand (kWh)
should always be available, along with the sampling time (day, week, month, year).
Alternatively, if no information about the total or fractional thermal loads is available, one
should try to estimate these using some key pieces of information such as:
Building materials used (to estimate an overall heat loss coefficient)
Number of people living/using the building on a daily or weekly basis.
Average monthly ambient temperature and humidity (if available).
By following the procedure outlined above, one will be able to determine what fraction of
electricity used in the system serves non-high-grade (high entropy) energy purposes (e.g.
heating and cooling). Note that using PV-generated electricity strictly for high-grade energy
purposes is the first important step towards more optimally designed PV-systems.
A comprehensive set of appliance performance monitoring parameters is described in
Questionnaire Q7.
2.2.8 User satisfaction and adaptation to technology (Q8)

There is a range of allied factors that promote the concept of renewable energy programs as
distinct from a project. The consequence for any monitoring exercise is that the performance
indicators that would normally be associated with a project directed at installing and
commissioning a system are no longer adequate. Program assessment requires a new set of
performance indicators that measure user adaptation to the introduced technologies.
Questionnaire Q8 provides a list of questions dedicated to monitor the extent of satisfaction
and adaptation of the user to the system.

3 Monitoring Equipment
The choice of data acquisition hardware is not easy. It is modulated by numerous factors.
The purpose of this document is to give some tips and tricks to build precise and reliable
systems while avoiding the frustration of falling in typical pitfalls of data acquisition.
3.1 Level of autonomy

Level of autonomy in this case implies three things:
1) Power supply;
2) Number of data points that can be stored; and,
3) Reliability.
The accessibility of the site where the data acquisition (DAQ) system is to be installed,
whether it is just next door or on a mountaintop you can only access once a year, will dictate
the level of autonomy you require. This is modulated by the fact that people can be located
near the site, but it is also greatly affected by their competency and level of interest in picking
up data once in a while. This section will describe how to deal with these three elements of
autonomy.
3.1.1 Power supply

For stand-alone systems, there are three solutions. For short term monitoring with few
readings and without external signal conditioners, a primary battery can be used to power the
unit for the time of monitoring. Otherwise, power must be produced on site. The best
situation is if you can have your own power supply for the monitoring system but it implies
some cost. However, with advance monitoring technology today, it is possible to have units
that draw very little power and thus won’t affect the system in a significant manner. But, if the
system can't suffer any power draw or if the logging function is too power consuming with
respect to the site supply, a separate power source must be designed for the DAQ system.
3.1.2 Number of data points

The number of data points grows surprisingly fast. Also, on many systems the date and time
are also datapoints. There are four ways to deal with the volume of data:
1) Print them;
2) Store them on a data card that you can collect once in a while;
3) Store them in the DAQ system and get them with a laptop once in a while; or,
4) Use a dial-up connection.
The first three imply that you have to go get the data on site. For short term monitoring, this is
fine but for long term, it must be integrated to the scheduled maintenance of the system,
otherwise it becomes prohibitively expensive to make a special trip to the site to get the data.
Whenever possible, a dial-up connection is preferred, but accessibility and cost of phone line
may render this impractical. Cell-phones may also be used for dial-up in areas where the cell
phone coverage is adequate.
3.1.3 Reliability
Reliability depends on:
1) How you choose your equipment; and,
2) How you use it.
One important rule: “One must not expect a piece of equipment to do what it's not meant for.”
This seems straightforward but this rule is often breached. Look at the specifications,
temperature, voltage, common mode range (which is often overlooked), etc., and choose

wisely. Also, try not to be too impressed by bells and whistles. While some features can be
very useful for a specific application, other features may draw power that you can ill afford.
Trust systems that have been out there for a while. You don't want anything to go wrong, but
if it does, you want a quick answer.
The second point is what you do with the equipment. Here is a little list of tips that are often
overlooked and that cause the death of many systems.
Mechanical/environmental considerations
Put your DAQ system in an enclosure to avoid dust build-up;
Ensure sufficient cooling;
Use silica gel whenever possible to avoid moisture build-up in your DAQ system;
In dry regions (poles or deserts), sometimes you have to shunt your high impedance
inputs a bit to the ground (with a 10 kΩ - 100 kΩ resistor) to avoid accumulation of static
electricity;
Measurement techniques considerations

Use a shunt whenever possible to measure DC current;
For AC current, choose current transformer with low hysteresis;
Make sure that your measurement technique does not significantly affect the value you
are trying to measure;
Use differential inputs whenever possible;
Electrical considerations
Keep wires short;
If you need isolation when measuring DC current, closed-loop hall effect sensors are a
good alternative;
Watch for current loops,
Use single point-earthed shielding,
Do not share the same wires to carry power and transmit analog signal,
Avoid multiple ground connections;
Use the least vulnerable form of signal transmission available,
Convert to digital as soon as possible,
Use current transmission whenever possible since series transmission is less
vulnerable to noise,
Boost low voltage signals close to the source if you have to transmit it on a long
distance;
Data quality considerations

Never have less than two weeks of slack with your data collection agenda. It is not
always possible to collect data on the exact date. Think about holidays, summer
vacations, sick days, travel etc.; if you want to download data every month, have at least
45 days of storage;
Find precisely the location of the site (longitude, latitude, altitude) to be able to compare
your measurements with statistical values;
Prepare and follow a calibration schedule;
Keep a logbook of the monitoring system that includes maintenance, failures, manual
interventions and changes of configuration or components.
Troubleshooting considerations
Choose every component on quality not cost (even a tiny resistor). In a remote location,
the real cost of replacing a failed component may only become evident when it fails;
Keep an up-to-date copy of the plans and layouts;
Take pictures of the system so you may help somebody repair it if something fails;
If you have sufficient resources, keep a spare of each of your sensors so you can
undertake quick repairs and avoid downtime while performing calibrations.
Have a comprehensive documentation of your DAQ system.

3.2 Choosing the logger

One may think that when he knows what he has to measure, the choice of a logger becomes
straightforward… Wrong! That is just the beginning of the compromising phase. The figure
below shows the block diagram of a typical data logger system.
Data
Storage
Analog MUX
Signal
Analog Cond.
S/H ADC
Inputs
CPU I/O Interface

Counter 1
Digital MUX
Counter 2
Digital
Inputs
Figure 1. Block diagram of a typical data logger
The structure may change quite a lot depending on the features of the logger but the main
blocks will remain:
ADC (Analog to digital converter) The ADC is the core of the logger unit. The ADC is
the circuit that converts an analog voltage signal to a
digital value of n bits.
CPU (Central Processing Unit) The CPU is the brain of the logger. The functions
performed by the CPU may range for a simple
interface with the ADC for computer driven loggers to
a full-featured programmable unit for autonomous
loggers.
Data storage Again, the distinction between computer driven units

and autonomous units is very important. Computer
driven units will usually have a buffer of only a few
samples because it is assume data will be stored on
the hard drive of the attached computer providing
virtually infinite capacity. Autonomous units on the
other hand store the data on board or on memory
cards where millions of samples storage capacity
can be achieved.
MUX (Multiplexer) Since loggers usually have only one ADC a

multiplexer is used to select the desired input source.
Multiplexers are also used in controller units to
activate outputs.
S/H (Sample and hold) While the signal at the input of the data logger may
vary with time, the conversion from analog to digital
is never instantaneous. Consequently, the reading
may be affected by this variation. To avoid this, a
sample and hold circuit is used to maintain the signal
constant during the conversion.
Some sophisticated units dedicated to perform

simultaneous readings of multiple channels will have

S/H chips on each channel to ensure that all

channels are sampled at the same time.
S/H
Analog MUX
S/H
S/H
S/H
Signal Conditioner The real-life signal can be anything (voltages,

currents, resistance, temperature, etc) while the ADC
circuit only accepts a fixed range voltage signal. The
role of the signal conditioner is to convert the real life
signal to the appropriate voltage range for the ADC.
When selecting a data logger unit, the exact internal behaviour is not important but one still
need to know what happens to his or her data in order to do a judicious choice. To do so, a
certain number of parameters are provided by the manufacturer to help the user makes a
choice. The following list gives an overview of these features/characteristics and what must
be looked at when choosing the logger:
Specifications pertaining to analog signals
Wiring method Voltage reading analog channels can be either

specified as single-ended or differential. The single-
ended wiring method assumes that all sensors can
share a common connection point for one of their
signal transmission wires. In many cases, this point
will be the ground of the DAQ system. In differential,
the DAQ system measures the difference between
the two wires, allowing both wires to be at any
voltage within the common mode range of the unit.
A more detailed description is provided in
section 3.5.2. Other sensors such as RTD and strain
gauge will require different wiring methods that
should be investigated prior making a choice.
Number of Channels The number of channels is unit specific. Loggers

often allow converting differential channels into 2 or
even 3 single-ended channels hence doubling or
tripling the channels available.
Channel Expansion Allows the use of other modules to increase channel

capacity.
Common Mode Range This is the maximum voltage any single input can
take without impairing the operation of the unit. A
high common mode range can be useful if a shunt
has to be placed on a live wire or if a DC voltage has
to be measured on a point of high voltage.
Common Mode Rejection Ratio Often written as CMRR. This is the ability of the
logger to ignore the common mode voltage. This
value is expressed in dB. The error (Verror) cause by
the common mode voltage (Vcm) can be calculated
as follow:

⎛ CMRR ⎞
−⎜ ⎟
Verror = Vcm ⋅ 10 ⎝ 20 ⎠
e.g.: for a data logger having a CMRR of 90dB,

measuring a shunt (50 mV) at a voltage of 5 V above
ground will introduce an error of 0.158 mV on the
reading or 0.3% This may seems low but at 5% load
(2.5 mV), this shunt reading will be off by 6%. This
shows well that for small signals, it is important to
have a high CMRR and to keep the common mode
voltage low.
Type of input Many types of inputs can be available: voltage,

current, resistance, thermocouple, strain gauge, etc.
Depending on the logger, different types will be
available and configurable via software or
dipswitches. Not all logger inputs are bipolar so care
must be exercised when connecting the sensors.
Resolution It is an important feature for the accuracy of data

collected. It usually ranges from 8 to 15 bit. If all bits
are significant, the resolution, in measured unit, can
be estimated by full-scale value/2n_bits. For example,
a 5 V channel on a 10-bit logger will have a
resolution of 5/210 i.e. 4.9 mV.
Sample rate The sample rate is expression in number of

conversion per second. This is the speed at which
the data is collected.
ADC method The analog to digital conversion method can be is

closely tied to the sample rate and resolutions. The
table below summarize the characteristics of the
different ADC types:
Table 1. Comparison of different analog to

digital conversion methods
Typica Sampling Typical
Type Noise Rejection
Rate Resolution
Successive
50kHz - 1MHz 8-16 bits Low
Approximation
Integrating below 30Hz 12-24 bits Very Good
Flash above 1 MHz 4-8 bits None

Wide range
Delta-Sigma 8-20 bits Good
50Hz - 250kHz
Where speed is not an issue, a Delta-Sigma or

Integrating ADC will be excellent. The advantage of
integrating ADC is that it allows to adjust the
integrating period to a value equal to the power line
period (20ms for 50Hz or 16.7ms for 60Hz) thus
improving the rejection of this major source of noise.
Specifications pertaining to digital signals
Digital Channels – Input/Output Useful to generate digital transition events to trigger

data acquisition or to control relays. Those channels

are often configurable in input or output mode.

Otherwise, they are presetted as input or output.
Counter Channels – Fast/Slow Speed and range of the counter are configurable.
The logger counts even when logger is in stand-by
mode reducing power consumption.
Other specifications
External sensor power-up Some loggers provide an output to power external

sensors. This is usually linked to the stand-by mode
feature of the logger. This feature allows to power
signal conditioners and sensor only when a
measurement is about to be taken. This is
particularly useful for low power systems. Special
care must be taken to choose sensors that can be
used intermittently with a short start-up time. If the
power drawn by the sensors is higher than the
allowed power from the logger or if the voltage level
is inadequate, a relay can be used.
Communication interface A lot of interfaces exist today: Serial (RS-232, RS-

485, USB) Parallel (SPP, EPP, ECP), GPIB
(IEEE 488), Ethernet, etc. For autonomous loggers,
the RS-232 is still a convenient interface. It is found
on most computers and it allows direct cable
connection or dial-up connection (with a standard
modem) to download data.
In all cases, interfaces should always be isolated in

some sort otherwise, if there is a failure of the DAQ
system, the attached computer or other device may
be damaged.
Removable Data Card Allows data storage on a card that can be replaced.
This is very useful to transfer data from a local
without communication means other than somebody
that pass by once in a while.
Internal Battery Allows the logger to continue acquisition when

external power supply is not available. Coupled with
the power requirements, this will define the
autonomy of logger when operating in stand-alone
mode.
Power Requirements When power is not available from utility, the logger
needs to have very low power consumption.
Usually, power consumption will vary with respect to
the number and rate of channels acquired.
Stand-by mode The logger can have a built-in “sleep mode” that
allows it to limit the energy consumption of the
system.
Temperature Range Some loggers can support ambient temperature

variation from –55°C to 70°C.
Different kinds of loggers are available in the market. You can find full-blown models with
multiple features for specialized application or models with fewer features but that offer a low
cost alternative. Whatever your needs are, chances are you will find a data acquisition

system that can fulfil them. However, whatever logger is chosen, it will always be a
compromise between cost, features and ease of use, so cost must not always be the decisive
factor. It is also advisable to choose with future needs in mind. Learning all the nuts and
bolts of a new logger model is time consuming, so sticking to a logger type/model will often
save time and avoid surprises such as no data for six months due to a missing configuration
parameter.
3.3 Choosing the sensors

Choosing the right sensor is important because that is where your first measurement error
gets in the loop and no matter how precise the data logger is, once in, the error will stay.
Sensors have to be chosen alongside with the data logger. There are various ways to
measure a single value, but the choice made will affect the logger requirements and the
method of measurement may require adjustment to fit within the cost and power limits
established.
Before looking at measurement techniques, here are the basic rules of data acquisition.
Limit impact on the system operation;
Avoid intermediates;
The simpler the better;
Add some redundancy;
Match sensor and system ranges.
3.3.1 Temperature
Common sensors to measure temperature are thermistors, thermocouples and RTDs
(Resistor Temperature Devices). The following table describes the different sensors and their
pros and cons.
Table 2. Temperature Sensing Elements Comparison
Sensor Thermocouple Thermistor RTD
Active element Two different metals A semiconductor plate A platinum wire
coupled together
Output A small voltage Resistance vary with Resistance vary with
proportional to the temperature temperature
temperature
Linearity Good Bad Excellent
Approximate Low (US$8) Medium (US$45) High (US$85)
Cost (probe unit)
Sensitivity Low High Low
Precision Medium (0.5-1°C) Medium (>0.5-1°C) High (≈ 0.1°C)
with a good equation
Distance 100 m > 300 m (4 wire conf.) >300 m (4 wire conf.)
Application Excellent for general Excellent on a very small Excellent for high precision
temperature range of temperature (to measurement on a wide
measurement for a trigger an alarm for range of temperature (∆T
limited range of example) but bad for DAQ >100°C).
temperature on a wide range.
(∆T<60°C).
Most of the time, thermistors are not appropriate for DAQ applications due to their lack of
linearity. They often need a special signal conditioner, which is cumbersome in DAQ
applications. RTD on the other hand are very precise, but their high cost often deters users.
Moreover, they will require a DAQ unit that supports RTD, which is not as common as for
thermocouples. To conclude with RTDs, let's say that the sensitivity is very low so to

measure the resistance accurately, a 4-wires measurement is preferred. 3-wires can be used
but be warned that not all manufacturers fully support the 3-wire compensation. The following
figures show the configuration and compensation for 4 and 3-wire RTD measurements.
I
R I
R
V*
DAQ DAQ
V RTD RTD+R*=V/I RTD
RTD=V/I RTD=(V-V*)/I V
Assuming R*=R
R* R*
4-wires 3-wires
Figure 2. RTD Connection Arrangements
Usually, a thermocouple is sufficient to provide a good resolution. For PV monitoring

applications, a T-type thermocouple is preferred since it has a temperature span going from
-270°C to 400°C for thermocouple grade, and -60°C to 100°C for extension grade covering
whole spectrum of applications. K-type thermocouples are more commonly found but their
precision is lower and the extension grade is not suitable for sub zero temperatures, which
may be a problem for cold regions.
One drawback with thermocouples is that they need special extensions and connectors to
allow the continuity of the metal used. This is a minor problem because extensions are easy
to find and cost less than US$1 a metre for good quality wire.
Temperature sensing elements are also offered in different shapes to accommodate the type
of measurements to be taken. For example, to measure the temperature of a PV module, a
flat sensor, solidly fastened to the back of the PV module will be preferred. For free air
measurement, a radiation shield, a bulb that covers the sensor while allowing the air to pass
through freely, should always be used to avoid the effect of rain, wind and direct sun.
3.3.2 DC Voltage
Voltage measurement is probably the easiest measurement of all since all monitoring units
have that capability. Most of the time, a simple pair of sensing wires will be sufficient. If the
voltage is too high for the input of the data logger, a voltage divider is used. Here are a few
rules of thumb about voltage dividers though:
The impedance of the output resistance of the voltage divider shall be at least 100 times
lower that the impedance of the data logger input;
The power dissipated through the resistance of the voltage divider shall not exceed 10%
of their nominal power to limit heating it up;
Metal film, low temperature drift (<25 ppm/°C) resistors are preferred.
If direct measurement is impossible, modular isolated signal conditioners are available on the
market with multiple output types.
3.3.3 DC Current
Measuring DC current can be done either with a shunt or with a hall-effect sensor.
Shunts are simply low value resistors that will output a voltage proportional to the current
flowing through them. While shunts will offer the highest precision, they also have two
drawbacks. First they offer no isolation, and second they need you to break the current path.
Since they are not isolated, caution must be exercised with the placement of shunts if they
are intended to be measured directly by the logger. A common mistake is to place shunts on
live wires leading to a high common mode voltage (Vcm). If direct measurement of shunts is

intended, all shunts should be placed as close to the neutral as possible to avoid errors (see
common mode rejection ratio) and potential damage to the unit. The following figure illustrate
the issue with the common mode voltage and the placement of shunts. If the system is
floating, in the sense that there is no neutral point, signal conditioners shall be used to
transmit the voltage to the logger.
48 V
Data Logger
Right Vcm=20 Vmax
Shunt
50 mV
48V 48V
Load Load
50 mV
Shunt
50 mV
Data Logger
Vcm=20 Vmax
Wrong
Figure 3. Common mode voltage issue with shunts
Hall effect sensors, though often less precise than shunts, offer multiples advantages. Two
main categories of hall-effect sensors are marketed, open-loop and closed-loop sensors. The
difference between these two resides in the way the hall-effect element is used. With open-
loop sensors, the hall-effect element is used to give a linear reading of the magnetic field,
which is then converted into current. In closed-loop sensors, a counter-current is used to
nullify the magnetic field and the hall-effect element is then used to determine if magnetic field
is present at all, thus leaving out any non-linearity the element may have. Generally, closed-
loop sensors are more precise, but they also are more expensive. The main drawback of this
type of sensors is that they need external power. Hall-Effect sensors are relatively easy to
find. Split-core sensors on the other hand may cause more trouble
Table 3. DC Current Sensors Comparison
Shunt Hall Effect
Sensor
Need external power N Y
Isolated N Y
Error <1% 1-5%
Temperature effect low Medium
Break current path Y N
Exist as clamp on devices N Y
3.3.4 AC Measurements
Most of the time, AC measurements will be used to get the power or energy output of the
system. Consequently, there is little interest in the voltage and current values themselves. If
voltage and current values are required, individual signal conditioners can be used to convert
the values into acceptable inputs for the logger. Current will most of the time be picked-up by
current transformers. Shunts can also be used but they do not provide isolation.
Moreover, energy meters can be useful to control the consumption for village power for
example.
WARNING: The output of current transformers must be loaded at all
time during operation, otherwise dangerously high
voltage may arise. Never place quick disconnects or

attachment plugs between a current transformer and its

load.
Wattmeters will take a current input, directly or from a current transformer, and a voltage
input. Depending on models, wattmeters will provide one or more of the following:
Real power (W)
Apparent power (VA)
Reactive power (VAR)
Power factor
Energy count (Wh per ticks)
Here are a few tips to choose watt transducers:

Range of wattmeters is usually given in VA, so the power factor of the equipment must be
taken into account when choosing the transducer;
All wattmeters do not have the same frequency response. True RMS wattmeters must be
used at the output of a modified square inverter or other noisy equipment (including
loads).
3.3.5 Solar Radiation Measurement

The choice of a sensor for PV radiation measurement will depend on the investment one is
ready to make for such a measurement. The following table gives of overview of the different
technologies available.
Table 4. Solar Radiation Sensors Comparison
Thermopile Photodiode Reference cell
Pyranometer Pyranometer
Principle of Two thermopile, one A semiconductor A shorted solar cell is
operation exposed to sunlight device that will used. The current
and the other one generate a current generated is
hidden to measure proportional to solar proportional to solar
ambient temperature. radiation. radiation.
The voltage
difference between
the two is
proportional to the
solar radiation.
Error due to < 0.2%/°C < 0.5%/°C < 0.5%/°C
temperature
Acceptance angle High Moderate Low
Error due to Low High High
spectral shift
Stability over time High Moderate High
Typical error < 1% < 5% < 5%
Cost (approximate) US$1800 US$300 US$100-2000
Precise solar radiation measurement is generally required only for research purpose in which
case, a thermopile pyranometer will be the tool of choice. Otherwise a simple photodiode
pyranometer or a reference cell can do the job. If reference cells are used, caution should be
taken to match the technology of the reference cell with the technology of the photovoltaics
used.
Average solar radiation falling on a place can be estimated mathematically if you have the
latitude, longitude and tilt of that plane. The equations are very well explained in Duffie and

Beckman “Solar Engineering of Thermal Processes” published at John Wiley and Sons3.
However, some factors as ground albedo or cloud cover are often difficult to estimate.
Measured averaged ground radiation data can be obtained from weather services of the
different countries. However, many locations of the globe do not have ground weather
stations, in which case, satellite data can be used. The NASA Langley Research Center now
offers to the wide public the results of their worldwide satellite measurements through their
website at http://eosweb.larc.nasa.gov/sse/. This is an excellent resource when ground data
are not available.
3.3.6 Wind Speed and Direction

Wind speed and direction can be an important parameter for thermal analysis of system
shelters and photovoltaic array performance. If only wind speed is important, a simple 3-cup
anemometer will do the job. These units cost around US$200-US$450. If both wind speed
and direction are required, there are four options:
Single propeller on a vane Economical unit. Precision is however impaired by
the fact that the vane and the propeller are disturbing
each other. (≈ US$900)
3-cup anemometer and vane Often the preferred option because wind speed is
measured independently of the direction while being
at a reasonable price. Moreover, units (anemometer
and vane) can be replaced separately. (≈ US$1200)
3-axis propellers Three propellers in X, Y and Z measure the wind in

all directions. This type of equipment is excellent for
measurement of turbulence. (≈ US$2000)
Ultrasonic anemometer This is one of the latest innovations in wind

measurement. Having no moving parts, these
require little maintenance and calibration.
(≈ US$2200)
3.3.7 Other sensors

Other sensors may be useful for scientific study of PV systems. Though humidity has little
effect on the equipment itself, it may change the solar spectrum, thus affecting the
performance of photovoltaic modules. Solid state humidity sensors are available on the
market, usually from companies specialised in meteorological instruments. Always use a
radiation shield with humidity sensors. Single probes measuring both temperature and
humidity are also available.
The temperature of crystalline silicon PV-array affects the overall performance of the system.
Simply measuring the temperature of the array may be sufficient to calculate the actual
performance but is not sufficient to predict the behaviour in other insolation, wind and load
conditions. The best case is of course if the overall heat loss coefficient (W/m2) and the
thermal capacitance (J/K-m2) is readily available, but this is seldom the case. Combined with
physical placement (air gaps), ambient conditions (insolation, wind) and load (how much
electric power is flowing out), thermal elements such as thin film heat flux sensors can be
used to characterize photovoltaic modules thermal parameters.
3
Duffie, John A. and William A. Beckman. Solar Engineering of Thermal Processes, Second Edition.
New York, NY: John Wiley & Sons, Inc., 1991, 917p.

3.4 Choosing the Signal Conditioners

Signal conditioners are versatile devices. They allow you to convert signals from one form to
another more suitable for data acquisition needs. For example, to get the AC power out of an
inverter, the following equation needs to be resolved: Vrms x Irms x cos φ. This may seem trivial
on a calculator, but loggers don’t support cosines and few will be able to measure the RMS
(root mean square) value of an AC signal. Most of the time a signal conditioner will be used
to do the job.
Another reason to go for a signal conditioner may be simply to amplify or convert the signal
into a less vulnerable signal. For example, if a pyranometer having a gain of 10 µV/(W/m2) is
located 300 m of the data acquisition system, the signal is likely to pick-up a lot of noise along
the line. It becomes desirable to “boost” that signal near the sensor with a 25 mV / 5 V
amplifier or convert it into a current loop with a 25 mV / 4-20 mA signal conditioner.
When choosing a signal conditioner, the following information must be considered:
Level of isolation (to avoid destroying all the system if one component fails);
Supply voltage (one must remember that a 24 V battery bank can go up to 28 V easily)
Warm-up time (if used intermittently);
Temperature of operation;
Temperature stability;
Field configurability (stay away from units that need a computer to get configured);
Polarity (many units are unipolar and will not be appropriate to measure current flow for
example).
3.5 Wiring Techniques

Wiring may seem evident but that is not always the case.
3.5.1 Choosing the Correct Wires

Instrumentation wires are available in a multitude of types and brands. Following is a list of
features and elements to look at.
Size Current used for signals is generally low so the size
is generally dictated by practical needs and perhaps
the current required to power the sensors.
Stranding This is the number of small strands or threads in

each single wire. The more strands you have, the
more flexible is the wire. Most of the time, you want
to avoid single core wires since they will tend to be
less reliable due to their tendency to break.
Material Copper is the most common type used. Aluminium

must only be used with active compression
connection. Tinned copper is often preferred since it
reduces the corrosion factor.
Type of cable Apart from standard cables where wires run parallel
to each other, there are twisted pair and coaxial
cables. Twisted pairs, as its name tells is simply
accomplished by twisting each pair of wires. This
allows higher noise immunity, especially if shielded.
Coaxial cables are built by having a central
conductor surrounded by the shield, which is also the
second conductor. This gives the highest immunity
but is very expensive and often impractical.

Shielding Shielding can be common for the whole cable or

individual for each pair of wires. Individually shielded
pairs will avoid cross-talk between pairs.
Isolation level Expressed in volt, this is the maximum voltage

allowed in the cable.
Isolation type This is the material used for the jacket of the wires
and cable. Three factors must be considered: 1)
chemical resistance, 2) mechanical resistance and 3)
UV stability. Different material will have different
resistances to chemical substances. For example,
nylon is not good near batteries where sulphuric acid
is present. Requirements for mechanical resistance
are defined by the exposure of the cable to human
interaction and weather. Finally, the UV stability is
very important for wires intended to be installed
outside. If not chosen properly, isolation will crack
and cause short-circuits.
Coding Wire coding in a cable is done either with colours or

with numbers. The choice is to user discretion.
Numbers are often difficult to read on wires but for a
colour-blind, this is a small price to pay.
Impedance Wire impedance is expressed in ohms. DC or low

frequency AC signals over a short distance are not
affected by wire impedance. However, for wires
used for communications (Ethernet, RS232, RS485
or analogue) the impedance becomes a key
parameter. If the transmitter/receiver does not
already have one, a termination resistor must be
added to match the impedance.
3.5.2 Wiring the System Together

Wires intended to carry power shall always have overcurrent protection. A fuse or breaker
will protect the wire, but also avoid bringing the whole system down if something goes wrong
with one sensor. The same rule should be followed with signals with potentially high current,
like battery voltage. In this case, protection can also be achieved by placing a resistance in
series with the wire. Care must be taken however not to alter the precision of the reading by
doing so.
As mentioned earlier, either single-ended or differential input can be used. Figure 4 shows
the connection difference for two thermocouples.
1+ 1+
1- 2+
2+ 3+
2- com
Differential Single-Ended
Figure 4. Comparison of differential and single-ended connection mode

Single-ended connections require less wires and simpler acquisition systems. However that
is where the advantages stop. They present also a lot of drawbacks such as:
Higher cross-talk between channels;
Higher vulnerability to noise;
Perfect condition to create current loops;
One wire must be common (grounded); and,
Decreased reliability since many channels depend on a single wire.
This does not mean that single-ended should not be used, but simply that the designers must
be careful in their use. Signals can be grouped under the single-ended method only if:
The common does not become a supply wire (total or partial);
All sensors are floating (no reference to the ground) or the common to all sensors can be
connected at only one place on the system;
The supply of the DAQ system, if connected to the DAQ common, has no direct
connection with the unit being measured, otherwise current may flow. This may be
corrected with isolated DC/DC converters.
Finally, special care must be taken in proper shielding of signal cables. Theoretically, best
shielding is achieved by grounding the shield at both end of the cable. Unfortunately,
practically, it is rarely the case, since ensuring that both ends are at the same potential is
impossible. Consequently, the best practical way of shielding is to ground the shield at the
data logger side. Many loggers have a ground screw or lug for this purpose.
The shield may not be used as a current carrying wire since it will inject noise. Also, the
ungrounded end of the shield must be isolated to prevent inadvertent contact with the
surrounding.
3.6 Tips for successful monitoring

To conclude this document on monitoring equipment and protocols, here are a few rules of
thumbs:
Use a shunt whenever possible to measure DC current;
If you need isolation when measuring DC current, closed-loop hall effect sensors are a
good alternative;
For AC current, choose current transformer with low hysteresis;
Make sure that your measurement techniques do not affect significantly the value you are
trying to measure;
Choose low power consumption sensors;
Watch for current loops;
Add some redundancy;
Whenever possible, use the less vulnerable form of signal transmission available;
Choose every component as you would do for a US$5000 piece (even the tiny resistor), it
may be the cost involved to replace it when it fails;
Keep a logbook of your system;
Keep an up-to-date copy of the plans and layouts;
Take pictures of the system so you may help somebody repair it if something fails;
Find precisely the location of the site (longitude, latitude, altitude) to be able to compare
your measurements with statistical values;
Prepare and follow a calibration schedule;
If you have sufficient resources, keep a spare of each of your sensors so you can repair
quickly and avoid downtime while performing calibrations.
4 Conclusion
The renewable energy industry is emerging from its “backyard” industry identity, to earn a
degree of respectability in corporate circles. This “coming out” process is accelerating, as
global environmental and social concerns raise moral issues and translate into higher
insurance premiums and social discontent.

The involvement of utility scale businesses in the industry is providing pressure for sensible
standards and performance guidelines, but the introduction of effective measures is fraught
with difficulties. The task requires that the performance of complex system technologies,
used in complex applications, be compared on an equitable basis.
As outlined though this document, equitable performance assessment of photovoltaic
systems required the combination of four key elements:
• Good knowledge of the system
• Accurate measurement techniques
• Quality procedures
• Reliable data gathering
Methods as presented in this document should allow to gather quality in order to achieve a
rigorous assessment of “performance” in all of its spheres, namely: technical, financial,
environmental and social acceptance. Nevertheless, these Guidelines are not presented as
the final answer to the problem. Persons using the Guidelines will no doubt encounter
difficulties with some of the prescriptions contained therein, and may well develop alternative
approaches that are more useful in their circumstances. When this happens, it is important
that they contribute to the process of refinement by contacting Task 3 with a description of the
improvement.

Appendix A
-
Questionnaire
Templates

Introduction
This appendix present a set of questionnaire templates to defined the required parameters in
order to perform exhaustive analysis of a system with respect to eight different goals namely:
Q1: General information
Q2: System details
Q3: Quality assurance parameters;
Q4: Commercial evaluation parameters;
Q5: Scientific evaluation of system performance;
Q6: Battery performance parameters;
Q7: Appliance performance parameters;
Q8: User-satisfaction and adaptation to technology.
In all cases, a general description will be required to at least what type of system is looked
after. The system details will be required whenever some technical details or classification
will be required.
Parameters defined in Q3 to Q8 will be used accordingly depending on the questions answers

are looked for. Here are a few examples:
1. System suppliers may want to use Q8 to assess user-satisfaction and Q4 to help
troubleshoot his systems.
2. Research organisation will most probably refer to Q5 for research purpose and Q6 is
heavily involved in battery performance issues.
3. Project funding agencies will want to assess the quality of the installation done by the
suppliers and will consider the Q3 subset.
4. The manager of a village power system may want to assess the impact of new
appliances on the local grid and could use template Q7 for this purpose.

Q1 General information
Location
Country:
Region:
District:
Site name:
Contact Person:
Name:
Company:
Address:
Phone:
Facsimile:
E-mail:
Phone:
Facsimile:
E-mail:
System Management:
Owner:
Installer:
Operator:
End user:
Maintenance:
Site co-ordinates:
Latitude:
Longitude:
Altitude:
Project type:
Aid:
Prototype:
Market demonstration:
Commercial

Q2 System Details
System details:
System Classification*:
Average electricity demand (kWh/d):
(design requirements)
Array capacity (Wp):
Make:
Model:
No of Panels:
Electrical configuration: (attach schematic):
Charge controller details:
Inverter details:
Commissioning date:
Array mounting system (short description):
Battery details:
Manufacturer (name and country):

Model and reference:
Rated Capacity:
C10 (Ah):
C100 (Ah):
Rated voltage (V):
Number of cells of monobloc:
In series:
In parallel:
No. of days of autonomy
Resource Data:
Radiation on array plane (W/m2):
Ambient temperature (oC):
Availability of low ion water:
* System classification key is outlined in the Guidelines for Selecting Stand Alone PV Power Supply
Systems (IEA-PVPS T3-).

Q3 Quality Performance Evaluation
Contractual Information:
List legal contracts:
Nominate who is responsible for system:

Nominate who is responsible for operation
& maintenance:
Warranty Details:
System operation warranties:
Individual component warranties:
Post commissioning system modifications:

Detail changes to generating capacity:
Detail changes to civil works:
Detail changes to power electronics
Detail changes to control system

software:
Outages:
Type:
Duration:
Cause, including the service record of the
failed equipment:
Action taken, with cost born by?
Performance assessment:
What parameters are measured?
What are the data management
provisions?

Q4 Commercial System Performance
Average daily load:

Average daily load:
DC (kW)
AC (kW)
Energy Flow:
Useful energy from the PV-array
(kWh/month):
Useful energy from each of the other
renewqable energy sources (kWh/month):
Energy from the auxiliary back up system
(e.q. a diesel) (kWh/month):
Energy to the load via the inverter
(kWh/month):
Energy to the battery (kWh/mounth):
Energy from the battery to the inverter
(kWh/month)
Power Factor: 0 to unity
Outages of the different system
components and loss of load
(duration/month)
DSM provisions:
Efficiency Measures:
Brefly describe any investment to improve
the thermal performance of buildings and
appliances:
Client status (Access to):

Project capital and operational funding
mechanisms:
Technology transfer arrangements:
Backup from system service technicians:

Costs:
Number of service users:
Capital costs:
Operation & maintenance costs:
Cost associated with managing the
service or program:

Q5 Technical evaluation
Resource Data:
Global/direct irradiation ((kWh/m2)/a):
Wind speed (annual average m/s):
Other relevant RE resource parameters:
Load Characteristics:
Daily demand curve dc/ac (power
vs.time):
PV contribution to meeting load (%):
Capture losses (Wh):
PV array:
Module average maximum operating
temperature (oC):
Array rated power (W):
Measured average peak performance
(W):
Array output delivered to inverter (kWh/a):
Array output delivered to batteries
(kWh/a):
Array outages (h/a). (attach schedule if
appropriate):
Additional Renewable Energy generation (eg wind power):

Energy output of each alternative sources
(kWh/a):
RE output delivered to load (kWh/a):
RE output delivered to inverter (kWh/a):
RE output delivered to batteries (kWh/a)
RE output to dump load (kWh/a)
Outages (h/a). (attach schedule):
Auxiliary “backup” system (eg Diesel):

Energy generated by aux (kWh/a):
Aux battery charging (kWh/a):
Aux output to load (kWh/a):
Aux fuel/energy input
Outages attributable to Aux. System (h/a)
(attach schedule) :

Energy efficiency provisions:

Building design (Briefly describe any
investment to improve the thermal
performance of buildings).
Appliance and tool load & efficiency

(Briefly describe results of the energy
audit):

Q6 Battery performance
Battery:
Rated capacity:
C5 (Ah):
C10 (Ah):
C50 (Ah):
C100 (Ah):
Performance @ 20oC and 45oC:

Cycle life (cut-off @ 80% rated capacity):
Self discharge rate (% of C5/5
capacity/month):
Internal resistance fully charged:
Predicted shelf life in dry charged state:
Physical arrangement:
Nominal voltage (volt):
Cut-off voltage (volt):
Number of cells of mono-block:
In series:
In parallel:
Battery type:
Solar:
Conventional:
General Technology:
Car or SLI
Traction
Stationary
Flooded
Sealed: AGM or Gelled
Plate structure
Tubular plate:
Flat plate: < 4mm or > 4mm
Grid alloy:
Lead-Antimony %:
Lead-Calcium %:
Rated specific gravity (fully Charged):

Sediment space (distance between
bottom of container and the bottom of the
plates, expressed as a percentage of
plate height):
Electrolyte reserve (the distance between
the top of the container and the maximum

electrolyte level, expressed as a

percentage of plate height):
`
Climatic conditions:
Temperature range:
Lowest temperature (oC):
Highest temperature (oC):
Other specific conditions:
Supply conditions:
Ready to use (electrolyte filled): Yes No
Dried-charged: Yes No
Conditions of installation:
Initial charge recommended: Yes No

Initial charge performed: Yes No
Conditions of maintenance:
Performed by the end-user: Yes No

Performed by technicians: Yes No
If yes, frequency? (Months)
Lifetime:
Date of installation:
Expected lifetime, if new or still in
operation (years of cycles):
Actual lifetime, in case of replacement
(years):
Reasons for having selected this type of battery:
Price:
Warranty period:
Expected lifetime:
Cycling performance:
Price / performance ratio:
Maintenance free:
Easy transportation:
Easy installation (no

electrolyte manipulation):
Local distributor:
Locally manufactured:
Other criteria(s):
Cost information:
Other observations and additional information (if necessary):
Protection against active material

shedding and plate shorting:
Steps taken to minimise electrolyte
stratification:
Electrolyte level indication:
Specific gravity indicator:
Automatic electrolyte recharge:
Catalytic converters:
Battery management system:
If a dedicated device is used, type of

regulator of charge controller:
Reference:
If no dedicated device, type of inverter or
power control unit:
Manufacturer:
Reference:
Battery management strategy:
Voltage control: Yes No

Overcharge protection: Yes No
High voltage disconnect (V):
High voltage floating current (A):
Deep discharge protection: Yes No
If yes, setpoint (V):

Periodic boost charge function: Yes No

Temperature compensation: Yes No
Current compensation: Yes No
State of charge of Ah control: Yes No
If details are known, please fill in details of
the algorithm:
Which monitoring data are available?:
Complete battery voltage profile: Yes No

Hourly voltage data: Yes No
Daily minimum and maximum voltages: Yes No
Input to batteries (Ah or Wh): Yes No
Battery output (Ah or Wh): Yes No
Other data (please specify):

Q7 Appliance performance
Lamps:
Alternative Current? Yes No
Number:
Nominal voltage (V):
Power (W):
Mean daily use (hour/lamp)
Television:
Size (cm diagonal)
Alternating Current? Yes No
Power (W):
Standby power (W):
Mean daily use (h):
Refrigerator:
3
Size (dm )
Power (W):
Yearly consumption @ 25oC (kWh):
Daily consumption @ 25oC (kWh):
Freezer:

Size (dm3)
Power (W):

Yearly consumption @ 25oC (kWh):

o
Daily consumption @ 25 C (kWh):
Video:
Power (W):
Standby power (W):
Mead daily use (h)
Radio:
Power (W):
Standby power (W):
Mean daily use (h):
Washing machine:
Capacity (kg):
Power (W):
Wash consumption @ 30oC (kWh):
Wash consumption @ 60oC (kWh):
Mean weekly use (time):
Dishwasher:
Capacity (meals):


Power (W):
Wash consumption (kWh):
Mean weekly use (times):
Other:
Type of appliance:
Alternating Current ? Yes No
Power (W):
Standby power (W):
Mean weekly use (times):

Q8 User satisfaction and adaptation to technology

Did the users seek the technology
installed from a supplier or were they
“pressured” into it? Please eleborate:
What was their reaction following

commissioning? Completely satisfied
Thankful
Apprehensive
Dissapointed
Has the new system improved their life

style? Please eleborate:
Has the new system enabled them to do

things they could not do previously?
Please eleborate:
Has the new system assisted them to

earn income? Please eleborate:
Do they perceive the new system as

being cost effective? Yes No
What would their advice be to a friend
proposing to invest in RE generation
capacity? Please eleborate:
How well is the system performing? Better than expected

As well as expected
Not as well as expected
A disaster from day one
Do they consider the system to be
reliable? Yes No
To what extent are they responsible for
system operation and maintenance? No involvement
Assist as and when directed
Regular operator responsibilities
Periodic maintenance
Responsible for all operations and
maintenance

Guidelines For Monitoring Stand-Alone Photovoltaic Power Systems

Uploaded by

Copyright:

Available Formats

Guidelines For Monitoring Stand-Alone Photovoltaic Power Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Guidelines For Monitoring Stand-Alone Photovoltaic Power Systems

Uploaded by

Copyright:

Available Formats

Guidelines for Monitoring

IEA PVPS T3-13:2003

Report IEA PVPS T3-13: 2003

<Guidelines for monitoring

Methodology and equipment

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 2

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 3

[2] Guidelines for selecting stand-alone photovoltaic systems. Under

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 4

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 5

The Guidelines supplement existing documentation dealing with performance monitoring of

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 6

1.3 Need for Guidelines

Quality assurance parameters;

2.1 Reporting procedure

2.1.2 Aggregate results

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 7

2.2 Data Requirements

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 8

2.2.1 General Information (Q1)

2.2.2 System Details (Q2)

2.2.3 Quality Assurance Parameters (Q3)

2.2.4 Commercial Evaluation Parameters (Q4)

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 9

2.2.5 Scientific Evaluation of System Performance (Q5)

2.2.6 Battery Performance Parameters (Q6)

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 10

2.2.7 Appliance Performance Parameters (Q7)

2.2.8 User satisfaction and adaptation to technology (Q8)

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 11

3.1 Level of autonomy

3.1.1 Power supply

3.1.2 Number of data points

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 12

Measurement techniques considerations

Data quality considerations

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 13

3.2 Choosing the logger

CPU I/O Interface

Figure 1. Block diagram of a typical data logger

Data storage Again, the distinction between computer driven units

MUX (Multiplexer) Since loggers usually have only one ADC a

Some sophisticated units dedicated to perform

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 14

S/H chips on each channel to ensure that all

Signal Conditioner The real-life signal can be anything (voltages,

Wiring method Voltage reading analog channels can be either

Number of Channels The number of channels is unit specific. Loggers

Channel Expansion Allows the use of other modules to increase channel

IEA PVPS Task 3 - Guidelines for monitoring stand-alone photovoltaic systems 15

e.g.: for a data logger having a CMRR of 90dB,

Type of input Many types of inputs can be available: voltage,

Resolution It is an important feature for the accuracy of data

Sample rate The sample rate is expression in number of

ADC method The analog to digital conversion method can be is

Table 1. Comparison of different analog to

Flash above 1 MHz 4-8 bits None

Where speed is not an issue, a Delta-Sigma or

Specifications pertaining to digital signals