Integrating Sphere Theory Apps Tech Guide

T E CHNI CAL G U I DE
Integrating Sphere
Theory and Applications
[email protected] www.labsphere.com
Integrating Sphere Theory and Applications
TABLE OF CONTENTS
1.0 Integrating Sphere Theory 1-4

1.1 Radiation Exchange Within a Spherical Enclosure 1
1.2 The Integrating Sphere Radiance Equation 2
1.2.1 The Sphere Multiplier 3
1.2.2 The Average Reflectance 3
1.3 Spatial Integration 3
1.4 Temporal Response of an Integrating Sphere 4
2.0 Integrating Sphere Design 5-8

2.1 Integrating Sphere Diameter 5
2.2 Integrating Sphere Coating Selection 5
2.3 Baffles and Detector Field-of-View 6
2.4 Flux on the Photodetector 7
2.5 Fiberoptic Coupling 8
3.0 Integrating Sphere Applications 9-16

3.1 Radiometers and Photometers 9
3.1.1 The Sphere Photometer 9
3.1.2 Laser Power Meters 10
3.1.3 Cosine Receptor 10
3.2 Reflectance and Transmittance of Materials 11
3.2.1 Substitution vs. Comparison Spheres 12
3.2.2 Measurement Geometry 0°/d vs. d/0° 13
3.3 Uniform Sources 14
3.3.1 Modifying the Source Radiance 15
3.3.2 Irradiance Uniformity 15-16
Guide To Optical Units 17
© 2017 Labsphere, Inc. All Rights Reserved

1.0 INTEGRATING SPHERE THEORY The fraction of energy leaving dA1 and arriving at dA2 is
known as the exchange factor dFd1-d2 . It is given by:
The integrating sphere is a simple, yet often
misunderstood device for measuring optical EQ. 1
radiation. The function of an integrating sphere
is to spatially integrate radiant flux. Before one can
Where q1 and q2 are measured from the surface normals.
optimize a sphere design for a particular application,
it is important to understand how the integrating
Consider two differential elements, dA1 and dA2
sphere works. How light passes through the sphere inside a diffuse surface sphere.
begins with a discussion of diffuse reflecting surfaces.
From this, the radiance of the inner surface of an
integrating sphere is derived and two related sphere
parameters are discussed, the sphere multiplier and
the average reflectance. Finally, the time constant of
an integrating sphere as presented is relevant to
applications involving fast pulsed or short lived
radiant energy.
1.1 Radiation Exchange Within

a Spherical Enclosure
FIGURE 2
The theory of the integrating sphere originates in the
Since the distance S = 2Rcosq1 = 2Rcosq2 :
theory of radiation exchange within an enclosure of
diffuse surfaces. Although the general theory can be
EQ. 2
rather complex, the sphere is a unique, yet simple
solution to understand.
The result is significant since it is independent of viewing
Consider the radiation exchange between two angle and the distance between the areas. Therefore,
differential elements of diffuse surfaces. the fraction of flux received by dA2 is the same for any
radiating point on the sphere surface.
If the infinitesimal area dA1 instead exchanges

radiation with a finite area A2, then Eq. 2 becomes:
EQ. 3
Since this result is also independent of dA1 :
EQ. 4
Where AS is the surface area of the entire sphere.

Therefore, the fraction of radiant flux received by
A2 is the fractional surface area it consumes within
the sphere.
FIGURE 1
1 © 2017 Labsphere, Inc. All Rights Reserved

1.2 The Integrating Sphere By similar reasoning, the amount of flux incident on the
sphere surface after the second reflection is:
Radiance Equation
Light incident on a diffuse surface creates a virtual light
source by reflection. The light emanating from the surface EQ. 7
is best described by its radiance, the flux density per unit
solid angle. Radiance is an important engineering quantity
The third reflection produces an amount of flux equal to
since it is used to predict the amount of flux that can be
collected by an optical system that might view the
illuminated surface. EQ. 8
Deriving the radiance of an internally illuminated integrating

It follows that after n reflections, the total flux incident over
sphere begins with an expression of the radiance, L, of a
the entire integrating sphere surface is:
diffuse surface for an input flux, Fi .
EQ. 5 EQ. 9
Where r is the reflectance, A the illuminated area and Expanding to an infinite power series, and given that r(1-f )
p the total projected solid angle from the surface. < 1, this reduces to a simpler form:
For an integrating sphere, the radiance equation must

consider both multiple surface reflections and losses EQ. 10
through the port openings needed to admit the input
flux, Fi, as well as view the resulting radiance. Consider
a sphere with input port area Ai and exit port Ae. Equation 10 indicates that the total flux incident on
the sphere surface is higher than the input flux due to
multiple reflections inside the cavity. It follows that the
sphere surface radiance is given by:
EQ. 11
EQ. 12
FIGURE 3
The input flux is perfectly diffused by the initial reflection. This equation is used to predict integrating sphere
The amount of flux incident on the entire sphere surface is: radiance for a given input flux as a function of sphere
diameter, reflectance, and port fraction. Note that the
radiance decreases as sphere diameter increases.
EQ. 6
Where the quantity in parenthesis denotes the fraction of

flux received by the sphere surface that is not consumed
by the port openings. It is more convenient to write this
term as (1-f ) where f is the port fraction and f = (Ai + Ae)/
As. When more than two ports exist, f is calculated from
the sum of all port areas.
© 2017 Labsphere, Inc. All Rights Reserved 2

1.2.1 The Sphere Multiplier 1.2.2 The Average Reflectance
Equation 12 is purposely divided into two parts. The first The sphere multiplier in Eq. 13 is specific to the case
part is approximately equal to Eq. 5, the radiance of a where the incident flux impinges on the sphere wall, the
diffuse surface. The second part of the equation is a wall reflectance is uniform and the reflectance of all port
unitless quantity which can be referred to as the areas is zero. The general expression is:
sphere multiplier.
EQ. 14
EQ. 13
It accounts for the increase in radiance due to multiple where; r0 = the initial reflectance for incident flux
reflections. The following chart illustrates the magnitude rw = the reflectance of the sphere wall
of the sphere multiplier, M, and its strong dependence ri = the reflectance of port opening i
on both the port fraction, f, and the sphere surface fi = the fractional port area of port opening i
reflectance r.
The quantity can also be described as
the average reflectance r for the entire integrating sphere.
Therefore, the sphere multiplier can be rewritten in terms
of both the initial and average reflectance:
EQ. 15
1.3 Spatial Integration

An exact analysis of the distribution of radiance inside
an actual integrating sphere would depend on the
FIGURE 4 distribution of incident flux, the geometrical details of the
actual sphere design, and the reflectance distribution
A simplified intuitive approach to predicting a flux density function for the sphere coating as well as each surface
inside the integrating sphere might be to simply divide the of each device mounted at a port opening or inside the
input flux by the total surface area of the sphere. However, integrating sphere. Design guidelines for optimum spatial
the effect of the sphere multiplier is that the radiance of an performance are based on maximizing both the coating
integrating sphere is at least an order of magnitude greater reflectance and the sphere diameter with respect to the
than this simple intuitive approach. A handy rule of thumb required port openings and system devices.
is that for most real integrating spheres (.94<r <.99 ; .02<f
<.05), the sphere multiplier is in the range of 10 - 30. The effect of the reflectance and port fraction on the
spatial integration can be illustrated by considering the
number of reflections required to achieve the total flux
incident on the sphere surface given by Eq. 10. The total
flux on the sphere wall after only n reflections can be
written as:
EQ. 16

The radiance produced after only n reflections can be where the time constant, t, is calculated as:
compared to the steady state condition.
EQ. 18
and r = the average wall reflectance

c = the velocity of light
Ds = the diameter of the integrating sphere
Time constants of typical integrating spheres range from

a few nanoseconds to a few tens of nanoseconds.
FIGURE 5
Since the integrating sphere is most often used in the

steady state condition, a greater number of reflections
produces the steady state radiance as both r increases
and f decreases. Therefore, integrating sphere designs
should attempt to optimize both parameters for the best
spatial integration of radiant flux.
1.4 Temporal Response of an

Integrating Sphere
Most integrating spheres are used as steady state
devices. The previous analysis of their performance and
application assumes that the light levels within the sphere
have been constant for a long enough time so that all
transient response has disappeared. If rapidly varying
light signals, such as short pulses or those modulated
at high (radio) frequencies, are introduced into an integrat-
ing sphere, the output signal may be noticeably distorted
by the “pulse stretching” caused by the multiple diffuse
reflections. The shape of the output signal is determined
by convolving the input signal with the impulse response
of the integrating sphere.
This impulse response is of the form:
EQ. 17

2.0 INTEGRATING SPHERE DESIGN

The design of an integrating sphere for any application
involves a few basic parameters. This includes selecting
the optimum sphere diameter based on the number and
size of port openings and peripheral devices. Selecting
the proper sphere coating considers spectral range as
well as performance requirements. The use of baffles with
respect to incident radiation and detector field-of-view is
discussed;and radiometric equations are presented for
determining the coupling efficiency of an integrating
sphere to a detection system.
FIGURE 6
2.1 Integrating Sphere Diameter The smallest sphere produces the highest radiance in
Figure 4 indicates that decreasing the port fraction has general. However, since the integrating sphere is usually
a dramatic effect on increasing the sphere multiplier. employed for its ability to spatially integrate input flux,
For port fractions larger than 0.05, one begins to lose a larger sphere diameter and smaller port fraction will
the advantage offered by the high reflectance coatings improve the spatial performance. Notice in Figure 6 that
available for integrating spheres. The first rule of thumb all three sphere designs converge on the same unit flux
for integrating spheres is that no more than 5% of the density as the reflectance approaches 1.0. Therefore,
sphere surface area be consumed by port openings. high reflectance integrating sphere materials such as
Spectralon can optimize spatial performance at only a
Real integrating spheres are designed by initially slight tradeoff in radiance efficiency.
considering the diameter required for the port openings.
Port diameter is driven by both the size of devices as 2.2 Integrating Sphere Coating Selection
well as the geometrical constraints required by a
sphere system. The sphere multiplier as illustrated by Figure 4 is extremely
sensitive to the sphere surface reflectance. Therefore,
Consider the case of a two port integrating sphere; both the selection of sphere coating or material can make
ports are of unit diameter. The relative radiance produced a large difference in the radiance produced for a given
as a function of sphere diameter, Ds, for an equivalent sphere design. Consider the diffuse reflectors offered
input flux is proportional to: by Labsphere known as Spectraflect® and Spectralon®.
Both are useful for UV-VIS-NIR applications in the
250 nm to 2500 nm spectral region. The typical spectral
EQ. 19 reflectance of each is shown in Figure 7 below.
The equation can be plotted as a function of

reflectance for different sphere diameters and
the resulting port fraction for each is shown in Figure 6.
FIGURE 7

Both coatings are highly reflective, over 95% from 350

nm to 1350 nm, therefore, intuitively one might expect no
significant increase in radiance for the same integrating
sphere. However, the relative increase in radiance is
greater than the relative increase in reflectance by a factor
equal to the new sphere multiplier, Mnew.
EQ. 20
The magnitude of this effect is depicted below:

FIGURE 9
Baffles can be considered extensions of the sphere
surface. Their contribution to the sphere area can be
factored into the radiance equation although it is not
usually significant. The fractional contribution of baffles
to the sphere surface area is usually quite small.
The radiance at the incident area, Li, is higher than the

average for the entire sphere surface by an amount equal
to Eq. 5 for the directly irradiated area Ai. The radiance
ratio for the incident to average sphere radiance is
given by:
EQ. 21
FIGURE 8
Although Spectralon offers a 2% to 15% increase in where fi = Ai/As.

reflectance over Spectraflect within the wavelength range,
an identical integrating sphere design would offer 40% to The radiance ratio increases rapidly with decreasing
240% increased radiance. The largest increase occurs in spot diameter. Considering the reciprocity of light rays,
the NIR spectral region above 1400 nm. the same considerations must be applied to the field-
of-view of a photodetector within the integrating sphere.
The fractional area fi can be converted into the detector’s
2.3 Baffles and Detector Field-of-View angular field-of-view.
In using integrating spheres, it is important that the viewed
radiance does not include a portion of the sphere surface
directly irradiated by incident flux. This will introduce a
false response.
Baffles coated with the same material as the integrating

sphere wall block the view of incident flux which has
not undergone at least two reflections from the sphere
surface. The optical system in Figure 9 cannot directly
view the incident flux. However, the baffle is positioned to
prevent first reflections from entering the field-of-view for
the photodetector.
FIGURE 10

In both cases, as either the area of irradiation or the 2.4 Flux on the Photodetector
field-of-view approaches total coverage of the sphere
surface, the radiance ratio approaches unity. As either The radiance of the sphere wall determines the total flux
parameter decreases, the radiance ratio rapidly increases. incident on a photodetector mounted at or near a port of
In applications where the port of an integrating sphere the integrating sphere.
is being used as a uniform radiance source, the result
is increased non-uniformity. When the sphere is used as
a collector to measure radiant flux, the result is increased
measurement error if incident flux directly enters the
detector’s field-of-view.
One method of providing a photodetector with a

hemispherical field-of-view is to attach a diffuser.
One of the best diffuser attachments is an auxiliary
or satellite integrating sphere. The port of this sphere
is baffled from direct view of incident flux.
FIGURE 12
By definition, the total flux incident on the detector active

area, Ad (m2) is:
EQ. 22
where: W = projected solid angle (sr)
A good approximation for W in almost all cases is:

FIGURE 11
Of course, Figure 9 (on the previous page) presents EQ. 23

another solution to the potential problem of small
detector field-of-view. In this case, the real field-of-view
on the sphere wall as defined by the photodetector’s In the case of imaging optics used with the detector, the
imaging system can be considered as a “virtual detector” angle q is subtended from the exit pupil of the system.
with a hemispherical field-of-view. The baffle placement The detector is the field stop of the system.
ensures this effect. The integrating sphere applications
shown in Section 3.0 present other examples of proper
baffle placement.
FIGURE 13

The f-number (f/#) of an optical system is also used Reflectance losses at the air/fiber interface must be
to express its light gathering power. Therefore: considered in determining the total flux accepted by
the fiber. If R is the reflectance at the fiber face, then:
EQ. 24
EQ. 27
The efficiency of the optical system, which is generally For a single strand fiber, Af is the area of the fiber end
a function of the transmittance and reflectance of individ- calculated for the core diameter. If a fiber bundle is used,
ual components, must also be considered. Therefore the this quantity becomes the individual core diameter times
detector incident flux is: the number of fibers in the bundle. The light emanating
from the other end of the fiber is a function of its length
(cm), the material extinction coefficient (cm-1), and the
EQ. 25 exit interface reflection.
where, e0 = optical system efficiency (unitless).
2.5 Fiberoptic Coupling

A similar equation is used to calculate the incident
flux gathered by a fiberoptic cable coupled to an
integrating sphere.
FIGURE 14
The numerical aperture (NA) of an optical fiber is used

to describe its light coupling ability. The projected solid
angle is:
EQ. 26

3.0 INTEGRATING SPHERE APPLICATIONS Photometers are a distinct type of radiometer which use
a quantum detector filtered to emulate the spectral
Integrating spheres collect and spatially integrate radiant response of the standard human observer. This specific
flux. The flux can be measured directly or after it has responsivity is known as the luminous efficiency function.
interacted with a material sample. The sphere as part of The primary unit of photometric flux is the lumen.
a radiometer or photometer can directly measure the flux The detector response function weights and integrates
originating from lamps and lasers or the flux density the spectral radiant flux to produce the lumen scale.
produced from hemispherical illumination. Perhaps the Photometry has the unique distinction of being the only
largest application for integrating spheres is in the measure- system of physical measurement based entirely on
ment of the total reflectance or transmittance from diffuse human perception.
or scattering materials. An alternative application utilizes the
port opening of an internally illuminated integrating sphere 3.1.1 The Sphere Photometer
as a large area source that features uniform radiance. These
sources can be used to calibrate electronic imaging devices
The oldest application for the integrating sphere is the
and systems or simply as uniform back illuminators.
measurement of total geometric luminous flux from electric
lamps. The technique originated at the turn of the 20th
3.1 Radiometers and Photometers century as a simple and fast method of comparing the
lumen output of different lamp types. It is still widely used
An integrating sphere combined with a photodetector of
in the lamp industry for manufacturing quality control.
the appropriate spectral response can be used to directly
The alternative method is a goniophotometer which would
measure the total geometric flux emanating from a light
need to rotate a photodetector in a complete sphere
source or the flux density of an illuminated area. The
around the lamp. Each discrete intensity point (lm/sr) is
geometric distribution of the light to be measured
then integrated over 4p steradians.
determines the appropriate integrating sphere design.
The spectral properties of the light source determines
the appropriate photodetection system.
In general, radiometers measure light in accordance with

the SI unit of radiant flux, the watt. Quantum response
photodetectors are most commonly used in radiometers.
Since their responsivity varies spectrally, it is more appro-
priate to tailor the response for a specific spectral region
through the use of optical filters in nearly all cases except
perhaps when the incident flux is monochromatic.
Thermal detectors respond equally to all wavelengths FIGURE 15

of light. This property also makes them susceptible to
In a sphere photometer, the lamp to be measured is
background thermal radiation. Most often, they need
mounted at the center of the integrating sphere and
to be temperature controlled and the input radiation is
baffled from the viewing port equipped with a diffuser
modulated for synchronous detection.
and photopic response detector. The baffle is usually
positioned at 2/3 of the radius from the sphere center.
Spectroradiometers measure light as a function of
Its size should be as small as possible yet large enough
wavelength. These feature a detector coupled to a
to screen the maximum dimension of the lamp.
spectral separation device such as a diffraction grating
monochromator or a Fourier transform interferometer.
The lumen output from the test lamp is determined by
first calibrating the photodetector signal using a lamp
The spectral dependence of the integrating sphere
standard of known luminous flux. The lamps are
multiplier modifies the relative spectral responsivity of
alternately substituted into the integrating sphere.
the photodetector. The sphere/detector combination
An auxiliary lamp can be permanently mounted inside
must be considered mutually in order to design or calibrate
the sphere to correct for the substitution error caused by
the measurement system for a particular responsivity.
different self absorption from the test and standard lamps.
Auxiliary lamps are usually mounted at the sphere vertex
diametrically opposite the viewing port. Lamp standards
originate from national standardizing laboratories and
are based on goniophotometric measurements.

The photopic response detector in Figure 15 can Integrating sphere power meters are extremely useful
be replaced with a spectroradiometer for direct for divergent and non-symmetrical beams such as those
measurements of spectral radiant flux. The ability to produced by diode lasers. These tend to overfill the active
obtain spectral information from the integrating sphere area of conventional laser power meters. Except in the
is advantageous for several reasons. With spectral case of high power diode laser arrays, the integrating
measurements, the spectral responsivity of the sphere sphere is once again utilized more for its ability to spatially
wall and the relative spectral responsivity of the integrate radiant flux than its ability to attenuate.
photodetector do not influence the luminous efficiency
function. Lumens is not the only quantity obtained for The baffle placement should be based on preventing
a particular lamp. Spectral flux is easily converted to direct view of the “hot spot” produced by the laser beam
yield important color properties such as chromaticity as depicted in Figure 16.
coordinates, correlated color temperature, and the
color rendering indices.
3.1.2 Laser Power Meters
The sphere photometer offers the advantage of total

collection and spatial integration. In the measurement of
highly collimated sources such as lasers, the integrating
sphere offers the advantage of signal attenuation. From
Eq. 12 and Eq. 22, it is evident that the fraction of flux
received by a photodetector mounted at the sphere
surface is approximately the fractional surface area
consumed by its active area times the sphere multiplier.
For a 1mm2 active area on a 100mm diameter sphere, FIGURE 17

the detected flux would be less than 0.1% of the incident
In the case of diode lasers, the field-of-view of the detector
flux. Even further attenuation is possible by recessing the
should not overlap the direct area of illumination. Although
detector from the sphere port as in Figure 12.
the design tendency in this application may be to select a
small sphere diameter to coincide with the device dimen-
sions, a larger diameter sphere more easily conforms to
the geometrical requirements and reduces the
measurement uncertainty.
3.1.3 Cosine Receptor

In the sphere photometer and laser power meter
examples, the integrating sphere is used for measuring
total radiant flux. The geometry of the radiant source in
both applications takes advantage of total flux collection
by the sphere. In the measurement of flux density,
FIGURE 16 photodetectors by themselves do not exhibit uniform
response. Diffusers are usually attached to provide the
The attenuation and insensitivity to misalignment provides needed uniformity. The integrating sphere is one
a laser power meter which employs a photodetector with such diffuser.
a smaller and faster active area. Integrating sphere power
meters have been used to measure industrial CW lasers
at kilowatt levels. The high reflectance of the integrating
sphere coating avoids direct damage from the first strike.
The utilization of faster response detectors supports real
time power monitoring in feedback control of industrial
laser output.

In either design, the angular response does not perfectly

correspond to the cosine function since certain incident
rays will first impose on the baffle before being distributed
to the sphere wall. The true angular response is usually
determined experimentally and then applied as a correc-
tion factor in high accuracy irradiance measurements.
3.2 Reflectance and Transmittance

of Materials
The single largest application for integrating spheres is

the measurement of the reflectance and transmittance
of diffuse or scattering materials. The measurements
are almost always performed spectrally, as a function of
wavelength. The one exception may be the measurement
of luminous reflectance or transmittance using a photopic
FIGURE 18 response detector.
The centrally located baffle prevents direct irradiation of In the ultraviolet, diffuse transmittance is used to
the detector. The entrance port becomes the effective determine the UV resistance of pharmaceutical
measuring aperture of the device. For regular irradiance containers, sun protective clothing, and automotive
measurements, the cosine angular response is required. paints. In the visible spectrum, the color of materials is
The irradiance of a flat surface E, is proportional to the quantified and controlled in industries such as paints,
cosine of the angle of incidence, q. textiles and the graphic arts. In the infrared, the total
hemispherical reflectance determines surface emissivities
EQ. 28 applied to radiant heat transfer analysis of thermal control
coatings and foils used in spacecraft design.
For spectral irradiance measurements in which a mono- A transmittance measurement places a material sample at
chromator is used, a 90° port geometry can be more the entrance port to the sphere (as shown in Figure 20).
accommodating. This design is commonly used for global
irradiance monitors since the integrating sphere provides
good spectral response from the UV to the NIR regions of
the atmospheric solar spectrum. A quartz weather dome
guards against environmental contaminants.
FIGURE 20
In reflectance measurements, the sample is placed at
a port opening opposite the entrance port. The incident
flux is reflected by the sample. The total hemispherical
reflectance, both the diffuse and specular components,
is collected by the integrating sphere.
FIGURE 19

Calibration of the reflectance scale is performed by

comparing the incident flux remaining in the sphere
after reflecting from a standard reference material.
FIGURE 21
The angle of incidence in reflectance measurements
is usually slightly off normal up to 10°. The specular
component can be excluded from the measurement
by using normal (0°) incidence or by fitting another port
in the specular path and using a black absorbing light FIGURE 23
trap to extinguish the specular flux.
The ideal measurement relationship is for the ratio of
Reflectance measurements at larger or variable incident radiance produced inside the sphere to be equal to the
angles are performed by placing the sample at the center ratio of the reflectance for each material.
of the sphere and rotating it about a fixed input beam.
Baffles are placed to prevent the photodetector on the EQ. 29

sphere from directly viewing the irradiated sample in
either measurement. In the reflectance geometry, a baffle
is usually placed between the portion of the sphere wall
where; rr = the reflectance of the reference material
that receives the specular component as well. It is best
to use a photodetector with a hemispherical field-of-view
The sample measurement quantity, rs, is properly
to reduce any sensitivity to the scatter distribution
known as the reflectance factor. The term refers to the
function of the sample.
fact that the incident flux was not directly measured as
the reference.
3.2.1 Substitution vs. Comparison
Spheres However, in the substitution sphere of Figure 23, the
average reflectance of the sphere changes when the
A unique integrating sphere error is attributed to the sample is substituted for the reference material.
designs depicted in Figure 20 and Figure 21 due to the The true measurement equation in a substitution sphere
sample effectively altering the average reflectance of the is, therefore;
sphere wall. Calibration of the transmittance scale is
usually performed by initially measuring the incident flux EQ. 30
before the sample is placed against the entrance port.
where; rs = average wall reflectance with sample

rr = average wall reflectance with reference
The average reflectance with the sample material in place,

rs, cannot be easily determined since it is also dependent
on rs . The magnitude of this error can be plotted for a
typical Spectraflect integrating sphere with 5% port
openings and a 1% sample port opening. The reference
material is Spectralon.
FIGURE 22

3.2.2 Measurement Geometry

0°/d vs. d/0°
The previous examples of reflectance and transmittance

geometries depicted a directional incident flux and
hemispherical collection by the integrating sphere after
interaction with the sample material. The geometry is
correctly defined as directional/hemispherical and
commonly abbreviated as 0°/d referring to near normal
incidence and diffuse collection. The 0° angle should be
replaced by the actual angle of incidence when describing
a particular integrating sphere instrument, for example,
8°/d. Reflectance factor is the quantity directly measured
in 0°/d geometry.
FIGURE 24 A reciprocal optical geometry can be used in both
In order to utilize Eq. 29, the proper sphere design reflectance and transmittance measurements. In the d/0°
is one that keeps the average reflectance constant. geometry, the sample irradiation is hemispherical and the
A comparison sphere mounts both the sample and sample is viewed from the near normal direction.
the reference simultaneously to port openings in the
integrating sphere.
FIGURE 26
The measurement quantity in the reflectance geometry

is properly termed the radiance factor. The radiance
FIGURE 25 of the sample under diffuse irradiation is compared to
the radiance of a reference material. The radiance factor
The comparison sphere features an average sphere wall is equivalent to the reflectance factor for reciprocal
reflectance that is constant, a function of each material geometries, i.e.- the étendue and angle from the
reflectance. In a single beam instrument, the sphere can sample normal is the same for the directional beams,
be rotated to alternately position each material in the a hemispherical detector field-of-view (0°/d) is replaced
incident beam. In double beam reflectometers, a baseline by a hemispherical input flux (d/0°). Instruments which
is initially established with a standard reference material use d/0° often employ an auxiliary integrating
mounted at the sample port opening in order to determine sphere source.
the ratio of incident flux in each beam.
Substitution error also applies to transmittance measure-

ments since the sample surface tangential to the port
opening will contribute to the average reflectance of the
sphere cavity. A comparison sphere is recommended for
transmittance measurements as well.

Tungsten halogen lamps are most commonly used

with integrating sphere sources. These lamps provide a
continuous spectrum, free of emission lines or temporal
instability when operated from a current regulated power
supply. The spectral radiance of the sphere source can be
estimated by combining the sphere radiance equation with
blackbody equations for the spectral radiant flux.
EQ. 31
FIGURE 27 and,
There are two main advantages of the d/0° geometry.

The incident flux is significantly greater since integrating EQ. 32
spheres provide total light collection, thus increasing the
signal-to-noise ratio for the instrument. Polychromatic
irradiation will stimulate photon induced radiance, such where;
as that due to fluorescence, which often needs to be
quantified in many color and appearance measurements. r(l) = spectral reflectance of sphere surface
The main disadvantage of d/0° instruments is sample Fi l = spectral radiant flux
heating which can induce thermochromic effects. Many Fo = rated wattage of the lamp
commercial color measurement instruments will utilize l = wavelength
flashlamps to reduce thermal measurement error. T = temperature of the filament
≈ correlated color temperature
3.3 Uniform Sources c1, c2, s = blackbody constants
In the previous applications, the integrating sphere is The radiance equation is multiplied by the number of
used as a collecting device for the measurement of lamps if more than one lamp is used.
radiant flux, either the absolute amount emitted from
a light source itself or the relative amount of flux
transmitted or reflected by materials.
The open port of an internally illuminated integrating

sphere can itself serve as a large area diffuse light source.
FIGURE 28
Lamps are placed inside the integrating sphere around
the perimeter of the viewing port. The lamps are usually
baffled from the port. The radiance of the sphere is a
function of the wattage rating of the lamp. Multiple lamps
can be used to increase the radiance as well as provide
a step wise method of attenuating the radiance level.

3.3.1 Modifying the Source Radiance 3.3.2 Irradiance Uniformity

If it is required to either modify the spectral radiance or Radiance is the flux density leaving a radiant surface
provide greater adjustability in radiance level, then the as viewed from a distance away from the surface.
lamp must be placed externally to another port opening. A Lambertian surface features a radiance that is
perfectly diffuse, independent of viewing angle.
Irradiance is the flux density falling on a surface and
is measured at the plane of the surface.
Integrating sphere sources are most often used to test

an imaging system. The desired effect is uniform radiance
within the field-of-view of the system under test.
FIGURE 29
The incident radiant flux is modified by the efficiency of the

optics used. The centrally located baffle is recommended.
Direct illumination through a side port would create areas
of increased radiance. The field of view into the integrat-
ing sphere should be confined to the baffle. The sphere
surface surrounding the baffle tends to be more brightly FIGURE 31
illuminated. If an optical filter is used to shape the spectral
radiance, care should be taken to ensure that the focused The source can be used to back illuminate a printed
light from the reflector does not damage the filter. or etched image such as photographic film for image
digitization or resolution targets for MTF testing.
When the field-of-view must be wider than the baffle Radiance uniformity to within 1%-2% is ensured by
will permit, a diffuse input at an entrance port opening using one of the three designs previously illustrated.
adjacent to the viewing port can be provided by an
auxiliary integrating sphere. It is sometimes desired to use the sphere source for
testing a non-imaging device such as a CCD or similar
array detector. In this case the desired effect is uniform
irradiance. The device under test is often placed
coaxial with, but at some distance away from the port
of the integrating sphere source. When used in this way,
the two important quantities to be determined are the
axial irradiance at the center of the object as well as the
irradiance at the off-axis edge.
FIGURE 30
In this case, the distance between spheres should be

minimized or the incident illumination becomes more
directional than diffuse. The system in Figure 29 is more
efficient than the auxiliary sphere design for a single lamp.
Multiple auxiliary sphere inputs can be considered to
increase the radiance of the larger sphere. FIGURE 32

In Figure 32, the axial illuminance, Eo is given by: In the examples illustrated Sfor a source diameter equal
to or larger than the object, the cos4f law predicts the
edge irradiance to within 1% for source to object
EQ. 32 distances at least two times larger than the source
diameter. At this distance, the uniformity is within 10%,
however, the irradiance is less than 5% of the value at
Even for a perfectly Lambertian, perfectly uniform circular
the plane of the port.
source, the uniformity of the irradiance across a plane
object at a finite distance will vary with the off-axis angle f.
It is important to note that Table 1 and Figure 33 display
The uniformity fall off is given in Table 1 where both
calculated theoretical values of uniformity for the ideal
the distance and the dimension of the object are
perfectly Lambertian source. Laboratory measurements
expressed as multiples of the sphere port diameter, x/D
of real integrating sphere sources correlate extremely well
and d/D respectively. Uniformity is defined as the ratio
with these predicted values. Therefore, the data provided
of the irradiance at the edge of the object to the axial
can be used as design guidelines in choosing the correct
irradiance, Ee/Eo.
uniform source for a particular application.
Examination of Table 1 reveals that the uniformity is
100% at the plane of the port. It decreases as the object
is moved away from the port for a short distance and
improves as the distance becomes sufficiently long.
This phenomenon can be illustrated graphically as
shown in Figure 33.
For small values of both q and f (<10°), the irradiance at

the edge is given by the commonly used cos4f law of
illumination where:
EQ. 33
FIGURE 33
TABLE 1

GUIDE TO OPTICAL UNITS
Photometric Quantities and Units

Quantity Symbol Units Abbreviations
Luminous energy Qv Lumen•second = talbot lm•s = talbot
Luminous density Uv Lumen second/m lm•s/m3
• 3
Luminous flux Fv Lumen lm

Luminous efficacy K lumen/watt lm W–1
Luminous exitance Mv lumen/m
2
lm m–2
Luminance (brightness) Lv candela/m cd/m2
2
candela/p ft2 = footlambert cd/p ft2 = fL

candela/p cm = lambert 2
cd/p cm2 = L
Luminous intensity Iv candela cd (lm sr–1)
Illuminance Ev lumen/m2 = lux lm/m2 = lx
lumen/ft = footcandle 2
lm/ft2 = fc
Radiometric Quantities and Units

Quantity Symbol Units Abbreviations
Radiant energy Q joule = watt–second J = W•s
Radiant energy density U joule/m3 J/m3
Radiant flux (power) F, P watts = joule/second W = J/s
Radiant exitance M watts/m2 W/m2
Radiance L watts/m2•steradian W/m2•sr
Radiant intensity I watts/steradian W/sr
Irradiance E watts/m2 W/m2
Spectral Response of the Normal Human Eye with Luminous to Radiometric Conversion
Wavelength CIE Photopic Photopic Wavelength CIE Photopic Photopic

(nm) Luminous Lumen/Watt (nm) Luminous Lumen/Watt
Efficiency Conversion Efficiency Conversion
Coefficient Factor Coefficient Factor
390 0.0001 0.13 570 0.9520 649.0

400 0.0004 0.27 580 0.8700 593.0
410 0.0012 0.82 590 0.7570 516.0
420 0.0040 2.73 600 0.6310 430.0
430 0.0116 7.91 610 0.5030 343.0
440 0.0230 15.7 620 0.3810 260.0
450 0.0380 25.9 630 0.2650 181.0
460 0.0600 40.9 640 0.1750 119.0
470 0.0910 62.1 650 0.1070 73.0
480 0.1390 94.8 660 0.0610 41.4
490 0.2080 142.0 670 0.0320 21.8
500 0.3230 220.0 680 0.0170 11.6
510 0.5030 343.0 690 0.0082 5.59
520 0.7100 484.0 700 0.0041 2.78
530 0.8620 588.0 710 0.0021 1.43
540 0.9540 650.0 720 0.0010 0.716
550 0.9950 679.0 730 0.0005 0.355
555 1.0000 683.0 740 0.0003 0.170
560 0.9950 679.0 750 0.0001 0.820

[email protected] www.labsphere.com
© 2017 Labsphere, Inc. All Rights Reserved
PB-16011-000 Rev.00

Integrating Sphere Theory Apps Tech Guide

Uploaded by

Copyright:

Available Formats

Integrating Sphere Theory Apps Tech Guide

Uploaded by

Document Information

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

Integrating Sphere Theory Apps Tech Guide

Uploaded by

Copyright:

Available Formats

T E CHNI CAL G U I DE

1.0 Integrating Sphere Theory 1-4

2.0 Integrating Sphere Design 5-8

3.0 Integrating Sphere Applications 9-16

Guide To Optical Units 17

© 2017 Labsphere, Inc. All Rights Reserved

1.1 Radiation Exchange Within

If the infinitesimal area dA1 instead exchanges

Since this result is also independent of dA1 :

Where AS is the surface area of the entire sphere.

1 © 2017 Labsphere, Inc. All Rights Reserved

Deriving the radiance of an internally illuminated integrating

For an integrating sphere, the radiance equation must

Where the quantity in parenthesis denotes the fraction of

© 2017 Labsphere, Inc. All Rights Reserved 2

1.2.1 The Sphere Multiplier 1.2.2 The Average Reflectance

1.3 Spatial Integration

3 © 2017 Labsphere, Inc. All Rights Reserved

and r = the average wall reflectance

Time constants of typical integrating spheres range from

Since the integrating sphere is most often used in the

1.4 Temporal Response of an

This impulse response is of the form:

© 2017 Labsphere, Inc. All Rights Reserved 4

2.0 INTEGRATING SPHERE DESIGN

The equation can be plotted as a function of

5 © 2017 Labsphere, Inc. All Rights Reserved

Both coatings are highly reflective, over 95% from 350

The magnitude of this effect is depicted below:

The radiance at the incident area, Li, is higher than the

Although Spectralon offers a 2% to 15% increase in where fi = Ai/As.

Baffles coated with the same material as the integrating

© 2017 Labsphere, Inc. All Rights Reserved 6

One method of providing a photodetector with a

By definition, the total flux incident on the detector active

where: W = projected solid angle (sr)

A good approximation for W in almost all cases is:

Of course, Figure 9 (on the previous page) presents EQ. 23

7 © 2017 Labsphere, Inc. All Rights Reserved

where, e0 = optical system efficiency (unitless).

2.5 Fiberoptic Coupling

The numerical aperture (NA) of an optical fiber is used

© 2017 Labsphere, Inc. All Rights Reserved 8

In general, radiometers measure light in accordance with

Thermal detectors respond equally to all wavelengths FIGURE 15

9 © 2017 Labsphere, Inc. All Rights Reserved

3.1.2 Laser Power Meters

The sphere photometer offers the advantage of total

For a 1mm2 active area on a 100mm diameter sphere, FIGURE 17

3.1.3 Cosine Receptor

© 2017 Labsphere, Inc. All Rights Reserved 10

In either design, the angular response does not perfectly

3.2 Reflectance and Transmittance

The single largest application for integrating spheres is

11 © 2017 Labsphere, Inc. All Rights Reserved

Calibration of the reflectance scale is performed by

Baffles are placed to prevent the photodetector on the EQ. 29

where; rs = average wall reflectance with sample

The average reflectance with the sample material in place,

© 2017 Labsphere, Inc. All Rights Reserved 12

1.1 Radiation Exchange Within

1.4 Temporal Response of an

3.2 Reflectance and Transmittance

3.2.2 Measurement Geometry