Integration Testing of the Compact High-Resolution Imaging by awm98253


									SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

       Integration & Testing of the Compact High-Resolution Imaging
                           Spectrometer (CHRIS)
                        M. A. Cutter, D. R. Lobb, T. L. Williams and R. E. Renton

                         Sira Electro-Optics Ltd., Chislehurst, Kent BR7 5EH, UK


The CHRIS instrument is a space-based imaging spectrometer that will provide 10nm spectral resolution over the spectral
range from 415nm to 1050nm The nominal spatial sampling interval will be 25m, however, larger sampling intervals are
possible. Band selection, spectral bandwidths and the spatial sampling interval will be programmable. The instrument is
planned to be launched on an agile small satellite of the 100kg class. This satellite will operate in a sun-synchronous, high
inclination orbit at approximately 830km. At this altitude the instrument can provide 19 spectral bands with a spatial
sampling interval of 25m at nadir. The field of view of CHRIS is 18.6km. Attitude control of the platform will allow
access to non-nadir targets, multi-angle observations of selected targets and improved radiometric resolution. This paper
describes the optical design of the instrument, including the telescope, spectrometer detector and in-flight calibration
hardware, as well as critical alignment procedures, with emphasis on spectrometer assembly and stray light control.
Results of performance and calibration measurements are presented.

Keywords: Imaging spectrometer, alignment, test, spatial resolution (MTF), spectral resolution, smile, frown, sun-sensor
calibration, radiometric calibration, stray light

                                                 1. INTRODUCTION

Sira Electro-Optics Ltd is currently completing the testing of a Compact High Resolution Imaging Spectrometer (CHRIS),
to be flown on a small satellite in late 2000. The CHRIS instrument has been designed principally to provide remote
sensing data for land applications, although its high spatial resolution provides potential for applications in coastal zone
monitoring. The first CHRIS instrument will provide the main instrument payload on the European Space Agency (ESA)
small satellite platform PROBA (Project for On-Board Autonomy). PROBA will be launched into a near-circular sun-
synchronous polar orbit, at an altitude of approximately 830km. It is a highly manoeuvrable small satellite, capable of
large, rapid rotations on pitch and roll axes, with fine control over pitch and roll rates.

On the PROBA platform, CHRIS will provide spatial resolution of 25m on ground, over a spectral range from 415nm to
1050nm. The nominal image size at nadir will be 18.6km square. On its first flight, the instrument will be used principally
to provide sets of images of selected target areas, at different pointing angles, forming a minimum of 5 images of each
target in a single overpass. The platform data storage and telemetry system will allow up to 4 complete image sets to be
transmitted to ground per day. The data will be used to analyse directional effects in the radiance of targets, with particular
emphasis on vegetation targets and aerosols.

This paper describes the optical design of the instrument, including techniques employed for reducing stray light and the
method of in-flight radiometric calibration using sunlight. The paper also describes the techniques that have been used for
aligning the instrument, calibrating it and measuring all aspects of its performance. Final testing is not yet fully completed,
but the paper presents the results of some preliminary measurements.

                                              2. CHRIS INSTRUMENT

2.1 General characteristics

The CHRIS instrument is an imaging spectrometer of basically conventional form, with a “telescope” forming an image of
Earth onto the entrance slit of a spectrometer, and an area-array detector at the spectrometer focal plane. The instrument
SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

will operate in a push-broom mode during Earth imaging. The detector is a thinned, back-illuminated, frame-transfer
Charge Coupled Device (CCD). CCD rows are assigned to separate wavelengths, and CCD columns to separate resolved
points in the Earth image.

The PROBA platform will be required to provide pointing in both across-track and along-track directions, for target
acquisition and multi-view imaging. The platform will also be required to provide slow pitch during imaging in order to
increase the integration time of the instrument. This increase in integration time is needed to achieve the target radiometric
performance, at the baseline spatial and spectral sampling interval, and to limit internal electronic bandwidths and data

The spectral waveband covered by the instrument is limited, nominally, to the band 415nm to 1050nm, which can be
achieved using a single CCD area-array detector. However, the design form selected is capable of extension to cover the
whole spectral range from ultra-violet through the short-wave IR, by addition of a short-wave IR detector array in a later
development. Addition of the spectral range up to 1700nm will enhance the capabilities of the instrument for a wide range
of applications, including monitoring of vegetation and soil. The range above 2000nm may be considered for
identification of minerals.

2.2 Optics and structure

The instrument optical design is shown in Figure 1. It includes a catadioptric telescope and a spectrometer. A catadioptric
design is preferred for the telescope. The axially-symmetrical design allows conventional construction methods to be
applied, it is very compact, and it can provide a very broad spectral range without aspheric elements. For an altitude of
approximately 830km, the focal length of the system is 746mm, and the aperture diameter is 120mm (f/6).

                                            Figure 1 Instrument optical design

The spectrometer is a design recently patented by Sira. It uses two Féry prisms (with curved surfaces) integrated into a
modified Offner relay – a system of two concave mirrors and one convex mirror. The spectrometer does not have a
common optical axis, but all surfaces are spherical, and all centres of curvature are in a common plane. These features are
important for minimising cost of optics, and for ease of alignment. Like the telescope, the system is very compact. The
spectral resolution of the spectrometer varies from approximately 1.25 to 11nm across the spectrum with the highest
dispersion at 415nm and the lowest in the near infrared at 1050nm.

The spectrometer design will provide registration to better than 5% of the pixel, in both spectral and spatial directions, with
spatial and spectral resolution limited essentially by the detector pixel size.

All refracting elements in the present design, including the three lens elements in the telescope, and entrance slit block and
the two Féry prisms, are made of fused quartz. The entire system will in principle operate from near ultra-violet through

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

the short-wave IR, though optics in the first development are coated to optimise transmission from 415nm to 1050nm. The
telescope secondary mirror, which is cemented to the first large refracting element, is also fused quartz, but the other
mirrors are made in a common optical glass (Schott BK7).

The telescope structure is based on a simple cylindrical cell, with precisely machined location bores and shoulders for
location of the optical elements. The primary mirror is cemented into a cell at the rear of the telescope barrel, which has
features for location of the central assembly comprising the entrance slit, two small corrector lenses, and a slot-shaped
baffle. The spectrometer optics are mounted on a rigid base plate. The telescope, the spectrometer base plate and the
detector are bolted to a common central bulkhead. These main structures and optics mounts are all in titanium. Selection
of similar thermal expansion coefficients for the structure and mirrors ensures good athermalisation (control over change of
focus with temperature) over a large temperature range. Aluminium is used for the immediate mounting structure of the
detector, for good thermal dissipation in this region, and aluminium is also used to minimise mass for the spectrometer
cover and the external baffle of the telescope.

2.3 Detector

The CCD detector is an area array EEV device (CCD25-20) with 1152 rows and 780 columns, and a 22.5 x 22.5µm pixel
size. The device is thinned and back-illuminated to provide good blue response. It operates in a frame transfer mode, with
image and masked storage zones. The opaque mask is extended along the sides of the image zone to provide transition and
dark reference pixels at the ends of each CCD row, which are used for dark signal and electronic offset calibration. The
mask leaves 748 unobscured imaging pixels in each CCD row. The spectrometer image fills <200 of the CCD rows, but
part of the nominally-unexposed area will be used to calibrate for stray light and CCD smear effects.

At the end of each frame cycle, the signal charges acquired in the imaging zone are rapidly transferred into the storage
zone (this process generating “smear” that must be calibrated). Signals in the storage zone are then transferred, row-by-
row into a horizontal shift register, from which they are clocked out through a single output port. There is considerable
useful flexibility in operation of the CCD. It offers the facility to sum sets of row-signals in the shift register, before read-
out – providing users with a facility to compose spectral bands of optimum widths. Signals can also be binned in pairs at
the output port, relaxing across-track spatial resolution by a factor 2, and integration time can be increased over a wide
range to provide control of spatial resolution along-track (in combination with control over the platform pitch rate).

It will be possible to read out 18 spectral bands during a nominal integration time of 12.7ms, plus one band assigned to
smear/stray light calibration in each frame. This spectral coverage will be associated with optimum spatial resolution and
maximum swathe width. However, it will be possible to read out much larger numbers of spectral bands with relaxations
of spatial resolution and/or swath width. Relaxed spatial resolution (in general associated with increased integration
periods) will also provide enhanced signal-to-noise ratios. The CCD incorporates a dump gate adjacent to the readout shift
register. This provides a facility for fast parallel dumping of charge for regions of the CHRIS spectrum that are not selected
for readout.

2.4 Radiometric and wavelength calibration in flight

The instrument will be calibrated in flight for radiometric response, using data recorded in dark and bright scenes of known
effective radiance. Full-field dark level calibration will be performed using dark Earth scenes. Dark field calibration
coefficients will be corrected for the effects of temperature variations (between recordings of dark-scenes and the required
Earth images) using dark signal data derived from masked pixels in all recorded frames.

A bright field for calibration will be provided by a solar calibration device (SCD). This device comprises a small
reflecting prism, with one lens surface, which will be located at the outer end of the instrument external baffle. When the
platform is over the Antarctic region, on the dark side of the terminator, imaging a dark Earth area at nadir, the SDC will
receive direct sunlight. The platform will be manoeuvred so that the device reflects sunlight into the field of the
instrument, with spread provided by it’s lens power. The SCD will fill the field of the instrument at a nominal radiance
equivalent to albedo 0.25 in direct sunlight. The SCD is not moved; it is fixed in the main instrument aperture area, but
occupies (and samples) only a small fraction of the instrument aperture area. The field of the device for receiving sunlight
is limited to 2° x 4°, and this field will be fully sampled, in pre-flight calibration and in orbit, to check for spatial
irregularities in transmission of the device and instrument optics over a small area.

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

Wavelength calibration will be checked in flight using atmosphere absorption features, including the oxygen absorption
band at 762nm.

2.5 Physical parameters

The CHRIS instrument has an envelope of approximately 200 x 260 x 790mm, a mass of less than 14kg and a power
consumption of less than 10W. An illustration of the instrument showing the telescope, spectrometer and the electronics
box is shown in Figure 2.

                                       Figure 2 CHRIS Mechanical Arrangement

2.6 Radiometric performance

The CHRIS signal chain is sized for a dynamic range just slightly greater than albedo 1. Typical signal-to-noise ratio for
land scenes at albedo 0.2 will be 200:1, at nominal spatial and spectral resolution (25m on ground and 10nm). Better
performance will be achieved at relaxed spatial and spectral resolutions.

                              3   ASSEMBLY AND ALIGNMENT PROCEDURES

3.1 Telescope alignment

All components of the telescope, including the cemented combination of front meniscus lens and secondary mirror, were
tested interferometrically before assembly to ensure that they conformed to specification. Measurements were also made
to check that mounting arrangements for the mirrors did not produce unacceptable distortion. The telescope is an axially-
symmetrical system, using only spherical components. There are no severe tolerances on relative locations of elements, so
that adequate alignment could be achieved mainly by use of machined locations for edges and surfaces. A special
cementing jig was however required to facilitate cementing of the secondary mirror to the large front meniscus lens.

A slot-shaped baffle at the front of the small lens assembly has an important function in preventing stray light from
reaching the entrance slit by direct transmission through the lenses (without reflections at the two telescope mirrors). We
are concerned here with stray light from areas of the aperture outside those edges of the (rectangular) secondary mirror that
are parallel to the entrance slit; stray illumination from these areas would be diffracted by the entrance slit onto the main

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

beam path in the spectrometer. In alignment of the telescope, it was therefore important to check the alignment between:
the entrance slit, and the edges of the slot baffle and the secondary mirror. (The external baffle of the telescope is useful in
thermal control of the instrument, and in excluding potential sources on contamination. It also provides a mount for a solar
calibration device. However, it does not have a critical function in control of stray light. The important stray light baffles
are the small slot baffle in the telescope, and a baffle in the spectrometer.)

3.2 Telescope focus and separate testing

Focus adjustment and evaluation of the performance of the telescope was done on an modulation transfer function (MTF)
test facility (Ref.1) as illustrated in Figure 3. The facility consists of an illuminated slit source (the object generator)
supported on a motorised micrometer stage that is at the focus of an off-axis paraboloid collimator mirror, the light path
being folded by means of a small plane mirror. The telescope is mounted by means of a bracket on a platform which can
be rotated about a vertical axis that passes approximately through the entrance pupil of the telescope and is positioned so
that its entrance aperture is central with respect to the light beam from the collimator. Rotation around this axis changes
the position of the object generator slit in

                               Figure 3 Arrangement for aligning and testing the telescope

the field of view of the telescope. A slit is mounted in the focal plane of the telescope. For initial testing this was a
dummy slit but the flight-model entrance slit, which forms part of the telescope assembly, was used for final alignment and
focusing. Behind the slit is the analyser head, consisting of relay objective of large numerical aperture that collects all the
light passing through the slit and focuses an image of the latter onto a large area silicon detector. The system is controlled
by a computer which scans the object generator slit so that its image moves across the slit in the focal plane of the
telescope and records the output from the detector. The object generator slit and the analyser slit are parallel to each other
for all measurements so that the output recorded by the computer is the line spread function (LSF). The MTF is obtained
by performing a Fourier transform of the LSF, with appropriate correction for slit widths. Two dummy slit arrangements
were used, for measurements on the LSF in orthogonal directions, at centre and edges of the telescope field.

It is important to focus the slit source precisely, to appear at infinity. This is done by auto-collimation off a large plane
mirror that can be positioned in the collimated beam, using a beam splitter and screen attachment at the object generator.
The telescope was focused by moving the primary mirror, using three micrometers temporarily attached to the back of the
telescope, with the mirror held against the micrometers by a temporary spring arrangement. After focusing and alignment
the front lens/secondary mirror and the primary mirror were cemented in position. The complete telescope unit was finally
assembled before the cement set, with nominal tightness on bolts, in order to reduce the chances of any distortion of the
optical surfaces occurring later as a result of imperfect mating of mechanical interfaces. After cementing, the MTF of the
telescope was again measured using both the dummy slits and the final flight slit.

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

3.3 Spectrometer alignment

The radii of curvature of all spectrometer optical surfaces, and the wedge angles of the Féry prisms, were measured
precisely before assembly, and the optical design was re-optimised for true component parameters. This allowed easy
tolerances to be placed on manufacture of the optics, including particularly the unusual prisms.

Fairly accurate alignment and positioning of the components of the spectrometer were then required in order to achieve the
desired performance. Alignment was achieved essentially by locating the centres of curvature of each optical surface in its
correct relative position, without direct reference to any mechanical features of the mirrors or prisms. Tolerances on
location of centres of curvature are typically in the order of ±0.05mm, in all directions, with respect to positions computed
from the re-optimised design data.

The three mirrors and two prisms were initially cemented into support frames. Each frame was then fixed to a separate
mount. Each of the five mounts has a base face locating on the spectrometer bench, providing a precise height datum with
respect to flat surfaces on the bench in a common base plane. Before fixing, the frames were adjusted on the mounts, to
place the centres of curvature (two centres for each prism) at the common nominal height above the base plane.
Subsequent adjustments on centres of curvature positions were then required only in a common plane parallel to the
spectrometer bench. These final adjustments were achieved by sliding the mounts on their base plane pads, using
adjustments built into the bases of the five mounts.

All adjustments on centres of curvature, with respect to the optical bench, were made using a co-ordinate measuring
machine (CMM) which could be fitted with either a mechanical probe or an auto-collimating microscope. A polished
sphere was used conventionally to retro-reflect light from the microscope autocollimator attachment, and as a touch datum
for the mechanical probe, in order to relate the readings from the two attachments. The mechanical probe was then used
critically only to locate datum faces of the spectrometer bench, defining a co-ordinate system for location of optics.

The auto-collimating microscope was used to locate the position of the centre of curvature of each concave surface, as
indicated in figure 4. Typically, the attachment was positioned to provide a focus at the correct nominal position for a
centre of curvature; the element was then located to retro-reflect into the attachment. A separate auto-collimator, with a
long focal length lens, was used to locate the centres of curvature of convex surfaces, as indicated in figure 5. Typically,
the CMM auto-collimating microscope was focused at the nominal centre location; the separate autocollimator was placed
to focus at the same point, by viewing it’s light though the CCM attachment; the optical element was then positioned to
retro-reflect into the separate auto-collimator. For the Féry prisms, it was necessary to use the separate autocollimator and
the CMM attachment together, focused respectively on the convex and concave surface centres of curvature, in order to
achieve the correct positions for both surfaces. In general, the location of optical elements on the bench was performed
with only one element on the bench at any one time. After alignment was achieved for one element, its correct location
was defined by a bracket placed against the element mount and then temporarily fixed to the base plate, before the element
was removed.



                                                                                            centre of

                                                                                                           test surface
              with graticule                                                   microscope

                               Figure 4 Auto-collimating microscope for determining centres of curvature

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

The spectrometer includes a baffle, located between the secondary and tertiary mirrors, at the image of the telescope
secondary mirror. The baffle has two blades, with edges orthogonal to the entrance slit, which were positioned just inside
the image of the (parallel) edges of the telescope secondary mirror. This spectrometer baffle is essential to block stray
light that reaches the entrance slit without reflections at the telescope mirrors, and is not stopped by the slot baffle in the
telescope small lens assembly.

                   source                                                                                    source
                              graticule                                             attachment
                                                                       convex                                graticule
                                                       centre of

        eyepiece                          microscope                                                                  with graticule
        with graticule                    objective
                                                                                 collimating lens

                            Figure 5 Method of locating the centre of curvature of a convex surface

3.4 Spectrometer separate testing

After all components had been aligned, the spectrometer was assembled with a dummy slit. Using both spectral lamps and
a white light source the performance of the spectrometer was checked visually using a travelling microscope to view the
spectrum formed in the plane that would be occupied by the CCD array. Critically, the visual checks were required to

    •       focus and aberration of the image,
    •       the straightness of spectral line images (departures from straightness are called “smile”),
    •       the straightness and parallelism of the line spectra of white point sources at various positions across the field,
            (errors are called “frown”), and
    •       position of the image with respect to the mechanical structure.

When these checks had been completed the components of the spectrometer were cemented in position.

The spectrometer alignment was completed with location of the detector assembly and adjustment of the rotation of the
entrance slit. This was done on the fully assembled instrument (spectrometer plus telescope plus electronics). The plane
of the CCD is first adjusted, in axial position and tilt on two axes, by shimming the detector assembly with respect to the
bulkhead. The optical arrangement for this adjustment is similar to that illustrated in figure 6, except that the object
generator is replaced by a spectral line source and a half plane is positioned immediately before the telescope aperture,
dividing the aperture in half along the slit plane. The output from the CCD array is recorded with the half plane
obstructing first one half of the aperture and then the other. The results are analysed to determine the exact row position of
the centroid of a selected spectral line in the two situations. The focus of the CCD array is adjusted until the row position
is exactly the same in both positions of the half plane. By making measurements at different wavelengths (row positions)
and different field (column) positions the tilt and focus of the CCD array can be correctly set.

The CCD is finally rotated in its own plane such that the line spectrum produced by a point white light source at the centre
of the field of view is imaged parallel to a column of the CCD. The entrance slit is then rotated such that its image in
monochrome light is parallel to a row of the CCD array. The methods for these adjustments are the same as those
described in the next section for measuring of “frown” and “smile” respectively.

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA


4.1 Evaluation strategy

Experimental evaluation of the complete CHRIS instrument is well advanced but has not yet been fully completed. In this
section we list the tests that have been planned for the instrument, describe briefly the test methods that are being used and
finally give the results of the measurements that have so far been completed. As part of the assembly and alignment
procedure, measurements of performance have been made separately on the telescope and the spectrometer sections of the
instrument. However, only measurements on the complete instrument are described here.

4.2 Measurements completed and planned

The following preliminary measurements have so far been made:

    •    Spatial resolution (MTF) in the along-track direction as a function of wavelength and field position and its
         variation with temperature
    •    Spatial resolution (MTF) in the cross- track direction as a function of wavelength and field position and its
         variation with temperature
    •    “Frown” (i.e. change in apparent field position with wavelength) and its variation with temperature
    •    Spectral resolution and its variation with field position
    •    “Smile” (i.e. change in apparent wavelength with field position) and its variation with temperature
    •    Variation of wavelength calibration with temperature
    •    Out of band stray light
    •    Noise and dark current

The instrument is currently undergoing thermal vacuum tests and these will be followed by vibration tests. During the
thermal vacuum tests, measurements are being made to monitor noise, dark current, radiometric response, wavelength
calibration and smile. After completion of these tests the performance parameters listed above will all be re-measured to
check the stability of the instrument. Other measurements that will be made on the instrument are as follows:

    •    Precise mapping of the full field of view on the detector array
    •    Spectral calibration over the full wavelength range
    •    Radiometric calibration in absolute terms
    •    Calibration of the sun sensor
    •    Out of field stray light

4.3 Measurement techniques

Most of the measurements have been made using a facility designed for measuring the MTF and other parameters of lens
systems (Ref.1). The arrangement for measuring spatial resolution (MTF) is illustrated in Figure 6. An illuminated slit (the
object generator) supported on a motorised micrometer stage is placed at the focus of an off-axis paraboloid collimator
mirror, the light path being folded by means of a small plane mirror.

CHRIS is mounted on a platform which can be rotated about a vertical axis that passes approximately through the entrance
pupil of the CHRIS telescope and is positioned so that its entrance aperture is central with respect to the light beam from
the collimator. The platform is provided with adjusting screws that allow CHRIS to be tilted about a horizontal axis also
passing approximately through its entrance pupil. CHRIS is mounted on the platform with the entrance slit of its
spectrometer in a horizontal plane, so that rotation of the platform about the vertical axis effectively alters the position of
the Object Generator slit in the instrument linear field of view.

For measuring MTF in the along-track direction the object generator slit is set parallel to the CHRIS spectrometer slit
with the micrometer stage arranged so that it scans in a vertical direction. A narrow band filter mounted in the object
generator is used to define the wavelength. The output from an appropriate pixel of the CCD array (i.e. a pixel
corresponding to the correct wavelength and field position) is monitored as a function of the position of the micrometer

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

stage. This gives a line spread function (LSF) for the instrument and the MTF is determined by taking the Fourier
transform of LSF, with a correction for the finite width of the object generator slit.

A similar procedure is used for measuring the across-track MTF, except that the object generator slit and the micrometer
stage are turned through 90°. In both cases measurements have so far been made at 3 wavelengths (411nm, 546 nm and
900 nm) and 3 field positions (on-axis and +/- 4.73 mm off-axis).

The data used for calculating MTF in the across-track direction is also used for assessing “frown”. The procedure is to
compare the object generator slit position corresponding to the centroid of the LSF for the different wavelengths and the
same CCD column. This was done for the three wavelengths and three field positions.

                            Figure 6 Arrangement for testing complete CHRIS instrument

In order to measure spectral resolution, the object generator was replaced by a motorised grating monochromator fitted
with a tungsten halogen light source. The monochromator was arranged with its slit vertical. Spectral resolution was
measured by monitoring the output from single CCD rows (appropriate to the wavelength used) and three groups of
columns of the CCD array (appropriate to the field position). Measurements were again made at three wavelengths for
three field positions. A similar arrangement will also be used for a final detailed wavelength calibration of CHRIS in
conjunction with the use of spectral line sources for accurate calibration of the monochromator.

For measuring “smile”, and for an initial wavelength calibration, a diffuser illuminated by a spectral light source
replaced the object generator at the focus of the collimator. An HgCd source and a Cs source were used. The procedure
was to record all pixel rows for a small number of columns corresponding to the field position being illuminated by the
source. Smile (i.e. a tilt or bowing of the spectral lines with respect to CCD rows) was assessed by determining the row
position (or fraction of a row) of the centroid of each spectral line at column positions corresponding to the centre and two
extremes of the field of view.

Out-of-band stray light has been assessed using a relatively large broad band source at the focus of the collimator in
conjunction with colour glass filters having relatively sharp transmission edges. Signal levels from the CCD array were
recorded with and without the glass filters.

All measurements have been made in a temperature controlled laboratory normally maintained at approximately 22°C. For
assessing how some of the performance parameters varied with temperature the laboratory temperature was lowered to
15°C and raised to 25°C to provide a 10 degree temperature change. Temperatures were allowed to stabilise for at least 16
hours before measurements were taken.

Absolute radiometric calibration of CHRIS and calibration of the solar calibration device (SCD) will be performed by
the UK National Physical Laboratory (NPL). Absolute response will be measured using a calibrated blackbody source,
operating at a temperature of approximately 4000K, illuminating a calibrated Lambertian diffuser. For calibration of the
SCD, a comparison will be made between signals measured when the instrument is (a) pointed at the black-body-
illuminated diffuser, and (b) oriented for the SCD to receive direct radiation from the black body.

 SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

 Measurement of out-of-field stray light will also be made at NPL. A collimated beam from laser sources of different
 wavelengths will be used and measurements will be made over a field of +/-45° and a range of azimuths.

                                                 5. RESULTS OF PERFORMANCE TESTS

 5.1 Spatial resolution – MTF

 The results of along-track MTF measurements for three field positions and two wavelengths are plotted in Figures 7 (a), (b)
 and (c). The results of cross-track MTF measurements, also for three field positions and two wavelengths, are plotted in
 Figures. 8 (a) and (b). Note that the MTF in the along-track direction is determined to a significant extent by the finite
 width of the spectrometer slit, whilst in the cross-track direction the MTF of the CCD array plays a significant part.
 In either case a change of 10°C in temperature had no measurable effect on the MTF.

                         MTF AT 411nm V FIELD POSITION                                                  MTF AT 900nm V FIELD POSITION

      1.00                                                                                 1.00

      0.90                                                                                 0.90

      0.80                                                                                 0.80

      0.70                                                                                 0.70

      0.60                                                                                 0.60
                                                                     -4.73mm                                                                         -4.73mm

      0.50                                                           On-axis               0.50                                                      On-axis
                                                                     +4.73mm                                                                         +4.73mm
      0.40                                                                                 0.40

      0.30                                                                                 0.30

      0.20                                                                                 0.20

      0.10                                                                                 0.10

      0.00                                                                                 0.00
             0                     10                         20                                  0                  10                      20
                                  c/mm                                                                            c/mm

                                 (a)                                                                                        (b)

                 Figure 7 MTF in the along-track direction (spatial frequency at the focal plane of the telescope)

                        MTF AT 411nm V FIELD POSITION                                                 MTF AT 900nm V FIELD POSITION

      1.00                                                                           1.00

      0.90                                                                           0.90

      0.80                                                                           0.80

      0.70                                                                           0.70
      0.60                                                                           0.60
                                                                   -4.73 mm                                                                       -4.73 mm


      0.50                                                         on-axis           0.50                                                         on-axis
                                                                   +4.73 mm                                                                       +4.73 mm
      0.40                                                                           0.40
      0.30                                                                           0.30
      0.20                                                                           0.20
      0.10                                                                           0.10
             0                    10                     20
                                                                                            0                   10                      20

                                        (a)                                                                               (b)

                 Figure 8 MTF in the cross-track direction (spatial frequency in the focal plane of the telescope)

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA

5.2 Frown

Frown was measured for three field positions (on axis & +/-4.73 mm off-axis) and three wavelengths (411 nm, 546nm &
900 nm). Preliminary indications are that at any field position the image of slit lies within +/- 0.1 of a column (pixel) width
over the full wavelength range.

5.3 Spectral resolution

Spectral resolution was measured at three wavelengths (approximately 467 nm, 547nm and 900nm) and for three field
positions (on axis & +/-4.73 mm off-axis). The results for the on-axis measurements are plotted as normalised curves in
Figures 9 (a), (b) and (c). Off-axis measurements gave similar results. (The curves have not been corrected for the spectral
resolution of the monochromator used for the measurements nor pixel-to-pixel non-uniformity.) The slits on the latter were
both set at widths equivalent to 0.5nm. Table 1 below lists the full width at half height (FWHH) for these measured curves.

                    1.2                                                                1.2                                                             1.2

                      1                                                                  1                                                               1

                                                                                                                                   NORMALISED SIGNAL
                                                                   NORMALISED SIGNAL

                                                                                       0.8                                                             0.8

                    0.6                                                                0.6                                                             0.6

                    0.4                                                                0.4                                                             0.4

                    0.2                                                                0.2                                                             0.2

                      0                                                                  0                                                               0

                    -0.2                                                               -0.2                                                            -0.2
                       405   410                   415       420                           535   540   545       550   555   560                          870   880   890        900        910   920   930
                                   WAVELENGTH nm                                                       WAVELENGTH nm                                                        WAVELENGTH nm

                                   (a)                                         (b)                                                                                             (c)
                                           Figure 9 Spectral resolution of the CHRIS instrument for 3 wavelengths

                                         WAVELENGTH                                                                     FULL WIDTH AT HALF HEIGHT
                                            411 nm                                                                                1.99 nm
                                            547 nm                                                                                4.33 nm
                                            900 nm                                                                               10.86 nm

                                                    Table 1 Measured spectral resolution for the CHRIS instrument
                                                            (uncorrected for the monochromator line width)

5.4 Smile

Measurements of “smile” were made at several wavelengths in the range 405nm to 900nm, by measuring the row position
of each spectral line at the centre of the field of view and at positions +/- 4.73 mm off-axis. For all wavelengths the
maximum departure from a straight line parallel to a row was less than +/-0.01pixels. When the temperature was changed
by 10°C no measurable change in smile occurred. The thermal vacuum measurements currently in hand will provide
further information about “smile” changes with the temperature and field of view varied over a much larger range.

5.5 Wavelength Calibration

An approximate wavelength calibration can be derived from data obtained using only the lines from a spectral source. This
is compared with the original theoretical data in Figure 10, where allowance has been made for changes in the exact
magnification of the spectrometer section of CHRIS. The spectral line source has also been used to determine the apparent
wavelength shift with temperature. The results for a 10°C change in temperature was a shift of approximately 1.2 rows
which is equivalent to 1.7 nm at 411 nm, 3.8 nm at 546 nm and 11.4 nm at 900 nm. We note of course that CHRIS has
been designed so that it can be accurately calibrated whilst in orbit once a suitable alogorithm has been constructed for
interpolating between spectral lines (a task that will be undertaken in the near future).

SPIE Vol 3753 July 1999, SPIE Conference Denver, USA




                                                    WAVELENGTH nm





                                                                           20          40          60       80         100             120      140    160     180
                                                                                                                 ROW NUMBER

                              Figure 10 Comparison of theoretical wavelength calibration curve (solid line) with
                                            some measurements using an HgCd spectral lamp

5.6 Out-of-band stray light

The results of measurements of out-of-band stray light using an RG610 filter and a BG39 filter are plotted in Figures 11 (a)
and (b) respectively. The plots show the signal level without the filter and with the filter. The latter have been corrected for
“smear” which is the small dc offset that arises as a result of the CCD read-out process. The source brightness was adjusted
for these measurements so that the peak signal from the CCD when the filter was out was close to the maximum usable
signal level without saturation occuring. Stray light appears to be at the level of noise and is certainly less than 0.1% of the
average signal in the illuminated region of the CCD array.
              4000                                                                                                                    4000

              3500                                                                                                                    3500

              3000                                                                                                                    3000

              2500                                                                                                                    2500


              2000                                                                                                                    2000
                                                                                                        No Filter                                                                                No Filter
              1500                                                                                      RG610 Filter                  1500                                                       BG39 Filter

              1000                                                                                                                    1000

              500                                                                                                                     500

                0                                                                                                                       0

              -500                                                                                                                    -500
                  400   500     600     700   800                   900         1000        1100                                          400    500   600     700   800     900   1000   1100
                                      WAVELENGTH nm                                                                                                          WAVELENGTH nm

                                              (a)                                                                                                                    (b)

                                      Figure 11 Assessment of out-of-band stray light using colour glass filters

5.7 Noise and dark signal

The noise floor has been measure as approximately 70 electrons RMS. The dark signal is <50 electrons per pixel at 0°C
rising to approximately 300 electrons per pixel at 20°C.

                                                                                       6. ACKNOWLEDGEMENTS

The authors would like to acknowledge the support of the British National Space Centre, for partial funding of the CHRIS
instrument, and the European Space Agency, for the opportunity to launch the instrument on PROBA and support for part
of the ground calibration phase.

                                                                                                    7. REFERENCES

1. T. L. Williams, “The Optical Transfer Function of Imaging Systems”, published by IOP, Bristol, UK, 1999


To top