Document text
A Service of
NASA-c#-/t,o,0s?
NAS A-CR- 160059
m/ ooo 'too!
NF01604
The NASA STI Program ... in Profile
Since its founding, NASA has been dedicated to ensuring U.S. leadership in aeronautics
and space science. The NASA Scientific and Technical Information (STI) Program plays
an important part in helping NASA maintain its leadership role.
The NASA STI Program provides access to the NASA STI Database, the largest collection
of aeronautical and space science STI in the world. The Program is also NASA’s institutional
mechanism for disseminating the results of its research and development activities.
A number of specialized services help round out the Program’s diverse offerings, including
creating custom thesauri, translating material to or from 34 foreign languages, building
customized databases, organizing and publishing research results.
For more information about the NASA STI Program, you can:
• Phone the NASA Access Help Desk at (301) 621-0390
• Fax your question to NASA Access Help Desk at (301) 621-0134
• Send us your question via the Internet to [email protected]
• Write to:
NASA Access Help Desk
NASA Center for AeroSpace Information
800 Elkridge Landing Road
Linthicum Heights, MD 21090-2934
3 1176 01365 4653
a&s&r'l ^ ?
AVHRR/1 - FM
ADVANCED VERY HIGH RESOLUTION RADIOMETER
FINAL ENGINEERING REPORT
PREPARED BY
ITT AEROSPACE/OPTICAL DIVISION
FORT WAYNE, iNDIANA
46803
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
GODDARD SPACE FLIGHT CENTER
GREENBELT, MARYLAND
20771
(.iASA-CK- 1 60 C5 9) A VH r.R/ 1— FM ADVANCED VEhK N81-12J32
iilGli RESOLUTION RADIOilETEE Final Report
(ITT Aerospa ce/Optica 1 Div.) .ill p
HC A 1 4/il F Au 1 CSCL 14b Unclad
S3/j 5 3D9bo
n n n n n n n n f'
TABLE OF CONTENTS
1.0
1.1
1 . 1. 1
1 . 1 . 2
1.1. 3
1.1.4
1. 1.5
INTRODUCTION
General Instrument Description
The Scanner Module-
Electronics Module
Radiant Cooler
Optical Subsystem —
Baseplate Unit —
2.0
2.1
2 . 1.1
2 . 1.2
2.1.3
2 . 2
2 . 2.1
2 . 2.2
2.3
2.3.1
2.3.2
SYSTEM SENSITIVITY-
Solar Channels 1 and 2
Detector — ■
Detector Responsivity *
Solar Channels System Sensitivity-
Channel 3 Detector and Sensitivity
Mercury-Cadmium-Telluride Detector
Channel 3 Sensitivity
Channel 4
Detector
Channel 4 Sensitivity
3.0
.1
O
• •*-
. 3
. 4
. 5
. b
. 6.1
. 6.2
..6.3
3.7
3 . S
3.3.1
3.8.2
3.9
3.10
3.11
3.11.1
3. 11. 2
3.11. 3
3.11.4
OPTICAL DESIGN-
General Description
Scan Mirror
Telescope Design
Channels 1 and 2 Lens Design
Channel 3 and 4 Lens Design
AVHRR Tolerance Analysis
Summary of Mechanical Tolerances
Surface Quality for Filters, Beamsplitters
Mechanical Adjustment Data (Inches)
Dichroics, Beamsplitters and Cooler Windows
Spectral Definition
Spectral Definition of Solar Channels
Spectral Definition of Thermal Channels
Channel Registration
Polarization Sensitivity
Scattered Sunlight :
Honeycomb Temperature Gradient
Sunlight Reflections from In-Flight Target
Signal Contamination —
Sun Scatter Test Results
4.0
4.1
4.1.1
4.1.2
4 . 2
4.2.1
4.2.2
MECHANICAL DESCRIPTION
Overall Instrument Configuration
Structure
Materials in Structure
Scanner Subassembly
Scan Motor
Bearings
PAGE
1-1
1-2
1-2
1-7
1-8
1- 9
1-10
2 - 1
2-1
2-1
2-4
2-5
2-12
2-12
2-16
2-21
2-21
2-21
3-1
3-1
3-4
3-4
3-9
3-13
3-18
3-18
3-22
3-22
3-22
3-25
3-25
3-35
3-53
3-54
3-57
3-58
3-58
3-61
3-63
4-1
4-1
4-1
4-4
4-4
4-4
4-10
-l-
TABLE OF CONTENTS
(CONTINUED)
i PAGE
4.2.3 Bearing Fits 4-10
4.2.4 Lubrication 4-13
4.2.5 Jitter 4-13
4.2.6 Life Test 4-14
4.2.7 Angular Momentum ' 4-14
4.2.B Venting of the Scanner Housing 4-15
4.3 Radiant Cooler Subassembly 4-15
4.3.1 Support Body 4-15
4.3.2 Detector Location 4-19
4.3.3 Deployable Earth Shield 4-19
4.3.4 Materials and Finishes in Cooler — 4-21
4.4 Optics Subassembly 4-21
4.4.1 Optics Outline 4-21
4.4.2 Materials and Finishes Used in Optics 4-21
4.5 Electronics Package 4-21
4.5.1 Electronics Package LayouL 4-21
4.5.2 Accessibility 4-21
4.5.3 Thermal Considerations-- — 4-25
4.5.4 Radiation Considerations ; 4-25
4.5.5 Materials and Finishes in Electronics 4-25
4.6 Weight Breakdown 4-2 5
4.7 Materials 4-25
5.0 ELECTRICAL SYSTEM 5-1
5.1 Electronic Packaging 5-1
5.2 Electrical Design Considerations 5-4
5.3 Video Scan Timing 5-4
5.4 Power Subsystem 5-6
5.4.1 General 5-6
5.4.2 Turn on Transient 5-6
5.4.3 Electronics Switching Regulator 5-6
5.4.4 Power Converter 5-9
5.4.5 +5V Regulators 5-9
5.4.6 ± 15V Regulators 5-9
5.4.7 Motor Power Supply Switching Regulator 5-10
5.5 Commands and Digital TM 5-11
5.6 Analog TM and Patch Control 5-17
5.6.1 Analog Telemetry r 5-17
5.6.2 Patch Temperature Control- 5-25
5.7 Motor Logics 5-29
5.8 Scan Count and Decode 5-30
5.9 Output Data Control 5-34
5.10 Ramp Calibration Generator 5-42
5.11 Auxiliary Scan Timing 5-46
5.12 Ch 3 Data Amplifier 5-50
5.13 Channel 4 Amplifier 5-55
5.14 Daylight Amplifiers 5-55
5.15 Multiplexer Board 5-59
5.16 Black Body -Mux 5-60
5.17 Motor Power Supply 5-60
5.18 Power Profile- 5-64'
5.19 Interface Connectors 5-64
5.20 Electronics Drawings 5-64
-ii-
TABLE OF CONTENTS
, (CONTINUED)
x _ PAGE
6.0 RADIANT COOLER 6-1
6.1 Field of View 6-3
6.2 Shield 6-5
6.2.1 Cover Temperature 6-5
6.2.2 Shielding and View Factors-- 6-7
6.3 Radiator 6-10
6.4 Patch 6-14
6.5 Solar Exposure 6-16
6.6 Anti-Contamination Provisions 6-19
6.7 Optical Port Loading 6-20
6.7.1 Optical Loading on the Radiator 6-21
6.7.2 Optical Loading on .the Patch 6-21
6.7.3 Absorptivity of the Instrument Patch Opening
(Theoretical Model) 6-27
7.0 CALIBRATION 7-1
7.1 Thermal Channels Calibration 7-1
7.1.1 Calibration Accuracy 7-1
7.1.2 Chamber Calibraticn Targets 7-7
7.1.3 In-Flight Calibration Target 7-19
8.0 THERMAL DESIGN 8-1
9.0 TEST AND CALIBRATION DATA 9_1
10.0 LIST OF DESIGN INFORMATION REPORTS 10-1
-iii-
1.0
INTRODUCTION
The Advanced Very High Resolution Radiometer (AVHRR)
was developed under Contract NAS5-21900. This program
covered the design, construction, and test oL" a Breadboard
Model, Engineering Model, Protoflight Model, Mechanical/
Structural Model, and a Life Test Model. Special bench test
and calibration equipment was also developed for use on the
p rogram.
The Flight Model program objectives under Contract
NAS5-22497 were to fabricate, assemble and test four of the
Advanced Very High Resolution Radiometers along with a bench
cooler and collimator. In addition, a group of parts known
as "common parts" were procured and delivered.
Initially, the instrument was to operate from a 906
n.m. orbit and be thermally isolated from the spacecraft.
The Breadboard Model and the Mechanical/Structural Model were
designed and built to these requirements. During the
Engineering Model assembly phase, the spacecraft altitude was
changed to 450 n.m., IFOVs and spectral characteristics were
modified, and spacecraft interfaces were changed. In addi-
tion, the final spacecraft design provided a temperature-
controlled Instrument Mounting Platform (IMP) to carry the
AVHRR and other instruments. The design of the AVHRR was
modified to these new requirements and the mod i f icat ions were
incorporated in the Engineering Model. The Protoflight Model
and the Flight Models conform to this design.
1-1
1.1
General Instrument Description
The AVHRR is a four channel scanning radiometer pro-
viding two channels in the visible-near IR region and two IR
channels. The instrument utilizes an 8 inch diameter optical
system. Cross-track scanning is accomplished by a con-
tinously rotating mirror direct-driven by a hysteresis
synchronous motor. The two IR detectors are cooled to 105K
by a two-stage passive radiant cooler. The data from the
four channels is simultaneously sampled at a 40 kHz rate and
converted to 10-bit binary from within the instrument.
A summary of the AVHRR characteristics is given in
Table 1-1. Figure 1-1 is a photograph of the Engineering
Model instrument and Figure 1-2 shows the outline con-
figuration of the instrument.
The AVHRR is comprised of five modules which are
assembled together into a single unit instrument. These
modules are:
Scanner Module
Electronics Module
Radiant Cooler Module
Optical Subsystem
Baseplate Unit
These modules are shown in the exploded view of Figure 1-3.
1.1.1 The Scanner Module
This module includes the scan motor, the mirror and
the scan motor housing. The scan motor design is based on
the motor developed for the. SCMR,..an 80 pole hysteresis
1-2
Table 1-1 Summary of Characteristics
Ch 1
Ch 2
Ch 3
Ch 4
Spectral Range (uM)
.58-. 68
.725-1.0
10.5-11.
5 3.55-3.93
Detector
Silicon
Silicon
HgCdTe
JnSb
Resolution (N.M.)
.59
.59
.59
.59
IFOV (MR)
1.3 sq .
1.3 sq .
1.3 sq.
1.3 sq .
S/N @ .5% Albedo
>3:1
>3:1
-
-
NETD @ 300K
-
-
. 12K
. 12K
MTF (1 IFOV/Single
Bar)
.30
. 30
. 30
.30
Optics - 8 inch
diameter
afocal cassegrainian
telescope
Scanner - 360 rpm hysteresis synchronous motor with beryllium
scan mirror.
Cooler - Two-stage radiant cooler, IR detectors controlled
at 105K
Data Output - 10 bit binary, simultaneous sampling at 40 KHz rate.
Commands - 23
Telemetry - 14 Digital, 20 Analog
1-3
Figure
synchronous motor. The motor has two power modes of opera-
tion. High power ( - 4-5 watts) will be utilized for driving
the scan mirror in air and low power ( - 3.8w) will be used
for nominal in-orbit operation. The scanner housing is an
integral part of the motor and is mad-e of beryllium. The
scan mirror is also made of beryllium and is -11.6 inches
across the major axis and 8.25 inches across the minor axis.
The scan motor rotates the mirror at the. 360 RPM to produce a
contiguous scan of the earth scene. The line-to-line jitter
is less than 17 microseconds.
1.1.2 Electronics Module
The electronics module is in two sections both of
which bolt on to the instruments inboard side panel. The
curved box (Reference Figure 1-3) is the motor power supply.
Twenty-five electronic modules are used to make up the elec-
trical system of the AVHRR . Nineteen of these are located in
the electronics box. The solar channel preamplifiers for the
solar channels and IR channel 3 are located in the area of
the secondary optics. The IR channel 4 preamplifier is loca-
ted on the rear of the radiant cooler housing.
Except for Channels 1, 2 and 4 preamplifiers, all of
the modules are accessible without the removal of the instru-
ment from the spacecraft.
The following is a list of the electronics modules:
1. Power Converter and Switching Regulator
2. Logics Regulators
3. + 15V Regulators
l
4. Command Relay #1
5. Command Relay #2
6. Command Relay #3
7. Patch Temperature Control and T/M
8. T/M Board #2
9. Motor Logics
10. Scan Count and Decode
11. Interface Logics #1
12. Interface Logics #2
13. Ramp Calibration Generator
14. Auxiliary Scan Logics
15. IR Post Amplifier
16. Daylight Post Amplifier
17. Multiplexer
18. Black Body MUX logics
19. A to D Converter
20. IR Preamplifier
21. Channel 4 Preamplifier
22. Daylight Preamplifier (Ch. 2)
23. Daylight Preamplifier ( Ch . 11.
24. Motor Power Supply
25. Switching Regulator
1.1.3 Radiant Cooler
The radiant cooler module is made up to four basic
assemblies. These are (1) the cooler nousing, (2) the firs
stage radiator, (3; the patch or second stage radiator, and
(4) the cooler cover. The first stage radiator is configured
in such a manner as to shade most of its 55.2 inch 2 area from
the earth by the cooler cover when the cover is deployed. A
"single shot" solenoid actuated, spring driven deployment
system is used to deploy the cover. Mounted on the patch are
the two infrared detectors. The patch has a 22.4
in 2 radiating area. The cooler housing surrounds the coder
on all sides except for the radiation area. The housing is
vacuum sealed so that when the bench cooler is clamped to the
front of the housing a vacuum can be pulled on the entire
cooler and the system permitted to cool as it would in space,
i.e. radiation to a cold target located in the bench cooler
(except that the cold target is at liquid nitrogen
temperature) .
Multilayer insulation thermally separates the first
stage radiator from the housing and the first stage optical
window is thermally isolated and heated several degrees
warmer that the 171K radiator temperature. The patch is
thermally isolated from the first stage by low emissivity
surfaces (gold to gold) and runs at 95K with no control
power.' During nominal operation the patch temperature will
be controlled at 105K.
1.1.4 Op tical Subsystem
The optical subsystem was designed by Ferson Optics,
a division of 3ausch and Lomb, to ITT Specification. (Ferson
1
fabricated the BBM ETM and PTM optics; however, the Flight
Model Optics are being fabricated by Perkin-Elmer , Costa
Mesa, California.) The subsystem consists of an afocal 8.0
inch aperture telescope (two coaxial confocal paraboloidal
mirrors) followed by secondary optics which split the radiant
input into four discrete spectral bands and focus them onto
their respective field stops. The spectral bands are:
Channel 1: 0.58 to 0.68 microns
Channel 2: 0.72 to 1.05 microns
Channel 3: 10.5 to 11.5 microns
Channel 4: 3.55 to 3.92 microns
The instantaneous field of view is i.3 milliradians in all
channels and is defined by an aperture plate in Channels 1
and 2 and by the detector active areas in Channels 3 and 4.
In addition the: optical subsystem has been designed to meet
the total system MTF requirements with the detectors
registered off axis by as much as 1.5 milliradians in
Channels 1 and 2 and 1 milliradian in Channels 3 and 4.
Polarization effects have been minimized ( <7% in
Channels 1 and 2) by orienting the polarization sensitive
elements in a predetermined way, thus having elements compen-
sate for other elements.
1.1.5 Baseplate Unit
The baseplate unit is tne common structure in which
all other modules are secured. Dowel pine are used to
establish and maintain alignment of the scanner and optics
modules. Alignment of the cooler to the optics is
established by shims.
1-10
2.0
SYSTEM SENSITIVITY
2 . 1 Solar Channels 1 and 2
2.1.1 Detector
Both solar channels use the same detectors with the
same operating characteristics as before the modifications.
The detectors used are Infrared Industries, Inc. silicon
detectors which are operated at -15 volts bias. The device
has an active area of 0.100 inch square and is packaged in a
TO- 5 can using a metallic hermetic seal. Some of the most
pertinent characteristics of the device are given in Table
: - 1 .
The detector will be used as a current source for an
op-amp preamplifier. In the current- to-voltage transducer
operating mode, the combination gives excellent sensitivity
and frequency response using a 4M ohm feedback resistor as
the effective detector load.
The detectors are used as energy collection devices
behind the 0.0238 inch square apertures which are the defin-
ing field stops for Channels 1 and 2. Optical analysis
showed that using a 0.100 inch square detector active- area at
an effective optical distance of 0.146 inch behind the. field
stop, resulted in a well over 99% of the rays, which passed
through the field stop, being collected by the detectors of
both channels.
2-1
Table 2-1 Solar Channel Detector Characteristics
Type Passivated, Planar Diffused
Silicon Pin Photodiode
Manufacturer
Infrared Industries,
Inc .
Active Area Sic
e
0.100 Inch Square
B ias
-15 Volts
Spectral Peak
900 + 50 nmeters
Responsivity -
Peak
0.62 Amp/Watt
Responsivity -
Ch 1 Avg.
0.3 7 Amp/Watt
Responsivity -
Ch 2 Avg.
0.54 Amp/Watt
Leakage Current
at -15 V
17 Namp Maximum
Capacitance at
-15 V
15 pf Maximum
Table 2-2 Responsivity Versus Temperature for Detector
Wavelength (A°)
Responsivity (Amps/Watt)
-10°C
+ 22°C
+ 50°C
4415
0.14
0.14
0.13
5013
G .28
0.27
0.27
5560
0.40
0.40
0.40
6000
0.41
0.43
0.42
6505
0.50
0.50
0.51
7010
0.59
0.57
0.59
7492
0.63
0.6 3
0.65
8011
0.72
0.74
0.76
8500
0.76
0.79
0.82
9000
0.73
0.78
0.8 i
2-2
Table 2-2
Responsivity Versus Temperature for Detector
( Cont' d )
Wavelength
(A°)
Responsivity (Amps/Watt)
-10°C
+ 22°C
+ 50°C
9500
0.70
0.79
0.87
10148
0.43
0.58
0.74
10612
0.14
0.24
0.39
11014
0.06
0.10
0.18
11512
0.01
0.02
0.05
2-3
2.1.2 Detector Responsivity Versus Temperature
1
The question was raised early in the AVHRR program as
to the effect of temperature upon the reqponsivity of the
silicon detectors. Infrared Industries .ran spectral response
versus temperature data on one of the. delivered units. The
measured data is of historical interest only since the latest
predictions indicate a maximum detector operating temperature
of about 24°C. The data is shown in Table 2-2.
The AVHRR temperature is more controlled by the
SPACECRAFT TCE and the maximum operating temperature is
within a few degrees of the nominal detector test temperature
of +24°C. The maximum overall temperature change under
various operating conditions is about 10°C. The detector
responsivity changes are insignificant over so small an
excursion and so the AVHRR output will not sensibly change
due to detector temperature variations.
2-4
2.1.3 Solar Channels System Sensitivity
For channels 1 & 2, the signal to noise ratio, S/N, is
given by
where ♦
S/N =
AX
AX
a NEP
= solar spectral flux incident on the detector
a = degradation factor due to electronic noise
pickup, 1 /f noise, etc.
NEP = effective detector Noise Equivalent Power
is given by
1
AX
2 1
I,, t p A (3 = — i-
AX *s o 4
2 2
i A , x d z e
AX so
where I
T
P s
D
c
e
AX
= solar spectral irradiance in the spectral band
incident on the Earth's atmosphere
= transmission of optical system
= scene spectral albedo
= diameter of the collection aperture
= instantaneous field of view
The NEP is given by
NEP
where j.
Vi
2 . 2.2
s + "d + "l
R
= shot ncise current due to signal flux on the detector
i. = photodiode leakage current noise.
i. = load resistor Johnson noise current
R = detector responsivity in ampere/watt
Using Thekaekara's Tables we find that the total solar
irradiance incident on the atmosphere in channels 1 and 2 (weighted
by the relative response in each band) is:
I)
I
1
AX "
4.24 x
IQ ' 2
oi/cm'
2
-2
AX =
3.00 x
10 z
w/cm'
2-5
This is the solar power in the 0.50 to 0.91 micron region (channel
1) and the 0.71 to 1.10 micron region of channel 2.
2 . 1 . 3 . 1 System Transmission
The elements affecting the system transmission can be
divided into two broad categories, those that are spectrally
variant and those that are not. The elements that are variant
are the scan, telescope, and folding mirrors, and the gold dichroic
beamsplitter. These elements are analysed in the section defining
the spectral response. Also analyzed are the bandpass filters
and dete-ctors; however, for S/N purposes, it is assumed that the
bandpass filters are invariant across the band and that the detector
has an average responsivity in the spectral band of interest.
Table 2-3 gives the spectral efficiencies of the mirrors
and gold dichroic used in these calculations. These are measured
PTM values and represent expected flight model values. Combining
the three mirror reflections and gold dichroic transmission for
Channel 1 gives an average transmittance of 0.60 for this channel.
Doing likewise for channel 2 (with one more mirror reflectance)
gives 0.43. These are the transmissions through the spectrally
variant elements only.
The elements which can be considered spectrally invariant
are the inconel beamsplitter separating channel 1 and 2, the relay
lenses in each channel, and each bandpass filter (the filter has a
relatively flat response across each band) . In addition the
obscuration caused by the secondary mirror and its support must be
considered. The values used in calculating the transmissions are
also shown in Table 2-3. There are three relay lenses in each
2-6
Table 2-3 Optical Efficiency of AVHRR Elements
(PTM Measured)
SPECTRALLY INVARIANT ELEMENTS
Reflectivity of Inconel beamsplitter = .225
Transmittance of Inconel beamsplitter = . 375
Transmittance of Chan. 1 filters = 0.85
Transmittance of Chan. 2 Filter = 0.90
Telescope Obscuration = 0.94
Lens Transmittance = 0.95
SPECTRALLY VARIANT ELEMENTS
SPECTRAL
POINT
MICRON
SCAN
MIRROR
REFLECTIVITY
TELESCOPE
MIRROR
REFLECTIVITY
CH 2
FOLDING
MIRROR
REFLECTIVITY
GOLD
DICHROIC
TRANSMISSION
0.50
.92
.88
-
.73
0.55
.93
.92
-
.772
0.60
.925
.93
-
. 80
0.65
.92
.93
-
.81
0.70
.90
.93
.84
.81
0.75
.90
.93
.81
.79
0.80
. 88
.92
.79
.765
0.85
. 875
.915
.79
.73
0.90
. 88
.93
. 80
.69
0.95
.89
.935
.815
.642
1.00
.89
.93
. 825
. 59
1.05
.89
.93
. 835
.56
1.10
. 885
.925
. 845
. 54
2-7
channel so that the combined transmission for the invariant
I elements is 0.154 for channel 1 and 0.272 for channel 2. Combining
these with the values for the variant elements gives:
T x = 0.092
*2. = 0.117
for the expected total system transmission for each channel.
Allowing a degradation factor for dirt, dust, etc. on each element,
will result in, perhaps, a more realistic overall sytem trans-
mission. It is reasonable that lens surfaces sealed from ambient
will not markedly degrade. For sensitivity calculations assume
that each channel will suffer a 50% degradation overall thus
x 1 = .046
x 2 = .058
2-8
2.1. 3.2 Detector NEP
The Solar Channel detector has a peak responsivity (ampere/watt)
of at least 0.62. The average responsivity across the channel 1 spec-
tral band is 0.37 ampere/watt while that in channel 2 is 0.54 ampere/
watt.
Several factors contribute to the noise. The shot noise due to
the detector dark current, the shot noise due to the detector scene
generated current, and the Johnson noise in the preamplifier feedback
resistor are the major noise sources. 1/f noise is negligible across
the 14.5 KHz bandpass of the electronic filter. Contribution of pre-
amplifier transistor noise is also very small compared to the above
sources and so can be ignored.
Both channels use a 4 megohm feedback resistance in the pre-
amplifier so that the Johnson noise at a 300K temperature over a
14.5 KHz bandwidth is 7.75 x 10 - ' 1 ' 2 ampere. The detector dark current
is 17 nanoamperes maximum in each channel giving a dark current noise
of:
1 2 1/2 - 1 ?
(lj) = (i*) = (2 e I d Af) ' = 8.88 x 10 ampere
The signal shot noise under minimum signal condition is based
on tiie current flowing in the detector under that illumination. The
minimum signal flux is calculated later to be 1.68 x 10~^ watt in
channel 1 and 1.43 x 10~9 watt in channel 2. The minimum DC current
out of the detectors then is 6.22 x 10“^® amp and 7.72 x lO - -'-® amp
for channels 1
and 2 .
The shot
noi
se then is
. 1
x s
1.70
.,,-12
x 10
amp
rms
i 2 =
s ■
1.89
, „-12
x 10
amp
rms
The total noise current in channel 1 is:
1/2
i T1 = ( (7.75) 2 + ( 8 . 88 ) 2 + (1.70) 2 ) x 10
i Tl = 1.19 x 10 11 amp
similarly for channel 2
i T 2 = 1.19 x 10 ^ amp
2-9
The noise is essentially the same in both channels due
to the predominance of i^ and i^.
A degradation factor a is included in the calculation. This
accounts for stray noises as well as degradations in operation of
the system. For this analysis a degradation factor of 1.6 based
on the measured BBM and ETM values was used.
The detector NEP then is
NEP^ = 1.19 x 10 ^ = 3.22 x 10 ^ watt
0.37
NEP 2 2 1.19 x IQ" 11 = 2.20 x 10 -11 watt
0.54
2 . 1 . 3 . 3 Signal to Noise Ratio Calculation
Using the equation previously given and assuming a minimum
scene as described in the AVHRR specification (p = 0.5%), a col-
lection aperture of 8.0 inches, and an IFOV of 1.31 milliradians ,
1.73 x 10 9 watt
1.54 x 10 9 watt
aNEP
1.73 x 10~ 9
1.6 x 3.22 x 10" 11 = 33:1
1.54 x 10~ 9
1.6 x 2.20 x 10 -11 = 44:1
we nave
and so
4 >
AA
2
AA
S/N
S/N.
S/N-
2-10
Table 2-4 Solar Channel Sensitivity
System Transmission (Degraded)
Detector Responsivitv
Detector NEP (25°C)
S/N Ratio at Minimum
Albedo (25°C)
Specified Minimum Signal
to Noise Ratio
CH 1
0.046
0.37 A/W
3.22 x 10 -11 W
33:1'
CH 2
0.058
0.54 A/W
2.20 x 10
44:1
3:1
3:1
2 . 1 . 3 . 4 Sensitivity vs Temperature
The foregoing calculation is based on a detector dark
current of 17 nanoamps. This is true at 25°C. At higher
temperatures, the detector dark current (and the load Johnson
noise to some extent) increases, thus increasing the noise and
increasing the detector NEP. Based on data from the manufacturer
and the results of the thermal math model which indicates very
little variation in detector temperature, the signal to noise
ratio for the solar channels will not perceptibly change in
operation.
2 . 2 Channel 3 Detector and Sensitivity
2.2.1 Mercurv-Cadmium-Telluride Detector
The mercury-cadmium-telluride detector is optimized for
best sensitivity between 10.5 and 11.5 pm wavelengths when cooled
to 105 Kelvin by the radiant cooler. The important characteristics
of the detector are summarized in Table 2.2-1.
The Hg Cd Te element is mounted in a small metal enclosure
shown in ITT-A/CD Figure 2.2-1. The aplanat lens is bonded directly
to the metal enclosure and also serves as the window through which
the optical beam passes. The alignment and spacing of the aplanat
lens with respect to the sensing element is accurately maintained
with this arrangement. The detector is tested by the manufacturer
both before and after attachment of the lens to assure a qualified
unit. The internal volume is filled with an inert gas and then
sealed by the manufacturer. The detectors are inspected both
before and after lens attachment by an ITT-A/OD Quality Control
representative who also witnesses the important acceptance tests
at the manufacturer's plant.
2-12
MERCURY -CADMIUM-TELLURIDE DETECTOR
ITEM
CHANNEL 3
Spectral Band
10.5-11.5 um
Width of Sq. Sensitive Area
0.0068 ± 0.0004"
Operating & Spec. Temperature
105 Kelvin
Field of View, Min.
100°
Minimum Resistance
10 Ohms
Max. Bias Power
1.2 M.W.
Avg. D* In Spectral Band At
105K, 2 KHz , 100° FOV,
1.2 M.W. , 1 HZ
» n ln 10 CMHZ
2 *° X 10 Watt
Avg. Spectral Responsivity , Min.
5> 500-~i|I_
Watt
Max. Change in Responsivity
Per Kelvin at 105 Kelvin
5%
Long Wavelength Response
<1% at 18 um
Time Constant
<1 us
Spatial Responsivity Uniformity
<2 to 1
I/f Knee Frequency
<1 KHz
Table 2.2-1 HgCdTe Detector Characteristics
2-13
NOTE'-
(T> THS PLFNC Of THE DETECTOR PACE IS TO COINCIDE WITH
SURFACE -B- WITHIN .OOI. THE CENTER Of THE ACTIVE AREA
OF THE DETECTOR SHALL BE LOCATED ON THE CENTER OF DlA -A-
WITHIN -OOI OIA. THE CENTER Or THE LENS SHALL Be CENTEREO ON THE
ACTIVE AREA OF THE DETECTOR WITHIN .OOI.
(INTOLERANCE FOR LOCATION OF .04VJ DlA. HOLES I l.OOO*
“■ TOLERANCE FOR LOCATION OF THKtAOEO HOLiSl 1.00*0
gN THE center line OF THE 04 VI OIA. HOLES SHALL
COINCIDE WITH SURFACE -B- WITHIN .002 ANO SHALL
BE PARALLEL To SURFACE B- WITHIN .OOI.
gN THIS DIMENSION SHALL NOT EXCEED the MEASURED LENS
HEIGHT (.IIG NCM.) PLUS THE MEASURED DIMENSION FROM
SURFACE -B- TO THE INDICATED SURFACE (.IBO NOM.^ .
[|N THESE DIMENSIONS ARE TO BE MEASURED
FROM THE CENTER LINE OF DlA. -A- .
[4N APPLY PI US (V) VOLTAGE MARKING AT PROPER PIN TO IMO'.CATE
OPTIMUM BIAS DIRECTION.
7 LENS TOSS SUPPLIED BV ITT A/oO.
gN ATTACH a" MIN. LENGTH OF *»l AWE COPPER WIRE, POLYTHCRMALEZE
COATED, TO EACH ELECTRICAL LEAD PER NASA DOCUMENT
NHB VJ0O.4 (SA"); MIN. OF TWO TURNS OF WIRE AROUND PIN.
*. ONE SIDE OF SGUARE SENSITIVE AREA OF DETECTOR
TO BE PARALLEL TO MOUNTING BASE WITHIN lot
RIVHI0N1 LIAQo-nt
LTM DMCmmOW P*T« Al
A AOOED TUC. LAST SEWTEMCS TO NOTE * I ; S-1V14 <B
n ‘THfe C6MT£R0f THE ItNi.lTC.*. fi
__ fcCM <b|4QQ- HA , s
ITEM 2 fT.N0.WA4 600fl»0fr-l *»*U*14 «
gCN C.IAOO-181
r IN MOIE 4 COINCIDENCE TOL. VI AS .OOI 4 »*tt-*»* f
aodeb paralielis* tol. ; IN NOTE 4 m
j.S42?.g& *a» .*41 :
AOOCO ITCH 4 TO UL
ttH_u«oo-iaw
fa-io-is 1
fiayf
st.tr
*»*U*14
lll-IE
iiLL
l VraJkI
ts
ecu &1400'
h* ,*4O4t.&004 tCNt DlA. RIF.
- SEALANT FOR LENS MUST HOT
EXTEND AfeOVC THIS PLANE
fcONO LEWS TO HOUftiNR * TQRR 4SAL EPoXT BC4IN
AND HARDENER. (VARt AN ASSO Cl AT g*. VACUUM 0»V.)
'k" '? T
/ 1 L_o} S max. ELECTRICAL LEADS ,2^
j ]»-'!£>
-THIS SURFACE TOE.E NON- REFLECTIVE
.ZZOt.OOi, DlA. - 7 -
.10 MAX -
EVACUATION TUBE
2 .OOI
g>.24
3 Pit 2_1
T" Vj IF 5
f n
.IBO — J I U\ S
1.00 1 \ N
.no— 1 1U- M
IT;
matl: kovar
FIG 2.2-1
0-80 UNF- 2 B
X.IOMIN. DEEP 2 HOLES
^ pOOOSj
X B . 00 0 ?|
Kef
LL
1?
1 6008507-2
WITR1CSS PLATe • APLAUAT
AR
*
'to** »IH*
‘ aohesne
REF.
£
|A|fc0087^^
SPECIFICATION, IP. DETECTOR
\
L*J_
UtADM tUTtH
detector
1
Z
i?
; SOOW32-I
lens
\
l
T
THIS DW6 >1
HOUSING
800 6 T /1 Ci 4 oo
Ml XT AUCHM.Y UUOOH
APPLICATION
MATERIAL
AS NOTED
-.0434/0448 DlA x .0) MIN DR
2 HOLES
SLIStL «£ .1 ** L*l LSi J rriMi H | part on
QUANTITY MW PROP* 1 *° 1 ■ I lORMTIFYlNQ NO. |
LIST OF MATERIALS OW PARTS LIST
PuU-«— OTWXRWIAXPKCiriKP I CONTRACT NO. i ‘ Tn ¥ lV j
I DIUKNttOM* ARC IN INCH** ANO HAS 8 "21*100
INCLUt* CHCUICAIAV APFUCD ...
OR FLATKO FIMAHC* APPROVAL!
TOLERANCE ■
j | LCCIMAL* }
APPROVAL*
DRAWN 1|
Q * oari K 'tL'kt* 7-1.
I AM M»u«e rn BTAFCT C Of#IW I MC« AM MUU MOT M KMOMKU. 04
WAIAM U«n AM TM UtM TO* T>M tUNUNCTbM M MALA or
1NMUM WrMWNMMMN.*
T m r il AEROSPACE /OPTICAL DIVISION
4 l.q. PORT WAYNE INDIANA. U.Q.A.
IIUIMTOMl UHHWI »> IIUIHW IMWUlN
HOUSING,, INFRARED
DETECTOR, CHANNEL 3
ASST OF AVHRR
c|T155o| 80087 31
8008791
The size of the sensitive area is determined by the
effective focal length (EFL) of the optical system and the
instantaneous field of view (IFOV) , which is obtained from the
specified resolution. They are related by the equation
w = 0 x E.F.L.
where w = edge width of square sensitive area
0 = angular width of square IFOV
For Channel 3, EFL is 132.3 mm (= 5.209 inches) and 0 =
1.31 milliradians which gives w equal to 0.0068 inch. A tolerance
of ± 0.0004 inch has been established as being a reasonable
amount consistent with detector fabrication capabilities and size
of the IFOV. In the electrode-to-electrode direction, the length
of the sensing element will be slightly greater (by 0.0005 inch)
to compensate for electrode end effects where the detector
normally has very little response.
A uniformity specification has been incorporated into the
detector specification in order to minimize non-uniformity of
response across the sensitive area. A total of nine equally spaced
readings arranged in a 3 by 3 grid will be taken across the sensitive
area and all readings must be equal to or greater than 50% of the
largest reading. The diameter of the test spot is about 0.0015
inch and the readings are spaced by 0.002 inch; the measurements
will be made on a standard "spot scan" test station by the manufacturer.
Acceptance tests are based on measurements made at the antici-
pated operating tesmperature of 105 Kelvin. Measurements are
made at 90, 95, 100, 110, 115 and 120 Kelvin in the event that the
cooler must be operated at the backup temperature of 107 Kelvin or
2-15
a cooler malfunction occurs. Details of what measurements are
made at the different temperatures are given in the procurement
specification for the detector.
2.2.2 Channel 3 Sensitivity
System sensitivity in the infrared channels (numbers 3 and
4) is expressed in terms of the noise-equivalent-temperature dif-
ference, NEAT (NETD) . The NETD is the difference in temperature
between two targets (several times larger than the instantaneous
field of view) which is required to produce a change in signal
voltage equal to the rms noise of the radiometer. The equation
used to calculate the NETD is given in Table 2.2-3, which also
defines the various parameters used in the equation. The de-
gradation factor, a, and the optical f-number have been made as
small as feasible whereas factors in the denominator of the NETD
equation have been maximized as much as possible. The detector
detectivity is the highest available from any vendor for the
spectral bands, operating temperature, field of view, etc. imposed
by instrument requirements. The field of view, element dwell time
(or electrical bandwidth) and change in scene radiance are all set
by instrument performance specifications. We will discuss in this
section only the degradation factor and the optical transmission .
since additional information on the other factors are given in
the Optical Section (3.0) of this report.
The degradation factor, a, consists of two factors for
the case under consideration, that due to 1/f noise from the
infrared detector, and a second factor to account for some ad-
ditional electronic noise. Diffraction effects are negligible
2-16
NETD
2 VT a fn
NETD' . =
d m
T
it D D* 10
o m
T
Noise Equivalent Temperature
Degradation Factor Including
Optical F-number
Diameter of Optical Entrance
Difference
1/f Detector Noise
Aperture
Average detector detectivity in spectral band
at the measuring frequency
Transmission of optics including obscuration
effects
Angular width of square instantaneous field of
view
Elemental dwell time ( = ^ — where Af is the
3 db bandwidth of presampling°f ilter )
Change in scene radiance for small temperature
change at temperature T
Table 2.2-3 Noise Equivalent Temperature Difference Equation
(Channel 3)
2-17
DEGRADATION FACTOR (a)
‘total a i/f • a 2
1/f
f
, f \ I
' In
( u 1
L u *
' f l / -
D 1/2
frequency where 1/f detector noise power equals
G-R white noise
upper cutoff frequency of the system
lower cutoff frequency of the system
= : 1 KHz
= 1 Hz
f = 14.5 KHz
u
ITEM
CHANNEL 3
1/f
1.29
(See text:
1.4
total
1.8
Table 2.2-4 Degradation F.actor
AVHRR OPTICAL TRANSMISSION
Item Description
Reflectance
Channel 3
or Transmittance
Channel 4
Scan Mirror
0.95
0.95
Telescope Mirrors (2 @ 0.97 Ea)
0.94
0.90
Transp. Gold B.S. (Dl)
0.88
0.82
Folding Mirror (M3)
0.96
0.95
Ge Focus Lens (LI)
Ch 3 (2 @ .905)
Ch 4 (2 @ .920)
0.819
0.846
Irtran II Windows Inner
Outer
0.92
0.82
0.95
0.95
Infrared Dichroic (D2)
(OCLI Guaranteed Minimum)
0.81 (T)
0.92 (R)
Ge Aplanat Lens (L2, L3)
0.94
0.94
Bandpass Filters (F4)
0.80
Telescope Obscuration
0.94
0.94
System Transmission
0.333
0.331
(Product of Above)
Values Given Above Are Measured PTM Values.
Table 2.2-5 Optical Transmission For Channels 3 and 4
2-19
since we are considering scenes which are several times larger
than an IFOV. The equation and parameters used to calculate
the degradation factors are given in Table 2.2-4. The degradation
factor for the electronic noise has been calculated by determining
the noise voltage level from the infrared detector and from measure-
ments on the BBM and ETM.
The apparent electronic pick up was somewhat higher on
the ETM than expected. The apparent, 02 , on the ETM was about
1.6 and the total degradation factor was about 2.1. Several
areas of potential pickup were redesigned on the PTM and a lower
a 2 is expected. A value for a 2 of 1.4 and a total a of 1.8 seems
reasonably conservative.
The optical transmission, x, was determined from the
measured transmittance or reflectance value for each optical com-
ponent in the PTM optical system. The value for each component
is listed in Table 2.2-5 for both Channels 3 and 4. Assuming
some optical degradation as in the solar channels we assign a value o
X J
SS
.20
u
=
.20
for use in the sensitivity calculations.
Table 2.2-6 gives the parameters used and the calculated NEiT.
As shown, the spec, value will be achieved.
2-20
2.3
Channel 4
2.3.1 Detector
The detector chosen for use in the channel 4 (3.55 to 3.93
microns) is a photovoltaic indium antimonide photodiode built by
Cincinnati Electronics Corp. The detector has a 0.007 inch square
active area and is used in a configuration identical to the silicon
detectors of channels 1 and 2. That is, it is a current source to
an amplifier used as a current to voltage amplifier with a 16.4 M ohm
feedback resistance. The detector is mounted in a Kovar housing
like the channel 3 detector with the aplanat forming a hermetic seal.
Photovoltaic InSb has better noise characteristics with a
small amount of reverse bias; therefore, a bias on the order of
-30 millivolts is used. The exact bias is determined by
C.E.C. during acceptance testing of the photodiodes. The method
used for bias generation and control is discussed in the section of
the electronics describing the channel 4 preamplifier. Table 2.3-1
summarizes the channel 4 detector parameters.
2.3.2 Channel 4 Sensitivity
The sensitivity of the InSb photodiode is specified in
terms of its quantum efficiency, n, and its noise output for a
given thermal background irradiance. Since noise sources external
to the detector-amplifier system are significant (i.e., background
flux and signal shot noise), this concept is most applicable. The
i use of D* implies a situation where system and detector noise
\
are the limiting factors.
The following analysis determines the sensitivity of the
system in terms of the detector output current for both the noise
sources and the NEAT. The inital section determines the system
noise. Following that are sections which define the detector
output for the NEAT, effects of an albedo signal in this spectral
region, and postamp gain and digitizer effects on the system
sensitivity.
2 . 3 . 2 . 1 System Noise
For Channel 4 there are four noise sources. These are:
1. Background flux noise
2 . Preamp noise
3. Signal shot noise
4. Stray pickup
These are discussed separately below. The stray pickup is handled
as a degradation to the total noise.
In the nominal 3.55 to 3.93 micron band, a 300K background
scene causes a signal current out of the detector given by:
x bs = K A - q n °B E <3- 1
K = optical filtering factor
n = quantum efficiency (average in band)
-19
q - 1.6x10
A O
A = detector area - 2.98 x 10 cin
Q = background photon flux
This assumes that the filter is cold and mounted on the detector.
The optical filtering factor accounts for the fact that when the •
detector is mounted in the system on a cold patch, the incidence
2-22
Table 2.3-1 Channel 4 Detector Parameters
Spectral Band 3.55 to 3.93 Micron
Type of Detector Source Indium Antinonide
Cincinnati Electronics
Operating Temperature 105K
Operating Mode Photovoltaic
Sensitive Area 0.007" Square
Quantum Efficiency in 0.75 Minimum
Spectral Band
Background Noise Level Maximum 10% over Theoretical
Level
Bias Voltage ^30 mvolts
Preamp Gain 16 x 10^
2-23
angle, 0, of the background is limited by the cooler windows and
reduced by the transmissions of the inner window (x ) , the band-
pass filter (x„) , and the aplanat (t ) .
r cl
. 2 ^
K = X_ .T . T . TTsm 0 Eq. 2
F w a ^
For x = x_ = x = .85 and 0 = 75°
w Fa
K = 1.80 ster
For a 300K scene, the photon flux, Q_, in the nominal
■D
14 -2 -1 -1
spectral band is 3.22 x 10 photon cm ster sec . Using
n = 0.75, we have
— 8
l'k s = 2.07 x 10 amp
The noise in this background signal is given by:
ib = / 2q X bs Af_ Eq * 3
The bandiwdth, Af, is 14.5 KHz and so
i^ = 9.80 x 10 ^ amp rms.
The second noise source (the preamp) really consists of
two sources. The input transistor contributes noise as does the
feedback resistance which contributes Johnson noise. Noise data
was measured by G.E. Sonnek on a preamp similar to that required fo
Channel 4. His measured data indicated that with a 16 Mohm feed-
-12
back a total rms preamp noise of 4.42 x 10 amp is obtained in
a 14.5 KHz bandwidth.
The third noise source, the signal shot noise, depends
on signal level. Since the NEAT spec is at 300K, this is the scene
level of interest. The maximum scene to be viewed is a 320K scene.
i '
■
The detector output current due to the scene is given by:
I =• N T R. Eq. 4
S S A
where N = scene radiance
s
T = system thruput
= detector responsivity (amp/watt)
The system thruput T is given by:
T = Aftx Eq. 5
where A = collecting area
Q = solid angle of view
t = system transmission
2 —6
For AVHRR, A is 324 cm , J2 is 1.72 x 10 ster, x is about 0.33;
-4 2
therefore, T is 1.83 x 10 cm ster. Using the minimum specified
quantum efficiency of 0.75 gives a of 2.26 amp/watt. The radianc
of a 300K scene in the nominal spectral band is 1.70 x 10 ^ w/cm^
ster. The detector output then, with no albedo contribution, using
equation 4, is:
I = 7.03 x 10 ^ amp
The rms noise, i s , is
-12
i = 5.71 x 10 amp rms
The last source, stray pickup is a source which is minimized
by design, but never entirely eliminated. We will assume a 1.60
degradation in the TOTAL system noise performance since ETM and PTM
noise degradation appeared to be less in this channel than in
channel 3. This includes a 4% 1/f noise contribution. The total
2-2
noise current, i , under these conditions is:
n
i n = 1.6 jj9 . 80) 2 = (4.42) 2 + (5. 71) J x lO -12 amp Eq. 6
i = 1.95 x 10 ^ amp rms.
n r
2 . 3 . 2 . 2 System Output for NEAT
The detector output caused by a 0.12K scene change at 300K
is determined as follows. Using the Lowan & Blanch "Tables of
Plancke Radiation and Photon Functions", we find that the photon
14 -2 -1
flux of 301K scene in our band is 3.36 x 10 photons cm ste'r
-1 14
sec . For a 300K scene it is 3.22 x 10 . The change in flux
due to a 1.0K temperature change in the scene is then 1.37 x 10 2 " 2
-2 -1 -1 -1
photon cm ster sec K . For a 0.12K NEAT, the Noise Equivalent
Photon Flux is :
12 -2 -1 -1
NEF = 1.64 x 10 photon cm ster sec
The noise equivalent flux on the detector is simply:
NEF D - T ' NEF
NEF q = 1.83 x 10 2 photon sec ~
The detector output current change caused by the scene temperature
change is:
I NEF = n q < NE V
I NEF = 2.19 x 10 -11 amp
By definition the NEAT of the system is the AT which causes
a I NEF equal to the total system rms noise i n . This analysis then
indicates that with the nominal spectral band, an NEAT of 0.12K is
achievable with a S/N ratio of
(S/N) 4 = 2.14 x IQ -11 = 1.12:1
1.95 x 10 -11
The effective NEAT is expected to be .107K as shown in Table 2.3-2
2-25
Table 2.3-2 Channel 4 Sensitivity Parameters
Collection Optics
8.00" Diameter
System Transmission
0.20
IFOV
1 . 3 rnrad
Bandwidth
14.5 KHz
Background Temp
300K
Degradation Factor
1.6
Quantum Efficiency
0.75
Specified NEAT
0.12K at 3 00K
Calculated NEAT
0.107K at 300K
2.3.2.. 3 Albedo Radiance
In this spectral channel, the sunlight reflected from
the Earth Scene contributes to the total scene radiance. Using
Thekaekara's table of solar irradiance we find that a 1.0 albedo
scene at noon at the subsatellite point has a reflected radiance
-4 2
in the nominal spectral band of 1.06 x 10 w/cm ster. According
to NASA personnel, the maximum actual albedo of a scene in this
band is about 0.10. Further, since the spacecraft is never over
a noon nadir, this value of reflected radiance is reduced by sin
67° (which is the maximum orbit normal to sun angle) . These factors
-“6 2
make the maximum reflected solar radiance 9.76 x 10 w/cm ster.
If W6i assume no atmospheric attenuation, this is the albedo
contribution on the day side of the orbit.
The maximum scene temperature is specified as 320K. The
-5 2
spectral radiance of this scene is 3.76 x 10 w/cm ster. The
albedo radiance is then about 25% of the maximum thermal radiance,
and the total scene radiance for a daylight 320K scene is 4.74
x 10 w/cm ster, if the maximum albedo is 0.10 in this band.
2 . 3 . 2 . 4 Post Amp Gain and Digitizer Effects
The post amp gain is determined by the output voltage
range and the maximum scene radiance level. As discussed previously,
~ 5
the maximum scene radiance for a 320K scene is 3.76 x 10
2 -8
w/cm ster. This causes a detector output of 2.04 x 10 amp
2-28
(using equation 4). With a 16 Mohm feedback, the preamp output
is a 0.326 volts. If 5.9 volts represents the output voltage swing
for a full to zero radiance change, then the post-amp gain must
be 13.1 .
Since the A-D converter digitizes to 10 bits (1024 levels) .
each level corresponds to:
6.39375 = 6,25 m volts
T5T3 —
The detector current change caused by the NEAT is 2.19 x 10 - ^
amp. At the digitizer input this is:
2.19 x 1 0 " x 16 x 10-^ x 18.7 =' 6.5 m volts
Thus each data bit will be about the same as the estimated
noise level of this channel.
2-29
3.0
OPTICAL DESIGN
3 . 1 General Description
The optical configuration of the AVHRR is shown in Figure 3.1-1
(the rotating scan mirror is not illustrated for simplicity) , The .
energy from the scene is collected by an 8.0 inch diameter clear
aperture afocal telescope, the primary mirror being the entrance
aperture. In the afocal design, the secondary mirror recollimates
an incoming collimated beam with the angular spread of the exit beam
being increased over that of the input beam by the ratio of diameters
of the input to the exit beams.
Dichroic No. 1 transmits Channels 1 and 2 (solar channels) and
reflects Channels 3 and 4 (infrared channels) . The latter two chan-
nels are then partially focused by a germanium lens doublet (LI)
which is axially adjustable for focusing. The beam passes through
a window on the cooler housing and a window on the gold box (or low
emissivity shield) , the windows being used to provide a vacuum seal
for bench testing the cooler and to prevent moisture from the multi-
layer getting onto a 105 Kelvin patch-detector assembly respectively.
Channels 3 and 4 are separated by dichroic No. 2 which is mounted
on the patch along with the detectors, bandpass filters and aplanat
lenses (L2 and L3) . Dichroic D2 is actually the Channel 3 bandpass
filter designed for the 45° incident beam and is used also as the
dichroic since it reflects the Channel 4 energy. The bandpass filter
are multilayer interference type filters which provide the spectral
response required by the AVHRR specification. Mounting of the two
infrared detectors and dichroic No._2 in a precision-machine mountinc
structure simplifies registration of these two channels to each other
3-1
>
>- 2 v.
tf o 01
<1 r -r V/I
- u O o
ji !
•7 w ,
5 Z U ■
J joj '
> c* £ <
^ H ^ o 4
o w o r
u- 5 0 di 1
0 C* O ,
UJ H ^ ,
5 of 2 I
- 4 ,0
0 ^ * u
ji ^ 7
£ 7 7 .Nl
>
JiSs:
3-2
Figure 3.1-1 WHRR OPTICAL LAYOUT
The beam which is transmitted by dichroic No. 1 is sub-
sequently separated by beamsplitter D3 which reflects Channel 1
and transmits Channel 2. Mirror M4 and beamsplitter D3 are used
to minimize polarization. Elements Fl and F2 are the spectral
r -
shaping filters for Channels 1 and 2 respectively; they both have
flat faces on both sides and produce no focusing effect since they
are in the collimated beam from the telescope. Lens assemblies L4
and L5 are re-focusing elements which form an image of the scene
at the "field stop aperture" (i.e. an aperture plate located in
the final focal plane of the system which contains an opening that
determines the field of view for these two channels) . The energy
which passes through the aperture opening is then detected by
silicon detectors are mounted in TO-5 housing with a flat glass
window hermetically sealed to the housing) .
The optical performance requirements are summarized in Table
3.1-1. In addition to the instantaneous field of view (IFOV) require-
ment, ITT imposed a larger, extended FOV on the optical subcontractor.
That is, the optical system is designed so that the minimum MTF values
given in Table 3.1-1 are obtained over a larger field of view than
would be required by the IFOV. The reason for this is so that the
*■
detectors can be moved individually in their respective focal planes
in order to register all channels simultaneously without losing MTF
performance. In Channels 3 and 4, only the focus lens assembly Ll
is designed for the EFOV since Channel 3 will be used as the reference
channel {i.e. it will be located on the optical axis and the other
detectors will be adjusted until aligned with it) . Since the tele-
scope operates over all four channels, it must cover the EFOV in all
channels. The spatial frequencies listed in Table 3.1-1 correspond
to the subsatellite target sizes given in the AVHRR specification,
i.e., 385 cycles/radian corresponds to a 0.59 nautical mile ground
target, etc. A complete detailed description of the optical sub-
assembly requirements is given to ITT-A/OD Spec. No. 8007907. A
more detailed description of the optical system and components and
the calculated performance is given in the following sections.
The initial design and fabrication of the development models
of AVHRR (BBM, ETM, PTM) was performed by Ferson Optics Div. of • •
Bausch and Lomb. The Flight Model optics are being obtained from
Perkin Elmer, Costa Mesa, California.
3 . 2 Scan Mirror
The scan mirror configuration is shown in Figure 3.2-1 (ITT-
A/OD Dwg. No. 8097928) . It is made using a waffle or egg-crate
construction to reduce weight while maintaining rigidity. The basic
material is HP21 beryllium with a precision elastic limit of 4000
PSI minimum. After machining of the blank it is electroless nickel
plated to provide a good polishing surface. After polishing to the
specified flatness, the flat surface is aluminized and overcoated
to give a high reflectance in all four spectral channels. The back
surface is gold plated for thermal reasons. The scan mirror is
mounted to the drive motor shaft and dynamically balanced as described
in Section 4.0. The scan mirrors are being procured from Applied
Optics Center, Burlington, Massachusetts.
3.3 T elescope Design
The telescope assembly is shown in Figure 3.3-1 (ITT Drawing
#8009383). The telescope collects energy from an infinitely distant
source (i.e., the earth) that subtends a solid angle of one IFOV
3-4
which is located within the extended FOV. The diameter of the
OPTICAL PERFORMANCE REQUIREMENTS
INSTANTANEOUS F.O.V.
ch_1 ql2 ch 3 . chJL
1,31 M.R. 1.31 M.R. 1.31 M.R. 1.31 M.R.
spectral band (um) 0,58-0.68 0.725-1,1 10,5-11,5 3,55-3.93
EDGE. WIDTH OF SQ, FIELD STOP 0,0238" 0,0238" 0,0068" 0.0068"
* «
EXTENDED FOV (RADIUS) 1.1 M.R. 1.1 M.R, 2.2 M.R. 2.2 M.R.
TELESCOPE— AFOCAL CASSEGRAIN (TWO COAXIAL/ CONFOCAL PARABOLOIDS)
TELESCOPE PRIMARY MIRROR— APERTURE STOP R ENTRANCE PUPIL
DIAMETER OE» ENTRANCE PUPIL 8.00 ± 0,01 INCHES
DIAMETER OF EXIT, AXIAL TELESCOPE 3UNDLE 1.00 INCH
MAX, SECONDARY OBSCURATION (iNC. BAFFLE) 2.00 INCHES
SPATIAL FREQUENCY ^-ES.
RADIAN
(ALL. channels)
19
257
385
£H_1
CH 2
QhJ.
OiA
0.96
0.96
0.96
0.96
0.91
0.93
0.90
0.91
0,88**
0.90
0,86**
0.88**
TELESCOPE MIRROR MATERIAL (OWENS-ILLINOIS) "CERVIT"
CHANNEL REGISTRATION SAME AS NASA/GSFC SPEC,
* Applies to focus lens only,
**DESIGN GOAL MTF IS 90%
Table 3' 1"1 - Summary of Optics Requirements
3-5
I
entrance beam into the telescope is 8.00 inches and the diameter
of the exit beam, which leaves through the center hole of the primary
mirror, is 1.0 inch. This reduction in beam diameter caused a
corresponding increase in angular extent, i.e. the IFOV of 1.31 mr
(milliradian) by 1.31 mr into the telescope leaves with 10.5 mr by
10.5 mr beam spread. The largest field angle over which the telescope
must operate is the extended FOV which has a 4.4 mr diameter in object
space '(without the extended FOV, the telescope would have to cover
a field of 1.85 mr) .
The telescope consists of two confocal, coaxial paraboloidal
mirrors which are called the primary (large) and secondary mirrors.
The primary mirror clear aperture is both the aperture stop and
entrance pupil of the optical system. It has an 8.00 inch diameter
clear aperture and a focal length of 1.0 inches giving it an optical
speed of 1.25. The intervertex distance (the axial separation of
the primary and secondary mirrors) is 3.75 inches. This gives a
1.0 inch diameter beam reflected by the secondary for a collimated,
axial beam into the telescope. Because of the extended field of
view, the clear aperture of the secondary mirror must be a minimum
of 1.040 inches. In order to • re-collimate the beam, the secondary
mirror must have the same optical speeid of 'the primary mirror, i.e.
1.25. The focal length of the secondciry mirror is therefore 1.25
inches .
A conical baffle is located around the secondary mirror to
prevent radiation within the EFOV from getting past the secondary
mirror and into the aft optics (i.e. the optical components behind
the: telescope). The telescope barrel is made of invar metal which
has; a very low coefficient of thermal expansion and therefore maintains
3-8
mirror alignment over the temperature range. The mirrors are
made of Cervit (made by Owens-Illinois) and has a low coefficient
of thermal expansion. After polishing, the mirrors are coated with
a Ugh reflectivity coating of aluminum which is protected by an
overcoat layer of silicon monoxide; this coating is used because- of
its uniform spectral reflectance in all four channels. Other
mechanical features of the telescope are described in Section 4.0.
A unique feature. of the AVHRR telescope assembly. is the absence
of any coma or astigmatism. Thus the angular alignment of the re-
flective, reimaging optics following the telescope is not critcal
with respect to the telescope exit beam. A tolerance analysis of
all optical elements was performed by Ferson Optics which showed
that the only sensitive elements are the primary and secondary
mirrors (i.e. the alignment with respect to each other). The optic
subcontractor not only fabricates the mirrors (as well as all lens
elements) but assembles and aligns the mirrors in the telescope
housing assembly shown in Figure 3.3-1. The primary mirror is first
potted in place and then the secondary mirror is adjusted by position-
ing the entire spider support in an annular ring at the front of
the telesdope) . The performance of the completed' telescope is
checked as a subassembly after potting has cured. .
3 . 4 Channels 1 and. 2 Lens .Design
The reimaging optics for Channels 1 and 2 are AR coated lenses
consisting of three elements. This triplet design is shown in Figure
3.4-1 which gives the radii of curvatures, axial thickness of each
element, axial spacing, minimum half apertures for each lens surface
3-9
and the material. The lens materials were selected to control
chromatic aberrations over the wide spectral bands, especially
channel 1 (o, 55 to 0.90 ym) . The triplet design has more than
adequate MTF performance over the extended FOV and has high
transmission and low manufacturing sensitivity. It has been
possible to achieve a single design which can be used for both
channels 1 and 2 even at the relatively fast effective f/number.
.The- focal plane is located, at position -NO. 7 in Figure 3.- •4-1.
The field defining aperture (or "field stop") is located at this
position; the size of the opening is 0.0238 inch by 0.238 inch
and produces, the 1.31 mr by 1.31 mr field of view in object space
in Channels 1 and 2 :
The optical performance data for Channels 1 and 2 is summarized
in Table 3.4-1. The optical design was performed by Ferson Optics
under subcontract to ITT-A/OD. Preliminary design was carried out
using computers at Ferson's plant while the MTF data for the final
design was obtained , 5 using -the Grey- Dif fraction MTF Program on a CDC
6600 computer. Five wavelengths were used for evaluation for each
channel and MTF data was determined for best alignment (on-axis)
and worse alignment (at the edge of the extended FOV) for all
channels. jiiote 'that the calculated MTF for. 385 cycles per radian*
at the ‘edge 'cf 'the EFOV is 97 . 4% for ' Channel 1 and 96.9% for Channel
2 and is much better on-axis. The required value at this optical
*385 cycles per radian corresponds to 0.831 line pairs per mm
in the focal plane and to 0.5 nautical mile target on the
earth's surface.
3-10
3
NO
RADIUS
MM
THICK
KM
HALF
APT. MM
MATL.
1
80.24
10.00
17.00
PS!< 53
2
-51.04
9.03
16 ; 59
V
.3
-20.70
4.50
13.33
SF 18
4
420.65
l.CO
13.57
5
26.15
11.38
13.92
PSK' 53
6
-1218.2
38.29
12.55
7
IMAGE"
NOTE: 1) Thickness between surfaces on axis
2) Half apt. shown are for efov edge ray.
Design unchanged by mods 6, 7 and 8.
\ ,
Figure 3.4-1 ’WHRR Triplet Design CH 1 & 2
ZT-
AVHRR OPTICAL SYSTEM MJF (PERCENT) .
ON AXIS
CM 1 fH 2 ch 3 ch 4
1 1 * — • L — ~
CYCLES/RAD
; 19
257
385
SPEC
96
93
90
CALC
>99
98.7
98.2
“■
. SPEC
96
\ 91
88
CALC
>99
>99
>99
SPEC
96
90
86
<
CALC
j >99
: 98
j 97
— w.
SPEC
96
91
88
CALC
>99
98.5
I 97,7
FIELD EDGE
19 ;;
.257
. 385- ' .
*
96
.. ..90.
>99
98,4
_ - -9L 6 .
96
91
88 _.
>99
98.2
97.3
96
90
86
, >99 ,
97
95
96
91
88
>99
98
■ 97
F.F0V EDGE
■ . t
IS
96
>99
>99
96
>99 "
96
>99
257
93
*
98.2
91
97,9
90
96
96
385
90 ,
■
97,4
88
96,9 •
86
94
88
94
Table 3.4--1 Calculated MTF Performance. ,
frequency for Channel 2 is 90% and for Channel 1 it is 88% with a
design goal of 90% for both channels (see Section 3.1). The high
optical performance achieved in the design permits comparatively
easy manufacturing tolerances for the lens elements and -tolerable
mechanical alignment tolerances (see Section 3.6 for tolerance
analysis results) .
3.5 Channel 3 and 4 Lens Design
The reimaging optics focus the energy coming out of the tele-
scope onto as small a detector as possible. The optics consist
of a focus lens assembly (doublet) made of germanium for Channels
3 and 4; the index of refraction of germanium does not change
significantly between 4 and 12 ym. The slight difference in index
is taken care of by locating each detector at the corresponding
focal plane. The focus lenses are axially adjustable to obtain
optimum focus by means of an adjustment external to the radiant
cooler; this also permits use of a smaller opening in the cooler
which lowers the thermal input thereto.
A third germanium lens, called an aplanat, is used just in
front of each detector to do the final focusing of the beam (an
aplana.t introduces no additional coma^or spherical aberration) . v£he
best signal-to-noise rat.io is obtained by using the smallest possible
detector area since detector noise depends somewhat on area in
Channels 3 and 4. The detector area can be found from the Abbe sine
law (see for example, R. C. Jones, Applied Optics, Vol.'l, p. 607,
1962 or D. Marcuse, Applied Optics, Vol. 10, p. 499, 1971).
3-13
where A,
2 ?
A 9 = A , it sm y
o d
area of collecting optics
0
A-
width of square IFOV
2
area of square detector = w^
maximum incidence angle of an axial ray at
the detector.
This can be simplified to D q 0 = 2 sin’ - y.
definition of f-number, i.e., f
2 sin y
Usin(y the sine
this becomes w, = f
d n
D q 0. Since 0 has been specified ( 0 = 1.31 m r) and we have
chosen D = 8.0 inches, we can only minimize . f n (maximize y) .
About the maximum feasible value of y is 50°; this gives f
= 0.653 and w^ = 0.0068 inch. The sensitive area of the infrared
detector is the field defining "stop" in Channels 3 and 4; in both
infrared channels the edge width of the square detector is nominally
0.0068 inch + 0.0004 inch. If we note that w^/0 = EFL (effective
system focal length) , then
f
n
EFL
D
which is the common definition of f-number.
The' apMnat has the ef'f.ect-rof optically magnifying thehsize
of the detector by the refractive index" (about' 4.0 for germanium
at 12 ym) so that the focus lens "sees" a detector image 0.0272
inch on an edge. Since the focus lens sees an angular field spread
of 10.48 x 10~ 3 radians from the telescope (Channel 3), it has a
focal length of about 2.6 inches. The speed of the focus lens is
also about 2.6 since the beam is 1.0 inch in diameter (neglecting
field spread) ; the actual clear aperture diameter of the focus lens
is large enough to accept the beam spread plus some allowance ' for
mechanical tolerance.
3-14
The detail design information for the Channel 3 and 4 re-
imaging optics is given in Figure 3.5-1. An enlarged scale drawing
of the optical elements mounted inside the radiant cooler is shown
in Figure 3.5-3., The sensitive area of the Channel 3 infrared .
detector is located at position 15 and that of the Channel 4 detector
at position 20. The position of the aplanat lens is held very
accurately with respect -to the detector sensitive area by mounting
the aplanat directly to the detector housing. This is shown in
Figure 2.2-2 (ITT-A/OD Drawing 8008791), the housing for the
Channel 3 infrared detector. The aplanat lens also serves as the
window for the housing since it is hermetically bonded in place.
The same type of construction has been successfully used by the
same detector vendor on the infrared detectors for other programs .
The MTF performance of the infrared bands for both on-axis
and edge of the EFOV is given in Table 3.1-1. The MTF data given
in the table includes the effects of diffraction. Since Channel 3
is used as the reference channel for registration, the detector-
aplanat assembly for this channel is positioned pn-axis. Achieve-
ment of the 86% MTF requirement for this channel is therefore not
expected to present a problem. The calculated MTF values for both
infrared channels is appreciably higher than the required values
imposed on the optical vendor by ITT-A/OD. Tolerance analysis
studies indicate that the required performance is achievable
without undue difficulty. The axial astigmatism in Channel 3 which
is caused by the tilted dichrcic (D2) was analyzed and was found
to be masked by diffraction effects at these long wavelengths.
..... . . . . , ,. .3-15
X.
v
AVHRR CH 3 & 4
ELEMENT PRESCRIPTION
*
NO,
RADIUS THICK MM
M M ,ON AXIS
HALF MA. L
APT MM . .. -
1
-185.15 j 10.00
19.66 „
r-, 6 E
2
-294.56 i 10,00
20.54
3
166.29 : 10.00
22.15 r
6 E
A
-3238.3 4,3.2
21.53
5
4,78
15.87 IRTRAN
/•
D
17,40
15.35 2
7
1,57
11.10 IRTRAN
8
- , 15.77
10.93 2
9
2,54
6 E
10
7.60
—
11
;■/ |v >
. FILTER
12
.
DELETED
13
l|.58 3.00
4,43 „
' G E
'14
2.80 2,48
2.42
15
IMAGE PLANE CH
3
9
11,5
—
16
1.01
4.18 r
- G E
17
0.25
4.12
18
3,90 j *3,00
3,52 _ _
G E
19
1,93 | 1,57
1,51
20
IMAGE PLANE' CH
4
Figure 3.5-1 Infrared Reimaging Optics Data
3. 6 AVHRR Tolerance Analysis
A tolerance analysis was run on the AVHRR Optical Design
using a Ferson Optics computer program. The program provides
information of changes in the OPD* (in Raleigh units) due to
perturbation of optical design parameters about “their nominal
values. From the OPD variations due to perturbations, the mechanical
tolerances for manufacture can be determined. The data presented
here is for individual elements as well as groups of elements
which are mounted on subassemblies. Figures 3.6-1 and 3.6-2
show concentricity and parallelism for group data in schematic
presentation. Mirrors and dichroics are not shown but must be
considered when distributing the parallelism and concentricity
tolerances. A summary of the recommended tolerances follows.
3.6.1 Summary of Mechanical Tolerances* *
Channel 1 and 2 - Focus Lens
a. Focus Lens Diameter - OD + 0.000 -0.002
b. Focus Lens Cell - ID +0.002 -0.000
c. Minimum Clearance for Thermal Expansion - 0.001
d. - -Triplet ’ concentricity relative to telescope optical
axis - 0.010 , ^
e. Triplet parallelism relative to telescope optical
axis - 4 mr
OPD = Optical P at ^ difference.
Linear dimensions in inches, angles in milliradians (mr) .
3-18
TELESCOPE
DOUBLET
APLANAT
.005 IN
CONCENTRICITY
*
*— .C02*IN
PARALLELISM
*0,002 INCH IS TOL. OF APLANAT WITH RESPECT
TO DETECTOR CENTER; TOLERANCE INDICATED
SHOULD BE 0,015 INCH,
CHANNEL 3 & 4 TOLERANCE DATA
FOR ELEMENT GROUPS
v
FIGURE 3.6-2
16-17*
17-18*
f . Spacer Data
Lenses
Length ±0.002 ±0.003
ID ±0.001 ±0.001
OD +0.000 -0.002 +0.000 -0.002
*See Figure 3.4-1.
Channels 3 and 4 - Focus Lens
a. - Focus Lens Dia. - OD +0.000 -0.002
b. Focus Lens CE11 - ID +0.002 -0.000
c. Minimum clearance for thermal expansion - 0.001
d. Doublet concentricity relative to telescope
optical axis - 0.005
e. Spacer Data
Length - ±0.003
ID - ±0.002
OD - +0.000 -0.002
Channels 3 and 4 - Aplanat
a. Outsider Diameter Aplanat - +0.000 -0.002
b. Diameter aplanat cell - +0.002 -0.000
c. Concentricity of cell diameter relative to detector
axis -• 0.002
d. Mounting Surface deviation relative to detector
* .
. 3-21
surface
± 0.001
3.6.2 Surface Quality for Filters, Beamsplitters
a. Scratch
b. Dig
c. Flatness
d. Irregularity
e . Wedge
Ch. 1 and 2
60
40
10 fr
1 fr
1 mr
3.6.3 Mechanical Adjustmerit .Data (Inches)
Ch . 3 and 4
80
50
2 fr
2 fr
1 mr
Back
Focus
Focus
Adjust
X-Y
Adjust
Channel
1
1.512
±0.050
±0.040
Channel
2
1.515
±0.050
±0.040
Channel
3
0.0976 ±
0.002
±0.050*
±0.040
Channel
4
0.062 ±
0.002
±0.050*
±0.040
*An additional adjustment of 0.0606 for Channel 3 and
0.0343 for Channel 4 toward the focus lens is required
for room temperature testing.
Position of Channel 4 aplanat relative to Channel 3 aplanat
± 0 . 002 .
3 • 7 Dichroics) Beamsplitters and Cooler Windows
Dichroics Dl and D2, Beamsplitter D3, the Irtran cooler windows
and the Channel 3 filter are being obtained .from OCLI (Optical Coating -
Labs, Inc., Santa Rosa, Calif.). The function of the two dichroics and
the beamsplitter is to separate the optical beam exiting the telescope
into four separate beams before final focussing in each of the four
channels. Dichroic Dl consists of a thin "Transparent" coating of gold
evaporated onto a flat glass substrate. This type dichroic reflects
3-22
radiation in Channels 3 and 4 with approximately 82% efficiency and
transmits Channels 1 and 2 with approximately 75% and 70% efficiency,
respectively. It does polarize the energy in Channels 1 and 2 and is
the reason why an additional folding mirror (M3) is used. More complete
details of this dichroic are given in ITT-A/OD Spec. No. 8009262.
Dichroic D2 reflects Channel 4 radiation and transmits Channel 3
This dichroic is the bandpass filter for Channel 3 which is designed for
a 45° incidence angle and the focused optical beam when cooled to 10 5
Kelvin. The " f ilter-dichroic" reflects energy in the 3.5 vo 4.0ym
spectral band. The reflectance for Channel 4 radiation is 90% minimum
and the transmittance for Channel 3 radiation is 75% minimum. The
substrate material for the dichroic is optical grade germanium; the
thickness was made as large as feasible to minimize bending when it is
cooled to operating temperature. More complete details of this dichroic
are given in ITT-A/OD Spec. No. 8008792.
A detail layout drawing showing the extreme optical rays at
Dichroic Dl, Beamsplitter D3 and folding mirrors M3 and M4 is shown
in Figure 3.7-1. The angle of beamsplitter D3 with respect to the
optical axis was minimized in order to reduce polarization effects.
Since the spectral bands of Channels 1 and 2 overlap each othef, D3
cannot be a dichroic beamsp'litter but must be-a neutral density
separator. The neutral density beamsplitter consists of a thin
evaporated coating of inconel on an optically-polished, flat glass
substrate. The minimum average reflectance is 22% for Channel 1 energy
and the minimum average transmittance is 34% for Channel 2. Complete
details for beamsplitter D3 are given in ITT-A/OD Spec. No. 8007932.
3-23
The cooler windows are optically-polished, flat circular
discs of Irtran 2 material (Eastman Kodak trade name) . They are
antireflection coated by OCLI to give a minimum transmission of
85% in both Channels 3 and 4. The coated windows meet standard
military specifications for .aun«j.cnce, hardness and humidity. The
outer cooler window (Wl) is comparatively thick since it must with-
stand atmospheric pressure when the cooler is evacuated for bench
'cooling tests.
The Channel 4 bandpass filter is a conventional multilayer
coating on a polished germanium substrate. The bandpass character-
istics such as cuton and cutoff slopes, tolerances on location of
the 50% transmission wavelength, etc., are given in Section 3.8.2.
3 . 8 Spectral Definition
3.8.1 Spectral Definition of Solar Channels
In order to determine the AVHRR system spectral response in
Channels 1 and 2, it is necessary to determine the effects of those
elements having a varying spectral characteristic. These elements
are :
a. Mirror Coatings
b. Gold Beamsplitter
c. Lens AR Coatings
d. Silicon Detectors
e. Spectral Bandpass Filters
Since the relative spectral ^response of the above elements affects the
system spectral response in the given bands, and since their absolute
response affects the signal to noise ratio, both requirements (spectra
and sensitivity.) must be considered together.. The spectral
3-25
characteristics of the above five groups will be discussed individually
below, followed by their combined effects.
3 . 8 . 1 . 1 Mirror Coatings
A complete study of the various potential mirror coatings was
done in the preliminary design of the AVHRR . It was decided that to
achieve maximum sensitivity in Channel 2 (where overcoated aluminum
mirrors have a dip in their reflectivity) silver mirrors would be used
for the telescope mirrors. 'For polarization compensation, however,,
the scan mirror and the folding mirror behind the solar channel beam-
splitter were aluminized. The silver coating chosen was the Muffelleto
Optics "low pit" coating which was supposed to survive the required
humidity and temperature extremes.
In the process of buulding the Breadboard and Engineering Model
telescopes, no silver coating was found (at reasonable cost) which
passed the humidity test. In all cases the mirrors degraded extensively
when placed in the humidity test. Therefore, the decision was made to
use aluminum mirrors which have a reflectivity equal to or greater than
that shown in Figures 3.8-1 and 3.8-2. The coating after which the
reflectivity curves were fashioned is the _ Evaporated Metal Films Corp.
"enhanced aluminum" coating.
3 . 3 . 1 . 2 Gold Beamsplitter
The dichroic to split Channels 1 and 2 from Channels 3 and 4
is the thin gold film type at 45° to the incident beam with the near
IR and visible energy being transmitted. Figure 3 . 8 - 3 . shows the
measured solar channel reflectivity of the gold dichroic used in the
KIOTO.' 30 '!• MiuiMUM @ 525 n. rn.
W ►*
u o
U < 2 ui
n x
“ a -> h
S o J *
»-»<«_;
* * 0 £
>-* < • w 2
Stg'S
“ u. “. 5 1
o “u ■ g
u i i u
£ m t | -
> 2 a ± D
<; o
n OJt»Z
g f 0 < £
§ 5 5 « 3
K w , * 3 H
*. “ £ <
5 • * *
S zo?
Sm-
i
w.
j_ . 6 °
0 ~ - u w
W> y _1
Ua ?§2
1 “ o Q O
sgsS-
3 *>gs
"S°86
iziii
S 3 *-i
2 a t m <
j Q 0 a z
I
FIG 3.6-1
.fcO .10 .80 .80 1.00 1.10
WftVt LENGTH ' Mlf..KOKit
ABSOLUTE SPECTBAL
ECFLECTIVITT BEQUIEEMcUT
CALL MlCeOES EXCEPT CHAW 2 FOLDIUG.)
CODE IDENT NO.
8008787
SHEET 3-21
IINCOOAOEVHIC C 03 f. IOIiixm 4 -?c.
IT TAOO □ tO 1
/
06
e&LArn
3 . 8 . 1 . 3 Lens AR Coatings
The anti-reflection coatings on the three relay optic lenses
used in each channel do not materially affect the relative spectral
response of the solar channels. A MgFl^ type single layer coating is
used and the total lens transmission will change from about 97% near
the band center to 95% near the edges. Even for three lenses, the
effect is insignificant.
3 . 8 . 1 . 4 Silicon Detectors
Because of the wide spectral bands covered by Channels 1 and 2,
the detector in each is an important factor in defining the band edges.
Quotes were solicited from six different companies, and all were
requested to define the spectral characteristics of their proposed
device. Selection of the vendor was based on both the relative spectral
response and the absolute sensitivity. The definition of the Channel 1
spectral band edges is complicated by the fact that the silicon detector
relative response is down to about 30% at 0.50 micron. This means that
above some relative transmission value (about 40%) the spectral bandpass
filter does not define the system response. The response is basically
defined by the detectors and the other spectrally variant elements.
This effect could have been- reduced by cutting— the*. detector peak
spectral response and lowering the system signal- to-rnoise ratio; however,
it was decided that the slow cuton of the system spectral response is
more acceptable than a reduced signal-to-noise ratio.
As stated previously, one detector design is used for both
solar channels. This approach provides excellent sensitivity in both
channels and simplifies (and so reduces the cost of) the detector.
The measured spectral response of one of the detectors , shown in figure
3.8-4, is typical of that measured on all detectors.
3-30
WAVELENGTH - MICRONS
Figure 3.8-4
3 . 8 . 1 . 5 Spectral Bandpass Filters
The spectral filters were purchased from Fish-
Schurman Corporation and have the responses shown in Figures
3.8-5 and 3.8-6. The filters consist of a two piece
laminated sandwich with one piece being a 3 mm thick Schott
glass filter and the other being a 2 mm thick clear substrate
upon which a multilayer coating is applied. The multilayer
is placed inside the sandwich and is sealed by the epoxy
sealant from ambient conditions. The problem of spectral
variations as a function of absorbed water is eliminated.
This approach was used in the ATS-F VHRR solar channel
filter. The epoxy sealant is Summers Laboratory C-59 which
meets MIL-3920 and has measured outgassing characteristics of
2.94% weight loss and 0.108% condensible materials. (This
information was supplied to ITT by NASA/GSFC) . This is the
same sealant used to seal and focus lenses in each solar
channel of the A VHRR .
The filter generates its spectral bandpass using a
Schott glass filter for the short wave cut-on and a multi-
layer coating for the long wave cut-off. The slopes of the
filters are all about 6% between the 5% and 80% response
points. Short wave out-of-band blocking is defined by the
Schott glass and is complete. Long wave out of band blocking
is defined by the multilayer and, while not as good as the
Schott glass, results in an out of band signal well below
that specified as discussed later (the detector has cut off
by the time the filter transmission comes back up) .
CHANNEL 1 FILTER
SERIAL 202
ANGLE OF INCIDENCE
NORMAL
X
PERCENT TRANSMISSION
3 8 . 1 . 6 System Spectral Response
Figures 3.8-7 through 3.8-14 show the measured chan-
nel 1 and channel 2 spectral response curves for the AVHRR/1
flight model instruments.
3.8.2 Spectral Definition of Thermal Channels
The spectral characteristics of typical bandpass
filters for Channels 3 and 4 are summarized in Table 3.8-3.
The filters are the dominant element in defining the spectral
response of the AVHRR instrument. The response of the indium
antimonide detector varies smoothly from 3.5 to 4.0 ym,
being about 5% less at the shorter wavelength side of the
band. The Hg Dc Te detector procurement specification con-
tains a requirement that the response at any wavelength
between 10.5 and 11.5 urn be no less than 80% of the maximum
in-band response. Therefore, the detectors should not appre-
ciably affect the system response. The complete procurement
specifications for the bandpass filters are contained in ITT-
A/OD Specifications 8008790 (Channel 4) and 8008792 (Channel
3) .
The germanium lenses have some effect on the response
of Channel 3 due to their increased absorption toward longer
wavelengths. This effect causes the Channel 3 response to be
somewhat skewed to shorter wavelengths in the band. Figures
3.8-15 through 3.8-22 show the measured channels 3 and 4
spectral response for the AVHRR/1 flight models.
3-35
TABLE 3.8-3
FILTER CHARACTERISTICS
CHARACTERISTIC
CHANNEL 3
CHANNEL 4
50%-of-Peak Cuton Wavelength
10 . 5±0 . 09pm
3 . 55± . 06pm
50%-of-Peak Cutoff Wavelength
11. 5+0 . 09pm
3 . 93± . 06pm
Cuton and Cutoff Slopes
£ 3%
<_ 3%
Response at 10.0 and 12.0pm
£ 1%
N . A.
Response at 3.40 and 4.12pm
N . A.
£ 1%
Transmission less than 0.1%
From 1.8 to
From 1.8 to
9 . 8pm and
3 . 2pm and from
from 12 . 2 to
4.3 to 7.0pm
18 . 0pm
In-Band Transmission
> 75% Avg.
> 75% Avg.
3-44
RELATIVE RESPONSE
RELATIVE RESPONSE
3.5 3.6 3.7 3.8 3.9 4.0
WAVELENGTH - MICRONS
Figure 3.8-18
3-48
WAVELENGTH - MICRONS
Figure 3., 8-19
WAVELENGTH - MICRONS
Figure 3.8-20
3.9
Channel Resigtration
The approach used to register the AVHRR channels con-
sists of first registering the two thermal channels at a
subassembly level. The patch, with both detectors mounted,
is placed in a test vacuum dewar and illuminated through the
relay optics, by a collimated beam. The collimator is a ger-
manium lens which has a slit target and cuopped radiation
source. The slit is moved while the outputs of the two chan-
nels are recorded. The difference between the 50% response
points is the misregistration. This is repeated with a slit
rotated 90° so that the distance that each detector must move
i o measured. The patch is then removed from the dewar, the
adjustment made, and the patch retested to verify alignment.
The following procurement is used for channel
registration.
a. Determine the physical location of the focal
points for Channels 3 and 4 with reference to the
optics subassembly during optics acceptance
tests .
b. Place the subassembly into the AVHRR and mount
the cooler to place the detectors in their proper
position.
3-53
c. With the solar channels centered in their EFOV,
determine the misregistration between all the
channels. If the error is more than can be
corrected by moving the solar channels , reposi-
tion the cooler to place the IR detectors in the
correct location.
d. Remeasure the misregistration and move the solar
channel detectors to complete the registration
procedure.
Step c. essentially eliminates the tolerance buildup problem
in the instrument by actually measuring the positions and correcting
accordingly. If the initial placement of the cooler is sufficiently
close to place the IR detectors within the solar channel EFGV, then
no cooler repositioning will be required.
The procedure has been used to register both the ETM and PTM
instruments well within the specified value,
3 . 10 Polarization Sensitivity
An analysis was made of the polarization sensitivity of
Channels 1 and 2 with the object of meeting the requirements of the
GSFC Specification.
P = T 1 ~ T 2 1-053
T 1 + T 2
This equation holds at nadir (or at 90° from nadir) as a
result of the orientation of the optical elements that contribute to
the polarization. (On the Flight Model instruments,, the maximum allow
polarization sensitivity is .07.) The transmittances obtained foj. the
parallel and perpendicular polarizations are the extremes for any
linearly polarized wave that travels through the optical fain.
3- 5
The theoretical and measured polarizations of the elements
are given in the tables contained in the DIR's. The channel transmit-
tances were calculated by giving a weight of 1/2 to the spectral end
points (,i.e., originally 0.5 and 0.9 ym in Channel 1 and 0.75 and
1.00 yra in Channel 2; revised to 0.55 to 0.9 ym in Channel 1 and 0.725
and 1.1 ym (silicon detector cutoff) in Channel 2).
The analysis performed during the design phase of the AVHRR
program considered several options including the angular position of
the dichroics and beamsplitters, the addition of reflecting surfaces
for compensation, and the types of coatings on reflective surfaces.
The complete analysis is detailed in AVHRR D.I.R. #6 and 7.
Several changes have been made to the orginal polarization
sensitivity design of the AVHRR. In particular, the gold dichroic has
been made by OCLI so that the percent polarization of that element is
considerably less than originally thought. The scan mirror coating is
the major compensating element for the polarization of the gold dichroic.
The high reflectivity coating used on the PTM and Flight Models also has
less polarization (because the reflectivity dip around 0.86 micron is
minimal) . The combination results in a measured PTI1 polarization
sensitivity as listed in Table 3.10-1. The Polarization sensitivity as
a function of scan mirror position was measured on the PTM as a matter
of interest. As can be seen the instrument performs quite well and it
can be expected that the Flight Model units will be well within the 0.07
allowed.
3-55
TABLE 3.10-
2ATION
CHANNEL 2
P.S.
.047
.0 49
.053
.047
.049
3-56
3.11 Scattered Sunlight
4
We have studied the following effects of direct sunlight:
a. The temperature gradient it produces across the
honeycomb of the in-flight target.
b. Its reflection from the in-flight target and the
resultant calibration errors.
c. Scattering and the resultant signal contamination.
Because the above occurs only during the nighttime portion of the
orbit, we need to consider only Channel 3 (10.5 to 11.5 ym) and
Channel 4 (3.55 to 3.93 um) .
From our studies to date, we conclude that:
a. The honeycomb gradient is not sensibly changed
from its value in the 906 n mi orbit.
b. Under the worst conditions, reflection of sunlight
from the in-flight, target introduces a significant
error (0.65K) in Channel 4. As a result, it may be
desirable to restrict the Channel 4 calibration
period, e.g., to the portions of the orbit when the
target is shaded by the earth, spacecraft, instruments
or sunshields (if added) . The corresponding error in
Channel 3 is negligible (0.002K).
The temperature errors quoted in b are for the preliminary TIROS-N
instrument/spacecraft layout shown in NASA/GSFC Drawing GDSK-4799.
Two separate instrument orientations are of interest, for scattering
from the telescope and from the scan mirror.
3-57
I
3.11.1 Honeycomb Temperature Gradient
This problem was analyzed in DIR #18 (Worst Case Honeycomb
Temperature Gradient in the In-Flight Thermal Calibration Target,
May 9, 1974; included in Section 7.1). The analysis was revised
new altitude of 450 n
mi and the
May 2
, 1974). The results are;
For 8
£ 27.83°, min.
6 = 62.17
For 8
27 . 83°, min.
9 = 62.17
(T o - Vl "
2.513°
ii
C-J
rJ
1
o
C . 6 39°C
Effective cell gradient = (T - T ) = 0.95°C
o t
Effective calibration gradient = 0.079 (T - T.) = 0.075°C.
o l
The effective calibration gradient is not sensibly changed from its
previous value of 0.077°C.
3.11.2 Sunlight Reflections from In-Flight Target
Sunlight reflected from the in-flight target introduces a
calibration error in addition to those errors we have previously
considered. The error is negligible in Channel 3, but can be
significant in Channel 4.
The solar exitance reflected from the target is given by
M = E o A s
s s
where
E
s
a
A
s
A
T
direct solar incidence (irradiance) in the wave-
length band and perpendicular to the sun's rays.
diffuse hemispherical reflectivity of the in-flight
target
exposed target area projected perpendicular to the
stun 's rays
effective target area viewed by radiometer
r ( 8 ) : ,/4 in : .
3-58
The values of solar incidence were calculated from the table given
by Thekaekara (Optical Spectra, March 1972, p. 32); the results are
Channel 3 (10.5 - 11.5 vim) , E = 1.71 x 10~ 5 Wcm -2
s
Channel 4 (3.55 - 3.93 pm), E s = 2.52 x 10' 4 Wcm' 2
The reflectivity p can be calculated from the limiting formula
of Treuenfels (J. Opt. Soc. Am. 53, 1162, 1963) or by interpolation
from the graph of Sparrow and Cess (Radiation Heat Transfer, Brooks/
Cole, 1966 p. 165). In either case, an assumed paint emissivitv of
0.92 (3M 401 black) results in a hemispherical cavity emissivitv of
0.980 and a reflectivity of 0.020. As shown in the graph of Sparrow
and Cess, the limiting emissivity of a cylindrical cavity is reached
at a length to radius ratio of about 2:1 when the paint emissivity
exceeds 0.7. The length to radius ratio of a honeycomb cavity is 8:1.
The ratio of flat to total target area is 0.045, so that the hemis-
pherical reflectivity is 0.0227. It would be desirable to have
spectral emissivity (or reflectivity) data on the 3M 401 in the bands
of interest. However, data could only be found for 3M 101 and;
therefore, an average 401 value of approximately 0.92 was used.
The exposed area A g normal to the sun's rays was determined
from scan cavity projections supplied by J . D. Crawford. The worst
case situation is shown in Figure 3.11-1. It occurs at a 6 angle of
27.83°, when the spacecraft leaves the earth's shadow, and an orbital
position 15° north of the plane of the ecliptic in the night-time part
of the orbit. The value of A is 10.9 square inches when shading from
the anticipated spacecraft is included. We then have (in the absence
of any sun shield)
M
(Ch.
3) = S . 4 3 x 10“ 8
Wcm'
s
M
iCh.
4) = 1. 24 x 10'*
Wcm'
3-59
The comparative effect of the reflected sunlight is shown in Figure
3.11-2. In Channel 3, the sunlight produces an exitance level that
is more than 4 orders of magnitude below that of the 295K self-emission.
On the other hand, the reflected sunlight in Channel 4 is only about an
order of magnitude below the self-emission.
The calibration error introduced by reflected sunlight may be
expressed as an effective temperature increase; it is given by
M
6T = s
s ' dM/dT
where dM/dT is the rate of change of in-band blackbodv exitance at the
nominal target temperature of 295K (22°C) . For the bands of interest,
we have
(Ch. 3) = 4.263 x 10 -5 Wcm - 2 K' 1 at T = 295K
dT
dM (Ch. 4) = 1.905 x 10~ 6 Wcm -2 K -1 at T = 295K
dT
The calibration errors for the worst case are then
6 T (Ch. 3) = 0 . 002°C
s
6T (Ch. 4) = 0 . 65°C
s
3.11.3 Signal Contaminacion
With the AVHRF. mounted as shown in Figure 2 of GSFC Specifica-
tion S-731-F— 118 (Rev. C) , there is a large effective sunshield (i.e.,
the spacecraft itself) that limits direct solar exposure of the instru-
ment to the night-time portion of the orbit. During this time, of
course, Channels 1 and 2 are not used. As a result, we need to consider
the effect of scattered sunlight only in Channels 3 and 4. In common
with the reflections from the in-flight target, the larger effect is in
Channel 4.
3-61
c If/r
We may define the required attenuation coefficient for
scattered sunlight as the normal solar incidence (irradiance) out-
side the earth's atmosphere divided by one-third of the minimum
radiant signal,
a = fs_
(1/3) (NER) Aft
whex'e NER = noise equivalent radiance = 1.6 x 10" 6
Wcm" 2 ster" 1 (Ch. 3), 8.0 x 10" 8
Wcm -2 ster -1 (Ch. 4)
Aft = throughput of optics = 5.43 x 10“ 4 cm 2 ster.
The NER values were supplied by R. J. Koczor for the nominal spectral
bands and a 0.12K NETD. We therefore require
a (Ch . 3) = 5.87 x 10 4 cm -2 = 3.79 x 10 5 in" 2
a (Ch. 4) = 1.74 x 10 7 cm" 2 = 1.12 x 10 8 in" 2
Attenuation coefficients for scattered sunlight were measured
as part of an earlier study on the HRIR (Final Technical Report, Solar
Sun Shield Study for the HRIR on TIROS-M Satellite, Feb. 6, 1967,
ITT-A/OD Report No. 14-16400). The HRIR had a 3.7 pm band channel and
was to operate in a 750 to 900 n mi sun-synchronous orbit over a sun
angle range from 3‘7.5° to 50°. The instrument had a modest external
shield and simple internal baffles with a Cassegrain telescope and a
reflective secondary optic. With the scan mirror looking 45° from
nadir on the sun side, attenuation coefficients of 0.6 x 10 8 in" 2 and
higher were measured, the value depending on the sun angle and orbital
position.
3.11.4 Sun Scatter Test Results
The BBM AVHRR was tested to determine the level of scattered
light getting into Channels 1, 2 and 4. The tests for Channels 1 and
2 were run tocether while that for Channel 4 required a modified set uc .
3-63
3.11.4.1 Channel 1 & 2 Test Setup
The BBM was placed on a 2-axis rotary table which rotated
the instrument about its pitch and yaw axes. A 9 inch aperture
Astrola telescope was used to illuminate the scan cavity. A 24 Hz
chopper and a tungsten iodide lamp (with a Lucalox diffuser) were
used as the radiation source at the Astrola focal plane.
Signals were taken out of the Channels 1 and 2 preamplifiers
(to reduce system noise) and fed to an electrical bandpass filter
(20 to 31 Hz) then to a Ballantine True RMS voltmeter. The BBM was
operated from the P.T.E.
3.11.4.2 Test Procedure
As discussed in Sectoon 3.11.3 the procedure is to calculate
an attenuation factor for the scattered light at various angles cf
irradiance. The detectors in the BBM were used for the test. They
were removed from the unit (mechanically not electrically) and
illuminated directly by the chopped collimator signal, thru the
appropriate spectral filter. They were then placed back in the unit
in their proper location to measure any scattered signal.
The collimator is used to irradiate the scan cavity at sun
angles (3 of 0° to 67° and orbit angles p s of 280° to 230°. Initial
readings were taken with the scan mirror positioned at nadir and ±45°
from nadir; however, no change was observed and so further testing was
done with a nadir-looking scan mirror only.
3.11.4.3 Test Results Channel 1 & 2
The data obtained in this phase of the test is given in
Table 3.11-1. The data is given as a function of sun angle then
orbit angle. The irradiance level signal measured at the scan cavity
was 6.40 volts in Channel 2 and 4.00 volts in Channel 1. As can be
3-64
Measured Irradiance Onto Scan Cavity
Channel 1 4.0 volts
Channel 2 6.4 volts
SUN
ORBIT
CHANNEL 1
SIGNAL,
CHANNEL 2 SIGNAL
'GLE
ANGLE
.58 mv
NADIR
-45°
+ 45
NADIR
-45°
67°
280°
.58 mv
.58 mv
.60 mv
. 53 mv
. 55 mv
. 53 mv
270
• 56
.58
.61
.52
.53
.54
260
.62
.63
.62
.51
.53
.53
250
.60
.60
.63
.52
.53
.55
240
.63
.60
.62
.54
.52
.57
230
.66
.63
. 66
.56
.57
.58
5 8°
280
.65
.60
.60
.55
.50
.52
270
.65
.65
.65
.50
.50
. 50
260
.58
.60
. 60
.53
. 46
.53
250
.56
.57
.57
.45
.48
. 52
240
.58
.62
.57
.51
.50
.50
230
.57
.57
.58
.55
.53
.53
50°
280
.62
.56
. 60
.52
.54
.62
270
.60
.61
. 59
. 54
.53
.52
260
.60
.60
.58
.52
.51
.51
250
.65
.63
.62
.54
.50
.53
240
.65
.62
.64
.55
.54
.54
230
.64
.67
.65
.63
. 62
.64
*1°
280
. 62
.55
270
.58
.52
260
.58
.52
250
. 60
.56
240
.62
.60
230
.63
.60
o
CN
280°
. 55 mv
.58 mv
270
.56
.59
260
.62
.58
250
.65
.58
240
.63
. 61
230
.63
.60
18°
280
.58
.55
270
.60
.57
260
.63
.57
250
.60
.60
240
.60
.60
230
.56
.56
9°
280
.63
.56
270
.61
.57
260
.61
.59
250
.62
.61
240
.60
.60
230
.61
.62
TABLE 3.11-1
CHANNELS 1 & 2 SCATTERED SIGNAL AND NOISE LEVEL
3-65
seen in Table 3.11-1, the measured signal and noise level Channel
1 was on the order of 0.65 m volts. This is a ratio of
4000
. 65
6150:1
Similarly in Channel 2 the ratio is
6400 = 11000:1
.58
There was no evidence of a measurable scattered light signal
in either channel at any of the measurement points. This was verified
by blocking off the source. The numbers given in the Table are only
the noise in the test setup.
Using the formula given earlier, we have
E , x .005 x 1.- x QA
S/TT /3
In the above case the coefficient gives the attenuation required for t
scatter to be equal to the specified noise equivalent albedo (1^ of 1
Using the coefficient we can calculate the required ratio of (colli
mator irradiance) to S 2 (scatt .r signal measured in system) as
s i/s 2 = A d x A
Where A^ is the test detector area. In our case AVHRR thruput is 5.43
10 '
cm 2 ster (including secondary obscuration) so that
a = 2. 2 4 x 10 7 in -2
The test detector has a 0.010" square aperture so that the ratio that
would be measured in the test set up if the scattered signal were
equivalent to the specified Noise Equivalent: Albedo is
S. . = (.01) 2 x 2.24 x 10 7 = 2240:1
i/b 2
3-66
In both channels our test ratio was at least 3x better than this.
Therefore, we conclude that if there is any scattered signal in
the orbits and at the angles measured (which are the computer-
projected "worst cases"), they are at least three times less than
the Noise Equivalent Albedo specified.
3.11.4.4 Channel 4 Test Set Up
The BBM AVHRR was in place on the two axis table with the
collimator in position as for Ch 1 and 2. However, an .027" square
active area InSb detector was placed at the focus of the IR Relay
lenses. A Channel 4 filter is set in place in front of the detector
so that only the spectral band of interest is viewed. This detector-
filter combination is placed at the scan cavity for incidence
measurement as above. The radiance source is a 900°C blackbodv sourc
chopped as before at 24 Hz.
The scan mirror was checked at several angles on various sun
positions, but no measurable effect was observed. All recorded data
is for the scan mirror at nadir.
The procedure was basically identical to that previously
described; however, since Channel 4 can be used at night, the orbital
angles which were checked were extended to a range of 230° to 310°.
Table 3.11-2 gives the measured results.
3.11.4.5 Test Results Channel 4
The maximum measured scattered signal plus noise was 85 micro
volts. With an incidence signal of .636 volts, we have a ratio of
ORBIT
ANGLE SUN ANGLE g
S S
Jl
18°
O
38°
o
00
o
IT
1
67
230°
20
20
15
15
20
15
25
240
20
20
15
15
25
20
30
250
20
15
15
15
15
30
25
260
15
25
20
20
25
20
30
270
25
30
20
25
45
50
85
280
15
25
25
20
25
35
30
290
20
20
30
15
20
20
15
300
20
15
15
15
30
25
25
310
25
15
15
30
20
25
20
Readings are in microvolts
MEASURED IRRADIANCE ONTO SCAN CAVITY IS .636 volt rms
1 NER = 24 microvolts
TABLE 3.11-2
CHANNEL 4 SCATTERED SIGNAL PLUS NOISE LEVELS
3-68
The required attenuation coefficient for Channel 4 has been
calculated to be 1.12 x 10 9 in - - for a scattered signal equal to 1/3
NER. The ratio S. /c for this channel is
1/ O ^
S. /c - (1.12 x 10 s ) (.027 in) 2 = 81650
1/ b ^
where .027" is the InSb detector width. This means that the maximum
scattered signal found is equivalent to
81650
3.6 NER
3 x 7480
SUMMARY
Measurements in the solar channels indicate that no measurable
scattering exists in these channels at solar incidence angles which
computer projections indicate are worst case conditions. In Channel 4,
the measured scattered signal is less than two NER's for orbits with sun
angles below about 48° and at all orbit angles except 270°. The maximum
measured any where is 3.6 NER's.
It is apparent that no sun shield is required for the solar
channels. Further because of the low level and limited extent of the
scattered signal in Channel 4, it is probable that no sun shield is
required in that channel. Further, assuming a reasonably sized shield
attached to the AVHRR, it is not possible to eliminate the scattered
signal completely since the irradiance is reaching the telescope throuoh
the shield area which would be cut out to allow an earth view at the
maximum scan angle.
3-69
4.0
MECHANICAL DESCRIPTION
4 . 1 Overall Instrument Configuration
The AVHRR instrument design provides for a basically
modular configuration. An exploded view of the instrument which
depicts the various modules is shown -in Figure 4.1-1. The basic
modules are :
1. Base plate with related cover plates
2. Scanner
3. Optics
4. Radiant Cooler
5. Electronics
An outline drawing of the assembled instrument is shown in ITT
Drawing No. 8008778 which is reproduced in Figure 4.1-2.
4.1.1 Structure
The baseplate can be seen in Figure 4.1-1. Provision is
made for locating the scanner and optics on the base plate by means
of dowel pins. The radiant cooler is positioned by means of shims
in order to align the Channel 3 and 4 detectors to the optical axis.
The electronics package attaches to the side of the instrument by
means of machine screws and dowel pins for rigidity.
The structural integrity of the instrument was proven
during the extensive vibration testing given both the Mechanical
Structural Model and the Engineering Model. The Engineering Model
successfully passed vibration tests per ITT Procedure No. 8120266
and acceleration tests per ITT No. 8120267 with the exception of
4-1
channel registration. The required registration stability was
achieved on the PFM after improved methods of securing certain
optical elements were incorporated into the design and pinning
and staking procedures were improved.
In order to prove the structural integrity of the radiant
cooler prior to vibration, a computer analysis of the configuration
was conducted by Computer Sciences Ccrp. and Butler Analyses, Inc.,
under NASA Contract NAS 5-24012 Mod. 10. Copies of the Final
Report of this analysis were reviewed and are on file.
4.1.2 Materials in Structure
The primary structural material is 6061-T6 Aluminum tooling
plate (Alcoa Type 200). The scanner housing is fabricated from
HP-20 grade Beryllium or equivalent. In some low stress areas
structural parts are fabricated from AZ31B magnesium.
The magnesium surfaces are finished with DOW-7, and the
aluminum is finished with Alodine 600. Where thermal control is
required, the surfaces are painted.
Beryllium is electroless nickle plated.
4 . 2 Scanner Subassembly
An assembly drawing of the scanner is shown in ITT Drawing
No. 8009201 which is reproduced in Figure 4.2-1.
4.2.1 Scan Motor
The 80 pole hysteresis synchronous scan motor was procured
from Schaeffer Magnetics, Inc., Chatsworth, California. The motor
is described in ITT Specification No. 8007929. Typical performance test
curves for the scanner motors are shown in Figures 4.2-2 through
4.2-5.
4-4
ID X to TO Tn: INCH 0703
I 1 li l»C“t * »»l« I» • 1 *. •
MUff tc ft IMIK CO.
ITsti /fjo r^^vooV-
/fjzyreK LOB£?\
Z-&- 7 Y \
: , j . _
140 150
160
180
200
220
CURRENT (MA)
Figure 4.2-3
TORQUE (IN. OZ)
Figure 4.2-4
4-8
140 150 170 200 220 240 260 280 300 320
CURRENT (MA)
Figure 4.2-5
4-9
4.2.2 Bearings
The motor shaft bearings are a set of DB duplex bearings
separated by 2 inch spacers. The bearings are specified in ITT
Drawing #8007937. Each bearing has a static radial non-brinell
load capacity of 602 lb. and a dynamic radial load capacity of
125 lb. for 17,500 hours of operation at 360 rpm.
An unusual feature of this bearing is the square ball
pockets in the retainer. It was determined during the SCMR program
that the shape of retainer ball pocket is a major factor influencir
jitter. Square ball pocket bearings exhibited less jitter than
round pockets. Based upon that, the AVHRR bearings were procured
with square ball pockets.
Tests have shown that bearings with improved surface finis
on the races (achieved both by diamond honing by the manufacturer
and running in at ITT) can perform within specification when
lubricated with the proper amount of lubricant. Tests have also
shown that performance is appreciably degraded when too much lubri-
cant is used.
4.2.3 Bearing Fits
Table 4.2-1 shows the shaft and housing fits for the
AVHRR.
4 . 2 . 3 . 1 Thermal Consideration of Bearing Fit
Bearing, shaft and housing materials and coefficients of
thermal expansion are noted in Table 4.2-2. The deviation from
nominal bearing fit over the temperature range, 0 to +40°C, is
considered negligible.
4-10
TABLE 4.2-2
THERMAL EXPANSION CHARACTERISTICS
OF BEARINGS
Coefficient of
Item
Material
Thermal Explosion
Bearing
440 C SS
5.6 (10)" 6
/°F
Shaft
Inconel X750
6.96 (10)' 6
/°F
Housing
Beryllium
6 . 4 (10) " 6
/°F
With all parts measuring nominal dimensions at
20°C, the total changes in fit at ±20°C from
nominal will be:
Change in Bearing to Housing Fit -4. 32(10)“ 5 IN +4.32(10) ~ 5 IN
Change in Bearing to Shaft Fit -6.13(10)" 5 IN +6.13 (10)" 5 IN
4-12
4. 2 . 4
Lubrication
The beairings of the scanner are lubricated with Krytox
143AB. This lubricant was selected after evaluating a number of
lubricants and upon the recommendation of GSFC.
A summary of the pertinent characteristics of KRYTOX
143A3 is shown in the following:
Weight loss % - 30 days at 50°c &10“ 6 mm Hg
Lubricity - cycles - A151550 Block
150 lb. @ 100 RPM
Viscosity 32F
Viscosity 77F
Viscosity 100F
Radiation Resistance - min. safe dosage
.023%
10 6 + cycles
140 CS
49 CS
36 CS
10 8 rads
The amount of lubricant in each bearing is critical for
achieving proper scanner performance. Nominally, 7 +1 mg of KRYTOX
-0
143AB is used in each bearing. To achieve this level, ITT
Procedure No. 8008007 is followed for cleaning and lubricating
each bearing.
4.2.5 Jitter
The current jitter spec calls for:
"scan line to scan line jitter as measured on the
leading edge of the synchronization pulse shall
be less than 1/2 of an IFOV for 98% of the data
points when data is taken every scan line for a
20 minute period.
4-13
"The jitter of the synchronization pulse between any
two scan lines within a 20 minute period shall be within an
I FOV (34 microseconds)."
Jitter character istics of the flight model instru-
ments were alL within the required specification. Details of
the tests can be found in the individual instrument test
reports .
4.2.6 Life Test
A nominal one year life test was run at ITT on the
Lite Test Model Scanner. The motor survived the life test as
indicated by its end-of-test performance characteristics .
LTM performance character ist ics are noted below.
Start-of-Test End-of-Tes t
Coast Down 1.21 min. 2.25 min.
Jitter-Line/Line 73.7% (within 3usec) 39.9% (within 8 s>
16 nsec)
Torque 3.3 in oz at Drop Out 4.0 in oz at Drop Out
4.2.7 Angular Momentum
Angular momentum of the ETM scanner is calculated to
be 35.8 in. oz sec based upon a measured moment of inertia of
.9502 in. oz sec-. Momentum vector direction is along y axis.
4-14
Moment of inertia was measured with a Model XR50 moment
of inertia tester manufactured by Space Electronics, Inc.
4.2.8 Venting of the Scanner Housing
Because of the limited torque margin of the scanner motor,
shaft seals are not used. A close clearance labrinth cap does
cover the shaft clearance through the housing for the purpose of
limiting the outgassing toward the scan mirror and optics. Details
of the cap are shown in Figure 4.2-8.
A vent hole is included in the motor housing to allow an
escape path for air during decompression to preclude large volumes
of air from rushing through the bearings and causing contamination.
Some concern has been expressed because open vent holes could allow
lubricant vapor to escape after the instrument is in orbit. While
this does not appear to be a problem when Krytox is used provision
has been made for a fitting to be attached to the vent hole as
requested by NASA/GSFC.
4 . 3 Radiant Cooler Subassembly
An exploded view of the radiant cooler is shown in Figure
4.3-1.
4.3.1 Support Body
The radiator is supported by nine support rods fabricated
of glass epoxy composite tubing (G10 Synthane) and stainless steel
inserts, or end caps:
-
pcs
. 25"
O.D.
X
.19"
I . D. x
1.31"
total
length
or
.98" thermal
ler.cth
2
pcs
.25"
O.D.
X
.19"
I.D. x
0.82"
total
length
or
.52" thermal
length
5
pcs
.38"
O.D.
X
.28"
I.D. x
3.40"
total
length
or
3.15" thermal
lenatl
4-15
The patch is supported by four rods of the same
materials as above:
.19" O.D. x .16" I.D. x 3.06" total length or 2.70" thermal lenath
4-18
4.3.2
Detector Location
The detectors for Channels 3 and 4 are located on the
patch. A layout of the detectors on the patch is shown in
Figure 4.3-2.
4.3.3 Deployable Earth Shield
The radiant cooler Earth shield is a "one-shot" device
which will be deployed by torsion springs located at the hinge
pins. A mechanical latch holds the shield in the closed (un-
deployed) position. Two rotary solenoids are utilized to release
the latch. Each is independent of the other and each capable of
unlatching the shield.
Switches located at both the open and closed positions
will indicate shield position.
A positive mechanical stop which is an integral part of
the radiator will stop the shield in the open position.
Hinges consist of music wire torsion springs rotating in
polyimide bushings. No lubricant is required.
The shield can be manually closed without the aid of
special tools.
In order to deploy the shield in a lg gravity field for
test, the torsion springs develop 162 in. oz of torque of which 40
in. oz are required to lift the weight of the shield and 2 in. oz
are required to overcome friction in the bearings. With torsion
springs designed to produce 162 in. oz torque, the torque margin in
orbit is greater than 80:1.
An analysis of the earth shield door motion was made and
reported in DIR #36.
4-19
(<**»*}
%
. *.y*
'■? s
- V
v
4-20
Figure
4.3.4
Materials and Finishes in Cooler
Generally, the major cooler components are fabricated
from 6061 T6 aluminum. Radiator surfaces are electroplated gold
and painted black with 3M401. The back of the patch and the
surfaces of the gold box are electroplated gold. The earth shield
is electroless nickle plated and polished optically. Then it is
coated with evaporated aluminum. Aluminized polyester film multi-
layer insulation is used inside the cooler with polyester mesh
separators .
4 . 4 Optics Subassembly
4.4.1 Optics Outline
The optics assembly is shown in ITT Drawing No. 8008030 -
Figure 4.4-1.
4.4.2 Materials and Finishes Used in Optics
Generally, 6061 Aluminum is used as the main structural
material. The corrugated tube in the telescope is invar. Graphite
filled polyimide is used as the separator material for lens elements.
4 . 5 Electronics Package
4.5.1 Electronics Package Layout
The Assembly Drawing No. 8009235 of the electronics package
is shown in Figure 4.5-1.
4.5.2 Accessibility
All electronics modules are accessible from the Earth side
of the instrument when it is mounted on the spacecraft with the
exception of Channels 1 and 2 preamps.
4.5.3 Thermal Considerations
Louvers on the spacecraft, platform are used for ther-
mal control. Heat sinks on each PC board conducts the heat
into the structure of the electronics package.
4.5.4 Radiation Considerations
Nominally 1/8 inch thick magnesium plates are used
for all external walls of the electronics package. This
thickness will provide adequate radiation shielding.
4.5.5 Materials and Finishes in Electronics
The entire electronics package is fabricated from
AZ31B magnesium with Dow 7 finish.
4 . 6 Weight Breakdown
A weight breakdown is shown in Table 4.6-1.
4 . 7 Materials
The Materials List has been issued as ITT Drawing No
8009466 .
HASS PROFILE
ITEM
WEIGHT
STRUCTURE
6,200
GMS
SCANNER
4,370
GMS
OPTICS
4,200
GMS
COOLER
2.575
GMS
ELECTRONICS
9,433
GMS
THERMAL BLANKET
425
GMS
TOTAL IN GRAMS (GMS)
27,198
EQUIVALENT IN PDS (LBS)
59
96
TABLE 4.6-1
5.0
ELECTRICAL SYSTEM
The basic function of the AVHRR electronics is to provide
outputs of four data channels in digital form and supporting
telemetry signals. Given inputs of power, a reference clock, and
commands, the AVHRR electronics provides the command storage, power
conversion and regulation, timing and control signal generation,
a signal amplification, and analog-to-digital conversion necessary
to perform its function.
A simplified block diagram of the AVHRR electronics is shown
in Figure 5-1. This diagram shows the basic functions and inter-
connections of the electronics.
5 . 1 Electronic Packaging
The majority of the AVHRR electronics are mounted on printed
circuit boards. The following is a list of the AVHRR boards:
1. Power Converter Assembly in a metal can
consisting of:
a. Power Converter and Switching Regulator
b. Logic Regulators
2.
± 15 Volt Regulators
3.
Command. Relay No.
1
4.
Command Relay No.
2
5.
Command Relay No.
3
6.
Scan Count and Decode Logics
7.
Motor Logics
8.
Auxiliary Scan
9.
Patch Temperature
Control and Telemetry
10.
Telemetry Board No. 2
11.
Interface Logics
No. 1
12.
Interface Logics
No. 2
13. IR Post Amplifier
14. Daylight Post Amplifier
15. Multiplexer
16. Ramp Calibration
17. Blackbody Mux
18. Motor Power Supply
19. Switching Regulator
20. Ch 3 IR Preamplifier
21. Ch 4 IR Preamplifier
22. Daylight Preamplifier
The first 17 cards listed are mounted as plug-in units in
the electronics assembly.
Items 13 and 19 are mounted in a metal can v.nich is located
between the electronics assembly and the scan motor. These two
cards are hard-wired inside the can. The entire can assembly is
a pluq-in unit with two connectors, one to the electronics assembly,
the second to the scan motor.
The Ch 3 IR Preamplifier (Item 20) is mounted in a metal can.
The assembly is mounted outboard of the electronics package.
The Ch 4 IR Preamplifier (Item 21) is located on the cooler
vacuum housing in the proximity of the cooler interface headers.
The preamplifier board is mounted in a metal can.
Two daylight preamplifiers- (Item 22) are located within
the optics package with the silicon detectors connected directly
to the PC board. The board is hard-wired within the optics module
and is provided with copper shieldinn.
5-3
Break connectors are provided between the electronics
package and the wiring required in the remainder of the instrument
making the electronics package in its entirety a plug-in unit.
T. final major assembly of the AVHRR electronics is the
analog-to-digital converter. This unit is located in the electronics
package with short connections to the post amplifiers and output
connectors .
5 . 2 Electrical Design Considerations
The electrical design is to the extent possible based upon
existing designs used on space qualified instruments. Parts
selection criteria include the use of parts on the NASA/GSFC
Preferred Parts List where possible and the use of previously
approved parts where the PPL parts are not suitable. Derating
of parts is in accordance with the requirements of the PPL and
consideration of the special environment of the AVHRR.
5 . 3 Video Scan Timing
Figure 5-2 shows the timing relationship of the scene as
viewed by the scan mirror and the sync pulse which is the time
reference point. Time intervals of 1 millisecond before and after
the sync pulse have been provided by insertion by the MIRP of
coded pre-cursor signals. The tolerance of the viewing times due
to spacecraft attitude control tolerance are indicated.
The times indicated on this drawing and times and frequencies
listed on all other timing diagrams and schematics are based on
a normal 1 MHz clock. To obtain actual times the numbers must be
multiplied by 1.0016. To obtain actual frequencies multiply
numbers of 0.9984.
5-4
VIDEO SCAN REPktSENTATION
i i i t i t i t j i i
IR VIDEO DATA
DAYLIGHT VIDEO DATA
I
[ i
|
i i i i" i i t ' \ i l
”1 l t
T
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
LINE SYNC SIGNAL
166.4
n r\ i
u-u.±
L1I1C 0 1 ilL
34.2
NADIR NOMINAL
0.5-1. 5
MIRP PRE-CURSOR TIME
62.6
SPACE START WORST CASE - EARLY S/C ATTITUDE
1.8
RADIOMETER SPACE VIEW START
63.1
SPACE START NOMINAL
1.9-3. 5
RADIOMETER SPACE SAMPLE
63.6
SPACE START WORST CASE - LATE
3. 5-4.0
RAMP CALIBRATION
65.6-65.8
IR TARGET TEMPERATURE
4.13
SPACE END WORST CASE - EARLY
65.8-66.0
117.1
PATCH TEMPERATURE
IR TARGET FULL VIEW START
5.3
SPACE END NOMINAL
117.6-118.4
IR TARGET SAMPLE
5.8
SPACE END WORST CASE - LATE
S/C ATTITUDE
119.0
165.0-166.0
IR TARGET FULL VIEW END
MIRP PRECURSOR TIME
5. 4 Power Subsystem
5.4.1 General
The +28V DC input from the spacecraft is regulated to
obtain isolation from spacecraft voltage variations, converted
to develop a system ground, and re-regulated to obtain noise
rejection and precision.
The schematic for the Power Converter and Switching Reg-
ulator Board is shown in Figure 5-3. The schematic for the
Logic Regulator is shown in Figure 5-4. Both boards are assembled
in a conetic can with LC EMI filters on all inputs and outputs
with the exception of the clock signals.
5.4.2 Turn on Transient
Turn on transients are suppressed by limiting the base
voltage rise time of the input pass transistor Q4 of the PC & SR.
The changing path is through T2 and R17. This circuit limits the
surge current into the input filtering. The input switching
regulators have long time constants on the reference inputs which
slows the build-up of all the circuits beyond that point.
5.4.3 Electronics Switching Regulator
Two input switching regulators are provided using Harris
HA2620 amplifiers as comparators: The circuits are self starting
and driven by a 62.4 Kh signal for synchronization with the space-
craft clock. A +34V boost voltage derived from the Power Converter
provides the drive voltage for the pass transistor to provide
good saturation for minimum dissipation.
5-6
5.4.4 Power Converter
Two DC/DC Converters are provided, one powered from each
Switching Regulator. These circuits establish signal ground
for the AVHRR and provide the proper voltage inputs to the
electronics regulators. They are driven converters being
synchronized with the same 62.4 Kh clock signal as the switching
regulators. The converters operate from switching regulator
outputs and therefore the maximum VCE on the converter transistors
is limited to 44V.
5.4.5 +5V Regulators
A switching regulator is used to provide +5 volts for logic
circuits. The output of the switching regulator directly feeds
the logic circuits required for motor frequency countdown and the
input clock circuits. The major portion of the logics is powered
through a switching transistor from the regulator output. This
transistor is turned on with the Electronics ON command.
5.4.6 t 15V Regulators
The 15 volt regulators are linear circuits utilizing the
Harris HA2620 I.C. as the voltage comparator. The schematic
for the ± 15V Regulators is shown in Figure 5-5.
Also on this board are the pass transistors for the channel
enable/disable commands and Electronics ON command.
The +15 volt regulator uses a +20 volt "boost" voltage from
the power converter to enable a low input/output differential
and minimize the pass transistor power loss. The circuit will
regulate under full load with an input/output differential equal
to the collector/emitter saturation voltage of the pass transistor
which is typically 0.2 volt.
5-9
5.4.7 Motor Power Supply Switching Regulator
The schematic for the motor power supply switching
regulator is shewn in Figure 5-6.
The circuitry is similar to the electronics switching
regulator, the exception being that the motor switching regulator
will regulate at one of two voltage levels selected upon command.
5 . 5 Commands and Digital TM
The following is a list of the presently planned commands
for the
AVHRR:
1 .
Scan Motor/Telemetry ON
2.
Scan Motor/Telemetry OFF
3 .
Electronics/Telemetry ON
4.
Electronics/Telemetry OFF
5.
Ch 1 Enable
6.
Ch 1 Disable
7.
Ch 2 Enable
8.
Ch 2 Disable
9.
Ch 3 Enable
10.
Ch 3 Disable
11.
Ch 4 Enable
12.
Ch 4 Disable
13.
Voltage Calibrate ON
14.
Voltage Calibrate OFF
15.
Patch Control High Mode
16.
Patch Control Low Mode
5-11
17.
Cooler Heat ON
18.
Cooler Heat OFF
19.
Scan Motor High Mode
20.
Scan Motor Low Mode
21.
Telemetry Locked ON
22.
Telemetry Unlocked
23.
Earth Shield Deploy
24.
Earth Shield Disable
25.
Patch Control ON
26.
Patch Control OFF
The
following is a list
executing
the listed commands.
1
Command No.
1. Scan Motor/Telemetry ON
2. Electronics/Telemetry
the functions performed by
Function
Applies Power to:
1 .
Electronics sw.
regulator
2.
Motor sw. regulator
3.
Power converter
4.
+15V regulators
5.
+5V logic reg.
6.
Clock receiver
7.
Motor logic
8.
Analog telemetry
circuits
9.
Patch temp, control
Applies Power to:
1 .
Electronics sw.
regulator
2 .
Power converter
3.
±15V regulators
Electronics/Telemetry - Continued
4. +5V logic regulator
5. +5V electronics circuits
6. Analog telemetry circuits
7 . A/D Converter
8. Scan timing logic
9. Clock receiver
10. Motor logic
11. Patch temp, control
5.
Ch
Enable
If
"Electronics ON" has been
executed - Applies Power to:
1 .
Ch 1 preamplifier
2.
Ch 1 Post Amplifier
7 .
Ch
2
Enable
If
"Electronics ON" has been
executed - Applies Power to:
1 .
Ch 2 Preamplifier
2 .
Ch 2 Post Amplifier
9.
Ch
3
Enable
If
"Electronics ON" has been
executed - Applies Power to:
1 .
Ch 3 Preamplifier
2.
Ch 3 Post Amplifier
11.
Ch
4
Enable
If
"Electronics ON" has been
executed - Applies Power to:
1. Ch 4 Preamplifier
2. Ch 4 Post Amplifier
13. Voltage Calibrate If "Electronics ON"
has been executed -
1. Deactivate IR & Daylight
detectors .
13. Voltage Calibrate (Cont'd)
15. Patch Control High Mode
16 . Patch Control. Low Mode
17. Cooler Heat ON
19. Scan Motor High Mode
20. Scan Motor Low Mode
21. Telemetry Locked ON
2. Provides simulated earth
scene and backscan video.
If "Telemetry ON" has been
executed
1. Sets patch temp control
point to 107°K
If "Telemetry ON" has been
executed
1. Sets patch temp control point
to 105°K
If "Electronics ON" , "Motor ON"
or "Telemetry ON" has been
executed - Applied Power to:
1. Radiator Decontamination Heater
2. Patch Decontamination Heater
If "Motor ON" has been executed -
1. Sets motor sw. regulator
voltage to HIGH LEVEL
If "Motor ON" has been executed -
1. Sets motor sw. regulator voltage
to LOW LEVEL
Applies Power to:
1. Electronics sw. regulator
2. Power converter
3. +15V regulators
4. +5V logic regulator
5. Clock Receiver
6 . Motor Logic
5-15
21. Telemetry Locked ON (Cont'd)
7.
Analog Telemetry Circuits
8.
Patch control circuitry
23. Earth
Shield Deploy
Applies Power to:
1.
Earth Shield Circuitry
24. Earth
Shield Disable
Removes Power from:
1.
Earth Shield Circuitry
25. Patch
Control ON
If '
'Telemetry ON" has been executed
1.
Applies Controlled Heat to Patch
Digital T/M
The
following is a list
of the
Digital T/M functions:
1.
Scan Motor Status
It It
ON
"0"
OFF
2.
Electronics Status
It ^ *»
ON
"0"
OFF
3.
Ch 1 Status
ii ^ >'
ON
"0"
OFF
4.
Ch 2 Status
ii ^ it
ON
"0"
OFF
5.
Ch 3 Status
11 2^ It
ON
"0"
OFF
6.
Ch 4 Status
il 2^ It
ON
"0"
OFF
7.
Voltage Calibrate
it ^ »»
ON
Status
"0"
OFF
8.
Patch Control Mode
ii ^ •*
107°K
"0"
105°K
9.
Cooler Heat Status
11 ^ 11
ON
"0"
OFF
10.
Scan Motor Mode
1
"0"
LOW POWER
5-16
11.
Telemetry status
It ^ It
ON
"0 H
OFF
12.
Earth Shield Status
it ^ »»
DEPLOY
"O"
DISABLE
13.
Patch Control
tt ^ it
ON
"0"
OFF
The schematics for Command Relay #1, 2 and 3 are shown in
Figures 5-7, 5-8 and 5-9. Command Relay's are P&B HL11D 12VDC
Latching Type. For Commands and Digital T/M "1" true or 0.0V Level
= False or +5.0V Level.
5 . 6 Analog TM and Patch Control
5.6.1 Analog Telemetry
The following is a list of the presently planned analog
telemetry points.
1. Patch Temperature
2. Fateh Temperature Extended
3 . Patch Power
4 . Radiator Temperature
5. Blackbody No. 1 Temperature
6. Blackbody No. 2 Temperature
7. Blackbody No. 3 Temperature
8. Blackbody No. 4 Temperature
9. Electronics Current
10. Motor Current
11. Earth Shield Position
12. Electronics Temperature
13. Cooler Housing Temperature
14. Baseplate Temperature
15. Motor Housing Temperature
16. A/D Converter Temperature
17. Detector No. 3 Bias Voltage
18 . Blackbody Temperature IR Ch 3
19. Blackbody Temperature IR Ch 4
20. Reference Voltage
The circuitry for telemetry points 1-16 is located on the Patch
Temperature Control and Telemetry Board shown schematically in
Figure 5-10 and the Telemetry #2 Board shown schematically in
Figure 5-11.
The following is a list of the range and resolution of the
analog telemetry points.
Range Resolution
1.
Patch Temperature
+0.2V
+ 5.0V
=
90.7 K
115. 5 K
99.4 K
316.0 K
0 . 1935V/Degree
2.
Patch Temperature
Extended
+0. 2V
+5.0V
0 . 02216V/Degree
3.
Patch Power
+ 0.2V
+5.0V
:
0.08 MW
50.0 MW
148. 7°K
317. 3°K
0.067V/mm at
32 mw out.
4.
Radiator Temperature
+0. 2V
+ 5. OV
_
0 . 0285V/Degree
5,
6,7,8. Blackbody Temp
1,2,3 & 4
+0.2V
+ 5. 0V
—
5 . 086°C
44.974 C
0 . 1203V/Degree
9.
Electronics Current
+ 0.2V
+5.0V
=
39.3 MA
982.5 MA
5. 088MV/Milliamp.
10
. Motor Current
+ 0.2V
+5.0V
_
12 MA
300 MA
16.6 MV/Milliamp
11
. Earth Shield Position
3 Levels
+5.0V Open
+3.0V In Between
+1.0V Closed
12
. Electronics Temperature
+0.2V
+ 5.0V
38.73
10.8 C
33 . 24°C
3 . 95°C
0.17173 5V/°C
13
, 14, 15. Cooler Housing
Temperature
+0.2V
+ 5.0V
0 . 129V/°C
Baseplate Temperature
Motor Housing
Temperature
5-21
16 .
A/D Converter
Temperature
17.
Detector #3 Bias Volt.
H*
CO
•
Blackbody Temp. IR
Ch 3
19.
Blackbody Temp . IR
Ch 4
20.
Reference Voltage
+0.2V = 84 . 49°C 0 . 12V/°C
+5.0V = 44 . 51°C
1.923 = -13V 0.23/V/V
2.384 = -11V
To be Calibrated*
To be Calibrated*
+0.2V = 0.266V 0.75 OV/V
+5.0V = 6.657V
*Ra,nge and value of calibration dependent on individual instrument
calibration. Equations derived from EM matched to Calibration
from 0°C to 40°C are
Ch 3 - °C - 1.1V 2 - 13.2 V + 330.2
Ch 4 - °C = ~ 3V 2 + 11V + 302.3
2000 Ohm ice Point Platinum sensors are used for the
following telemetry.
1 . Patch Temperature
2. Radiator Temperature
3. BB No. 1
4. BB No. 2
5. BB No. 3
6. BB No. 4
30K ohm Y.S.I. Thermistors, 0.1°C interchangeability, are
used for the following telemetry.
1. Baseplate Temperature
2. A/D Converter Temperature
3 . Electronics Temperature
4. Cooler Hsg. Temperature
5. Motor Hsg. Temperature.
Diodes have been provided on all operational amplifier outputs
to limit analog telemetry points voltage range to -0.7V to -*-6.0.
5.6.2 Patch Temperature Control
The circuitry for temp, control is shown on the Patch Temp
Control and Telemetry schematic, Figure 5-10.
It is a proportional control with a DC output (steady state)
to eliminate transients.
The temperature sensor is a platinum resistance unit, Rose-
mount Model. No. 146MA with an. ice point resistance of 2000 ohms.
This will provide the input amplifier with a low impedance, stable,
and linear source with a resistance change of approximately 8.6 ohms/
degree .
The heating unit on the. patch is a high rel. RNR55 2000 ohm
resistor.
The operational amplifiers are Harris devices, HA-2700. They
wove selected for the low offset voltage, typically 0.5 mv and
offset voltage as a function of temperature typ. 0.25 mv from 0
to 50°C.
5 . 6 . 2 . 1 Theory and Operation
The R3, R4 resistors are used to zero the 1st stage amplifier
offset voltage at +25°C.
The R6 , R7, R8 , combination is tailored to match the platinum
sensor resistance at 105 a.nd 107 Kelvin. R-2 limits the sensor
dissipation to the values in the following table and establishes the
input voltage change to the 1st stage amplifier at 12.5 mv/degree.
Sensor Dissipation/mw
95~ 1.31
Sensor Power Dissipation
-o
100
103°
110
323
1.41
1.50
1.59
3.15
5-25
R-5 is selected for a first stage gain of approximately 70 and the
amplifier output is 0.0 volts at the selected control point.
With the 0.0 volt inverting input on the second stage and
0.134 volt established by Rll, R12 on the non-inverting input,
the second stage amplifier output = (0.34 x 67.6) =9.07 VDC. With
a 0.6 volt drop across CRl the patch control output is 8.4 volts
at the selected temp, control point. With the 2. OK heating unit
on the patch, the patch control circuitry provides 35 milliwatts
at 105.0 or 107.0 degree Kelvin.
Figure 5-12 indicates the deviation in temperature of the
control point with variations in control heat requirements (35 mw
nominal) .
5 . 6 . 2 . 2 Circuit Stability
The resistor types used in the patch control are Vishay
S202 high rel units with shelf life and load life data available.
Calculations from these data indicate that the resistor stability
is a negligible factor.
The operational amplifier stability can be related to a milli-
volt/degree factor at the first amplifier input. The typical initial
offset voltage for the HA2700 is 0.5 millivolts which is compensated
for at board test. With the loop gain indicated on the schematic,
a long term offset voltage change of 1.25 millivolts would be
required to result in an error in temperature control of 0.1 degree.
The relative location of the detectors, temperature sensor and
heater is shown on ITT Dwg. #8008799 Detector/Frame Assembly,
Channel 3 and 4 Figure 5-13.
5-26
Figure 5-12
NOT t :
l. ASSEMBLE PER aOOT^en INSTRUCTIONS.
\i.^ fti'f / Pf m> or itsms eo to ot DETimwmtD by omc*
ALIGNMENT REQUIREMENT*. ITEM & TO BE CHOSEN FROM
e006T‘»ft-l.-*i <?. AND iTEVk «} FROM ftOOBTB B - 1. 4 C <*.
T6 iQCAT £ THE DETECTORS AT THE NOMINAL. POSITION,
.0»S THICKNESS IS REQUIRED FOR EACH CHANNEL. AFTER
ORT'CS ALIGNMENT. REPLACE UtM t. WITH ITEM SO AND
KLPlACE liEH S WilU ITEM M PER frSSl INSTRUCTIONS.
|S • AFTER OPTICS A» iCwMtHT , TAP *0-80 Tm*o ITEM <*
LOCATED F K OtA fcJUSTvNQ HOLE Ut ITEM »Z.
[i. -!^Yi&i 3
,v' '/
TUvi
*> TAKE TO FRAME (zTA
4 T 0 ‘^CME USIN <4 T
iUm 2 I Ai ft&ftO
U • QTT < PART NO. Of ITEMS lO 4 II TO BE DETERMINED l_
BY OPTICS ALIGNMENT REQUIREMENTS. ITEM 10 TO BE Jl
CHOSEN FROM BOOSTA^ -I ThRU- 4 AUD ITEM It J“
FROM 6Q0ai44 - l THRU - 4 . SPACERS MAT BE
MACHINED TO REQUIRED TWICKNESS- «OOe>‘T4 ?y 2/.OT T HA
AND 600BT44 - 5 {.oa THK^i are REQUIRED TO LOCATE
detectors AT THEIR NOMINAL POSITION.
[h AFTER OPTICS AUQNM6 NT, MACHINE A PILOT HOLE THRU
ITEM 11 AHO .14 MAX. DEEP INTq ITEM *. REAM TO .OCZO/
.OfciS OlA THRU ITU* 11 kMO.Cn Mm DEEP INTO HEM *) .
.-• after optics Alignment, machine, item 12 to
THIS MEASURED OtMENSlON.
[7 APPLY ADHESIVE TO SCREW HEADS ONLV, *
not to the threads.
^ a (splice INSULATSOU)
£y STAAE TO FRAME
USIN4 ITEM II
HtVUlONt Cl 4 O'
LTwJ HfCWfTWN
a ktHO 'TIM! n . n 16 17 ; nt * ia « a*
n fcO OC 4 ^«|| RvNi'HO ntPk »* HCNW.AllT *
A.DOtO ntMv iT PK.-loRlkLL'l * ADDED «!*'%
__ fXAOKAHS- CiAao-^1»
R nijrlTV PT >*Q "&uaVs» - » • HEM 10
_ P7 u p WAS RhSSQZOO iF, C<M feHup. »*n ,
£ M>OCD -5 TO ft H<J ft, -o -to HiH S; > tK.
u >nas n»*o -{►; Hereto mil* 1 4 4 r«*
KSCHI; Hw ftEQio-OO
Q AODCO tTfcM »oaijiUM6i1M«t>TiHi
w <-0010 LM.T 1 fc>* • L**Cl TO MOU 1
__ CCH M *OQ-fc 2 * /»*Q 1 q-|B( 4 >
C PL *<A* OK1HH t>W4k‘, OEa-ETEO ST C»*CL%
1\ A1 **wte*6' r «-'*' »*■> AhO A»»CO
I ijCH (.VOI O -TII
v!tt
1
IW™*]
tUT*
aaror
ccs
«arcr
S-II-1W
RiijKg
v'tvnT
V ‘f ?7 l ilWe T ° * iTti^O
\^y U 4 INO (TfcM 11 k^HlOO
j 10.00 MiN. LRTH
' leads PROM
Otic CTO*. V
^1
, 16 ;
[?->(»')
/, /
, h'v
\L 6 »)U-
-COMhCO SET EC To* LEADS
TO THC TERMtHAE BOARD
( STAKE TO MT MS<A-
u»iME new H
r @ .
!i - 'Of ‘-/
(?) [j^(»;Wi 4 i
■ CONNECT DETECTOR LEADS
to the tervmnal board
4 stake to det. mee.
USIN4 ITEM It
, , , M ■,
Figure 5-13
StE SEPARATE PARTS UST
T*M H PAWT on homknclatumc O*
wo. g lOCKTirYIMO WO MtCHirnOH
H T "t J ? 7 ' ! i'
eooBECj 4 ,lo 1 o
LOOSaoT "CH 4 QO
NBXT AIMMM.Y UMDOH
VJlRiNS DIAGRAM
■llctrr A* MA» •« OTMIWIM nKWMS ■' COMTVACT TM<M UUinMI
amo »f(cific*TioAi am TtnL NMin or i\i *. uvkkkI
m iwuk m rnnCrcoHriMMi mo *h*h mi m Mfnooucm o« i
cow«e on ukou tm« ma>« row tx« manu»actvim on «a ut on i
g 1 H 1 g? 1 ?* 1 °» 1 °g- l. ■■ IT*M H r*KT OH
QUAKTrrr fW owour wo g| iocwtipyiwo wo |
UlTOf MATERIAL* OW PAITTR L»«T
UNLIM OTHK AWUMt APtgrilO COffTPACT WO, Tf|yp /
DIMIMAIOM* ARKINIWCHESAM MAC A - Zl^QO .1-1-1. f
IHCLUOK CNIMICAU.Y AATUIO M ^ ^ CIlULI ||U||I|WI<
OR KEATKO FIHItHti APPROVALS 1 '■■
TOLERANCE! dhawn ~(£u£i — LUy DEL~FE.C.”T(
«IZc PCOtWAljJ ZiMXXXB ' j'ri.C i _S..LT II A 1 N l r
HMIMKIM * *_ AWOLM kmoH U l~\ fMM l N t
HAC* Kt-ACK ‘ “ ‘ 1 ”
•JWP€P ■ A.Ol A.OIO — —
*»4,WCL ^Oj^oic
Ovt* *4 | * . 0*1 A.oooj J^yj /r r' 1 - -A 1
DQ 1 WOT SC ALA FWIK *- ( ' f J 7 "
COW.. T O L TO STOCK I
•hot PAACTtcc. ss.ioi
rrmi AEHosPACtiomcM orw
■1 1 ,1. PORT WAYNE INDIANA, U V«.
miimm ii m im n i>» hii ihm titHtiiw
DETECTOR/ FRAME ASSY
CHANNEL 3 4 A
KVHRE.
WM COM > - T nO
C 31510 800&733
wn
MALI i
1
2
I
SJB 008793 Je [^1
The sensor and heater is located as close as possible to
the detectors to minimize sensor/detector/heater gradients and
provide the input circuitry with feedback which is a monitor of
detector temperature as opposed to the patch temperature.
5.7 Motor Logics
The clock receiver arrangement utilizing the Texas Instruments,
Inc. SN55107 clock receiver, as specified in the GFSC specification
for cin Advanced Very High Resolution Radiometer, is located on the
motor logics printed circuit board. An output of the clock receiver
is available on a connector pin for use in the interface logics.
The motor countdown logics convert the 0.9984 MHz clock
receiver signal into 62.4 KHz for use in the switching regulators
and 240 Hz 20 for use in the motor power supply.
The 0.9984 MHz input signal is first divided by 16 in a
ripple counter to achieve the 62.4 KHz signal at pin 9 of A7.
This signal is then buffered and is available for use in the
switching regulators.
To obtain the 240 Hz 20 signal, the 62.4 KHz signal is
divided in a series of counters. The first is a divide by thirteen
counter which produces a 4.8 KHz signal at pin 9 of A4 . This
signail is then divided by 10 to obtain a 480 Hz signal at pin 9
of Al. The inverted and non-inverted forms of this signal are
further divided by 2 in their respective J-K flip flops, and the
240 Hz 10 signal is present at pins 9 and 12 of Bl. A reset
circuit acts as a monostable and supplies a pulse to the clear
5-29
input of one of the J-K flip flops, B1 pin 2. The reset pulse
insures the proper 90° phase relationship between the 20 240 Hz
output.
A portion of the scan count logics is also included in the
motor logics circuit. The time base for scan counting is the
spacecraft 0.9984 MHz clock. A divide by 100 counter provides
a 100 ysecond (10 KHz) time base for the remainder of the counter
and decoding.
The counter is designed to count continuously and be reset
by the pick-up sync pulse. The reset logics for the entire scan
count circuit are on this circuit.
The final circuit on this board is the sync pick-up circuit.
There are two magnetic pick-ups in the motor housing which provide
the inputs to the sync pick-up circuit.
The HA-2700 operational amplifiers act as level detectors
and provide logic level pulses to the logic circuitry. The circuit
is designed for redundant operation such that if either pick-up
fails, a sync pick-up pulse will still be present at the output.
Drawing No. 8008045 is the schematic diagram of the motor
logics (Figure 5-14) . The motor logics timing diagram is shown
in Figure 5-15.
5 . 8 Scan Count and Decode
The scan count and decode board provides the basic count-
down of the scan period and the decoding of gating signals for
the AVHRR electronics. The schematic diagram is given on Drawing
8008049 (Figure 5-16) .
5-30
31550 \&0CS0J5
5-32
Figure 5-15 Motor Logic Circuit - Timing Diagram
A 10 KHz clock from the motor logics is the time base
for the counter. The first six flip flops are a conventional
binary ripple counter. The last four counting stages are a
divide by 13 ripple counter. The counter is designed to repeat
the scan period automatically but is normally reset by magnetic
pick-up initiated reset pulse.
The method of decoding is setting flip-flops at the
beginning and end of the desired time interval and decoding the
set condition of the first and the reset condition of the second.
The clock input is always the highest frequency of the decoded
time, which arrangement when used with a ripple counter, provides
unambiguous decoding.
The timing of the normal operation signals is shown in
Figure 5-17. Timing for the voltage calibration signals is
on Figure 5-18.
t
5. 9 Output Data Control
The logics for controlling the conversion of Analog to
Digital Data and the transmission to the MIRP is shown on Schematic-
Interface Logic No. 1 - 8009206 Figure 5-19 and Schematic Inter-
face Logic No. 2 - 8009209 Figure 5-20. The timing for the output
Data Control is shown in Figure 5-21.
By specification, the minimum timebetween samples from the
MIRP can be 25 micro-seconds. With minimum timing, the A to D
converter operates in the hold or convert mode for 16 microseconds
and in the track or sample mode for 9 microseconds. For longer
5-34
■ 1 • i ■ i '
■ 1 '
0 1.0 2.0 3.0 4.0 5.0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 166.4
0. 0-0.1 SYNC PULSE
1.9-3. 5 AMPLIFIER SPACE REFERENCE SAMPLE
3. 5-4.0 RAMP CALIBRATION
65.6- 65.8 IR TARGET TEMPERATURE
65.8-66.0 PATCH TEMPERATURE
117.6- 118.4 BLACK BODY TEMP - IR SAMPLE
LH
I
U)
cn
FIGURE 5-17 SCAN TIMING SIGNALS
i
r — ^ i i ill
0 10 20 30 40 50
’ i
l
I
t 1 I ’ i T l ’ i 5 i • *7 T i 1 i 1 i i i i
60 70 80 90 100 110 120 130 140 150 160 170
166.4
4.0-65.6 SIMULATED EARTH SCENE
117.6-118.4 SIMULATED BACK SCAN TARGET
Ln
I
u>
cr»
FIGURE 5-18 SCAN TIMING SIMULATED CALIBRATION
SPACECRAFT DATA PROCESSOR/AVHRR INTERFACE TIMING
CLOCK PERIOD = t = 1.001602564 x 10 SEC
clock jiRjinjuTjm^^ uiiumRrumJi^^
SAMPLE
PULSES
DATA
OUTPUT
10 BIT ~
PARALLEL
SAMPLE PERIOD
- 25t to 30t min.
10 millisec. max.
__rh.lL
5t J I
±300ns 1
y~ )
lOt ± 300ns-l^1
15t ± 300ns —
20t ± 300ns
r» —
r — .
St to lOt nlin
10 millisec 'max
CH 3 ^ CII 1
25t ± 300ns
CH 2 | CH 4 IcH 3
SAMPLE N-l — Urn
OUTPUT DATA
-M
I
-ijl N » 1
5t ! !
i300ns , I
1 Cl
-lOt ± 300ns-^-j
|5t to lOtnlin |
I To millise^ Imax I
I
CH 3 i CH 1 |
— SAMPLE N
OUTPUT DATA
SAMPLE
' OUTPUT
u»
I
u»
vx>
N+l
DATA
Figure 5-21
V intervals between sample pulses, the track time is increased.
The sample signal from the MIRP determines the transition from
track to hold.
The A to D converter operates on a basic 3 microsecond
conversion time. The time allowed for Channel 5 conversion
is increased to 4 microseconds to minimize the data storage
and gating circuits required.
The A to D converter provides for six input data channels.
The target five channels correspond to the five channels of the
specification. The sixth channel has multiplexed telemetry data
and is switched in under logic control in place of channels 3,
4, and 5 of the appropriate time in the line scan.
Three microseconds (4 microseconds for channel 5) after a
channel is selected a strobe signal is sent to the A to D which
stores the digitized data in the A to D internal register. The
data of this point is in Grey code. Within 0.5 microseconds
the Grey to binary conversion is completed and valid binary data
is present on the A to D output lines.
The data is later strobed into an intermediate storage
register and in sequence to the output register whose outputs
directly connect to the output drivers for the MIRP lines.
The following chart gives the timing data for the sequence
in microseconds with respect to the MIRP select pulse.
5-40
CH 1
CH 2
CH 4
CH 3 • »
CH 5 ** '
Analog to Digital
Conversion
0-3
3-6
6-9
9-12
12-16
Data in A to D
internal register
(Grey Code)
3-6
6-9
9-12
12-16
16-3*
A to D output
(binary)
3.5-6
6.5-0
9.5-12
12.5-16
16.5-3*
Data in Inter-
mediate register
( Reg . 1 )
4-7
7-10
10-15
15-20
20-4*
Data in output
5-10
10-15
15-20
20-25
25-5*
register
(MIRP Signals)
Time of next conversion cycle.
Channel 5 input is connected to the Output Data
Amplifier of Channel 3 .
5-41
5.10 Ramp Calibration Generator
The ramp calibration and simulated calibration functions
have been combined to have the same digital to analog converter
be the signal source for both and .have the signal control and
switching only a logic function. Drawing No. 8007963 sheets 1
and 2 is a schematic diagram of the ramp calibration generator
(Figure 5-22) .
The counter for the ramp calibration is a conventional 10-bit
ripple counter. The outputs are gated to the D to A converter.
The system is connected such that each successive scan wi3 1 give
a one-step voltage change in the direction of increasing radiance.
This means that the signal will be increasing in voltage in
channels 1 and 2 and be decreasing in voltage in channels 3 and 4.
Since ramp calibration signals of both polarities with
respect to zero radiance are required, a digital offset (Binary
Number 43) is maintained at all times other than ramp calibration
time. During space look, the data amplifiers will re-zero on
thio offset allowing ramp calibration signals of both polarities
from zero radiance. The following tables will, illustrate this
further.
Since the zero radiance voltages were specified at different
intervals from the limits of the ramp calibration voltage (+0.25
volts and -0.1 volt) and +6.2 volts and +6.4 volts) the difference
was split maintaining the 6.5 volt ramp calibration range giving
a 0.275 volt excursion below minimum radiance. This gives a binary
value of 1023/6.5 volts x 0.275 volt = 43 bits.
5-42
80073&3
•• The earth scene during simulated calibration is derived
from a divide-by-3 counter. The three states are decoded and
gated through the two most significant bits giving signals of 1.626,
3.253,, and 4.879 volts above minimum radiance.
A test connection has been added to allow three known values
of ramp calibration signal to be sequentially gated out when
calibrating. This signal must be matched to the amplifier gain
which is a function of detector sensitivity.
The clock signal for both counters is derived from the
scan timing logics.
The D/A Converter is a 10-bit current switching type
converter. The output of the converter is a positive going 0
to 10.23 volt ramp consisting of 1023 steps. The output of the
D/A converter is sent to a unity gain inverting amplifier located
on the multiplexer board. These two outputs provide both polarity
ramp signal for insertion into the data channels as required by
circuit configuration.
The ramp calibration generator also provides ± 6.2 volt
outputs for use as a stable reference voltage.
VOLTAGE CHANNELS 1 and 2
+6.475
6.39375
6.1
0.25
0
-0.025
Channel Ramp Calibration Limit - Binary 1023
A/D Converter Signal Limit
Maximum Scene Radiance
Zero Radiance - Binary 43
A/D Converter Signal Limit
Channel Ramp calibration Limit - Binary 0
5-45
CHANNELS 3 and 4
+6.475
Channel Ramp Calibration Limit -
Binary 0
6.39375
A/D Converter Signal Limit
6.2
Zero Radiance - Binary 43
0.3
Maximum Scene Radiance
0
A/D Converter Signal Limit
-0.025
Channel Ramp Calibration Limit -
Binary 1023
5.11
Auxiliary Scan Timing
The auxiliary scan timing circuitry is added to provide
continued operation of the AVHRR in the event of the loss of
the pick-up signal which synchronizes the scan timing with the
scan mirror position.
A block diagram of these circuits is given in Figure
5-23. The complete schematic is shown on Drawing 8008052 (Figure
5-24) . The timing for auxiliary scan operation is shown in Figure
5-25. Timing signals are generated on the Scan Count and Decode
Board.
The circuit operation is based on the fact that the minimum
output signal on Channel 3 occurs when the radiometer views space.
With the loss of the synchronizing pulse, the scan counter period
is the same as the mirror rotational time but not in synchronism.
Two samples of the output voltage are taken at time the scan counter
timing indicates that the instrument is viewing space. From this
point two different modes of operation ensue. If in the remainder
of the scan a signal level lower than the average of the two samples
is detected, the scan counter is reset. When the cold target
5-46
2 P£>
/-/£ J.J3
^PMP? 2
A/V&
A/ox a
j£> 05 /T'/Ort
xo<s/2s
&/
tts&r
5C/)//
C&VA/r^/?
Figure 5-23
poj/r/o//
1.06/CS
*z
, /*£>& 0A/P 4o#H.
00£JAfT t cWJ.s£
/M///0/7- 0A/&
40/&/Z
coo/vr polsp
BLOC/? D /AGP AM ,
AUX/L/APY SCAM T/M/A/G
-49
n nr
' I ‘ I • ! \ ' !
0 1 2 3 4 5
j 11 , , i — j — i — j — i — i — i- • i • i •* i • i • i • i ! i • i • i ' l ’ • 1 >
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
0 - 0.1
SYNC PULSE
1.8-1. 9
FIRST SPACE SAMPLE
3. 4-3.5
SECOND SPACE SAMPLE
51.2-76.8
TRAILING ED6E SPACE WINDOW
160.0-12.9
SPACE WINDOW
FIGURE 5-25 AUXILIARY SCAN TIMING
U»
which causes the reset is space, the reset places the counter
within a few milliseconds of the correct point. This operation
is performed only every 128th scan to allow the data amplifier
to settle with the new reference. When the scan counter is near
synchronism, a logic gate inhibits this circuit while the mirror
is viewing space on the trailing edge of the earth view.
For finer positioning, the relative amplitudes of the
two samples are compared. An up-down counter sums the net number
of differences over sixteen scans. If a sufficient unbalance
exists, the logic will cause a 40 KHz clock signal to be added
or omitted, depending on the direction of unbalance. The counter
limits are unbalanced in an attempt to cause the samples to go
to the early portion of the space view as in the normal scan cycle.
5.12 Ch 3 Data Amplifier
The schematics depicting the Channel 3 amplifier chain
are shown on Drawing 8008101 (Figure 5-265 and Drawing 8008078
(Figure 5-27) .
The preamplifier has a positive and negative voltage
regulator to provide additional regulation for the detector bias
and to provide power supply isolation from the remaining electronics
for these low level signals.
The input stage of the preamplifier is a differential pair
consisting of transistors Q5 and Q7. Transistor Q4 provides a
constant current source for Q5 and Q7. The detector signal is
inserted at the base of Q5. Stage feed-back, zero-reference feedback,
ramp calibration signals, and a DC offset signal are all summed at
the base of Q7. This transistor circuit provides an open loop
gain of approximately 2500. Amplifier U3 is included in the
first stage to provide a single ended output for the signal.
5-50
» , 1 kteiogoottl
Amplifier U4 is an additional stage of voltage gain.
Transistors Q3 and Q6 perform the function of shorting
out the detector bias and the offset bias when operating in
the Voltage Calibration mode. The removal of bias from the
detector effectively eliminates the detector signal. The offset
bias is simultaneously removed to allow the zeroing reference
circuit to operate at close to normal operating conditions.
The first two post-amplifier stages comprise the pre-
sampling filter. This filter is a four pole transitional Butter-
worth Thomas filter. The relative frequency response curve for
the filter is shown in Figure 5-28. The two filter stages give
a DC gain of approximately 9 to the chain. An additional stage
provides additional gain and a relatively wide band output stage.
A unity gain amplifier is used to provide a buffered output to
the test connector.
The final amplifier output (inverted) is sampled through
a FET and fed back to the preamplifier. This signal is stored
in capacitors C14 through C18 . This signal is fed back to the
preamplifier input stage through amplifier U5 which is an FET
input amplifier operating as a low pass amplifier. This amplifier
is operated as a non-inverting amplifier providing feedback to
the input during the sampling interval.
Sampling time constants are selected such that the amplifie
output can be restored to zero during the 2 millisecond sampling
time. The DC gain of the U5 amplifier stage allows correction
Relative Response
of DC offsets with very low DC shifts in the output signal.
A significant amount of tailoring is required in the
Ch 3 amplifier to match the possible detector responsivity and
bias currents. Provisions are made in several of the amplifier
stages for a systematic adjustment by tailoring of gain with
fixed resistors. The ramp calibration signal must then be
tailored inversely to the overall amplifier gain.
Individual amplifier stages are designed -or essentially
flat frequency response (except for the filter stages) through
the cut-off frequency. For board test the combined gain of the
preamplifier and post amplifier will be set at approximately
50,000 which is more than any anticipated requirement. Any
tailoring will then be a gain reduction which will not have an
adverse affect on frequency response.
5 . 13 Channel 4 Amplifier
The schematic for the channel 4 preamplifier is shown
on 8009217 (Figure 5-31) . The channel 4 post amplifier is
physically located with the channel 3 post amplifier (Figure 5-27).
The channel 4 preamplifier is a current to voltage converter
with a high (16 meg) feedback resistor for low noise operation.
The detector is maintained with a lew back bias for minimum
detector noise output. This bias is tailored to match each detector
The Post Amplifier is identical to the Daylight Post Anplifie
5.14 Daylight Amplifiers
The schematics of the daylight amplifier system are shown
in Drawing 8008130 (Figure 5-29) and Drawing 8008096 (Figure 5-30) .
The amplifiers for channel 1 and channel 2 are identical except for
the tailoring required to match detector responsivity.
The preamplifiers are located in the optics module with
the detectors directly connected to the preamplifier terminals.
g ggn
veoeoos’
31550 BOO0 2.M
The detectors are operated in biased mode as a current source.
A 4 megohm feedback resistor is used to minimize noise current.
The post amplifiers for the daylight channels are essentially
the same as the post amplifier for channel 3. The space sample
for the zero reference is fed back to the input stage of the post
amplifier. The ramp calibration signal is summed in at first post
amplifier stage. A FET is used to short out the preamplifier
signal during the Voltage calibration mode of operation.
5.15 Multiplexer Board
>
This printed circuit board contains the multiplexer for
the 6th input channel of the analog-to-digital converter, the
blackbody temp. IR TM circuits, and over-temperature monitor
circuits for the patch and radiator.
The multiplexer consists of five FET switches feeding amplifier
U3. The logic timing signals for control of the FET's are generated
on Black Body Mux board. The four Black Body TM signals are
sequenced in the same time position on a line scan basis. During
the fifth line scan, none of the signals are gated providing a 0
volt marker in that time position. The low range patch temperature
is the other multiplexed signal.
The blackbody temp - IR telemetry signals are two identical
sample and hold circuits (amplifiers U12 and U13). The sample logic
signal is generated on the Scan Count and Decode board and is to
provide a sample of the two IR channel outputs at the time the
scan mirror has a full view of the internal calibration target
(internal blackbody).
5-59
The patch and radiator temperature TM voltages are fed
i
to amplifiers U6 and U8 operating as comparators with a reference
voltage. The amplifier outputs drive switching transistors in
the patch and radiator heater circuits to remove power if the
temperature exceeds the allowable figure (currently 40°C) .
The schematic diagram for this board is shown in Figure
5-32.
5.16 Black Body Mux
A logic divide by five counter of the BB Temp (logic)
signal with four of the counter states decoded provide the logic
control signals for the time multiplexing of the four Black
Body TM signals. The fifth undecoded state provides the marker
scan.
Individual transistor Q2 and Q4 are used to drive each Earth
Shield deploy solenoid. Feedback from the shield position switches
remove transistor drive current through amplifier U6. The Earth
Shield Disable removes all power from these circuits. An R-C
delay in the base circuit limits the input current surge. See
Figure 5-33.
5.17 Motor Power Supply
The motor power supply converts +22 volts DC (high power
mode) or +18 volts DC (low power mode) into a two-phase 240 Hz
square wave to drive the 80 pole scan mirror motor. The circuit
is a conventional bridge switching inverter. Drawing No. 8007941
is the schematic diagram of the motor power supply (Figure 5-34) .
The input signals from the Motor Logics circuit are AC
coupled for ground isolation. The HA2620 operational amplifiers
act as comparators and provide sufficient drive for the primary
circuit to insure saturation of the inverter transistors.
5 — 6 (
31550 &OQ e,\-i."2>
5.18 Power Profile
Figure 5-35 depicts the power usage under the mode
of operation indicated.
5.19 Interface Connectors
Table 5-1 lists the connections cind type of connectors
for the AVHRR Electrical Interface.
5.20 Electronics Drawings
Table 5-2 is a list of the drawings for the electronics
printed circuit boards.
5-64
Function
POWER
ANALOG TELEMETRY
1. 89W
A-D A ELECTRONICS **
12.91W
SCAN MOTOR HIGH
4. 48W
SCAN N< TOR LOW
7
3. 67W
MOTOR LOGIC
1.3W
CHANNEL 1
. 84W
CHANNEL 2
. 84W
CHANNEL 3
1.5W
CHANNEL 4
1.16W
COOLER HEATER
24. 6W
COOLER COVER DEPLOY
56. 9 W*
STANDBY HEATER
2 2 . Ort* * *
* Required only once for a period of appro>
** Measured PFMvaluea. ETM total was 2 ‘
* * Supplied from TCI?
not from +28V
r
Table 5-1 Interface Connectors
NO.
FUNCTION
GSFC STYLE
DESCRIPTION
J1
Command
311P405-4P-C-12
37
Pin Male
32
Digital TM
311P405-3S-C-12
25
Pin Female
J3
Power
311P405-3P-C-12
25
Pin Male
34
Analog TM
311P405-4S-C-12
37
Pin Female
35
Clock
311P405-1P-C-12
9
Pin Male
36
Data Processor
311P405-2P-C-12
15
Pin Male
37
Test
311P405-5S-C-12
50
Pin Female
333
Pulse Load Heater
311P405-18-C-12
9
Pin Female
Ul
i
C7\
J1 COMMAND
PIN NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
2 b
FUNCTION
Elec/Telemetry On
Elec/Telemetry Off
Motor /Telemetry On
Motor/Telemetry Off
Telemetry Not Locked On
Telemetry Locked On
Channel 1 Enable
Channel 1 Disable
Channel 2 Enable
Channel 2 Disable
Channel 3 Enable
Channel 3 Disable
Channel 4 Enable
Channel 4 Disable
Motor Low Power
Motor High Power
Patch Low
Patch High
Patch Control Off
Patch Control On
Earth Shield Disable
Earth Shield Deploy
Cooler Heat Off
Cooler Heat On
Voltage Calibrate Off
Voltage Calibrate On
5-67
J1 COMMAND
PIN NO.
28
29
(Continued)
FUNCTION
37
Chassis Ground
J2 DIGITAL TM
PIN NO.
1
2
3
■!
5
6
7
8
9
10
11
12
1 3
14
FUNCTION
Earth Shield Status
Patch Control Status
Patch Mode Status
Motor Mode Status
Voltage Calibrate Status
Cooler Heat Status
Electroni.cs/Telemetry Status
Motor/Telemetry Status
Telemetry Lock Status
Channel 1 Status
Channel 2 Status
Channel 3 Status
Channel 4 Status
Chassis Ground
J3 POWER
PIN MO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
FUNCTION
+ 28V Buss
+28V Buss
+28V Buss (Motor)
+28V Buss (Motor)
Power Ground
Power Ground
AC 28V Return
AC 28V Return
+10V Buss
+10V Buss
+5V Buss
+5V Buss
Interface Power Ground
Interface Power Ground
Signal Ground
Signal Ground
Chassis Ground
Chassis Ground
5-70
J.| ANALCC, TM
1M_N NO.
1
■I
i >
10
1 1
i .:
! i
! 4
1 <>
i ;
! S
1"
0 0
1
FUNCTI ON
Radiator Temp. TM
Patch Power TM
Patch Temp TM Lew Ranee
Pat eh Temp TM Kxt Ranee
Black Body ft 1 TM
Black Body #2 TM
Black Body M TM
Black Body #4 TM
Motor Current. TM
K 1 e c t . Current T M
Kart.li Shield Position TM
Fleet rentes Temp. TM
Base Plate Temp. TM
A C Cor.v . Temp. TM
Motor Hsu . Temp TM
Cooler Hsu. Temp. TM
tV too tor Bias Volt Ch . >
BB
Temp .
1 R
Ch.
.1 TM
BB
Temp.
1 R
Oh .
4 TM
Cl t set Voltaee TM
! '
Chass-
is ilrouttii
J5 CLOCK
PIN MO.
1
2
3
4
FUNCTIO N
Clock - Ref
Clock
Clock Shield
Chassis Ground
5-72
J6 DATA PROCESSOR
PIN NO.
1
2 9
MIRP
FUNCTION
Data
2
C30
(N
MIRP
Data
3
2 7
MIRP
Data
4
? 6
MIRP
Data
5
2 5
MIRP
Data
■ 6
2 4
MIRP
Data
7
2 3
MIRP
Data
8
2 2
MIRP
Data
9
2 1
MIRP
Data
10
2°
MIRP
Data
11
Chassis
Ground
12 Sample Pulse From MIRP
13
Chassis Ground
14 Sync Pulse
15
Chassis Ground
J7 TEST
PIN NO.
1
4
5
o
7
8
9
10
II
i:
n
14
15
17
FUNCTION
Test - Pick-up loss sim.
Ramp Cal. 3 level
Ch . 3 Test
Ch. 4 Test
Ch. 1 Test
Ch. 2 Test
REf. V Test Point
Pick-up #1 Test
Pick-up #2 Test
-15V Test
+ 15V Test
Clock Rcvr Test
Solenoid +28
Solenoid +28
+5V Test
Sync Pulse
50
Ground
5-74
I
PIN NO .
1
2
3
4
5
6
7
8
9
J33 - PULSE LOAD HEATER
FUNCTION
Pulse Load Heater
Pulse Load Heater Ret.
Temp. Control Sensor
Temp, Control Sensor
Chassis Ground
5-75
Ta \j 1 o ~ 2
AVI if
I yii../ ^ . 6 k v<«.
3 0 0 7 j 7 0
logics p^tc^mgps
6 h <j 7 j 7 ,
± 15V PLGCLA70PS
8 0 0 7 j 6 0
MV LW . P.L'J .
8 0077 4 -5
go: •mall relay »i
3008807
comma. ;d el lay =2
8008704
comma:;:, relay »3
6 0 0 0 213
patch temp co:;t. l t/m
8 00 81 2 0
*‘/M POA.Pij ?f 2
8 0 u 8 1 2 7
MOTOR 000 ICS
8008045
sc a:. - cgu:.t a dec.
8008040
imtlp.face logics =1
8 0 0 5 2 0 C
IMVLEFACL LOGICS #2
8000200
KAMR CAL. GEL.
8007062
AU>: SCA.; LOGICS
8008052
Cii. 3 PPL All P
8008101
ch. -i p he amp
3000217
IP POST AMP
8008078
DAY L IGL'T PREAMP
8008130
DA. / l. I G li V PO 3 i.’ AMP
8008000
MULT I p LEXER.
8008133
LH
MO T 0 P P OLE P. L P p L Y
8007941
-76
Li LA V L ; i 'j j / ' i ^ X
8008250
L V.*. ;0 5
DPAV.IMV
PRCCEDC? '.
8 0 3 7 971
8007972
800323 0
8008273
<j C 0 ? ; 7 C
8007977
8006231
8 0 9 S 2 7 9
5 007 :> 3 1
8007962
8008232
8 0 j 3 2 8 2
8 0 '3 7 0 4 5
6007 3 4 6
8008233
3006297
8 0 0 8 801
8008802
8008234
8005260
8008795
8006796
8008235
d 0 0 S 2 rf 1
800879 5
8008796
8009224
8009227 j;
8001 121
6008122
8008236
8008290 j:
8008128
6008129
8008237
3G08291 j;
8008046
8008047
8008238
6008263 li
8 0 0 80 5 0
6008051
8008239
6008284
3009207
8009206
8008240
8008293
8009210
8009211
8009225
8909226
8007964
8007965
8006241
8008285
8008053
6006054
8008242
8008286
8008102
8008103
8008243
8008294
8009218
8009219
0009226
8009229 ;
8008079
8008080
8003244
8008288
80GS131
8008132
8008245
8008295 |
8006097
8006098
8008246
8008289 j
6 0 0 81 34
8008135
8008247
8008292
8007942
8007943
8003248
8008296 |
8 00 82 51
6008252
8008240
8008287 1
6.0
RADIANT COOLER
The radiant cooler has a simplicity of design (Figure
6.0-1) that reflects the advantageous orbit and the absence of
spacecraft extensions in the anti-sun direction. It is a
conservative design that employs proven hardware and techniques;
there is ample cooling margin for both temperature control and
possible thermal degradation. The specific requirements imposed
on the detector cooler are as follows:
a. The cooler shall be mounted on the anti-sun side
of the space vehicle and shall look into a hemi-
sphere (of cold space) except for the solid angle
subtended by the earth. The cooler must be shaded
from earth radiation.
b. The nominal temperature of the infrared detectors
shall be actively controlled at either 105K or 110K
(selectable by command) . At the beginning of orbital
life, the uncontrolled temperature shall be 95K or
less.
c. During acquisition, the cooler field of view can
sweep through the sun at a rate; of 0.5 rpm for an
undeterminate period. The cooler must operate with
no degradation after this exposure.
We have assumed the earth shading requirement (a) applies to
the patch or second stage of cooling and that the margin require-
ment (b) applies to the nominal orbit (450 n mi altitude and
6-1
SHIELD COVER
F1G>. 6.0-1 BASIC DESIGN OF RADIANT COOLER (DIMENSIONS IN INCHES')
CT\
I
to
1.00
37.5° orbit normal to sun angle). The earth shield will be
used as a protective cover during acquisition: exposure is limited
to the shield cover, which reaches a maximum temperature of about
0°C during a continuous 0.5 rpm sweep through the sun (Section 6.5).
The design of the radiant cooler is described in detail
in Sections 6.1 through 6.4. The performance is analyzed for
the nominal condition and for variations in the orbital parameters.
The anticontamination provisions to protect both thermal and
optical surfaces are described in Section 6.6. Because of the
relatively large amount of thermal loading through the optical
port to the cooler, we have established detailed models that have
been verified by thermal tests (Section 6.7).
The nominal in orbit characteristics of the cooler are
summarized in Table 6.0-1 and 6.0-2.
6 . 1 Field of View
The hemisphere above the cooler patch is cold space except
for the earth. As a result, any point on the patch can be shaded
by a semicircular shield that matches the angle subtended by the
earth and any vertical line by the semicircular shield for its
top point. The smallest shield for the entire patch is then
obtained by translating the latter shield along the horizontal
dimension of the patch. We have modified this shield ^nd increased
the shading of the radiator by the use of small vertical sides. The
maximum angle subtended by the earth is the solution to sin
t? ra =: + where ^ is the equatorial earth radius (3444.3
n mi) plus the tropical atmospheric height (9.1 n mi) and h is
4
Table 6,0-1 Nominal Characteristics of the Radiator
Temperature ^
170.6 K
Power radiated ^
1.66 W
Radiating area
55.2 in 2
Conductive input ^
0.195 W
11.7%
Insulation input ^
0.592
35.6
Earth input
0.210
12.7
Covers input
0.388
23.4
Optical port input ^
0.275
16.6
dT/d4> (a) (b) (c)
2.05 K (0
.1 W)“
(a) For housing at 25°C
(b) Rate of change of temperature with input power
at temperature shown.
Table 6.0-2 Nominal
Characteristics of
the Patch
Temperature ^
105 K
Power radiated
96.6 mW
Radiating area
22.4 in 2
/ y. \
Conductive input '
10.0 mW
10.4%
( c )
Insulation input
15.4 mW
15.9
Joule heat input
2.3 mW
2.4
Optical port input
25.3 mW
26.2
Shield input
9.1 mW
9.4
Control power
34.5 mW
35.7
dT/d4>
0.31 K (mW) 1
(a) Nominal control point.
(b) Including effect of support shields
(c) Radiative decoupling
6-4
the spacecraft altitude (450 n mi). The solution is B = 62.20°.
m
The actual shield completely covers 63.79°, which leaves a margin
of 1.59° for spacecraft wobble (1°) and cooler alignment.
6.2 Shield
Figure 6.0-1 shows the radiant cooler shielded designed
for use at an altitude of 450 n mi. The shield completely
shades the patch (second stage of cooling) from the earth.
6.2.1 Cover Temperatures
Both the vertical and horizontal earth shields are
insulated from external inputs by shield covers. The three
optically polished shields are thermally and mechanically
connected. However, the two vertical covers are not connected
to the horizontal cover. The cover temperatures are listed in
Table 6.2-1 for the range of sun angles (S ) at the 450 n mi
altitude (see memorandum of May 2, 1974).
The thermal balance equation that determines the temperature
T of a cover is given by
c 4
C T
= F (eW+ctW)+ctS
ce c e c r c o
< sm i >
where
= emissivity = 0.72 (silvered Teflon)
a c = solar absorptivity = 0.08
2
F = view factor from cover to earth; sin
cs e
for a horizontal cover and — — (S 'cos
it e
3 ) for a vertical cover, where 8 (62.17°)
e e
is the mean angle from nadir to the earth-
tangent line.
6-5
Table 6.2-1 Shield Cover Temperatures
6 s
Vertical
0°
167. 8K
27.83
187.4
37.5
190,. 1
67
195.6
Table
6.2-2 View
Element
Weight*
a
0.1046
b
0.1046
c
0.1345
d
0.1345
e
0.05575 ,
c
0.12485
g
0. 3412
Radiator
1
* Relative area.
Horizontal
232 . OK
241.9
238.6
238.7
Factors from Radiator to Earth
0.011 832
0.012 734
0,013 825
0.075 471
0.002 673
0.037 845
0
0.019 454
6-6
<sin i>
= orbital average of the solar incidence
angle, taken to be zero when the cover is
shaded from direct sunlight;
sin 8
(1 + cos Au e ) for a horizontal surface and
2tt
surface.
sin 8 (1 + cos Ay ) for a vertical
s e
Ay = arccos (cos 8 /sin 8 ) ; zero when sin 8 < cos 3 »
e e s s— e
i.e., when the spacecraft is in direct sunlight
throughout its orbit.
-2 -2
W = infrared exitance of earth = 2.1 x 10 Wcm
e
W r = reflected solar exitance of earth =
-2 -2
1.68 x 10 sin 8 S Wcm
_2
s = solar constant - 0.14 Wcm
o
The equation for W r is derived in Appendix I to Part I
of the Final Report on Contract NAS5-10113 (Dec. 1967).
6.2.2 Shielding and View Factors
Figure 6.2-1 shows the projection of the shielded onto
the scan earth disk as seen from an upper corner of the patch
(i.e., from the most difficult point to shield). The shading
from the earth is complete up to a disk angular radius of 63.79°.
The View factor F from the black radiator to the earth
re
was determined by dividing the radiator into 14 elements, as
shown in Figure 6.0-1. By symmetry, there are seven elements
are shown in Figure 6.2-2. The view factors were calculated
by means of the contour integral technique (R. V. Annable, Applied
Optics, Jan. 1970 and July 1972) . The results are given in Table
6 . 2 - 2 .
6-7
PROJECTION
OF SHIELD ED§S.
FI Q. 6.2-1 PROJECTION OF SHIELD ONTO MEAN EARTH DISK AS SEEN FROM
UPPER PATCH CORNER (POINT C IN FIGURE S.O-t)
2.0 1.5 t.O 0.5 - 0.5 -UO -1.5 -2.0
Fl(q. <0.2-2 PROJECTIONS OF SHIELD OKI EARTH DISK AS SEEN
FROfv\ RADIATOR. ELEMENTS.
The view factor F from the patch to the shield was
ps
calculated in two parts. First, the view factor to the
horizontal shielded was calculated using view factor algebra
and the formula for two perpendicular rectangles with a common
boundary (E. M. Sparrow and R. C. Cess, Radiation Heat Transfer,
Brooks/Cole, 1966, Section 4-3). The result is
F . = 0.37161
psl
Secondly, the view factors to the vertical shields and their
specular images in the horizontal were calculated from the
center of the patch by the contour integral technique. As shown
in Figure 6.2-j, each vertical shield and its ccmplte specular
image can be seen from the patch center. By symmetry, the view
factors are the same to either vertical shield. The results are
F _ = 0.00359 to shield
ps2
Fp S 2 = 0.00339 to shield image
Finally we determine an effective view factor that can be
used directly in the design equations. It is given by
F = F . + 2F t + 2 (1 - e ) F 1
ps psl ps2 s ps2,
where (1 - e ) - 0.965 is the reflectivity of the horizontal
shield .
The result is
F = 0.3853
ps
6 . 3 Radiator
The radiator and earth shield have a temperature that
is thei solution to:
6-10
i '
\
where
e r oA r T r
f + <t> + $.+$.+
er cr i k
ab
= radiant pc Jer from a that is absorbed in b
e
r
$
earth (infrared and reflected sunlight)
black radiator, c - shield covers
input from instrument through a
multilayer insulation; k =, supports and wires;
o » optical port
er
cr
F (e W + a W ) A
re r e r r r
ctA 4 4
— (T c -T r ) + K c (T c -T r ) / for each cover
ab
view factor from a to b; A, = area of a
d
£ = emissivity of a; a = solar absorptivity of a
d
-2 -2
= earth infrared exitance = 2.1 x 10 Wcm
e
W = earth reflected sunlight exitance
-2 -2
= 1.68 x 10 sin 6 Wcm ; 0 = orbit normal
s s
to sun angle
2 2
A (horizontal) - 104 in ; (2 vertical) = 7.5 in
A_
» 55.2 in 2 ;
re
= 0.01945 (Section 6.2)
= a r = 0.97 (honeycomb cavity array covered
with black paint)
2
= -1; e = emissivity of aold platina
e c
c
on facing surfaces of shield and cover = 0.035.
6-12
K„
thermal conductance of supports between shield
and cover
2.88 x 10 3 WK ^ (horizontal),
1.44 x 10 3 WK (2 vertical)
$ .
i
oA.
i
(T
S i
h
-T r ) ; = 115 in ; S j _ = 50
= instrument temperature
= K (T. -T ) ; K = thermal conductance between
K r h r r .
- 3 — J "
h and r * i.l4 x 10 WK
* = 0.257 (T h = 20°C) 0.275W (25°C) , 0.313 W (35°C) ;
Because the thermal time constant of the radiator-
shield is much greater than an orbital period, orbital averages
are used for the earth inputs. The insulation factor of 50
is the minimum expected and was achieved on the ETM.
The thermal conductance K r between the housing (main
instrument) and the first stage of cooling consists of 1.35
10 '
.-1
x 10 3 WK 3 from the synthane support tubes and 0.18 x 10 3
WK * from the following electrical connections:
Quantity Diameter (inch) Material
2
3.9
X
10~ 3
Copper"''
4
3.0
X
10~ 3
Chromel
2
5.0
X
10" 3
Chrome 1,
4
3.0
X
-3
10
Chromel
2
3.9
X
io” 3
Copper
4
2.0
X
10~ 3
Nickel .
To radiator
components
To patch
components
All the electrical connections have a conductance length of
3.15 inches.
6-13
Table 6.3-1 shows the radiator temperature range of the
450 n mi altitude. The housing temperatures (T^) are estimates
based on the thermal analysis .
Table 6.3-1 Radiator Temperature Range
6 S
T h
T r
0 °
20°C
165. 6K
27.83
35
173.7
37.5*
25
170.6
67
20
170.1
* Nominal orbit (8:30 AM or 3:30 PM).
6 . 4 Patch
The thermal balance equation for the patch is
at A T = <t>
P P P s
$. + + $ . + i
k i 3 c
where p
s
i
j
o
-i
* .
j.
patch
earth shield (upper side)
thermal conductance including the influence of
radiative inputs from the support shields
gold-to-gold radiative insulation
joule heat of detectors and temperature sensor
optical port
a e t A F T ^
p s p ps r
M K (T -T ) ; M = dual mode multiplier
P r p
<ja.
i
s.
1
(T
- T p > ; S i
C i
6-14
4>j - 2.3 x 10“ 3 W
$ Q = 2.30 x 10 -2 w (3 S = 0°); 2.83 x 10~ 2 W (27.83°);
2.53 x 10 -2 W (37.5°); 2.41 x 10 -2 W (67°);
Ep = 0.97 (black paint on honeycomb cavity array)
e s = 0.035 (vacuum deposited ciluminum)
= 0.035 (gold plate)
Ap = 22.4 in 2 ; =32.6 in 2
F = 0.3853 (Section 6.2)
ps
K p = 1.1875 x 10“ 4 WK -1 ; M = 1.28
The thermal conductance K between the radiator and patch
P
consists of the following connections:
Quantity
Diameter (inch)
Length (inch)
Material
4
3/16 x 5/32
2.70
G-10 synthane
4
0.002
2.70
nickel
4
0.003
2.70
chromel
2
0.005
2.70
chromel
The dual heat mode multiplier M was calculated using the approach
described in the memorandum "Combined Radiative and Conductive
Heat Transfer in the Patch Supports", from Contract 5-21651. A
shield emissivity e g of 0.035 was used in the calculation; and the
analysis was carried out for the case T s = T r (i.e., the thermal
gradient was evaluated a x = l ) .
The above equation does not include the input present
during chamber testing as a result of reflection of the radiator
power from the cold space target. Based on thermal tests and
6-15
their analyses, we estimate this input to be given by
4 4
T
7.5 x 10
-3
W
164
for the 22.4 in black radiating area.
The solutions to the patch thermal balance equation
are listed in Table 6.4-1. In the nominal orbit, the tempera-
ture margin is 10. IK at the 105K control point. The corresponding
control power is 35.7% of the total patch load. Or, stated
another way, all other thermal inputs would have to increase
by 55.5% in order to use up the margin.
6 . 5 Solar Exposure
If the cooler sweeps through the sun during acquisition,
both the patch and radiator will be protected by the shield/cover.
The exposure is therefore limited to the (horizontal) shield
cover (which is in a vertical position during storage) . The
worst case is when the sun sweeps through a plane perpendicular
to the plane of the shield cover. Even then, as shown by the
following analysis, the shield cover has a worst case average
temperature of only -0.5°C and a worst case maximum of -0.2°C.
During this period, the cooler temperature willthen be regulated
at 40°C by the outgassing power circuit.
We may treat the shield cover as thermally isolated from
the instrument. Its thermal balance equation is then
dT
dt
ct
€
H (t) + W + a W
s e — — rm
6-16
Table 6.4-1 Patch Temperature Range
0 °
27.83
37.5*
67
* Nominal orbit.
T p (space)
92. 8K
96.7
94.9
94.3
Tp (chamber)
95. 5K
99.6
97. 8
97.2
Table 6.4-2 Other Patch Parameters
AT /A4> 0.34K/mW @ 94. 9K, 0.3lK/mW @ 105K
zr
Nominal margin 10. IK, space; 7.2K, chamber
Maximum margin 12. 2K, space; 9.5K, chamber
Minimum margin 8.3K, space; 5.4K, chamber
6-17
where e = 0.72 is the enissivity of the cover and a = 0.08 the
solar absorptivity. The cover is treated as a plate of thickness
5, density p, and specific heat c. H s (t) is the solar irradiation
as a function of time. For the sun in a plane perpendicular to the
radiator, H g (t) is a train of cosine pulses given by
H s ( t) =
5 ° [
1/2 + Re E c exp (jwnt)
n=l n
where S Q is the solar constant, Re is real part of, and
-1 (- 1 )
n/2
- V4 , C n -
2 ,
n -1
, n = 2 , 4 , 6 . .
27rn
u)
= frequency of rotation (0.5 rpm) .
n
W (3 is the infrared exitance of the earth (Section 6.2.1), and
W is given by (Appendix
is the maximum reflected solar exitance
I to part I of the Final Report on Contract NAS 5-10113, Dec. 1967)
W
cos
rm
L -SIN B,
sin B s S o A,
where A = 0.4 is the average solar reflection factor for the
earth .
The worst case orbit is then B s = 67 , so that we have
W = 4.86 x 10 -2 W
rm cm
-2
The steady state solution to the above differential
4 a
equation is <3 , T Q = S Q and yields the average radiator
temperature of T q = 212. 1Y. - -0.5°C. For small temperature changes
about T q we may now linearize the equation by means of T - T q + (i'T
6-18
™ rs a y
WffWJf!
and extract the equation of the transient solution
3 d (AT)
4cT AT + a dt
o
2a
S, Re Z C exp (jw t)
ire o n ^ J n
where a
AT (t)
6 /e .
pc'
a S_
The solution is
Re Z
2iraeT.
C n exp (jw n t)
1 + jW n T
where x = a/4oeT Q is the thermal time constant. For a 0.03 inch
aluminum plate (p = 2.7 gms/cm^ , c - 0.895 joules/gm°C) , we obtain
x = 10.34 mins. Taking the real part of the sum, we obtain
S C
o “ n
AT (t) = 3 Z T~ 2 (cos t +w T sin n
4TraeT^ n = 1 1 + w x
o n
The maximum value of AT(t) occurs at t = 0.19 P when t = 0 is the
point of zero solar incidence angle. The resultant value is
AT == 0 . 29°C
The peak difference from average is reached about 55° of rotation
after normal incidence and is very small compared with the average
temperature. The maximum temperature, in fact, is only -0.2°C.
6 . 6 Anti-Contamination Provisions
The radiant cooler is designed to prevent optical or
thermal contamination by either the cooler components themselves
or by the instrument/spacecraft atmosphere. Specific provisions
are (1) conditioning and de-contamination, (2) elimination of
internal outgassing paths, and (3) positive protection of sensitive
areas .
6-19
Emm* T wyww;*'.
flstp3jpyaij»igpggW^^
To outgas the cooler and prevent condensation of
external contaminants from the instrument and spacecraft,
the cooler will be heated to approximately 40°C for an extended
period after acquisition. The same heaters will be used for any
subsequent decontamination. The times required for a complete
decontamination and for the cool down to operating temperatures
are estimated in DIR No. 19. The cooler will be outgassed with
the shield/cover closed and decontaminated with it open. In the
open position, we require a power level of 14.0 watts on the
rcidiator and 6.4 watts on the patch to achieve a 25°C temperature
Internal outgassing paths are eliminated by windows that
seal the openings between the instrument and radiator and between
the radiator and patch. The volumes within the cooler can outcas
only by paths that lead directly to space.
To provide positive protection for sensitive areas, the
window on the radiator is heated a nominal 8K above the radiator
temperature and protected by a cold trap at the radiator temperature.
The design and analysis of the window heater is covered in DIR.
No. 15. The optical elements on the patch and the low emissivity
rear areas of the patch are protected by a cold trap at the patch
temperature .
6 7 Optical Port Loading
Three thermal tests were run on a feasibility model cooler
and two tests run on the BBM. These tests were used to establish
a mathematical model for the loading from the optical port opening
onto the radiator and the patch. The opening is shown schematically
in Figure 6.7-1. This is the configuration of the instrument final
design.
6 - 20 )
6.7.1 Optical Loading on the Radiator
The optical loading on the radiator is
i 4 ' '
'or * S l” T h A 1 F 12 < F - P 2» + (1 -F 12 ’
2
where equals tt r^ and is the fraction of room temperature
radiation passed by the inner window. For an. inner window of
Irtran 2, P 2 i- s about 0.5 (corresponding to a cutoff wavelength
of 14 pm). The emissivity of the opening is 0.9 in the instrument,
where an outer germanium window is added. Using the data given
in Figures 6.7-1 and 6.7-2 plus Table 6.7-1 and 6.7-2, we obtain
the calculated optical port loadings on the radiator in the final
instrument design are then:
$ (T. = 20°C) = 0.257W
or h
$ (T, = 25°C) = 0.275W
or h
$ or (T h = 35°C) = 0.313W
6.7.2 Optical Loading on the Patch
The optical port loading on the patch consists of inputs
from the instrument and the radiator. We will restrict the
radiator input to that from the inner window itself; radiation
at the radiator temperature (ie., the cold trap) is efficiently
absorbed in the window. The input from the instrument to the
patch is given by
4
*13 = e l aT h P 2 t 2 F 31 y 3 A 3
where x 2 is the transmittance of the inner window for greybody
radiation at the temperature T h , is the effective aborptivity
of the patch opening.
6-21
r l
S 13 S 23 S 12
Instrument 0.630 0.570 0.415
0.855
0.07 0.685
Instrument
Opening
Radiator Patch
Opening Opening
Germanium
Irtran 2
Optical elements
& gold baffle
Figure 6.7-1 Characteristics of the Optical Opening to the
Radiant Cooler (Dimensions in Inches)
6-22
Table 6.7-1 View Factors and Blockage Within the Optical Opening
32
0.875
31
0.319
f
12
0.303
Table 6.7-2 Temperatures Within the Optical Opening
(Instrument Test)
Instrument
Radiator
Patch
20°C ,
35°C
173. 6K
181. 7K l
105K (b)
25°C ,
20°C
178. 6K f (a)
178 . IK/
(a) Heated Window
(b) Control Point
6-24
1
The effective absorptivity y^ is given by
Y j = a 3 + (1 -a 3 ) P 2 F 33» a 3 + ^ x
+
2 .2
p 2 .(P 33 »)
a-
1 -p 2 (1 ~ a 2 ) F 33 '
where is the actual absorptivity of the patch opening, p 2
is the (specular) reflectance of the inner window, and is
the view factor from the patch opening to its own specular image
in the inner surface of the window. The value of y 3 was experimentally
determined from; a theoretical model of a 3 is described in Section
6.7.3.
The input from the inner window attached to the radiator
is given by
°23 = CT T 4 e 2 F 32 y 3 A 3
w
The window temperature T w is equal to the radiator temperature
plus 8K . For an inner window of Irtran 2 and for greybody radiation
in the range from 160K to 17 5K, the window emissivity e 2 is
approximately 0.6 and the window reflectance p_ approximately 0.2.
f
Using the theoretical ct^ value of 0.675 (Section 6.7.3) and test
values for p 2 (0.1), and F^' (0.743), the theoretical value of y
is 0.692. The experimental y^ is 0.773 and is based upon test data
from a feasibility model cooler.
Using the experimental value for and the instrument
parameters listed in Figures 6.7-1 and 6.7-2 and in Tables 6.7-1
and 6.7-2, we calculated for the final instrument. The results
are given in Table 6.7-3.
6-25
Table 6 . 7-3
Calculated Patch Loading
Port (Instrument)
from Optical
a
T.
$
s
h
op
0°
20°C
0.0230 W
27.83
35
0.0283
37.5
25
0.0253
67
20
0.0241
6-26
6.7.3 Absorptivity of the Instrument Patch Opening (Theoretical
Model )
To complete the consideration of optical port loading, we
constructed a theoretical model for the absorptivity a 3 of the
patch opening in the instrument cooler. A view of the patch
opening as seen from the inner window is shown in Figure 6.7-3.
The area b^ occupied by the Channel 3 spectral filter and its
specular image b 2 in the infrared dichroic are effectively black.
Thus in area b^, some radiation is absorbed in the Channel 3
filter or transmitted by the filter and absorbed in the cavity
below; the remainder is reflect to b 2 , where it is either absorbed
or transmitted by the dichroic and absorbed below.
The effective absorptivity of the area d (dichroic minus
area b 2 ) is igven by
a d ' = a d + (1 "V a g’
where a d = actual absortivity of area d = 0.4
ct ^
g = effective absorptivity of gold baffle g
This equation assumes that radiation not absorbed or transmitted by d
(the fraction a is specularly reflected to the gold baffle. The
value given for is an estimate for the infrared dichroic on a
germanium substrate. Thus for radiation at the instrument temperature,
only about 0.5 is transmitted through the inner Irtran window. For
a dichroic with a separation wavelength of 9.0 pm, about 0.4 is
transmitted and therefore absorbed. For radiation at the radiator
temperature, about 0.2 of the total is reflected if the dichroic
reflects over the band from 9 pm to 16 pm. If the dichroic has
6-27
the properties of a germanium substrate beyond 16 vim, the
remaining fraction of 0.8 has an absorptivity equal to ap-
proximately 0.47, the transmittance of the substrate. The
value of for radiator emission is then 0.8 x 0.47 or about
0.4.
In order to calculate the effective absorptivity a , of
the gold baffle, we assumed that the gold is in the form of a
diffuse gold plating. We then have
a ~ + d “<*„) F ct + F . . + F . -
g g g gg g gbi gb2
where is the infrared absorptivity (0.035) of the gold plating
and is the view factor from area i to area j . This equation
I
can be solved for a ; we then have
g
+ (1 -O F cb + F„,, + F_, (a, + <1 -ct A > )
gb- gd d
‘d db.
1 - (1 -a ) F + F , (1 -a.) F ,
g gg gd d dg
The view factors F^ ^ were approximated by
13 =
A.
_L
E A.
This equation is strictly true only for a sphere. For the components
in the patch opening of the instrument, we have
2
A, = 0.636 in (0.9 inch diameter opening)
bl
2
0.126 in (0.4 inch diameter)
A.
b2
2
a A bl = 0.178 in
. _2
it 0.7 x 1 ■Aj 3 2 = 2.021 in
6-29
t
A_ = 1 u 0.45 x 0.75 - A, , = 0.404 in 2
g —x — bl
ZA. =
3.365 in
F ij
0.12
F.,
lb i"
0.04
F ib 2 =
0.05
H
xs
•rl
Pu
0.60
H-
0
n
0.19
We then have for the effective absorptivities
a ' = 0.435
= 0.66
Finally, the value of can be calculated as the area-
weighted average
A,
A A a + A. + A, + A. a ,
g g g b. b„ d d
where the primes on the areas denote projection into the opening A^
0.210 in 2
A b 1 = 0.041
. 2
in
b2
0.126 in
0.259 in'
For the above values of a and , we then have
= 0.675.
6-30
7.0
CALIBRATION
7.1 Thermal Channels Calibration
The infrared calibration signal to the AVHRR is the
difference in signals from a calibration blackbody of known
temperature and a blackbody of essentially zero exitance. The
zero level source is provided by deep space during in-flight
calibration and by a liquid nitrogen cooled black cavity during
chamber calibration. Other calibration requirements are listed
in TcLble 7.1-1. We have interpreted the temperature calibration
accuracy of the chamber targets (B.b.) to mean the equivalent
temp«irature accuracy determined by all error sources.
7.1.1 Calibration Accuracy
The absolute radiometric calibrationof the instrument is to
have an accuracy of ± 0.5K throughout the calibration range of each
channel. Before we consider how this requirement can be met, let's
define some terms. In particular, we can differentiate between
precision (or sensitivity) and accuracy by means of the following:
Precision: A measurement is regarded as precise if the
dispersion of values, i.e., the standard
deviation a, is small.
Accuracy: A measurement is regarded as accurate if
the values cluster closely about the correct
value.
By accuracy of an individual measurement or of an average
of measurements is usually meant the maximum possible error (constant
and/or random) that could influence the observed value. It is fre-
quently thought of in terms of the number of significant figures
to which a value can be regarded as correct.
7-1
TABLE 7.1-1 Requirements for Thermal Channels Calibration
{
A. Inflight Blackbody
a. Measured with platinum resistance thermometers
Appropriately arrayed to adequately define the
temperature .
b. Temperature sensor instrumentation accuracy of
±0. IK.
c. To be compared with the chamber targets during
thermal vacuum calibrations.
B. Chamber (Standard) Blackbodies
a. Greater emissivity, temperature stability, and
temperature sensor accuracy than inflight target.
b. Absolute temperature measurement accurate to
±0. 5K.
7-2
The precision is limited by the instrument noise, i.e.,
by the sensitivity. In order to measure the noise of the instrument
amd to reduce its effect on the absolute radiometric calibration,
a set of n measurements will be made at each calibration point.
The precision (or sensitivity) is then given by the best estimate
of the standard deviation (see for example, D.C. Baird, Experi-
mentation: An Introduction to Measurement Theory and Experiment
Design, Prentice - Hall, 1962) .
a = fl (x -x j /(n -]}J ,
where x is an individual measurement and x the average of n measure-
ments. On the other hand, the standard deviation of the average
of n measurements is given by
CT n = a/n ^
The influence of the instrument noise and of all other random
errors can therefore be reduced to the point where the accuracy
of a calibration is determined by the systematic errors in the
calibration target itself. This could be done during both the
chamber (Section 7.1.2) and in-flight (Section 7.1.3) calibration.
There are 10 calibration points (elemental dwell times) during
each scan of a target, and so for a 1 minute period, we have
n = 3600.
Before we consider the errors in the calibration targets,
however, let's list all the components and procedures that can
limit the accuracy of a calibration. We can identify two major
7-3
areas, within which there may be important subareas. They
are as follows:
A. Calibration target
a. Temperature: Measurement
Gradients or uniformity
Control (chamber)
C Sensor
^Instrumentation
b. Non-blackness
B. Electronics
a. Noise (NETD)
b. Signal processing: Digitization
Recording
(In-flight)
Transmission
Ground Processing
We will limit ourselves to errors from sources A and B.a. We
are therefore assuming that the experiment is so designed that
the errors contributed by B.b. are negligible by comparison.
If the difference in surround between chamber targets can
be made zero or very small, the non-blackness errors will be
greatly reduced and the accuracy of the calibration limited only
by the errors and uncertainties in the calibration target tempera-
ture. Accuracy estimates for the chamber calibration are given
in Table 7.1-2 and for the in-flight calibration in Table 7.1-3.
We see that we have met our objective of 0.5K for the
total channel calibration error throughout the temperature range
of both channels.* In addition, the total in-flight error at 295K
* We have arbitrarily set the lower limit in Channel 4 at 25 OK,
where the noise equivalent temperature difference is approximately
IK.
7-4
Table 7.1-2 Accuracy of Chamber Calibration
Sensor
±0 . 0 5 K
Measurement
Instrumentation
±0.05
!
V.
Control
±0.05
Gradients '
base (uniformity)
±0.10*
Honeycomb ( IK) *
±0.06
{ Diff from box (±2X) *
Wall «;
±0.0026
( Gradient within (5K)*
±0.00045
Accuracy (max of errors + uncertainties)
ilackness
0. 32K
Net from standard-cold space difference
(10K difference
in surrounds) :
Channel
T
-2
o T
3
185K
+0. 018K
3
320
-0.046
4
250
+0.027
4
320
-0.016
accuracy (including noise for n = 3600)**
Channel
T
5 T
3
185K
0. 35K
3
320
0.37
4
250
0.37
4
320
0.34
* Actual value of gradient.
** Based on specified NETD of 0.12K at 300X.
Table 7.1-3 Accuracy of In
Temperature:
Measurement
Gradients
Non-b lackness :
Total accuracy (including noise
flight Calibration For T = 29 5K *
r
i Sensor
+ 0.
, 0 5I<
\ Instrumentation
+ 0.
, 10
Base
±0.
,08
, Honeycomb (IK)
+0 .
,08
±0.
31
Channel
6 T
3
-0.080
4
-0.029
for n = 3600)
Channel
6 T
3
0. 39K
4
0.34
★
Exclusive of errors from scattered sunlight (Section 3.11)
has a comparable value when the calibration is made in the
absence of direct sunlight (See Section 3.11). In the following
sections, we consider in detail how we obtained the estimate
listed in Tables 7.1-2 and 7.1-3.
7.1.2 Chamber Calibration Targets
_2
Errors and uncertainties in the exitance (emitted Wcm )
of the calibration target arise from its temperature inaccuracies
(Section 7.1. 2.1) and its deviation from blackness (Section
7. 1.2, .2). Because a calibration signal is equal to the difference
in signals from a calibration target and a cold space target,
the inaccuracy from non-blackness is greatly reduced by making the
two targets the same form and exposing them to the same surround.
7 . 1 . 2 1 Temperature Uncertainty
The accuracy of the calibration target temperature is
limited by measurement errors, control stability, and gradients.
The uncertainty in a temperature measurement relative to the
international practical temperature scale (IPTS-60, which is
essentially identical to the absolute thermodynamic temperature
scale) is +0.10K. About half of this is the calibration accuracy;
the remainder is produced by the sensor, bridge, and power supply.
The latter produce errors that are largely random in nature, as
do the readout device and temperature controller. The readout
device introduces an error that can be kept small, about +0.01K
for an integrating digital voltmeter. The stability of the
controller is about +0.05K.
7-7
The base gradient or uniformity can be held to 10.10K.
We have included this variation as part of the calibration
error. In fact, the base temperature will be measured with
an array of calibrated platinum sensors. The average of this
array should then provide a measurement whose gradient error
is less than the actual gradient. The gradient through the
honeycomb can be estimated from the measurements on a similar
target (A. R. Karoli, J. R. Rickey, and R. E. Nelson, Appl.
Opt. 6, 1183, 1967). The honeycomb gradient was 1.6K in a 290K
target that had a view factor of about 0.5 to a warm surround at
25°C. If the target temperature were reduced to 210K, the
gradient would increase to about 2.5K. However, the view
factor to the warm surround is reduced to 0.2 in our design,
so the gradient is about 1.0K.
Moreover, the corresponding increase in the radiance
temperature is much less than the gradient because most of the
normal emission comes from the base and walls near the base.
When the instrument views the calibration target at normal
incidence during a calibration, the nominal cavity emissivity of
0.999327 (Section 7. 1.2.2) may be divided between the base honey-
comb and the cavity walls. For the nominal paint emissivity of
0.92, the base has a normal emissivity of 0.996696. Therefore
0.002631 of the normal cavity emissivity arises in the cavity
walls (and is seen by reflection in the base) . The emissivity
of the base may, in turn, be divided among emission from the
base bottom, the flat top area of the honeycomb, and the walls
7
of the honeycomb. The fraction of flat area of 0.025, so that
the emissivity from the top is 0.92 x 0.025 and from the bottom,
0.92 x 0.975. The remainder of base emissivity, 0.996 696 - 0.92 =
0.076 696 arises in the sides of the honeycomb. We will assume
that the walls of the cavity emit at the bottom temperature of
the base and that the base honeycomb sides have an exitance equal
to the average of the basebottom and honeycomb flats. The effective
exitance M £ of the target seen at normal incidence is then
0.999 327 M e = 0.061 348 + 0.937 979 Mg,
where F denotes base flats and B base bottom. Now a calibration
temperature will be the measured value of the base bottom B. In
both Channels 3 and 4, we find that a honeycomb gradient T -T_ =
F o
±1B results in a calibration error T_ -T_ = 10.061K (see Table
E o
7.1-6 in Section 7. 1.2. 2).
The temperature errors introduced by deviations in the
cavity wall temperature were analyzed for Contract NAS5-21651
(HIRS for Nimbus F)*. This analysis shows that the wall tempera-
ture deviations (difference from the base and internal gradient)
introduce a temperature uncertainty of only about +0.003K. As
a result, the total temperature uncertainy in the chamber target
is approximately 0.32K.
7 . 1 . 2 . 2 Deviation from a Blackbody
The uncertainties in the calibration of the instrument
are expressed as absolute temperature errors in blackbody sources
*Merao from R. V. Annable "Deviations in the Wall Temperature
of the Chamber Calibration Target", dated 6-28-72.
7-9
within the calibration range of each channel. The principles
and practice of absolute radiometry are explored by R. E.
Bedford and A. R. Karoli in Volume 14 of Advances in Geophysics
(Precision Radiometry, ed. by A. J. Drummond, Academic Press,
1970) . We have already covered the uncertainty in the temperature
of the calibration target (Section 7. 1.2.1). We now wish to
consider the uncertainties produced by non-black calibration
and cold space targets.
The real problem here is not the small decrease in
target emission below that of a blackbody, but the reflection
of the higher temperature surround. The calibration target
consists of a honeycomb array with a length to width ratio of
4:1 (or its emissivity equivalent in another geometrical form)
housed in a tube whose length is equal to the aperture diameter
(Figure 7.1-1). The tube is covered on its inner wall with a
honeycomb array whose length to width ratio is 2:1. In this
way, we obtain a second, large cavity in addition to the array
of small cavities. It is also equivalent to controlling a
large fraction of the target surround. The tubular enclosure
must not be thermally attached to the base calibration target;
this would induce significant thermal gradients in the target.
In addition, the cavity mouth and base must be sufficiently
large that only the base is seen by the instrument during a
calibration.
Because a calibration depends on the difference in
signals from the calibration and cold space targets, the accuracy
can be further increased by making both targets in the same form
and exposing them, as nearly as possible, to the same surround.
We have an estimate of the residual non-black error based on a
…[truncated]