IRIS Instrument Description

Index

Click for the text version of this file.


Instrument Description

The Voyager IRIS instrument consists of a Michelson interferometer for measurements in the thermal infrared and a single channel radiometer that operates in the visible and near infrared. The two components of the instrument share a 50 cm Cassegrain telescope with an effective focal length of 303.5 cm. The angular field of view is 0.25 degree. Light passing through the telescope is divided into two beams by a dichroic mirror, with that longer than about 2.5 micrometers going to the infrared interferometer and radiation between 0.33 and 2 micrometers going to the radiometer. The effective spectral range of the interferometer is 180-2500 cm**-1 (4-55 micrometers) and the apodized spectral resolution is 4.3 cm**-1. The beam splitter of the interferometer consists of a multilayer dielectric coating applied to a cesium iodide substrate. The moving mirror is mounted on one end of a motor shaft; the moving mirror of an auxiliary reference interferometer is attached to the other end of the shaft. A 0.5852 micrometer neon line source is used for the reference interferometer; the signal from this unit is used by a phase comparator and digital sampling circuit to control the motor speed and to quantize the analog signal from the main IR detector. The latter is a low impedance, Schwartz-type thermopile with a noise equivalent power (NEP) of about 2E-10 Watt/Hz**1/2.

The spectral response of the radiometer, which is designed to measure the broadband reflected solar radiation, is controlled by the dichroic mirror and an additional coating on the radiometer side of the mirror. The radiometer detector is an eighteen-junction thermopile with an NEP of 4E-10 Watt/Hz**1/2. A sapphire window is placed in front of the detector to reject unwanted long-wave radiation that may result from temperature gradients within the instrument. The signal from the radiometer detector is fed to a low noise, low-drift dc amplifier with a time constant of approximately 2.7 seconds, and then to three different output circuits. The first circuit integrates the radiometer signal over the 45.6 seconds it takes to record an interferogram, thus providing the average of the reflected sunlight during the time the infrared spectrum is recorded. The second and third circuits provide the analog signal from the radiometer as well as the signal amplified by a factor of 8. These latter channels are sampled every 6 seconds and digitized. The timing of the samples, and their relationship to events in a single-frame read out of an image are given in Table 1.

                 Table 1. IRIS and ISS Events

        Event                         ISS line count
        ____________________________  ___________________
        ISS frame start               0 (0 seconds)
        IRIS frame start              167
        IRIS interferogram start      179
        IRIS radiometer samples       217,317,...,717,
                                      and 017, 117 of
                                      the following frame
        IRIS pointing information #1  350
        IRIS interferogram peak       375 +/- 5
        IRIS pointing information #2  750
        ISS shutter close             767
        ISS frame end                 800(48 seconds)

In order to visualize the effects of quantization uncertainties directly, it is sometimes useful to convert the radiometer watts at the detector back into data numbers (DN). The relationships are:

        For Voyager 1:

           Sampled data (normal gain):
              DNn = (WADn * 1e+08 - 2.51937) / 3.39426
           Sampled data (high gain):
              DNh = (WADh * 1e+09 - 1.55472) / 4.26206
           Integrated data:
              DNi = WADi / 1.42145 * 1e+09

        For Voyager 2:

           Sampled data (normal gain):
              DNn = (WADn * 1e+08 - 6.33020) / 3.40070
           Sampled data (high gain):
              DNh = (WADh * 1e+09 + 6.57589) / 4.24459
           Integrated data:
              DNi = WADi / 1.42047 * 1e+09

Pairs of baseline levels are averaged at the beginning and at the end of the 45.6 second integrations, occasionally resulting in half-integral DN values for the integrated radiometer data.

Despite the stability of the calibration data (see below), a small systematic error may remain. Comparison of spatially resolved IRIS observations of Jupiter and Saturn shows that reflectivities derived from Voyager 1 and Voyager 2 IRIS instruments are in the ratio 0.883:1. An investigation was made of albedos of various objects (Jupiter, Io, Ganymede, Callisto, and Saturn), as determined using the two instruments. When compared to groundbased determinations, no systematic differences were apparent for either instrument, within the uncertainties. Thus, a systematic error on the order of 10% may exist in the calibration of one or both of the IRIS instruments’ radiometer systems.

The temperature of the instrument is passively and actively controlled to operate at 200 +/- 0.5 K with a maximum drift of +/- 0.1 K/day. A thermal radiator mounted on the interferometer cools the instrument by radiating to deep space. Three sets of proportionally controlled heaters provide fine thermal control for the primary telescope mirror, secondary mirror, and the interferometer. It is necessary to maintain temperature differences between the three components to less than 0.1 K. In addition, high-powered ‘flash-off’ heaters are available for increasing the instrument temperature by approximately 70 K. These are controlled by command. These have been used to warm the instrument during cruise periods to reverse a gradual change in the elastic properties of a silicone compound in the Michelson motor dampers and beam-splitter mounts.

Back to the index

Science Objectives

The scientific objectives of the Voyager Infrared Interferometer Spectrometer (IRIS) investigation include the following:

 (1)  Identification and determination of the abundances of
      gaseous atmospheric constituents;
 (2)  Determination of the helium abundance of the
      atmospheres of the giant planets;
 (3)  Determination of the energy balances of the giant
      planets through measurements of their total thermal
      emissions and Bond albedos;
 (4)  Determination of the three dimensional thermal
      structure of the atmospheres of the giant planets and
      Titan;
 (5)  Inference of information on atmospheric dynamics;
 (6)  Inference of the infrared optical properties of
      clouds and hazes;
 (7)  Determination of the temperature, composition, and
      structure of the surfaces of satellites without
      atmospheres;
 (8)  Determination of the thermal characteristics of
      Saturn's rings.

These objectives are accomplished through the analysis of measurements of thermal emission spectra and broad band reflected solar energy.

Back to the index

Operational Considerations

The IRIS instruments on Voyager 1 and 2 operated normally throughout the mission. The infrared interferometers of both instruments experienced a temporal decrease in responsivity, believed to be due to a gradual stiffening in silicon compounds used in the mirror mounts and motor dampers. However, this was partially reversed by using the flash off heaters to raise the temperature of the instrument periodically during cruise between encounters.

Back to the index

Calibration Description

In-flight calibration of the data from the interferometer is accomplished using periodic observations of deep space along with a precise knowledge of the instrument cavity temperature. Use of the instrument temperature itself as a calibration point was necessitated because the large size of the telescope made the use of an on-board blackbody target impractical.

The calibration is carried out for each wavenumber interval independently using I = B*(Tinst)*(C2 C1)/C2 where I is the calibrated planetary radiance, B*(Tinst) is the Planck radiance at the instrument temperature Tinst, and C2 and C1 are the spectral amplitudes measured while observing the planet and deep space, respectively. This calibration technique requires that the interferometer and all elements within its field of view, including the telescope mirrors, apertures, and baffles, be at precisely the same temperature. This condition is insured through the use of thermostatically controlled heaters. To minimize systematic errors from possible small changes in the instrument responsivity, the calibration must be updated as a function of time during the encounter.

During an encounter, the deep space spectra for each day were averaged and a time corresponding to the average time was assigned to each average. In addition to the daily averages, a grand average of all deep space spectra was calculated. For each day a ratio spectrum consisting of the daily average power spectrum divided by the grand average power spectrum was used to scale the grand average instrument response to obtain the daily response. Individual spectra were then calibrated using a linear interpolation. The calibrated spectral radiances are expressed in Watt/cm**2/sr/cm**-1.

Radiometer calibration consists of a verification of instrument stability by repeated determinations of t(target plate), based on observations of a diffusely scattering target plate mounted on the spacecraft. The calibration conversion to Watts at the detector takes into account the detector response and electrical gains. Observations of the target plate were carried out before and after each encounter with the exception of after the Voyager 2 Saturn encounter when jamming of the instrument scan platform caused the maneuver to be aborted.

The long term behavior of the target calibration data for the two IRIS radiometers is presented in the following tables. The observed target signal, corrected for offsets, and normalized for changes in heliocentric distance, has remained constant within quantization uncertainties throughout the mission. In the absence of compensating changes in the target plate and the instrument, this implies excellent radiometer stability. The arithmetic average of all target measurements, with a 1.5% uncertainty, is adopted as the normalized target signal for purposes of absolute calibration: 30960 +/- 460 DNi x AU**2; this differs slightly from the published value 31152 +/- 312 DNi x AU**2 [PEARL&CONRATH1991]. The target signals for the integrating, x8 sampling, and x1 sampling radiometer data are in the ratio 24:8:1.

Table 2. Target plate calibrations - integrating radiometer
         (integrating:high_gain:normal_gain=24:8:1)

                             VG1 Jupiter     VG1 Jupiter
Calibration Quantity         Pre-encounter   Post-encounter
____________________________ _______________ _______________
Space before (DN)            9.5 +/- 1.0     17.0 +/- 1.5
Space After (DN)             9.0 +/- 0.5     16.0 +/- 0.5
Space weighted mean (DN) [1] 9.1 +/- 0.4     16.1 +/- 0.5
Raw target (DN) [2]          1157.5 +/- 5.8  1029.5 +/- 5.1
Adjusted target (DN) [3]     1148.4 +/- 5.8  1013.4 +/- 5.2
Distance to Sun (10e+8 km)   7.67024         8.10236
DN x AU**2                   30190. +/- 150. 29730. +/- 150.


                             VG1 Saturn      VG1 Saturn
Calibration Quantity         Pre-encounter   Post-encounter
____________________________ _______________ _______________
Space before (DN)            19.5 +/- 0.5    16.25 +/- 0.25
Space After (DN)             17.5 +/- 0.5    16.5 +/- 0.5
Space weighted mean (DN) [1] 18.5 +/- 1.0    16.3 +/- 0.2
Raw target (DN) [2]          415.5 +/- 2.1   328.5 +/- 1.6
Adjusted target (DN) [3]     397.0 +/- 2.3   312.2 +/- 1.7
Distance to Sun (10e+8 km)   12.93381        14.66979
DN x AU**2                   29680. +/- 170. 30020. +/- 160.


                             VG2 Jupiter     VG2 Jupiter
Calibration Quantity         Pre-encounter   Post-encounter
____________________________ _______________ _______________
Space before (DN)            59 +/- 2        49.0 +/- 0.5
Space After (DN)             51 +/- 2        48.0 +/- 0.5
Space weighted mean (DN) [1] 55 +/- 4        48.5 +/- 1.0
Raw target (DN) [2]          1315.0 +/- 6.6  992.5 +/- 5.0
Adjusted target (DN) [3]     1260.0 +/- 7.7  944.0 +/- 5.1
Distance to Sun (10e+8 km)   7.43461         8.60397
DN x AU**2 [4]               31120. +/- 190. 31230. +/- 170.


                             VG2 Saturn      VG2 Saturn
Calibration Quantity         Pre-encounter   Post-encounter
____________________________ _______________ _______________
Space before (DN)            51.0 +/- 1.0          [5]
Space After (DN)             47.0 +/- 1.5          [5]
Space weighted mean (DN) [1] 49.0 +/- 2.0          [5]
Raw target (DN) [2]          455.5 +/- 2.3         [5]
Adjusted target (DN) [3]     406.5 +/- 3.0         [5]
Distance to Sun (10e+8 km)   13.08739              [5]
DN x AU**2 [4]               31110. +/- 230.       [5]


                             VG2 Uranus      VG2 Uranus
Calibration Quantity         Pre-encounter   Post-encounter
____________________________ _______________ _______________
Space before (DN)            52.0 +/- 0.5    53.0 +/- 0.5
Space After (DN)             51.5 +/- 1.0    53.0 +/- 0.5
Space weighted mean (DN) [1] 51.9 +/- 0.4    53.0 +/- 0.5
Raw target (DN) [2]          143.5 +/- 0.7   138.0 +/- 0.7
Adjusted target (DN) [3]     91.6 +/- 0.8    85.0 +/- 0.9
Distance to Sun (10e+8 km)   27.34484        28.88645
DN x AU**2 [4]               30610. +/- 280. 31690. +/- 320.


                             VG2 Neptune     VG2 Neptune
Calibration Quantity         Pre-encounter   Post-encounter
____________________________ _______________ _______________
Space before (DN)            53.0 +/- 0.5    52.5 +/- 0.5
Space After (DN)             53.0 +/- 0.5    52.5 +/- 1.0
Space weighted mean (DN) [1] 53.0 +/- 0.5    52.5 +/- 0.4
Raw target (DN) [2]          95.8 +/- 0.5    86.3 +/- 0.8
Adjusted target (DN) [3]     42.8 +/- 0.7    33.8 +/- 0.9
Distance to Sun (10e+8 km)   39.6584         45.2993
DN x AU**2 [4]               30040. +/- 500. 30940. +/- 800.

____________________________________________________________
Notes:
[1] Where the difference between pre-encounter and
    post-encounter measurements of space exceeds the random
    baseline variation, the average is taken as the
    arithmetic mean, with the difference as the uncertainty.
[2] Considers the larger of 0.5% (estimated maximum possible
    nonlinearity) or the baseline variation as the
    uncertainty.
[3] All uncertainties are considered uncorrelated in
    calculation of the adjusted target signal.
[4] For Voyager 2 only, the adopted mean value of 30960 +/-
    460 is entered for all cases in the calibration files.
[5] For Voyager 2 Saturn post-encounter, the instrument
    platform jammed so the calibration maneuver was aborted.

Because the spectral transmission of the radiometer is not flat across the bandwidth, further calibration requires knowledge of the relative spectrum of each object observed. The absolute reflectance of the Voyager 1 and 2 target plates are 0.502 +/- 0.018 and 0.497 +/- 0.018, respectively. A detailed description of a technique for calibrating the radiometer is given in [HANELETAL1981A].

Back to the index

Instrument Detectors

The detector of the broadband reflected solar radiometer is an eighteen- junction thermopile with a 3 second thermal time constant and is operated at a temperature of 200 K. The thermopile used with the broadband radiometer has a noise equivalent power of 4E-10 Watt/Hz**1/2.

The detector of the infrared interferometer is a low-impedance Schwartz-type four-junction thermopile with a thermal time constant of 12 millisecond. and is operated at a temperature of 200 K. The thermopile used with the interferometer has a noise equivalent power of approximately 2E-10 Watt/Hz**1/2.

Back to the index

Instrument Electronics

The bulk of the analog and all of the digital circuitry is located in an electronics module separate form the interferometer module. The power supply is located in still another box. Only elements that had to be close to detectors to minimize noise, such as the sensors for temperature measurements and controls, were located in the optics module. The electronics box contains timing and control elements, mirror drive circuitry, housekeeping monitors, thermal controllers, analog-to-digital converters, and spacecraft interface circuits. The power module contains one primary and two auxiliary power supplies that convert spacecraft primary ac into dc voltages required to operate the instrument. One of the auxiliary supplies is always on to maintain the instrument at the proper temperature. The primary power supply is on only when the instrument is on during tests, calibrations and observation periods during planetary encounters. Because of the requirement to pass through the Jovian radiation environment, hardened electronics components were used where possible. The bulk of the instrument electronics consists of radiation- hardened integrated circuits. A description of the electronics, along with a block diagram, is given in [HANELETAL1980A].

Back to the index

Instrument Optics

The Cassegrain telescope has a parabolic primary mirror 50 cm in diameter with a 7.62 cm hyperbolic secondary mirror, and an effective focal length of 303.5 cm and an F number of 6.07. A dichroic mirror divides the beam from the telescope into two separate beams, one going to the infrared interferometer and the other to the broadband reflected solar radiometer. The telescope mirrors and support structure, like most of the other mechanical parts of the instrument, are made of optical-grade beryllium. The primary mirror is gold coated over a nickel plating. The secondary mirror has an aluminum coating over the nickel plating, and this in turn is overcoated with silicon monoxide for protection.

Back to the index

Instrument Offset

The instrument and the associated electronics and power supply modules are bolted to the scan platform. The telescope is approximately bore-sighted with the wide and narrow angle television cameras, and with the PPS and UVS instruments.

Back to the index

Instrument Parameters

The following instrument parameters are measured.

Sampled Visible Radiance - Series of 8 radiometer samples taken during a 48 second data frame with the high gain channel. The quantity given is power at the detector in Watts. (-1.0 indicates data off the planet).

Integrated Visable Radiance - The broadband, reflected solar radiometer signal integrated over the 45.6 seconds that IRIS data are taken within the 48 second data frame. The quantity given is power at the detector in Watts.

Thermal Radiance - Radiance (W cm-2 ster-1) within a 4.3 cm-1 spectral interval.

Back to the index

Instrument Modes

The instrument possesses only one operating mode. When turned on the instruments acquires one interferogram every 48 second data frame. In addition data from the reflected solar radiometer are also obtained every data frame. The latter consist of a measurement integrated over 45.6 seconds as well as 8 samples of both high and normal gain measurements distributed at 6 second intervals throughout the data frame.

Back to the index

References

Hanel, R. A., D. Crosby, L. W. Herath, D. Vanous, D. Collins, H. Creswick, C. Harris, and M. Rhodes, Infrared Spectrometer for Voyager, Applied Optics, 19, 1391-1400, 1980.

Hanel, R. A., B. J. Conrath, L. W. Herath, V. G. Kunde, and J.A. Pirraglia, Albedo, internal heat, and energy balance of Jupiter: preliminary results of the voyager infrared investigation, J. Geophys. Res., 86, 8705-8712, 1981.

Pearl, J. C., and B. J. Conrath, The albedo, effective temperature, and energy balance of Neptune, as determined from Voyager data, J. Geophys. Res., 96, 18, 921-18, 930, 1991.

Back to the index