2. System Description
2.1 Optics/ Structure2.1.1 Telescope
WUPPE is a 0.5-m Cassegrain telescope. The telescope consists of the primary mirror, telescope tube, secondary mirror, and sunshade. The front of the tube is closed by the aperture door, which also holds an ultraviolet test lamp on the inside. The backend of the telescope, i.e. behind the Cassegrain hole, is where the detectors are mounted. The light passes through the aperture/ filter wheels to the zero-order detector and the spectrometer detector. The WUPPE electronics is also located at the rearend of the structure on the radiators. Figure 2.1-1 presents an overview of the WUPPE components, which will be discussed in considerable detail below.
Figure 2.1-1 : WUPPE structures overview.
The WUPPE telescope is a classical Cassegrain, its optics is depicted in Figure 2.1.1-1. The telescope primary mirror is a 0.5 m F/3 parabola, and the secondary mirror provides an F/10 beam, for a focal-plane scale of 41.3 arcsec/mm. The good image field is approximately 10 arcmin (15 mm). The mirrors are coated with aluminum and overcoated with magnesium fluoride, optimized to enhance reflectivity at 2000 Å and to suppress Lyman Alpha.
Figure 2.1.1-1 :The telescope optics.
WUPPE Instrument Characteristics
Telescope Aperture 0.5 m (net area 1800 cm2) Offsetting FOV 20 armin diameter Focal-plane scale 41 arcsec/mm (F/10) Images 0.3 arcsec (center), 1.1 arcsec (10 arcmin offset) Spectropolarimeter Focal-plane scale 27.5 arcsec/mm (F/15) Zero-order images 2 arcsec (acquisition), 1 x 5 arcsec (observing) Field of View (FOV) 3.3 x 4.4 armin (in locate beam) spec aperture (in observe beam) Detector S-20 and CCD Sensitivity 0 to 15 mag/ 1sec Spectrum Images < 0.7 x 6 arcsec Dispersion 78Å/mm, blaze at 1900Å Range 1350Å to 3250Å Detector CsTe and dual Reticon Resolution 5Å FWHM (stellar) FOV 50 arcsec perpendicular to dispersion Dynamic range 1000 (one exposure) S/N limit 500 (photon limited) Photon rates 5 x 10-2 < F < 2 x 10-4 detected photons/ sec x Å Sensitivity (system peak Q.E.) 5 % (estimated) Polarimeter Accuracy P > 0.03 % (photon limited) Lyot mode (Q,U,V) resolution 40Å (all photon rates) Wave plate mode (Q,U,V) resolution 5Å (F > 0.1 photons/ sec in line)
Figure 2.1.1-2 : Expected WUPPE integration times (note that a half orbit of the shuttle, i.e. one pointing, equals to about 2700s).
The WUPPE structure is 3.8 m (149 inches) long and fits within a 91 cm square cross-section envelope. It weighs 360 kg (800 lbs) and is fabricated almost entirely of aluminum. The main structure is acylindrical telescope tube, 1.7 m long by 0.6 m in diameter, and closed at the rear end by the square primary-mirror bulkhead and at the front by the aperture door. The telescope is supported at two points on the primary bulkhead and one point on a hardpoint ring just behind the secondary-mirror drive-ring at the front of the telescope tube. The aluminum honeycomb sunshade attached to the aperture-door jamb eliminates scattered light from neighboring instruments on the IPS and provides radiator surface. Baffles are located on the inside of the tube, on the secondary mirror, and in the hole of the primary mirror. The spectrometer and filter/ aperture wheel are mounted on the rear of the primary bulkhead. The WUPPE electronics are mounted mainly on two radiators attached through thermal isolators to the primary bulkhead. The WUPPE structure is descibed in detail in section 2.2. The optical characteristics of the WUPPE telescope are summarized in Table 2.1-1.
2.1.2 Mirror-Offset Mechanism
The secondary mirror may be articulated by means of stepper motors attached to the three secondary vanes. Driving the motors in the same direction provides an on-orbit refocus capability; the maximum typical focus correction is 1 mm. Driving the motors in opposite directions provides pointing trim and offsetting capability with one step corresponding to 0.15 arcsec at the focal plane. The mechanism is shown in Figure 2.1.2-1.
Each 200 step/revolution motor drives a "harmonic drive" gear system with 80:1 reduction; this drives a cam with a linear ramp. The end of the secondary vane rides on a cam follower, and a linear potentiometer senses the position of the cam follower. The motors are driven simultaneously at a constant rate of 333 steps per second. One step of the motors yields a motion of 0.913 Ám, and there is a mechanical range of 14 mm. The linear pots change by 1 bit per 103.4 Ám. The centered position of the mechanism is aligned to place an on-axis object 4 arcmin off-axis in the pitch direction at the focal plane, halfway between the on- and off-axis apertures; this insures the maximum offset travel available from both apertures. The position of the motors is tracked as the number of steps away from an initial position determined from the linear pots. All motor motions
Figure 2.1.2-1 : Mirror-offset mechanism.
are pre-checked by comparing the predicted pot positions with the known readings at the mechanical limits; motions beyond the mechanical limits are not harmful, but knowledge of precise mirror position is lost. The maximum practical offset is 15 arcmin (22 mm), set by the decollimation image degradation: telescope images are approximately 0.3 arcsec on-axis and 2 arcsec with a 15 arcmin offset. With a 1 mm focus travel, the guaranteed offset range from both apertures set by the mechanical limits is 15 arcmin in the yaw direction and 11 arcmin in the pitch direction.
2.1.3 Image-Motion Compensation
An Image-Motion Compensation Actuator (IMCA) system is incorporated to improve the effective quiescent and disturbed pointing excursions of the IPS. Pointing errors with frequencies of 0.1-10 Hz are sensed by the Astro-IMCS system and corrected at the WUPPE secondary mirror by an IMC actuator system. Lower freqency collimation drifts are compensated using the mirror-offset mechanism described above. An IPS/IMCS system pointing stability of 0.7 arcsec is required during exposure periods. Stability of 2 arcsec over 30 minutes and 10 arcsec absolute pointing accuracy is expected.
Figure 2.1.3-1 : WUPPE Image Motion Compensation Mechanism.
The IMCA system (Figure 2.1.3-1) consists of two Piezo-Electric Transducers (PZT's) driven by two IMCA High-Voltage Power Supplies. These cause the secondary mirror to rock against a universal pivot. All are preloaded in compression to withstand launch loads. Two mirror position feedback Linear Voltage Displacement Trancducers (LVDT's) complete a hardwired feedback loop which stabilizes the mirror in pitch and yaw. The travel range of the PZT system is 11.4 arcsec (in the telescope focal plane) in both directions, with one bit of correction giving 0.09 arcsec.
The spectrometer is modelled after a Monk-Gilleson design in which a 600 line/ mm plane reflection grating is illuminated by a converging bundle from a spherical relay mirror. The coma introduced by having a grating in a converging beam is compensated by using the relay mirror off-axis. Figure 2.1.4-1 illustrates the principle of the Monk-Gilleson design, while Figure 2.1.4-2 shows the spectrometer optics.
Figure 2.1.4-1 : Spectrometer principle.
Figure 2.1.4-2 : Spectrometer optics.
The grating is blazed in first order for maximum efficiency at 2000Å, with measured values of 20, 65, and 40 % at 1400, 2000, and 3000Å, respectively. The spectrum is split perpendicular to the dispersion into two parallel orthogonally polarized spectra by a Wollaston beam-splitting prism made of MgF2 to give UV throughput, and the two spectra are recorded simultaneously by the spectrometer detector. At the spectrometer focal plane the beam is F/15: the spectral dispersion is then 77.6Å/mm and the scale in the spectrometer focal plane is 28 arcsec/mm. The spectropolarimeter has a spectral range of 1400-3300Å, limited at the short end by the MgF2 prism and the grating blaze function, and at the long end by the detector.
2.1.5 Zero-Order Image
Field identification, offsetting, focussing, and pointing trim are accomplished by relaying the zero-order image of the spectrometer grating to a blue-sensitive zero-order detector (ZOD) camera (see section 2.1.7). Pictures from the ZOD camera are stored in Dedicated Experiment Processor (DEP) memory and relayed to the aft flight deck CCTV.
For field verification and offset procedures the zero-order of a 3.3 x 4.4 arcmin locate beam, displaced 8 arcmin off-axis in pitch (parallel to the spectrometer dispersion) in order to miss the beamsplitter, is relayed to one half of the ZOD camera by a flat. To use this beam the telescope secondary mirror is articulated to place the area of interest into the locate beam. Then a ZOD picture (2 arcsec images) is assembled by the DEP and relayed by CCTV to the crew.
For focus and pointing trim operations the zero-order of the on-axis observing beam is relayed to the other half of the ZOD. The central 1 mm (30 arcsec) of the image is available to provide an oversampled image shape which is centroided to provide pointing errors once per second and a CCTV image for instrument performance verification. The focus of the telescope/spectrometer assembly is adjusted by minimizing the zero-order image size (0.8 arcsec along the dispersion and 5 arcsec perpendicular to the dispersion) in the direction of the dispersion. Since the spectrometer and zero-order observing images are detected simultaneously, pointing monitor information is available during spectrometer exposures, allowing for data culling during post-mission analysis.
2.1.6 Aperture/ Filter Wheels
An aperture wheel followed by a polarimetric analyzer filter wheel at the telescope focal plane select the spectrometer observational modes. The 16-position filter and aperture wheels (see Figure 2.1.6-1) are driven by 24 step/revolution stepper motors under control of the DEP. With a 25x spur-gear reduction, the distance around a wheel is 600 steps, and the distance between positions is 37.5 steps. The motor rate is 142 steps/sec. The wheels may be driven an arbitrary number of revolutions in either direction; the shortest distance between two positions is chosen.
2.1.6-1 : Aperture/ Filter Wheel Mechanics.
The wheel position sensors each consist of five photodiode/phototransistor pairs mounted on opposite sides of an encoder cup with the following hole pattern: one photodiode (the fiducial) has a small hole at each position and four have larger holes in a binary code. When the fiducial is not seen (between positions) the displayed position will be "99"; otherwise the binary code 0-15 is displayed. Both wheels are fine-detented by permanent magnets; the wheel motor power is left off except during motions.
The available observing apertures are listed in Table 1 of the Payload Flight Data File (PFDF). An MgF2 Rochon polarizer occupies position zero. A star illuminating this aperture provides a 100 percent linear polarized signal at a known position angle, and small offsets in yaw (perpendicular to the dispersion) allow a simulation of the degraded resolution obtained through the wider apertures. A 40 arcsec diameter on-axis aperture provides a setup position before observing. A stellar (4 arcsec hole) and various diffuse (1.3-50 arcsec rectangles) apertures are provided for observing. The spectral resolution for stellar objects is 5Å, limited by the imaging; spectral resolution for diffuse objects is limited by the slit width. An alignment target occupies the final position. Most on-axis aperture positions also have corresponding 3.3 x 4.4 arcmin off-axis apertures. The off-axis hole in aperture position 1 is used for acquisition in the nominal mode. The off-axis apertures in the other positions allow for guiding on off-axis stars while observing faint or diffuse objects in the on-axis apertures. To prevent contamination of the ultraviolet spectrum with light from the off-axis holes, glass filters limit off-axis light in the spectrometer to the visible (>3500Å). Finally, an off-axis hole which filters out light shortward of 5000Å is provided for finding red objects.
184.108.40.206 Polarimetric Analyzers
The analyzer filters (also listed in Table 1 in the PFDF) are used to modulate the polarimetric Stokes parameters before passing through the spectrometer beam-splitter. One type of analyzer is a "Lyot"-type filter which consists of a stack of MgF2 thick wave plates (see Figure 220.127.116.11-1). This filter modulates the three Stokes parameters in frequency, thus relieving the need for rapid time modulation. This mode is used for objects fainter than about 11th magnitude, where the spectrometer detector must have long integration times in order to overcome readout noise. Lyot analyzer #1 allows a maximum polarimetric resolution of about 18, 36, and 80Å at 1400, 2000, and 3000Å, respectively, simultaneously with spectrophotometry with 5Å resolution. A spectral resolution at least 5 times better than the Lyot polarimetric resolution is required in order to resolve the polarimetric modulation pattern. For diffuse objects, therefore, use of a wide slit requires analyzers with lower polarimetric resolution: Lyot analyzers #2 and #3 give polarimetric resolutions 0.86 and 0.44 times that of #1. Also, placing the filters at 45 degrees (positions 1, 3, and 5) reduces the polarimetric resolution by another factor of two (allowing a doubled slit width), at the expense of giving up one of the linear Stokes parameters.
For bright objects, a rotating wave-plate analyzer is provided by filter positions which contain wave plates at various angles (see Figure 18.104.22.168-2). These obtain one Stokes parameter at
Figure 2.1.6-1 : The Lyot plates.
Figure 22.214.171.124-2 : The half-wave plates.
a time, at a resolution comparable with the spectral resolution. The rotating-plate mode involves a modulated sampling (one spectrum per 16 seconds) through orthogonal pairs of wave plates, which are located in neighboring filter positions. Filter positions 6 and 7 obtain the Stokes parameter corresponding to linear polarization parallel or perpendicular to the spectrometer dispersion. To obtain full position angle information for linear polarization requires in addition the 8/9 and 10/11 filter pairs.
Circular polarization is obtained with the 12/13 filter pair.
2.1.7 Zero-Order Detector
The ZOD (Figure 2.1.7-1) is an intensified RCA (SID501) 320 x 512 two-dimensional CCD array. The intensifier is an ITT (F 4111) 18 mm proximity-focussed channel intensifier with P-20 phosphor, quartz faceplate, and S-20 photocathode for blue-light sensitivity. The array is uncooled, and is typically read out every 1 sec. A dynamic range from 1-16 is obtained by gating the image tube for exposures of 1-500 ms and varying the tube gain from 10 to 20000.
Figure 2.1.7-1 : The Zero-Order Detector box.
2.1.8 Spectrometer Detector
The Spectrometer Detector (SPD) is an intensified Reticon 1024 S dual self-scanned array (see Figures 2.1.8-1 and 2.1.8-2).
Figure 2.1.8-1 :The spectrometer-detector box.
Figure 2.1.8-2 :WUPPE Detector Layout, conceptional details.
The two parallel spectra, 25 mm long and separated by 3.5 mm, are detected simultaneously by the two Reticon arrays. The Reticon is preceded by an ITT F 4112 25 mm proximity-focussed channel intensifier with a MgF2 faceplate and CsTe photocathode. The intensifier output, a P-1 phosphor, is optically coupled to the Reticon through a fiber optic bundle. The spectrometer spectral resolution is limited by the intensifier resolution to 50 Ám (5Å, 1.3 arcsec, or 2 Reticon pixels), and the field of view perpendicular to the dispersion is limited by the Reticon pixel length and spectrometer images to about 50 arcsec. The Reticon is operated in an analog integration mode; with an intensifier gain of 20000 and Reticon exposure times of 0.1 to 256 seconds, input photon noise will be larger than the readout noise for stars from 0-18, assuming a system peak quantum efficiency of 4 %. For diffuse objects in a 100 squarearcsec aperture, this corresponds to a faint limit of 23 mag/arcsec (continuum) or 1 Rayleigh (line). Thermal noise in the Reticon is suppressed by cooling to -45°C by a four stage Thermoelectric Cooler (TEC) and heat-pipe.
2.1.9 Test Lamp
A Test Lamp (Figure 2.1.9-1) is located in the aperture door. The lamp is a Deuterium/ Molybdenum hollow cathode, providing a continuum from 1400 to 7000Å with a peak at 2200Å. The lamp beam is condensed and collimated into a 5-cm beam, illuminating the primary mirror near its edge. The image is about 10 arcsec, and the collimation repeatability is about 1 arcmin, set by the repeatability of the door closure. The lamp is about 5th magnitude in the UV and 7th magnitude in the visible. To observe the test lamp, the door has to be closed and latched. Test Lamp observations are carried out during ground-based testing and alignment.
Figure 2.1.9-1 : Test-Lamp Optics.