DEIMOS
Quarterly Report
Number 23
January
1 to March 31, 2000
The camera was completed and testing is in progress. All the initial tests are very encouraging. A pinhole projected into the camera with a collimator is slightly smaller than the error budget would predict, the back focal distance is correct, and tests with the interferometer show only about 2 waves of spherical aberration.
Tests of the dewar and detector system are in progress. By the end of the quarter we had readout of 7 of the 8 engineering-grade chips on the engineering mosaic. The 8th chip has a problem that will need to be investigated during the next dewar warm-up. We have taken flat fields and images with the engineering mosaic, and at the end of the quarter were starting to do CTE tests in order to determine the optimum operating temperature. The temperature of the dewar varies by about 10° C during a fill cycle, as the cold finger in the LN2 can becomes exposed. This will have to be replaced at some point, but we can live with it for now. More serious is that the ion pump produces a bright glow owing to the fact that the vacuum is not low enough. Work on these and other challenges continues into the current quarter. We plan to have the dewar ready to go into the instrument by June.
The dewar window mounting has been redesigned and tested. The original design for the dewar window called for it to be held in place by RTV. It was discovered that the RTV was outgassing water that was then condensing onto the CCDs. The new design uses an O-ring. By the end of the quarter it seemed to hold vacuum well. It will be tested on the dewar early in the current quarter.
We are preparing to assemble the science mosaic, doing the metrology of the CCDs, molybdenum mounts, and back plane. All the fixturing is now fabricated, and we currently plan to assemble the science mosaic by the end of April.
Testing of the grating system has started, resulting in several minor modifications. Testing is scheduled to continue now through the middle of the current quarter. We expect to complete the last two grating sliders by the end of April.
Testing of the slit mask system is also in progress. We have been able to select and install masks in all gravity directions. The final version of the silt mask form is currently being designed and is planned to be installed on the instrument in May.
We received the fabricated parts for the filter wheel, and it is currently being assembled.
2.1 Optics
The DEIMOS Camera optics were assembled into their cells and the cells assembled into the camera bodies this quarter. We succeeded in getting nearly all alignment errors below 0.001 arc seconds, as required. The camera was then moved to a test cradle on a Newport air table structure and tested in 3 modes. A back focal measurement was taken first using a collimated light source and a measuring microscope. The measurement agreed well with the design value. Several images were then taken with various pinhole sizes at different focus positions and field locations. Again, good agreement with the design image size and quality was seen. Finally, a double pass interferometric test was performed on axis, filling the central 10.2" aperture of the camera. A dominant error of approximately 2 waves of spherical aberration resulted, which was also within the design tolerance of the camera performance.
Additionally, the invar backplane that the moly blocks mount to was flattened, shined, and measured to be flat within a few microns.
2.2 Mechanical
Structure
We mounted the camera mount in the structure and rebalanced the spectrograph to account for this.
The cable wrap was mounted, tested and is ready for cable installation.
Camera
The camera mount was installed into DEIMOS for a fit check. It was then removed. The counterweight for Side 'B' was also installed for a fit check. It, too, was removed.
Filter Wheel
The filter wheel is being assembled and will be ready for testing and installation in the current quarter.
Collimator
The collimator cell was test-fitted into DEIMOS, and the collimator installation fixture was completed and sent out for painting.
Gratings
The grating slide drive was operated under computer control for the first time. Limit switches were installed to prevent damage, but software limits and a fiducial switch have not yet been installed. Several real-time measurements of motor current were made at low speed. The system was adjusted to minimize friction and to balance the gratings with the opposite-moving counterweight. Not all position angles have as yet been studied. The assembly of Slider #3 and Slider #4 (these hold and rotate the 6 by 8 gratings) continued. They are almost ready to be assembled onto the slide. Chain guards for the grating system were fabricated.
The assembly of the grating installation tooling progressed during the quarter.
Slit Mask
Testing of the slit mask system resumed during the quarter; it works well in most rotational positions. The slit mask tends to fall into the form too quickly when inserted from above; modifications are underway to correct this.
The final slit mask form is being redesigned to better match the focal plane, remove vignetting, and provide points to attach the FC fibers. We expect to complete this design and fabricate the new mask form in the current quarter.
Dewar
The Fe55 X-ray source was fabricated.
The engineering mosaic was installed and cooled in the dewar three times during the quarter. We were able to control the temperature of the detectors with the heating resistors on the mosaic. A full set of darks and a set of exposures with the Fe55 X-ray source to measure CTE were taken.
The dewar window mounting of RTV proved to have an outgassing problem, and the mounting was redesigned to include the use of an O-ring to seal the window. The design was tested with a dummy window and works well. The coated dewar window is being removed from its RTV mounting and will have the chamfer ground and polished to accept the O-ring sealing scheme. The new mount will be tested in the dewar in the current quarter.
Several smaller issues have come up in the course of testing, including the lengths of some of the wires in the dewar and an O-ring seal around the FC connector plate. These are being rectified before the science mosaic is mounted in the dewar in the current quarter.
Bright glow is seen in the dewar during “dark” measurements. This comes partly from CCD #8 (which has a glowing amplifier) and partly from the ion pump. The ion pump glow could be serious. We placed a light baffle around the ion pump, which reduced the glow by a factor of 100, but it still can be seen. We are hoping that the glow will decline sufficiently as the dewar outgasses.
CCD Mosaic
A third invar backplane was fabricated and measured in preparation for installation of the science CCDs. It is satisfactorily flat.
A method was perfected to measure the heights of the science CCDs and mounting packages with the microscope. The moly blocks for mounting the CCD packages were measured on the profilometer, and the Instrument Lab machined the pads to place all the mounting surfaces in the correct plane for the package assembly. QA of the mounted CCDs on the packages showed that 6 out of 7 were exactly as expected. The 7th was considerably different than expected and needed to be reworked; it is now OK.
The 8th science CCD was returned to Gerry Luppino for re-bonding of a gold pad to the AlN. This device is expected back in May.
We expect to build the science mosaic in the current quarter.
TV Guider
No progress this quarter, as expected. The drawings exist, and we plan to start fabrication in the current quarter.
2.3 Detectors
The first distribution of MIT/LL lots 9 and 10 occurred on February 18, 2000. There were fifteen devices in this distribution, and Keck's share was four devices (University of Washington received six and Subaru received five). These are all considered science-grade devices. From the standpoint of DEIMOS, the distribution results were very favorable. We had ranked all fifteen devices based on individual CCD characteristics and on DEIMOS needs, and three of the four devices we received were among our top-four-ranked CCDs.
With these four CCDs plus the four science-grade devices DEIMOS already owns, we can proceed to populate the DEIMOS science mosaic with blue-sensitive CCDs. The four high-rho CCDs for the red side of the array will be arriving later.
Since the first of the year we received and tested four more MIT/LL high-rho lot 14 CCDs. Two of these devices (14-7-2 and 14-7-3) are from the second half of lot 14. As reported in the previous Quarterly Report, this half-lot was processed with a proper anti-reflection coating, correcting the bad coating that was applied to the first half of lot 14. Our tests show the AR coating on these devices is good and the measured QE is back to normal. In addition, lot 14 does not show any of the charge-transfer efficiency problems seen in lots 9 and 10. All of lot 14 is now ready for dicing at MIT/LL so we expect to see many more lot 14 devices in the near future. A first distribution of lot 14 devices may occur in late April or early May.
2.4 Software
The DEIMOS slit mask design and database processing specifications were reviewed, corrected, and enhanced. Test implementations of the entire ensemble of FITS tables describing a slitmask were created, exchanged, and interpreted between Steve Allen, De Clarke, and Drew Phillips.
The slitmask milling procedure was revisited. We began talks and correspondence with Greg Wirth of CARA about floor space requirements within the Keck building.
The real-time image capture protocol was explored and documented more fully in an effort to discover concepts which are ambiguous in the context of a mosaic readout. A scheme was outlined for modifying the current image capture code such that it can evolve to be compatible with figdisp, the NOAO mosaic display tool, and DS9 without requiring excess or redundant effort.
Test and evaluation of the DS9 image display software (from CFA Harvard) continued. Negotiations with CFA were completed regarding specification of features needed to meet DEIMOS image display requirements. CFA agreed to add these features into DS9 no later than May 1.
A redesign of the keywords for controlling the calibration lamps was reviewed and a final design selected for implementation.
Work began on the CCD mosaic thermal servo control loop, which should be operational early in Q24 to support tests for determining CCD CTE as a function of temperature.
Images were successfully read from 7 of the 8 CCDs in the engineering mosaic, both with the dewar warm and with it cold. The readout noise from the cold images was approximately 3.5 electrons and did not show any fixed pattern noise. Dark current measurements could not be obtained due to the glow problem inside the dewar.
2.5 Electronics
Work was done on the web-based electronics manual, and upgrades have been made to the shutter controller. The ion pump controllers that failed on ESI have been repaired (they are also DEIMOS units). We purchased and received 90 degree connectors for the dewar umbilical cables. Ion-pump interface wiring has been checked out and works fine.
UPS’s were received and are being installed.
2.6 Flexure Compensation
Design of FC fiber system was started, including the light source, fiber system, and the placement of the fibers in the focal plane. We expect to complete the design and be well along with fabrication in the current quarter.
2.7 Alignment
No report this quarter.
Analysis was completed of a grating reflection problem seen in LRIS and its implications for DEIMOS. The problem is caused by light reflecting off the CCD to the grating, then reflecting again off the grating as though it were a mirror and coming back to focus in the CCD at a point rotated 180 degrees about the center of the detector. This reflection occurs when the grating is approximately perpendicular to the camera axis. In LRIS this occurs primarily with the 830-line grating.
LRIS data confirmed the above model. The intensity of the reflected spots are about 0.5% of the origined intensity, in agreement with rough estimates.
This problem is expected to be worse in DEIMOS, owing to its wider camera acceptance angle, 11.5 degrees versus 6.4 degrees in LRIS. This means that the reflections will be visible over a wider range of grating tilts. Thus, not only will the problem affect the 830-line grating throughout its range, it will also be visible for the 600-line grating at red tilts and the 900-line and 1200-line gratings at the blue tilts. In short, a large fraction of observers will be affected, unlike the situation with LRIS. Additionally, the amplitude of the reflected spots could be as much as twice as bright because the AR coatings on the DEIMOS CCDs are probably not quite as good as those in LRIS.
The brightness of the night-sky OH lines is such that their reflected images will be distinctly visible. If a typical OH line has a peak count of about 2000 electrons, its reflection will have roughly 20 electrons. This is large enough to disturb sensitive measurements. Hence, we foresee that the effort will have to be modeled or otherwise removed. Exactly how to do this depends on how stable the flexure is. The flexure compensation system may become a vital component of the solution.
We intend to request funds from the National Science Foundation to write software to fix this problem. The purpose of this report is to alert the SSC to a significant anticipated operational issue that may require additional programming support to resolve.
[Tables and figures are not available via the web. Please contact Heather (heather@ucolick.org) for more information]
The project budget and spending are summarized in Table 1. Details are shown in Tables 2, 3 and 4. At the end of the quarter we had spent $6,022,634 on the project, or approximately 96% of the approved budget of $6,298,820 million.
Table 3 summarizes the expenditures on manpower. Approximately 77,563 hours of effort have been expended on the project to the end of the quarter, or about 92% of the total budgeted. Figure 1 graphically shows the expenditure of manpower.
Expenditures on materials and supplies are summarized on Table 4 and on Figure 2. By the end of the quarter we had expended $2,423,458 of the $2,533,583 materials budget.
During the quarter $300,000 additional funds were added to the project budget, as approved by the SSC and the CARA board.
The project schedule is shown as Figure 3. A more detailed milestone schedule is shown as Figure 4. The critical path schedule is shown as Figure 5, and the complete schedule is Figure 6. A schedule analysis was presented to the SSC on March 14, 2000, using the October 11, 1999 schedule as a baseline. During that 5-month period, we slipped about one month, including our contingency. Applying this error to the entire schedule predicts a total slip of 15 weeks to the Pre-ship Review, which would place it near March 1, 2001.
The following is a list of milestones for this quarter from the last Quarterly Report, together with the progress made on them:
Milestones for Quarter 23:
1. Complete camera assembly and start testing in the optics tunnel. Complete
2. Start cold tests of the engineering mosaic. Started
3. Complete fabrication of the filter wheel assembly. Complete
4. Complete design of the calibration and FC light sources. Delayed
5. Complete design of the slit mask form. Delayed
6. Complete fabrication of grating slider #4. Complete
7. Complete fabrication of chain guards for the grating drive system. Complete
8. Start testing of the grating system. Started
9. Analyze the LRIS ghosting problem. Complete
10. Finalize CCD height-measuring procedure. Complete
11. Test procedure to tune heights of molybdenum pads. Complete
12. Design and build the Fe 55 X-ray source. Complete
13. Fabricate grating alignment jig to install gratings in cells. Complete
14. Test-fit collimator cell in DEIMOS. Complete
15. Install UPS’s. Started
16. Test the FC signal chain. Delayed
17. Design website. Started
18. Complete mosaic descrambling software. Delayed
Milestones for the next quarter:
1. Complete design of the calibration and FC light sources.
2. Complete design of the slit mask form and focal plane assembly.
3. Complete the grating system and test it.
4. Complete cold tests of the engineering array.
5. Complete assembly of the science mosaic, install in dewar, and test.
6. Aluminize the tent mirror.
7. Complete camera testing and prepare for installation in DEIMOS.
8. Install cables and glycol lines in the cable wrap.
9. Complete fabrication of the TV system.
10. Install collimator in cell.
11. Hold distribution of Lot 14 CCDs.
12. Complete installation of UPS’s.
13. Test the FC signal chain, including CCDs.
14. Complete the website design.
15. Load the cable wrap.
16. Purchase components for the rotation drive system.
17. Order test lamps for the calibration system.
18. Design the fiber feeds for the FC system.
19. Finalize keyword design and operate at least one system under computer control.
20. Finalize FITS header-writing software.