DEIMOS
Quarterly Report
Number 21
July
1 to September 30, 1999
Element #8 has been received back from Coherent with a good optical coating.
The doublet comprised of elements #1 and #2 has been mounted in body segment #1 of the camera body and the doublet filled with the optical couplant. No leaks are evident.
The CaF2, optical element #3 of the camera, has been mounted in segment #2 of the camera body.
Element #6, which is part of the triplet, has been mounted in body segment #3, and the assembly of the remainder of the triplet is in progress.
Final fabrication of body segment #4 of the camera is about to begin. This is the last body segment that needs to be completed.
We have decided not to build a mosaic of Orbit CCDs, but to use only MIT/LL devices - we currently have 4 MIT/LL devices that could be used in the Mosaic. We expect to have an adequate number to start assembling CCDs onto the backplane by mid-November when they are required. These additional devices would be from the distribution of lots 9 and 10 that is currently being organized.
A new backplane that supports the CCDs in the dewar has been designed and prototyped. Two final backplanes are currently being fabricated so that testing of the dewar system and mounting of CCDs in the mosaic can continue in parallel.
Modifications to the dewar thermal circuit were finalized and parts are being fabricated for a complete assembly and test in mid-October.
The focus and x-stage actuators are currently being bench tested.
The rebuild of the TV/front hatch support structure was completed.
The cable wrap has been completed and is being installed.
2.1 Optics
Camera element #1 and element #2 were mounted into body #1 and sealed with RTV. The bond around element #1 developed several spots of separation, none of which reached from top to bottom of the sealed edge but were a concern. After temperature cycling the spots worsened. We believe the problem is in the primer application method used and have modified it for future elements. We decided to continue bonding element #2, mounting both lenses and testing for leaks of fluid around the bond separations. A special pre-stressing ring designed to maintain positive compressive stress on element #2 down to -20° C worked well. We filled the gap between the lenses with the coupling fluid and have watched for leaks. None have developed as of this report (duration = 50 days).
We mounted the calcium fluoride lens element #3 into body #2. The final installation went smoothly and took about 1 day. This is a singlet with no RTV or optical fluid. However final mounting was complicated by "potato chip" deformation in the lens landing area due to improper machining at Danco. Alterations to the mounting design delayed mounting by almost three weeks.
We continued preparing the procedure for assembling the triplet of element #4, #5 and #6 into body #3 and started that installation late in this reporting period.
2.2 Mechanical
STRUCTURE
The TV/front hatch support structure was modified to locate its attachment points
on the outer part of the main drive disk. This was done to reduce variable forces
on the inner part of the disk, where the grating and camera supports are. The
parts were welded, fit-checked and sent out for painting. They will be installed
next quarter. The slit mask support structure was re-designed and partly fabricated
during this quarter to accomplish the same mechanical isolation. Part of the
TV support structure which holds the grating slide drive still needs modifying,
and part of that same structure which defines and holds the forward skin is
still being fabricated.
We discovered that the main hatch opening was slightly too small to accommodate the CCD dewar housing with X and focus actuators. However, the problem was solved by making small enlargements in the hatch lip area. A full size model was used to help decide how this should be done. The modification will be completed soon.
The cable wrap hardware was finished and installed. It works fine and will be disassembled for painting when the next paint order is set up. The measured travel is 855°, which exceeds the minimum specification of 760°.
After ESI left with all our counterweights, we ordered new lead bricks, cut them to shape, drilled holes in them and had them painted. DEIMOS was then balanced.
A review was conducted of the interface between DEIMOS and the Keck II Nasmyth platform. A height error of about 0.5 inches will be corrected by installing thicker "tires" on DEIMOS.
CAMERA
We test-fitted body #1 to spacer #12 and body #2. The pieces went together well,
meeting both centering and leveling specs. Bladders as received from the manufacturer
had oversized outlet tubing, necessitating rework to the bladder mounts on bodies
#1, #3 and #4. A solution was found, but the rework delayed camera assembly
by nearly two weeks. For other progress on the camera, see Section 2.1.
FILTER WHEEL
The drawings for the filter wheel have been reviewed and sent out for bid. We
expect to contract the fabrication of the filter wheel in October.
GRATINGS
The roller chain sprockets for the grating slide drive were NC-machined by Jeff
Lewis and sent out for cadmium plating. They are back and look very nice. They
now fit the new ISO roller chain for the grating slide drive.
The design and detail drawings for the grating fixtures were sent to Danco for fabrication. These parts were received by us, cleaned up, test fit and sent out for plating. They are now back.
SLIT MASK
A new support structure for the slit mask system was designed and fabricated
as part of the TV/hatch rebuild. The new support moves the mounting location
to the outer part of the drive disk, as for the TV/hatch supports. The SM system
will be installed next quarter.
DEWAR
Work progressed on the extensive redesign of the detector backplane, thermal
standoffs and cooling circuit mentioned in QR 20. The end-to-end specifications
for the system were blocked out including design of procedures to attach the
CCDs to the backplane the thermal spider connection to the underside of the
molybdenum support pods, and the backplane to the focus stage in the dewar.
Designs were developed for a flipping assembly table to hold the backplane during
mosaic assembly and a lucite transparent cover to both hold and protect the
mosaic as it is mounted in the dewar. Fabrication of these parts will continue
next quarter, with the goal of assembling the "engineering" mosaic by Christmas.
The hold time of the LN2 tank was measured to be 50 hours bare, not connected to the dewar. This is shorter than hoped for under no external load and gives urgency to full-scale thermal tests scheduled for next quarter.
CCD MOSAIC
During Q21 we created and tested a microscope facility for measuring the planarity
of individual CCDs and mosaics. In Q22 we will clean this facility and move
it into our clean room for use in the assembly of an "engineering" mosaic. A
detailed task list and procedures for achieving these assemblies is being created.
The fixtures for carrying out these assemblies have been designed and are being
fabricated. This work was greatly helped by a visit by Kirk Gilmore to Hawaii
to observe Gerry Luppino's process for assembling his 8K x 12K mosaic camera.
We experimented with methods of machining the molybdenum block pads that the CCD packages mount to and have developed some strategies to deal with the height and parallelism of the blocks.
TV GUIDER
Our suggestion to CARA that the Photometrics TV cameras used by ESI and DEIMOS
can be operated satisfactorily in shutterless mode was tested at Keck and worked
well on ESI. This mode will reduce wear on the costly and unreliable shutter
mechanism by about a factor of 10.
2.3 Detectors
The Detector Lab continued to test devices from MIT/LL lots 9 and 10. A distribution of the first half of these lots will occur in November ( Did not occur until February ). The latter half of both lots will be delayed by work on high-rho lot 14 devices at MIT/LL. The first lot 14's were tested and found to be generally good but with peak QE of only 60% whereas 80% was expected. MIT/LL traced the problem to improper processing of the AR coating, which had too little O in the TiO layers. Fabrication of lot 14 has stopped while the problem is being investigated.
Tests of the FC devices have started and will continue through next quarter.
2.4 Software
The only DEIMOS-specific software-related activity in this quarter was Allen's completion of the assembly and test of the DEIMOS RAID disk array. This array is now fully operational and connected to the DEIMOS instrument computer (hostname celeste).
During Q21, most software effort continued to be directed towards ESI commissioning, documentation, and providing operational support to CARA staff on a variety of ESI software and hardware integration issues. This effort is finally winding down and should end with the last ESI engineering run in late November. Following that run, we plan no further work on ESI software other than to provide emergency bug fixes for any serious software problems that might arise. The DEIMOS software effort should begin to ramp up in November, with the bulk of our effort directed towards DEIMOS by December.
A portion of the Q21 ESI software effort was relevant to DEIMOS, including upgrades by Allen to the Keck Task Library (shared by all Keck instruments), and related upgrades to the make files and general build procedures. Also, Clarke made significant upgrades to the automated instrument test software which will also be used for the DEIMOS instrument test suite. The portion of the ESI user interface (UI) which displays the current exposure number and disk file names was extensively revised to operate correctly with exposures generated from external scripts, and that UI logic is shared with DEIMOS.
2.5 Electronics
Only a small amount of work on electronics occurred during the quarter, mostly on checkout of the previously fabricated assemblies for motion control. Wiring for the CCDs inside the dewar was ordered and is expected in November.
2.6 Flexure Compensation
FC CCDs are being tested.
2.7 Alignment
No report for this period.
No report for this period.
[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 $5,434,045 on the project, or approximately 91% of the approved budget of $5,998,820 million, leaving $564,775.
Table 3 summarizes the expenditures on manpower. Approximately 66,872 hours of effort have been expended on the project to the end of the quarter, or about 86% of the total budgeted. Figure 1 graphically shows the expenditure of manpower.
Expenditures on materials and supplies are summarized in Table 4 and in Figure 2. By the end of the quarter we had expended $2,362,762 of the $2,533,583 million materials budget.
The project schedule is shown in Figure 3. A more detailed milestone schedule is shown in Figure 4. The critical path schedule is shown in Figure 5 and the complete schedule is Figure 6.
The scheduled date for first star light in the instrument has slipped by about 3 months since the last report. This slip is primarily due to two facts: that the effort on ESI did not end as soon as hoped, and that the development of the detector and dewar system are going to take longer than thought. We will not have a detector/dewar system ready for the spectrograph until May of 00. In the previous schedule we had anticipated completing this in March. Also a little more time has been added to the post pre-ship review period, based on our experience with ESI. (The schedule shown was presented to the SSC in early October 1999.)
Milestones for Quarter 21:
Milestones for the next quarter: