The Collimator Blank was ordered this quarter, and is scheduled to be delivered to Kodak for generation in mid July. We expect to receive it in Santa Cruz by mid August. The mechanical design of the collimator cell is progressing and we had a design review in early June. The final design is in progress and we expect to start fabrication of the cell in September.
Design of the grating slide and slit mask handler is also progressing. We have adopted a linear slide design for the gratings capable of holding two 8x12 inch gratings, one 6x8 inch grating and a 7x9 mirror. Both of the 8x12 openings will also be capable of handling 6x8 gratings.
A prototype slit mask handler is currently being fabricated to evaluate the proposed mechanism. We have received quotes to cut demonstration slit masks and expect to award a contract to two separate sources to be able to evaluate the quality of both of them. It is apparent that we will not likely be able to get the edge smoothness we originally specified (0.00015 inches) and are currently evaluating the effect of loosening up this requirement. Another possibility being investigated is using a high rotor speed NC router to make the slits (26,000 RPM). We can get tools that are small enough, and there are machines available that may be adequate for our needs. A slit mask was made using our NC machine that had the best slits we have seen to date (edge smoothness 0.0002 inches) but required considerable manpower and time. A slightly different machine may be able to reduce both considerably.
Work has started on the design of the Tent Mirror support and flexures. Prototypes have been fabricated and are being tested. We ordered a Piezo electric translator from Physik Instrumente to drive the Tent Mirror and expect to have delivery in the current quarter. A prototype support and drive system will be fabricated for tests.
Design of the structure is progressing. The drive disk design has been changed to one which is two inches thick to allow for proper contact stresses at the drive and encoding wheels. This considerably stouter disk will also allow the disk to become the backbone for mounting the grating slide and slit mask handler, rather than having to mount to the shell. Frank Melsheimer continues in his role as consultant, giving advice during the development of the various designs.
Terry Ricketts returned to Lick for a short time while we conducted the search for a new Electrical Engineer. During this period, Terry largely completed the design of the servo and electrical control functions in the spectrograph. We purchased one complete control system from Galil which we are currently testing using the HIRES Image Rotator. Initial tests indicate it works well. The system is an updated one from the HIRES design, and is used with a terminal server. It will allow us to have all communications to the instrument through an ethernet line rather than multiple conductors. This will greatly simplify the cable's into the instrument. Generation of a test boule of CaF2 has begun. This boule was lent to us for optical tests, particularly coupling to other optics. The boule has a 10.8" diameter. The test program is being conducted by Harland Epps.
Steve Allen is working on the image storage systems, and has contacted several sources of information. He will be putting together a package of materials for review. For the overall software effort, we have decided to delay the Software PDR until January 1996. This will be the earliest time that the scientists working on the specifications for much of the data acquisition, display and processing software will have time to accomplish this task. Work on other aspects of the Software effort is continuing with Steve Allen committing a significant part of his time to this. The instrument control software is very similar to the software used on other instruments, and work will begin on it as soon as the schedule allows.
2. Status of Risk Items:
2.1. Flexure Control System:
We have ordered a FC translation system from Physik Instruments which will be delivered during the current quarter. A prototype flexure pivot has been fabricated for evaluation, and we expect to fabricate a complete prototype of the Tent Mirror support with translators for testing in the fall quarter.
2.2. Optics Fabrication:
Nothing new to report this quarter on high-risk items.
2.3. Slit mask Handler:
We are currently constructing a prototype of the handler. It will be installed on a large wooden drive disk, which was in turn installed onto the rotating model structure (96" diameter x 2" thick). This prototype is not exactly identical to the one we expect to design for the instrument, but is intended to allow us to experiment with how the slit masks are handled on the chain drive and how the transition from the handler to the focal plane will work. It is a ten position device. A large counterweight will be attached to simulate the weight of the second side handler.
2.4. Detector Mosaic:
The CCD lab has test bonded a silicon chip to an Aluminum Nitride wafer with an epoxy sheet at an elevated cure temperature (95 C). The bonded assembly was then mounted in a dewar and cooled to -120 C. At -120 C it had a deviation from perfectly flat of 9 microns. This is getting close to what is required for DEIMOS (10m), and it probably can be improved on by varying the bonding technique.
2.5. CCD Data handling:
Work has started on image storage, and this will be reported on in detail in the Software section.
2.6. Total Software Burden:
Steve Allen is now spending a significant amount of his time on DEIMOS. It is concerning from a scheduling point of view that the Software PDR has had to be delayed into 1996. We are investigating hiring a 1/2 time Research Astronomer experienced in observing with multi-slit spectrographs to assist with the DEIMOS software effort, particularly with respect to requirements definition, multi-slit astrometry, and instrument calibration and testing.
2.7. Slit mask Cutter:
We have requested quotes from three firms who commercially laser cut materials. We have also asked a supplier of an NC router to cut us a mask. If the router works, it has the advantage of being less than a third the cost of the Laser cutter, and uses less "hazardous" technology.
2.8. Total Budget Overrun:
We are continuing to avoid increasing the instrument capabilities and tightening the specifications, meaning that the budget projections in this quarter remain stable.
3. Reports on Specific Areas:
3.1. Optics:
Equipment that had not been received last quarter has now arrived. This includes the two Strasbaugh polishing machines, the Zygo f/.65 Transmission Sphere for the interferometer and the Mildex spherometer. Additionally, interferometer shims have been fabricated by the Instrument Lab to facilitate height adjustments for on-axis testing of the larger camera lenses.
An order has been placed with Schott Glass Technologies for the 46.3" diameter collimator mirror blank. The Zerodur blank will be shipped from Schott/Germany to Eastman Kodak in Rochester, New York in mid-August, where they will diamond-generate the concave radius of curvature and machine an annular flat on the front face perpendicular to the optical axis for ease of alignment in the instrument. We expect to receive the mirror blank from Kodak in mid-September for grinding, polishing and aspheric figuring here.
All of the optical glass has been ordered from Ohara Corporation, and the order is progressing normally. Delivery is expected sometime after November 1995. This could be delayed by six months or more if a melt failure occurs. This is not expected to happen; however we must realize that the pieces we have ordered are very large and thick, somewhat more extreme than any that Ohara has ever made in these glass types. We just have to wait and see what happens.
A 10.8" diameter calcium fluoride blank provided to us by Optovac is being processed into a finished lens to be used in coupling and bonding experiments with an existing glass lens, using our proposed Syl-527 couplant material. What we learn in this experiment will be applied to the camera lens bonding of the doublets and triplets.
The collimator-mirror optical design has been finalized and a location for the slit-mask plate some 3.0 inches beyond the nominal telescope focal location has been chosen and verified to be acceptable with regard to anticipated image quality. It has been verified by ray tracing that a 7.0-inch diameter hole in the collimator mirror will not vignette rays at any field location. A hole of this size will provide enough "extra" glass so that anticipated mechanical distortions will not affect the optical image quality. A collimator mirror blank of 4.0-inch thick Zerodur has been ordered. Computer-controlled aspheric machining experiments will be conducted to improve our ability to cut high-amplitude aspherics accurately. Wear rate experiments will also be conducted so as to try to compensate the initial curve such that the correct aspheric shape is attained on the tool just as the aspheric grinding on the glass is completed. This will eliminate the need to recut the tool near the end of the aspheric glass-grinding process.
We are returning the original 14.5" diameter boule of calcium fluoride that Optovac sent us in an exchange agreement with them wherein they will provide us with the shaped lens blanks of CaF2 instead of whole boules. This reduces risk of breakage at Lick during initial shaping of the boules, and avoids the cost of tooling up for these early specialized machining operations. Optovac has the equipment and the experience to do this and has agreed to take the risks.
Some good quality cast iron has been ordered and received for use as aspheric plunge tools. The aspheric tools will be cut in mid-July and tested for stability. The aluminum radius tools for processing the camera lenses are being specified and drawn up prior to releasing the job to the Instrument Lab for fabrication.
3.2. Detectors:
3.2.1 Lick CCD development Effort:
The first wafer run of our new CCD mask set is scheduled to be out of fabrication at Orbit Semiconductor on July 19. The mask set produces on each wafer two 2Kx4K CCDs as well as the 600x1200 CCDs to be used in the flexure compensation system. We will begin wafer probe testing of the wafers as soon as they become available. Two additional wafer runs have been ordered. These POs are on hold at Orbit until we can test the devices from the first run. 3.2.2 Lick Thinning: Work is continuing at UC Davis on the thinning effort. One of the first, and most critical, steps in the thinning process is to bond a handle wafer to the CCD wafer before thinning begins. We have performed an extensive series of tests on various bonding techniques and adhesives. A paper is now in press describing some of our work. We believe we have identified a technique and an adhesive which will give us a void-free bond with uniform thickness.
The next step after bonding the wafers is to mechanically thin the CCD wafer to within about 10 microns of the epi layer (the active CCD layer in the wafer). We are evaluating several commercial suppliers of this type of service. We have sent two sets of bonded wafer pairs to one vendor, and as soon as we can establish an appropriate blanket PO we will send two sets of bonded wafers to a second vendor.
After the wafer is mechanically thinned, we complete the thinning with a chemical etch which will stop at the epi layer. Characterization of the chemical etch rates, etc., is continuing at the UC Davis facility.