DEIMOS Meeting Minutes Thursday April 20, 1995 NS-143 9:30 am Present: Neal Jern, Terry Ricketts, Barry Alcott, David Cowley, Eric James, Jack Osborne, David Hilyard, Dean Tucker, Deanne Lago, Steve Allen, Sandy Faber, Marlene Couture. Detector Progress The detector project is progressing. A press is being fabricated in the labs which will be used to laminate the wafers. DEIMOS Structure The cylinder design is progressing, the bulkheads are being inserted now. A three dimensional solid drawing has been made with all the beams going through DEIMOS. The drive disk will be placed at Station +13 (re: datums, the origin is located at the center line of DEIMOS, and the focal plane of DEIMOS, which is Datum A-Station 00. Datum B is the center line and the optical/ mechanical axis etc. As we go away from the telescope we are going in a positive Z direction so everything gets larger measured in inches. Station 13 is 13" downstream from the focal plane). We have been defining the stay-out zones and adding structures at various places where they may be needed. The bulkheads will be inserted to hold up the heavy components. Preliminary studies have been done on the collimator mirror cell. We have assigned space for some of the new elec- tronic components that are now onboard the instrument rather than in the electronics vault. This will require a miniature electronic vault that will also live onboard. The cylinder is actually two cylinders (one on each side of the drive disk) that are different diam- eters. The one on the inside of the telescope bearing is smaller than the one on the outside. The only item on the telescope side are the TV cameras. The grating is right next to the drive disk. These are just concepts at this time that will be studied in detail. Ultimately an FEA will be done on any concepts that are determined to be usable. This will be the first instrument with a solid light path model. Slit Mask / Slit Mask Handler / Slit Cutter / Grating Handler A draft RFQ for the slit cutter is being developed by David. Jack has sent a letter to a company in San Jose who is interested in cutting slit masks for us out of 5-mil stainless. They are interested in cutting them forever, they believe they think it would take approximately 1-1/2 to 3 seconds per slit. After his slits are analyzed we will determine if it is feasible to transport them to Hawaii, etc. This may allow us to not purchase the cutter until we are well into development of the instrument or actually after science begins. This method is a departure of our original plan which was to actually cut the masks at the mountain. We are considering a 5 position grating slide (copy of drawing attached). The slit mask handler is a device that lives in the space between the drive disk and the telescope. This grating slide is now approximately 8 feet long which will use some type of caterpillar method. The mirror will be the home position. The two big gratings will need to be closest to the center. This design will no longer fit inside the cylinder of the telescope, it sticks out whether four or five gratings are used. This afternoon we will discuss the number of positions and the shapes, whose concepts will be developed further. Each grating will have its own motor. The slide is non-rigid but once it is at a location there is a set of kinematic mounts that are clamped against it, this will enable it to rotate effectively. Frank Melsheimer's recommendations were reviewed when developing this plan. Electronics Galil has a new 8-channel stand alone controller and by using just two of them we would be able to control the entire instrument. This system is very similar to what was done on the MOS system except that we can get twice as many motors controlled in a smaller box. This will allow us to remote the entire instrument so nothing will need to go into the vault except some things that have to control items not on the rotating instrument, (i.e. the lamps on the telescope, the motor that actually rotates the instrument, etc.). By moving things out to the instrument, we were able to cut down on the amount of heat because there is no longer a large VME chassis with a big CPU inside the vault anymore, just the controller itself. We are also able to eliminate almost all of the control cables going onto the instrument to just one coax. This will be an extreme change from what was discussed at the PDR, but it could turn out to be much cheaper (approximately $23.5K per beam plus associated labor). A thermal enclosure will need to be purchased for the instrument which will be big enough to handle both this controller and the CCD. Terry has designed a board that can plug into some of the digital outputs to provide 8 analog outputs. The work has been partitioned into two sections; Barry will start on the design of the actual motor stages and Terry will work on the overall connection and getting things into the box. The other change being made is that manual controls for each motor will plug directly into the controller but mounted onto the outside of the instrument. Software is designed that once reset it will respond only to the manual buttons until another command is sent from the computer to disable it. A terminal server will be mounted inside the enclosure and the thinnet ethernet will come up directly into the enclosure. RS-232 lines can come straight off from that terminal server into the boxes. If for some reason we can't go through the terminal server we can run the RS-232 down through the cable wrap. Preliminary design has been done to assign the signals to the various pins and a board has also been designed which will be used to mount beside the controller to take some of the digital lines to produce extra digital outputs and eight analog outputs. The design is simple enough now so that by the end of May the electrical design should almost be completed. (Not including the CCDs.) We are trying to have the major part of the instrument designed before the new Electrical Engineer comes on board. There will be an instrument disable switch, when thrown, everything must be done manually (almost like lockout/tagout). Collimator Three materials have been investigated for the collimator mirror: 1) Corning - ULE; 2) Schott - Zerodur; 3) Ohara - E6. To date the best option is ULE from Corning ($50K for a generated blank, plus tax & shipping), slumping would be $30K more. The blank itself is $40K and another $10K to generate it. Kodak has been contacted for a quote to do the generating if we decide to go with Zerodur from Schott. A mechanical cell will be needed for the collimator so it can be tested, so this item will be moved up in the schedule. We are anticipating holding the mirror in the center much in the way Frank Melsheimer recommended. We would like to approach the SSC to gain approval for the purchase of this glass assuming Brian's work shows no problem with the optical design. Camera The camera glass has now ordered within the grace period for the quote provided by Ohara. We have also started the paperwork also for the purchase of the CaF2. The design for the camera bar- rels will be started later this year when Matt becomes involved in the DEIMOS project. Software Some initial investigation has been done to look at how thick each wire needs to be between the components (readout gear, disk drives, observers). Depending on how the outcome of the driftscan analysis, we are considering storing the images as four separate pieces. In the current orientation of the chips all the science from one observation is on only two chips and they are independent of the other chips while doing spectra. This will keep the image size down to only 32mb which means our existing equipment can interact with them in a responsive time (on the right computer). While interacting with direct images we will have only 64mb because only four chips are used instead of eight. Spectral images and direct images will probably have different formats. We should begin planning the Software PDR, who is invited, when will it be, what should be cov- ered. Final documentation should be completed by the first week in June with a PDR date for sometime in the first two weeks of July. A separate meeting will be held next week to further plan for this upcoming event. Schedule David is revising some of the details in the schedule. He is bringing the collimator and grating design forward. He will publish this new version when the draft of the Quarterly Report #3 is published (approximately 1 week). As it stands now this project is still on track to be completed sometime in the second quarter of 1998. The CDR is tentatively set for November 15, 1995. Budget The revised budget has been circulated. It has been revised three times since the last quarterly report: 1) Optical glass funding; 2) Electronics estimate; 3) Fund the assembly and test (previ- ously this time has been included in the design & fabrication allotments). At the bottom of the report is an explanation of why each of these things was done. We need to determine if there are funds enough for the mosaic process. A meeting with Richard Stover, Jack Osborne, Sandy Faber and David Cowley will be scheduled. Prospectus The prospectus is evolving quite well. The campus development office is working closely with us, we are following their format with gusto and our prospectus should be the longest and the most ambitious that they have done to date. This push is being driven by a visit to a gentleman in Idaho who will be shown two UCSC projects for his choice. The prospectus format is a page of text then pictures etc. The text discusses the overall project, gives a summary, UCO/Lick and our work with optics & instrumentation, Keck telescope, the DEIMOS spectrograph drawings, the DEEP science project, budget and schedule. Quarterly Report #3 The draft is in progress at this time, it will be mailed the second week in May.