DEIMOS Team Meeting Tuesday May 16, 1995 NS 143 9:30 am Present: David Cowley, Barry Alcott, Deanne Lago, Erich Horn, Eric James, Jack Osborne, David Hilyard, Brian Sutin, Steve Allen, Dean Tucker, Richard Stover, Sandy Faber, Bob Kibrick, David Koo, Marlene Couture, Drew Phillips. Detectors Lincoln Labs has informally started on their CCD development project. No contract has been signed to date because financial commitments still need to be made with all the consortium part- ners. No timescale is in place to determine when we may begin to see devices, however this is a one year development effort so it will be at least six months before anything can be hoped for. Gerry Luppino has inquired via Richard Stover about the packages for the devices and mentioned that he would like some engineering assistance with this process. Gerry however does not have any money to pay for this service so is willing to provide us first pick of whatever devices come out. David will contact Gerry regarding the engineering time needed and Sandy will contact Gerry formalizing this arrangement. Richard has ordered six pieces of aluminum nitrite that are slightly larger than 2K x 4K CCDs (0.080 thick) which is what would be used for the substrate, so that some experiments could be performed on them (i.e. gluing them to a CCD cooling and check the warping, etc.). Orbit's first run is in and should be out in a couple of weeks (two 2K x 4K CCDs). Richard has the P.O. for the next two wafer runs submitted and should now be at campus central purchasing. The first run is standard epi wafers, providing that run is acceptable the next two runs will begin immediately. In the following two runs part of the wafers will be of boron, which will be a new experiment. Orbit is backed up with orders at this time so they have put a hold on R&D for at least six months (high resistivity wafers, etc.). They are willing to do anything that does not require changing their process which would interrupt their production line. This may mean doing a moderately high restivity CCDs with a thicker epi layer. Richard does not want to start that until the results of the second wafer runs are in. We are making process with our work at Davis regarding packaging and thinning. Within the next month or so we will actually thin a wafer with a CCD on it. The lab being built upstairs is going to be changed. Unfortunately it will never be a lab where chemical thinning is done. The Campus EH&S and Campus Facilities people looked closely at what we were planning to do and gave us much higher cost estimate to do it all. Our fume hoods in the mechanical chase in NS-II are not adequate for the chemicals we want to use. Our chemical bench requires 3/4 inch of water pressure at its manifold exit to operate properly. Our building only has 1/2 inch but if the motors were high powered enough (by changing pulleys, blades, etc.), they could develop 3/4 inch of pressure. However, that would in turn unbalance everything else on the stack and an outside contractor would have to be brought in to re-balance the other two dozen items, at approximately $100 each. An approved chemical list for the fume hoods was cir- culated and all of them were ok except hydrofloric acid which is used in the etching process. We are using very low (100/1) concentrations in very small amounts. We were informed that if we go ahead anyway because the amounts were so small they would do periodic inspections and if any deteriorations were found we would be subject to replacing the entire stackin NS-II. We do have the use of the Davis facility and are now pursuing their ok to use it long term. We are planning to proceed with the clean room upstairs but not do the chemical thinning. This will allow us to do the packaging, bonding, gluing of the wafers etc. We will simply transport the wafers to Davis for thinning. Richard is thinking of proposing to the SSC to fund a graduate student to work on the project at Davis (a Davis student). This will tie us in closer to their operation and it also gives us an on sight person. We are also exploring transferring the Intel donated equipment (chemical bench etc.) to them. Richard has ordered a Leach S-bus interface card (which was being considered for use in DEI- MOS) for a different project, this will allow us to test one and see if it will be viable for DEIMOS. Structure/Gratings/Grating Handler We are proceeding with the cylinder approach, however the cylinder on the forward end of the instrument idea has been abandoned. After the gratings and slit mask handler was inserted there wasn't room for one. Most of the items on that end will be attached directly to the drive disk any- way so there will be no reason to have a high integrity/low flexure element out there. The only element out there now will be the hatch which does not pose a concern. The drive disk is now larger and is up to an 8 ft. diameter which is as large as we can get in one piece. This will provide more room to start attaching things to it. (It is actually becoming the "spinal column" of the whole structure.) This is advantageous because we don't have to start directing the forces through the skin and can send them directly to the drive disk. It can then be balanced because the cameras are on one side of the disk and they will balance the items on the other side of the disk. A question was raised regarding how close the hatch can be made towards the focal plane. The closer it is, the shorter DEIMOS will be because it is the only thing out there inside the telescope. The shorter DEIMOS is, the easier it will be to take in and out of the tele- scope. If there is no specific criteria that will define this except the window we will place it as close as possible. We do need to take into consideration however, where the internal comparison lamps shine because that drives how close the hatch can be. The lamp illumination needs to be fairly uniform over the inside of the structure. At the present time DEIMOS is almost outside the telescope so the backward shift is only approximately 5-6 inches. We still need to determine if optics are needed on the lamps, a diffuser is definitely needed. Can we put the lamps in the mid- dle of the field or can they be off to one side. The cost is $1500 for each lamp (each card runs 4 lamps, but the power supply can only run 2) a separate meeting will be held regarding this issue. We also need to decide the approximate lamp number. We have been working with the TV system and it is now much more defined. It has moved to within 10" of the drive disk and one of the folding mirrors has been eliminated. Our optical design for the TV has been put on hold, but Sandy would like to start roughing out a design to at least get a sense on what will/will not work. A lot of the electronics are moving on board DEIMOS so space is needed on the outside of the cylinder (not far from the collimator mirror) to put them. Placing them on the outside has several advantages: 1) Lots more room out there which provides a much better fit; 2) Easier access to the electronics components; 3) Will keep the generated heat from the electronics outside of DEIMOS it will still be cooled, but it is less critical. We have established a method of getting the collimator mirror in and out of DEIMOS fairly eas- ily, it is now a straight shot right out the side, it does not need to be rotated or move in any way it just slides out a hatch. The tent mirrors will be able to come out through any man-hatch that is installed. The silver coatings willlikely need to be re-done at least every three years. A piece of aluminum was passed around which is the size of a grating (8 x 12) for review. This will also have a cell and other things around it. These will also be used as "dummies" when designing the actual cell. Our first engineering report was sent to Frank Melsheimer this week which included some of the following information: · The handler is now a linear slide with four positions (see drawing attached), three of them have motors and encoders. There are a few positions which could possibly get into trouble when rotating the instrument. However, it will be possible to move the grating slide without homing anything. If gratings are to be replaced we will have to go to a position for taking gratings in & out, which shouldn't be confused with homing. Last month we discussed having all four posi- tions being able to take the large grating and rotate in any direction, but we now know that this won't work. It will be impossible to crash the system because it will not be possible to place a large grating in each cell, only in positions 1 & 3. All four stations are the same width which is why the mirror in position 2 is so wide. With the exception of position 2 the other positions will each be independently rotatable and have their own encoders. The grating slide will weigh 800 lbs (one side), with the gratings installed. If a grating is removed a "dummy" will need to be inserted for balancing. · Frank had also suggested a caterpillar for slit mask storage (also shown on the right side of the drawing). This is a very tight package with not much room for structure, ten masks and four gratings will be possible. At the time these packages are being designed, we are also running FEA's on them to make sure they are usable. We are also building a full size plywood mock-up of this item to mount on our model which can be rotated to give us a better idea what the package will be like. We have ordered one of the Teledyn Gurley encoders which is intended for use in DEIMOS for evaluation and for use in testing. They are not rated for the cold temperatures experienced at Mauna Kea(but are used on Mt. Hamilton) so we will want to test them here. We expect them to be OK. The instrument encoder will also be of this type. We will be test driving the Image Rotator which will be giving a Teledyn Gurley encoder some use which we can also evaluate for DEIMOS. Slit Mask/ Cutter/Handler We have received a test sample masks from Directed Light and also ART and received a few from CFHT (one of which was actually used). The CFHT masks give us a sample of what others are actually using for comparison (even though our edges to not have to be quite as good as theirs). A slit evaluation form is in development. We have a microscope set-up which can be used to take a hard copy looking through the jaws to check the edges. We here at Lick made some slit masks for LRIS which were wonderful, but these would be to costly to think about. Brian Sutins's report can be adopted for slit mask cylinder, radius and angle. Electronics The schematics are being developed at this time. We are approximately 50% completed with the design for the instrument control. We are not looking into the CCD area, because it still needs to be a little better defined. We do know how to build a basic Leach controller and won't have much problem with this. We need to wait to know if we're using the new Leach-II design boards or the original models. The Galil controller is being used to run the prism rotator for HIRES and it seems to be pretty standard to us now. Terry has designed an extra IO board which will give us both analog and dig- ital bits to play with which will be addressed via the Galil box. Pictures have been circulated by Gerry Luppino of the mosaic that he has assembled which is in similar geometry with what is planned for DEIMOS (2K x 4K devices, 1 amp per CCD, using old style Leach electronics to read them out). Each half of the mosaic will be reading out to a differ- ent timing board chain. They could not run both controllers simultaneously because of a serious cross talk problems, essentially the two controllers are not synchronized in their readout. They are still trying to investigate where the problem is occurring. Camera/Collimator The optical materials are in various stages of purchasing. A purchase order to Schott for Zerodur has been written for a 46.3" diameter 4" thick blank for just under $40K. (Corning's price for ULE was $40K with an additional $10K for generating.) Kodak will generate the Zerodur for just under $2K. Some of the new machinery has also arrived. Software/Software PDR We have done some investigations of terminal servers to allow us to get the Galil controllers onto the ethernet and to minimize the RS-422 cabling that has to go into the wrap. This was also driven by some problems with MOS (noise from the RS-422 signals coupling into the CCD read- outs). We have evaluated terminal servers from Sun, Anixter, Lantronics and Stern Electronics. We eventually selected one from Lantronics which will fit our needs. Our plan now is to use RS- 232 which will be fine. We have placed an order which will be here for our evaluation within the next few weeks. They do not seem to have any altitude problems either, but they will also be tested in the freezer. We have also ordered a Leach S-bus board which would allow us to eliminate the VME crate for the CCD sub-system. Steve Allen has had the most time to work on DEIMOS since the last meeting. He has been look- ing into disk displays/formats etc. He has written a description of a scheme by which we can store the images on disk as one large chunk or many separate pieces. For ease of reduction and the ability to handle such large images it would be best to store the spectral frames as four sepa- rate pieces and imaging frames as one piece. Brian will verify the current scheme and zero point which is using a mirror tilt of 22-1/2. We have been using 44.5 as the spectrograph angle, but this depends on what is called the center of the field of view. This used to be 5 minutes and is now 4.5 minutes. These are the items that need to be nailed down now. We want to trace this beam from the center of the slit mask and hit the center of the mosaic. A working group will get to gather to decide the numbers. Steve has also started thinking about the target selection algorithm inputs and outputs. At this time it looks like it will be a cross between MOS and LRIS. Bob is concerned that the materials won't be ready for a July PDR. Sandy mentioned that the astronomers have not finished their assignments. This is a very busy time of year for everyone, Sandy will be gone in June, David Koo in July and Drew Phillips is writing his Thesis among other things. One of the main reasons for having a PDR now is to nail down the requirements that are needed. We have already decided that the three areas of software that need to be addressed: 1) CCDs is pending because we don't know what equipment we are using; 2) Instrument control which is easy, we have done this before; 3) Astronomer Interface which is what is really needed. We decided that there will probably be no PDR in July. Schedule There has been a major change with the camera optical material which will take approximately 5 months for deliver. Because of this the collimator has been brought forward along with the colli- mator cell development. We are currently on track for the CDR in November with or without the Software PDR. Budget The budget that was summarized for the quarterly report had three major revisions: 1) Optical Glass; 2) Funding of the Assembly and Test from the various places it had been included; 3) the electronics portion was re-estimated. This means about $30K went back into the contingency. Quarterly Report #3 The report has been sent and copies should have been received. We are now starting on Quarterly Report #4. Prospectus The prospectus is coming along and should be finished soon. We will be meeting with Jack Halli- day a fundraiser to discuss DEIMOS in the next couple of weeks.