DEIMOS ##Starting with item 28 I think everybody knows that we have CDR comments from Dan Faber- camden's committee and I've written a response to those and passed them on to Sandy, and once Sandy and once she goes through the draft we'll circulate it to the team and I propose to circulate it with the draft quarterly report 6 which I've written the first few pieces to it, so even if you don't have substantial comments to responses which will be an attachment to quarterly report 6 you might get little sections, so get that done. The next SCC meeting minors getting is about March 22 or 23 thats when we'll be headed for and I'd like to get this under control by mid-Feb- ruary if we can. ##Our response? Yea if we can. I'd also like to finish the quarterly report. Mar- leen's working on the budget parts of that. ##How is banner these days, I heard it was totally paralyzed.##It's doing better, Marleens been working on it, or at least trying to her quarterly bud- get from that rather than using 2 sets of books. So the responses are fairly extensive, even though Dan's comments look easy to respond to he bounces around alot I've found out when I reviewed? So in our response I've tried to accumulate all text about certain topics under those topics. Let me check the cover of the topics Dan is talking about everywhere in his comments. Sometimes he talks about things 2 or 3 times in different places. ##Everybody knows that the main thrust of the CPR report was that we hadn't done enough design work in order to start fabricating this structure and I think everybody knows that our response to that is to basically agree and to have a very focused mini CDR in April. ##We're planning to have a mechanical design report out in the end of March. Didn't we cover this in committee. What we're planning to do is have a number of very focused reviews. One which would be on the mechanical topics including the floatmask CAM work grade handler and?? The structure?? Certain mechanical things won't be designed with that, tv camera systems; much of the mechanical part of the Fletcher control system probably won't be designed it will be concepted though. If necessary we'll have another mechanical detail review probably toward the end of the year. What I'd like to do after the April review to be able to start fabricating the structure, and that's the entire focus of that. In getting to that point there's alot of design work to e done. We won't probably have the software PDR by that time. The detectors and doer review will be later in the summer. What's happening is instead of having one global review which the CDR recommendation seem to indicate we should do; we're thinking we'll have another very small focused review, with detailed drawings and?? After have two major global reviews the chances of getting substantive and detailed comments on drawings is very remote. The two global reviews did a number of beneficial things, but I think we're past that at this point. ##SOFTWARE: We've tried to have a couple of meetings with the software group for DEIMOS since our last team meeting here. We had one on 12/11 in which we developed an outline for the software DR and allocated that for. I believe this is what you used Dave for drawing these gant charts. In reviewing these, Steve and I come up some things that don't match up that we'll need to iterate with you on. We've been shooting to have meetings each week on Thursday after- noons.##This I didn't know!! ##It ended up conflicting with the ESI meeting that Joe Miller scheduled at last weeks meeting I was ill for that Steve and Drew met and Steve maybe you can comment on where things stood as of that meeting. ##Drew and I exchanged documents that we have been preparing and tutorials for each others?? How we're going to store information and saw the requirements where to run in, certain requirements neccesitate what language it makes since to use. ##So we'll be planning to meet again on Thursday at which point I'll have a much better handle on where we are. That's tomorrow at 2:00 pm, at Carr 68. Based on a meeting between Steve and I about 1/2 hr before this meeting. We went over what we knew we had on here and how far each of us have gotten and I'm a little bit pessimistic about an end of February date. That would mean having all this stuff done roughly from one month from now if we're going to have it out in time for the review. We'll have a better handle on that tomorrow but, I think that's going to be tight given other thing. Another thing I tried to do today was to go through based on the tele- scope schedule and everything I have here and tech. Marking off other things that we have to deal with between here and only through August, and the other thing we're up against right now is high resemitrotator we have Al Conrad and John Gothwright coming out on the 29th and 30th to do software integration on that. Dean and I have a fair amount of things we need to do that will consume a good chunk of next week. I have for this week tomorrow and Friday pretty much blocked off as DEIMOS days. I hope that will also be true next week. And then we have yet another MOS run coming up in mid-February, caspuldemer comes up after that which I have to be involved with some minor things. So it looks tight, but I think tomorrow we'll try and go through and I'd also like to resolve the discrepancies we have on the gant charts and see if we can make sense of that. ##What's the story on hiring another software person?##Bob and I haven't had time to get together to talk about it. Maybe we can talk about it tomorrow. ##Who was the person you were introducing to us? ##That is not directly related to DEIMOS, it's a position to support Mt. Hamil- ton software, the hope of off-loading my time and Richards and these who are trying to support a piece of the Mt. Hamilton software effort and not providing the coordinated response that they need. That recruitment has also moved at a glacial pace. An offer has been made, and we're in this very slow loop of trying to resolve moving expense issues with Joe Miller being in Hawaii and our candidate being in the Netherlands and Human Resources only dealing via telephone. I'm hoping we'll finally be settled this week. ##Did we turn up anybody else in that search that looked interesting? ##For DEIMOS?? ##Yes, suppose DEIMOS wanted to hire somebody. ##I think we would have written a rather different ad. ##I'd like to make it a very high priority issue tomorrow to decide whether or not we're going to host a DEIMOS software I'm becoming concerned that we don't have enough manpower. ##It's fairly obvious. ##That's all I have to report. ##I'd like to add to your report a little bit. Drew and I had a meeting and we started thinking about the perfor- mance requirements for the acquisition software. Drew's begun mapping out what a screen would look like and the kinds of functions it would have. A hand paddle mimiced by a mouse button for example. We're thinking about laying out the control room where the OA's turf is, where the sci- entists turf is, where some sort of shared turf is along the lines of the layout of 120 inch control room. If we get a picture in our minds I think it would be very advantageous to run it past the CARA people as quickly as possible to find out if it's even within the realm of consideration, because it's quite different from what they had sent up before. We may be in a big negotiation with them over it, and we lose. The second thing I'd like to say is David Coo really wants to be the Manager from the Science side of the software effort, and has volunteered to do that. He couldn't be here today, but he would like to be the person who simply walks around between all the software people and asks the right questions and ties the effort together on a regular basis. I was very glad to hear him say that. ##We called and your software PI come in like that?? Yea, and Joe MIller asked his to do that a long time ago and he just wanted to get it straight with me, whether or not this was really a good time for him to really jump in and I said yes. Unfortunately he is going on sabbatical around March. But he says between now and then he's got alot of time. I wouldn't kick his offer out of bed, even if he can only do it for that period of time it's worth it. ##Thru March 1. ##So he needs to know about this software meeting tomorrow if he doesn't already. I just heard there's a software meeting regularly every Thursday afternoon. Yes, it used to be Monday but with all the Monday holidays we moved to Thursday, for a while. ##Well we haven't decided anything else, so I was keeping it as Thursdays that seem to work out in terms of where observing runs and other things seem to be falling. That there were fewer conflicts this month and next, later on we get into collisions in April, then Thursday's not going to work for a while. ##I don't think David is apprised of the meeting so, I'll make sure he his. ##In terms of your wanting to run things by CARA people, although it is a small subset and I'm not sure they have any great authority in this matter. The fact that Al and John will be here the 29 and 30 the point of their visit is high resimator rotator software integration, on the other hand both of them will be involved with DEIMOS. ## Excellent idea!! ##If your going to be available and Drew can slot out some DEIMOS time within that visit. ##You know the rotator won't be secure. ##We have our fake rotator. ##Can you take charge of making sure we have a date for that meeting. Or maybe Deanne. Deanne schedule that meeting for us ok?? ##The 29 and 30? Whos coming,## Godfried and Al Conrad. ##Are you responsible for their time those two days? ##Yes. ##Drew you really need to be at that meeting. ##I said I will. ##Let's take a vote on the 29 or 30. ##I can't I have to go home and look at my calendar. Sorry. ##ok. DETECTOR PROJECT Richard: I have given Orbit a written ok to start on the DEIMOS CCD wafer project. We'll see something out of that in 6 weeks if guess. ##Why did you do this? ##There's nothing more we can do in terms of correcting the problems from the previous runs. And they're going to do what they can to get the correct implants for these next runs. ##The last team meeting you said they were going to go off and they would make some taste probings. ##They were going to do some addi- tional work on this backup run they had already started which already had the bad implants in it. And try and do things to see if they can overcome the problem. We got that run, they didn't over- come it. They didn't do anything bad, but they didn't do anything positive either. ##Did they know before a run whether the implants are bad? ##They don't know before we get it. ##That's what I mean, it's not as though you can take a bunch of wafers, test those wafers to find out if they've been properly processed, and if they haven't you just don't go ahead. ##I believe that this is probably possible. ##That's the gist of my question. When you were talking to us last, what I remember, maybe it was your suggestion not theirs. What they ought to do is see if they could detect this problem before they made a run. The net result is they are not doing that. ##Well I don't know what they're doing. My feeling is it's their debt. They're already redoing two??. So their putting alot of money into this screw up. It's costing them alot of money. So they have a big financial incentive to do it right. There's not that big of a financial incentive on DEIMOS runs, those were only $20,000 runs, besides those two they're going to do 3 other runs at $50,000. ##Do they guarantee to replace all of these because their bad because of this problem. ##In previ- ous to the $50,000 runs they are replacing them, they agreed to do that. So they're redoing those 2, plus they're starting another $50,000 for Princeton. They will replace all of those. They have this 10 major minimum guarantee.They agree none of ours is reasonable. I haven't gotten clear from them what they're going to try to do, or try to get their implants service to do. Because it`s the implant service that's screwed up. ##I'm just curious about your reasoning.Maybe there's a subtly here I'm not appreciating. Given the fact that there's 3 more $50,000 groups in line ahead of us. #I don't think it makes any sense to do these one at a time, and say this one is good, let's go on to the next one.##You don't?? ##No because you don't know what's going to happen between the time you do one and the time you start the next one. ##I See. ##It would be very nice if given that they have 5 runs to do now, that they got the implant to spend the extra effort to actually calibrate at the level we're doing this, to make sure they all come out right. ##I see. ##But it doesn't make sense for them to calibrate for one, then hope that it's going to be the same for the next ones. ##Because this problem could come and go. ##So if your going to do one, do them all, and I think that gives them a bigger incentive to make sure they do it right. I tried to make that clear in fax to them. We don't want five more wafer runs screwed up. ##How would you test for the proper implant level? That's not something we have the ability to do here is it? ##We wouldn't do it, we have no capabilities, and it's not easy to do apparently, that's why they don't do it. They have some crew test it at high lev- els of for the implant service. We're going ahead!! Test on Lincoln Lab device. At our last meeting did we talk about the test we did on the Lincoln Lab device in terms of the cross tolerance. We had some interesting results, and I xeroxed them. It's there and it's pretty bad. ##How do we see this as negative images. ## No actually positive images. ##It's not quite like a low RISC chip. ##It's a different amplifier configuration, totally different. You cannot use the schemes I heard people bouncing around about doing feedback from one environment to another to correct the problem. It's not going to work because the cross talk we see has some plots here, it's pretty linear up till and sags right out, is function of the charge accumulated, not a function of the?? amplifier. On these devices thes3e are very different things because the amplifier saturates at 10% of full well. The amplifiers upswing up, but the cross talk just continues to increase. ##You have no way to measure that charge because the amplifier's only saturated at 10%. ##Furthermore you can't correct the picture after the fact, because all you know is that pixel was saturated, you don't know how badly saturated it was. ##Right. ##How were you supplying the bias levels and clocks, were they separately driven. ##This was done with our old Lick controller. All of the amplifiers are run together. ##Yesterday Way started wiring modifications on our dueler, so we can run everything separately. There'll be separate supplies and things for both amplifiers on the chip. ##We don't have this device here anymore though?? ##We have it through January, I have to send it back in February. So that's why were doing this right now. ##Do you have much hope for that? ##I don't know what to predict at this point. We'll wait and see. It's not a function of the aplitude of the cross talk is not a function of the gain in the amplifier. There's some confluent there. But we changed the gain in the amplifier beta factor of 6 and the amplitude of cross change by a factor of 2. So it may not be a gain, so much as an offset. So exactly what the nature of what's happening we don't know. ## What does Lincoln say?? ## They haven't said anything, I haven't discussed this with them. I don't think anyone done this test on them before. I asked Jerry, and he'd never looked at this device. We got these toys from him. He only ran it with 1 amplifier. This is next line of investigation to try to understand what's happening here, and what if anything can be done to affect it. ## Did they make us any kind of guarantees what kind of performance they expected to get in this particular respect. This is purely a best effort kind of thing. ##I don't think there was any kind of a spec put on it. Of course if we get devices with the kind of ampathize these have, we almost don't need to run. ##Richard, do you have any final tests for the leach based system before it has to go back? ##Maybe, but I don't think its particularly relevant, that if we understand what the problem is, what's the difference of running on one control or another. It may not work when you put it on the beach controller or it may work differently, but there should be some way to modify the leach controller to get at least as good as we've been getting with??? ##Not every place do you have enough??voltages with the Lick system to totally isolate the two. I asked Pino what he's doing in terms of control Leach control with Mosaic, how the work, he says it's working fine, and he's trying to stay with each one of these Mosaics, he still has the Cross Talk problem he's trying to mold each system, he can't run it all but one. He seemed satisfied with each one. He's said he's going to do in conjunction with CFHT, there trying to do a 8k x 10x mosaic, using Lincoln devices, and he' planning on sticking with each Leach 1 controllers.##Did you get the details on how many Leach 1 systems he's got multiplexed, and therefore how many amplifiers he...##For his AK I think he's go t 2 separate controller systems, he's having terrible Cross-Talk problems, but he keep on the boat at the same time. ##I think we understand why that is. ##Is it curable?? ##Not easily. ##What does this mean for his present 8k mosaic, he has affec- tively...##He reads out 1/2 of it with one controller, then he reads out the other 1/2 from the other controller. ##Sequentially. So it's like having one amplifier for an entire 8k x 8k. ##No. There are 4 amplifiers being read out simultaneously, which gives your 1/2 of the Mosaic. Then you do the other 1/2 with another 4 amplifiers. In terms of time, you have effectively 4 amplifiers going at one time? ##That's correct. ##So it takes 10 minutes to read this thing out. ##Probably something like that.10 minutes if you go with the typical read rates we have now, if you had to link the devices with the low noise amplifiers you could conceivably have much lower crapixel time. ##We need to go over this arithmetic. ##That's what he's hoping for. ##I think that's all I've got. ##How about your thinning. ##It's moving forward, I was must tell- ing Bill Brown this morning that I think we need to push very hard and have a working NCCD for the March SSC meeting. ##Why do you think you need to do that? Because I'm going to need to ask for more monty if were going to continue this and make sure it succeeds. ## What do you need to do in order to do that? ##Put all the pieces together, we have one more thing to do, and that's to produce the mask which exposes the bon pads of things on the back side of the device. It allows us to hedge through the Silicon. We've done the gluing of our CCD wafer, the handle wafer successfully. After the CCD wafer is glued and the handle wafer is mechanically ground successfully, then it's mechanically thinned, and we've done that successfully now. These are all just test wafers, they don't actually have real devices, so this week or next we're going to include some real DEIMOS wafers with CCE's on them. The ones with quite a bit more or less fail time which most are doing self imaging, so we have alot of test places to use, So we have a bunch of pieces together, but I don't think this mask is, without the design for the mask will be ready. ##The mask from which you'll etch? ##That's right. ##The graduate student that we have work- ing for us, is a super user on the mask making machine out there, so he's prepared to go in there an do that. It's just a matter of priorities. Why we don't have the mask ever to get the ?? ready. ##What happened to your Micron peak to valley curve. Are you gluing now on a bent piece of glass? We aren't working anymore on that, we'll work in it later. For Mark I don't even know if it will have a mounted CCD. We may still just have the CCD in the wafer form. We can image the thing using our wafer program. If we get to that stage, it won't be hard to cut the device out. It's just one extra thing, which we don't actually have to do to get an image out. ##How are you going to cut these devices, are you sawing them? ##Yes, That's part of our mask that exposes, also etches through the CCD, and then we have to come back and cut through the handle wafer. But the CCD wafer itself will be etched out.It's an extra week to cut it out and glue it to a sub-straight then wiring and all that. I just want to get a picture. ##That's very exciting, great. ##Before we leave the Detectors Controller question, you had some conversations with Bob Leach regarding progress on Leach 2, and we need to come to some decision as to what they decided to proceed or abandon. ##We had a meeting yesterday which we talked about some of those issues. Leach said don't come down for 6 weeks until you have something. The feeling at the meeting yesterday was, "That's not a good idea" We should go down there now, and see what his designs are, he claims to have designs on everything and some of the things he's sending up are circuit boards. So we ought to be able to evaluate what he's doing, and see if it's really going to meet our needs. See if he's stressed all the sorts of issues that have come up in his old system. ##We decided to do 3 things yesterday. We decided to recommend to you that we do 3 things: 1. Exactly what Richard said, namely we think there should be a visit down to find out what Leach is doing, to get some sense of is he being more sensible this time. To see if his hardware is more robust than it was the 1st time around. 2nd, see if there's any information about slow digital sky data controllers. So I wrote both Map and Chrone yesterday, and I'll see if I get an e-mail back. 3. We should sit here and ask ourselves is there even a ghost of a chance of doing a controller in- house. It seems to be the last possible moment we could take that step if we want to do it. /si we thought we should revisit that question. We don't know what the outcome's going to be. Richard feels if Leach 1 is any guide, there's going to be alot of labor and time spent in making the Leach system work, and we have to weigh that against how much labor and time it would take to do it ourselves. Chris obviously needs to be consulted with soon. That's what we thought we should do. Do you have any reactions? or do you want to think about it and we'll talk about it tomorrow Jan, go down to see Bob Leach, fairly soon, maybe next week, make sure to include Chris. Optical Tools The interesting news on flask hating was we got most of our glass from O'hara, but some of its' the wrong diameter, too small, so we've reordered. The have agreed to replace these three pieces on a three month time schedule. They also broke one of the Elligate 12's that we are not going to get a replacement on, but that was an extra piece anyway. So were pretty much waiting for the final meltsheet data for that, before we can go ahead. Do you want to talk about anything about the ?? The tools are being prepared on the CNC. We'll make a program as soon as we have com- plete data as to the dates. We'll make them one at a time. We have all the material now, though we still need about a few feet of aluminum. About 90% we have in house now. Call Meter The Call meter has been grinded and fused a hard non-flexible lap on it to polish for 3 hrs. and I polished with a flexible lap for 11 hrs. The intent is to spot any non- symmetric dips or hills, commonly called roller coaster on the surface. We've set this mirror up to a clokal test and romkey test and we've determined that both these tests aren't sensitive enough to spot anything every small. We're actually seeing a whole base sphere departure in these tests. My latest scheme is to use the thermometer with spherical wave front, looking at the surface and moving it at different radiuses from the mirror to look at different angular zones, and to try to access whether these are in the same plane. I've made some pictures that might need more reviewing than our discussion before ???##Did you get the precision spacer in place. ##No, the problem with this test, I've only spent about 4 hrs with this test, is vibration. We have a large county, its almost a 20 ft. county between the inner-thermometer and the mirror, and sitting on the floor, and both are not isolated, so their dancing at different frequencies, therefore the fringes are dancing all over the place. The first cut at trying to eliminate this is to use a board reaching between the two. That generally puts them in some sort of phase. At this point I've done some very preliminary tests, that I believe will tell us what we need to know. It needs some work, its at a pause right now because of the John Hopkins being here. After we finish with Hopkins people, I'll set up the board again and try again. ##You've got pictures, do you want to pass them around? ##Sure, this a circle wave front from the transmission sphere at different points of focus. This is looking at it from 0-7" out from the center of the 14" diameter. As I go through these the prints get worse and worse because we're coming up on the ace sphere, and it's getting narrower and nar- rower. This is the region we were looking at the other day when you came in. This is between 7- 10" radially. So I've moved focus back and I'm picking up that area now. What's on the surface of two little square post-its, so I can see what I had. This is between 10-12, and the idea is to see straight trenches, or relatively straight trenches in this angular area. If I see dips or on one side where's there's straight trenches on the other, that tells me I have a roller coaster. My hope was to put this through the fringe analysis software that I have, but that needs ???. ##Have you thought of putting this on that vibration isolation bed we made for the secondary. Will that be real compli- cated to do that. It's long enough it seems like. ##yes. ##My other question was we were playing with this inner thermometer jar two weeks ago, and it used to be brighter and darker than this, wasn't the contrast better than this in the old days? Is it too burned out or something. ##yea we've got some annoying interference on these from the monitor too, but that could probably be iso- lated. ##So can we do something about the contrast here? ##I'll look into it, I may need to replace a monitor or something. ## noticed in one of them, the mirror is actually moving and you can see it, it's real high contrast, and then you have this low contrast. ##That says your changing from the aline mode to the view mode. I'm not sure what's happening there, but I'll see what I can do. ##Neither do I, I'm just hoping it can be improved, cheaply.##That intersum, which was the first one is the best, and it certainly didn't look like there was any roller coaster in there. ##I've removed essentially the full amount of the aspheric project from that center. Optics The main thing I've been working on. I have been feeding to Jim Burdge and ORA, writing them long letters and memos and sending files describing our optical designs, and the long and short of this is ORA seems poised now to pick up our project and we're preparing a purchase order for them. I got an e-mail from them last night saying "we'll be back to you today" and we're all set for the start on the tests. The first thing they're going to do is simply enter the data for the current design and trace it and make sure it focuses and gives spots like, spots that we've already seen. I believe they have already done that, when getting acquainted with our project. ##We're going to be using a standard tool cull code 5, it shows the robustness of our design. ##Our designs have already been traced with 3 separate coats. I don't think there's any doubt that it's the design that focuses light. Then the next thing they're going to do is, it's really very telling, and should be coming now pretty soon. They're going to put in the new melt sheet data. I told them to go ahead us8ing the meltsheet data for the small piece; just as a test, to find out if they can re-balance our design and re-optimize it. Pick it up sort of mid stream and carry it to conclusion, because some of the folks who came for our CDR said we might have trouble in this hand off from Harlans code to other peoples code. Can you rebalance a design in mid-stream. Obviously were going to be changing our criteria for optimizing, it's inevitable when you change codes. So we're going to find that out I would say within two weeks whether or not they can get on and ride this design. If they can do that then their next task is to study the tolerance question and give us their cut at a fab- rication scheme. At this point were going to start iterating you and see if we can set up some com- munication that will carry us through the fabrication period. As I think I've told you before Dave, they are currently doing this with the blue side of low res. Which has aspheric being made by Harold Johnson and an aspheric being made were not quite sure, where that's being made. They plan on using the same kind of pickup scheme approach that we've been using. ORA is familiar with that concept, and they don't seem to have any problem with it, at least they told me they don't have a problem. I also gave them one more task, so lets walk over the order they're doing things: 1. Putting our design on their system and retracing it to reproduce formurna. 2. Prepare a short study and set of recommendations for us for opticom coupling fluids. As part of that they are going to do a couple of retraces to see if we're very sensitive to the exact choice of fluid. They will vary the index of refraction among a few obvious fluid choices and I hope will show in that it doesn't really matter within a wide range indexes, what we pick there. I think there will be some effect, but maybe by just a slight refocus we could pick that up and show that it doesn't matter.If that's true then we can go forward and do the rest of the design and still hold open our choice of fluid. Also as part of this report I'm asking them to recommend to us which fluids they would choose. And what their various considerations are. 3- Once we understand something about the sensitivity of our problem to the choice of fluid, they'll do this test rebalance for the melt sheet, and when that's done they will start the tolerance analysis and then sooner or later the real blast will come in and if necessary we'll do the final, final rebalance. I'm asking them when they do the rebalance using the current indexes which is perhaps just an exercise, because we're still whack- ing the real indexes of the refraction for the one glass that we don't have. They are going to tell us whether or not in fact they have to do a complete new rebalance or where or not we could actually start fabrication and maybe they would be able to pick up that uncertainty index and surfaces that are further down stream. It may be that we can start fabricating before that final piece of glass comes in and we have a final, final design. So alots going to be happening within the next 2-3 weeks. Now Burge sent me a message, Burge is the young man from Arizona, and it was my feel- ing originally that we probably should do somethings in parallel with ORA and Burge. He's been reading all of our materials. He says "are you sure you want me to reproduce all of this work", "I'm kind of busy, maybe it would be better to wait and see if ORA are making good progress in the next few days. I'll hold off and we'll see what they do. If you have any particular questions for me, maybe we can sit down and find some tasks that I'm particularly well qualified for." I think that's good advice. We'll just wait with him and see what ORA produces in the next couple of weeks. I do want to call Burge and ask his advice about coupling. ##What's ORA's estimate on cost for something like this. ##The first three tasks put the design on the system and reproduce current results; write a report on coupling fluid and rebalance using new meltsheet data. They didn't quote on two of those tasks. We're both estimating it's around $8,000-$9,000 to do those three things. A cost estimate for a full tolerance study around $10,800. Altogether, by the time they're done with this I'd be surprised if be couldn't get by with something less than $30,000. ##Sandy, does a complete tolerance include all elements or just the camera. ##Just the camera. ##Not the collimator or the slipmask, shape. ##The slipmask shape and all that was studied by Brian. So were not going to do all that over again. I think Brian and I can manage the calumniator. The challenge there is we know we can get very smooth surfaces. But we can't hit the radiuses that are back in the real design. We may have a perfectly wonderful surface that's within a fraction of a wave, but the radiance tolerance is such that if the edge were off by at least 10 waves, that's a very big aberration in comparison to the local irregularities of the surface. And thats really as I understand the fabrication, that's where the art is. These gross shapes are off by many waves, that's why you have to pick up from somewhere else, by actually changing the design. One other thing on the couplant is Dan Cavert came and ??? perhaps another TV type couplant might work. I guess personally I'm kind of skeptical. I see it as much different from the gel that we tried. None the less they have our TV and are asking if were interested in testing on ?? We have a brass and glass pair. I don't know if we ever cut the brass to match the glass. ##we did. ##We're thinking of testing it with that pair. Somebody ought to test our tv with our??? In this case I'm not sure what's going to happen. Dave is saying the FDA analysis shows that it won't. We know with working with the gel it's a very complex thing to do. ##I'm not sure we have that much to offer them. They can do a brass/glass test too. There's nothing magic about that. They don't intend to do that for a year plus. They're in no big hurry. ##Looking at the material copies our TV is quite a bit stiffer than the gel we used. My understanding of the process that happened is you have to conserve the ?? in there. ##I wonder if they've done a volume analysis. Maybe we should make sure they're aware of that. I need to send them a note anyway. My feeling is the big unknown with this RTV is its transmission properties, and I don't feel like getting into another 6 month or year long test of ultraviolet transmissions, the way we did with the silguard. My feeling is we have gone far enough along the fluid route that we should stick with that for most of our own bases. We do have one interface which in another week or so we're going to know more about. That's the famous element 6, which is rather thin. It is conceivable that we might want to do one of these surfaces. I would prefer to avoid it, but we might want to do one. If we were going to do a test, I'd really want to do a test that's geared very much to the particular geometry of that element. I'm not interested in doing theoretical tests. ## That element has a very much radius of curvature. Thing like the silkguard.It helps somewhat, it doesn't help by an order of magnitude. I think it's a factor of 3 or 4 better than the silguard standpoint. I would suggest we do nothing until the FDA test and had a chance to think about what's going to happen. Let's do that then think about whether we want anything to do with RTV. ##I wouldn't be against doing a test of the RTV on the pieces that we have, in that it's a decent backup to the fluid. At worse we're going to sacri- fice that piece of glass. I'm not against doing the test. Eric's finished design of a test belt, checked the RTV support fluid coupling of a bacalcium fluo- ride test element that we have. We're waiting for Phil for that. We have the materials, they came in yesterday. It'll take a week or two to make it. My goal is to have it done and have some mileage on it by the April review. We put silver samples in the dome last September or October. 6 samples, 3 bare silver and 3 over- coated with a thin layer of magnesium fluoride. I think two pair went various places in the dome and 1 pair went in the high res enclosure. Mike VeDitorio wrote me last week saying the bare samples showed degradation, some quite substantial. The coated samples, to his eye, showed no degradation. Dave's going to look at these next week. I felt a little sad about this, I thought all this myth that their silveroncia is incredibly long lived. We didn't find that to be true. ##Have you done any actual measurements? ##No that's the issue should we bring them back and start mea- suring. My feeling is, if the bare sample shows visible degradation in something like a few months, forget it! It's not a viable plan. There's no point in measuring it. We're not going to use it. ##Now the do have a reflectometer there. We could use that and see what... ##I think he went ahead and put these out without measuring them first. He had some troubles with that reflectome- ter as I recall. ##Thanks for mentioning that, I'll ask him if he's been following the samples with that though. Now the issue is, the other ones show no degradation. Does that mean we should bring 1 or 2 of them back for testing. Your going to take a look, so we'll discuss it when you get back. DEIMOS STRUCTURE It gets more interesting as time goes on. ##I don't like the sound of this. ##Amongst other things this weekend I'm playing with the DEIMOS structure Binel Aldman and we're looking at the var- ious holes in it. So far I haven't been able to model structure that meets our pressure criteria at all. The hole that goes perpendicularly to the structures we can pull the colenader out of it from it very neatly. Now everything I've tried gives us about 10,000 worth of pitch in the drive disk when you dig up its hat. And the that's way too much. You use light and harder tin more than sand. Maybe a hundred times more. So we're playing with that. I'm also modeling the disk and just today I've started thinking about decoupling the radius of the disk which is driven by the dry rollers and the sander part of the disk which is holding the optical part. The radius, sort of disk part of it can eas- ily support the weight of the rollers and that sort of thing. The trouble is if you put any load axi- ally on it you can get a cluck rate through the center of the disk. If you can decuple the center of the disk phneumatically then thats it, it moves very very little. ##Actually very little lateral motion of the thing to keep the pitch. So if you can phneumatically mount it perhaps we can stabi- lize the center of the instrument. So we're in a jelly stage, we're looking at all different things and trying different things to model. ##I don't understand the last problem and the second problem. ##The disk? ##yes. ##As a monolithic disk anything I do to this disk affects all other places on the disk. And one of the things I do to this disk is I load it up here. So I'm disk starting it both radially and axially. There are forces from the drywall that will go axially. You can see the affect of those things in the center. So what I'm thinking if you can kitimatically mount the center disk on this other ring, perhaps we can better control it. Any point on this disk actually moves very lit- tle. ##It's the axially load that's causing the problem. ##Some problem. Probably 10x water gain than I'd like to see. ##What's happening, the whole thing is moving in and out like this?? ##yes. Of course if you remove patches of layers, your affecting everything. ##What's themagnitude of the axially load you expect to see from the dryboard? ##400-800 lbs. ##That's 10% of the weight, that's the maximum load. ##There's ways we can manage that down to much less than that. ##It slips after that is what your saying? ##It could be 0 if your real good. ##Right if you could get the steel adressment of the roller really tiny, maybe you wouldn't have such a load. ##But none the less, almost any load you put on that other disk can be seen by the thatngs that are holding the optics up. ##I'm not sure that's a problem if everything just moves in and out. ##That's the point, it doesn't move in and out hardly at all, but it does change slope. I hope to have more control by the end of the week. ##The only thing I have to add is regarding 9interfacing with the electronics. We're doing some electr9onic packaging for the onboard electronics. Report on list of anticipated components that will be on a rotating part of it. So we'll see CCD Electronics and Motor Control electronics. They look like they fit in there very nicely. I'm just starting now to integrate the elec- tronics we were going to have sitting in the vault, into a compartment under DEIMOS that will actually move along with DEIMOS, so we can have any other electronics on deck. i haven't got- ten too far, but it looks like we have plenty of room to do that as well. That was something that was suggested in CDR. ##Since our last team meeting you've done alot of work on the rear bear- ing. ##I did a preliminary design of the rear and front rings. We're trying to get as far along as we can for the March CDR. I've also done some work to the kinematic mounts to the platform. Hans Volksgard has looked at that and he kind of likes what we've done by using some camflowers and shearing up some platform ??, for those platforms to be positioned permanently on the deck, so they wouldn't move. So they roll up there and it stops. Hans has done a little bit of work on con- nection to the platform. We've also been in contact with the people at Verspeck, because they will using kinetic mounts, no exactly the same mounts, but located very close to ours, so they need to know where ours are. That's coming along pretty well. I've also done some preliminary designs on a carriage. That's starting to take shape. We're doing quite well for our March target. ##I talked with Hans this morning, he said he went to a preliminary meeting on that transport scheme, and he said everyone liked it. He's not sure where the money's coming from but ... GRADING WE have drawings in the machine shop, and their working on them right now for the grading han- dler/rotator slider test magnacy. Which will test motors, kinetic mounts, structures, bearings and coater, and as many things as we can think of. Mostly to eliminate alot of the uncertainty in those huge errors we've predicted that showed up in the error budget. We actually want to get a handle on what they might be. So that's moving right along. We're building parts. Part of the CDR report said to have a nice formal proposal on how to test these things, and we're moving ahead on an informal sort of idea of how we can use some of the equipment we have in the Optical Shop from Nytec to measure deflection. It's not quite done yet, but we intend to start measuring stuff. These are going to be angular motions as you rotate a grade as you would in DEI- MOS. Hopefully we can bolt this stuff onto DEIMOS to do the test. ##I should probably stop by and hear about this scheme, and what your thinking about doing. ##You've designed the slider/til- ter part. In our scheme I though this whole thing was going to be enclosed by some kind of rigid platform on which probes would be mounted. Have you designed that? ##No, we're working on the inner part first of the tilter. We have a big packaging problem, so that's why we start there. We're going to bump into the telescope, this week we found out we're going to bump into it 6" earlier than we thought last week. Cap 2 is different from Cap 1. I've been working with Hans on the telephone, and getting a little bit of ?? He's going to the mountain tomorrow and take a look. Dave and Bruce will be there next week taking pictures. So that should help find out how tight it's going to be in the grading slider. We have a big uncertainty, and that's how to encode position 3. If we have enough room we can actually put an encoder on axis for position 3, that would be the best outcome of all of this. ##To summarize, the present situation is a disaster. If a solution can be implemented, it will not only solve that problem, but will also make for room for the grading/slide mechanism, and maybe enough room for an encoder for a third grading, which would solve a huge problem with that grading and drive mechanism. So there may be a silver lining; or it could be a horrible disaster Two obvious choices are to go to an 8 mask handler or move the instrument rack another 6". ##I don't think that's optically not viable. ##So the cable tray has kind of a shelf we have to live inside of, it's a certain width 600mm. That wasn't enough wire, so they bought the next size big- ger of the two and didn't tell anybody. So now we have to fit inside of that, but we can't be up close to the telescope in the focus direction. We've lost 6". Hans thinks we can make that larger in diameter, so it appears it's like the same as the floor. So anything we build clears the floor, will clear the telescope. If he can do that, that's nice. He's going to take Kevin Ho up to the mountain and take a look. ##This is after showing us the Cat 1 drawings for the fabled Cat 2. ##All changes are minor changes. I don't know if anybody wrote that down. ##We've asked so many times for the Cat 2 drawings. SLOW MASS Regarding Jim McCarthy's request in the CDR asking for a slip mask insertion is a good idea. It'll probably be implemented hopefully before March. We plan to order a Slip Mask cutter. Sandy just signed off on it. ##I gave it to Marlene. ##The CDR committee had no problem at all using an NT machine or buying it early. We want to learn to use it and also use it to actually fabricate DEIMOS itself. It'll take a few months to inte- grate the thing in the shop. It'll cost around $2,000. ##What sort of interface does it have? Do you load files into it? Is it a PC based system? ##yes. RS232 communications between it and the maskhandler. It's an industrialized PC unit. Let's make sure we have a slot for a PC card for net- working. It's going to take months to get this thing in here. It's an $80,000 system. The PC costs $49,000, plus a spindle @ $10,000. ##Electrical or air? ##Electrical, just 1 and tool changer. The tool changer goes to a wind/wine it will allow us to have several spindles in there loaded with tools, so that someone very skilled at this job can load those tools in; which is not a trivial thing. Let's say we break one in the middle of the night, we won't need to come up and change it. ##How many can you put in there?? ##So it's $80,000 for a complement of spindles? They're $10,000 each. We also have $1,000 ones. ERROR BUDGET Nothing PLUSHER CONTROL Sandy you were going to write a report for April. ##Yes I am, ##We'll have full Plusher Control System fleshed out. That will include everything from our feeding lights to our light source or the optical path ault material detector. An idea of how the software would work in fact drive the con- troller. ##Your the person I'm looking to for filling in major gaps of my thinking, because I don't know where this software lives, I don't know what machine is running it. I don't really know what the signal chain look like from those detectors. I think I can fill in most of the rest of it. ##Do we know whether we are going to be doing anything with the CCD clocks for the main part of the detector during the time your integrating? You had proposed some schemes for shifting charge around, as a scheme for dealing with fringing. If we do, it impacts the hardware and soft- ware. This is how we do the fletcher control. ##I'll think about it, but I think it will be no. ##So the fletcher control would have it's own analog boards. ##In any case it's going to have it's own analog boards, but it doesn't need it's own timing. ##I just did an update on schematics to keep them current. ##Erics's talked with you about the Contactor maybe having an 19" rack cutting... Seems like a nice standard thing to do. ##yes, or in some instances, I'm still not convinced or spe- cially the gallow equipment rack, is not inherent at a 19" rack, not with equipment. It might be a better use of the space to have that compartment with one 19" rack, and the rest of it ??? I'll have to think about it a bit. Bird where do we stand now in the process of getting miscellaneous Iopids assigned to their...##I generated some drawings that have those bits on them and I have some updated schematics for you that have my first go around. I'm going to ask you to look at it and see if it makes sense to you. I've got a couple of things on my end. I've compiled a quotation on a barcode scanner we're look- ing at. It's a 2x2 little square that's an industrial version and more durable than the pistol grip. I'm going to go ahead and place a PO for it this week. I will send it down along with some accompa- nying software that allows us to create our own barcodes via a laser jet. It's going to be around $1,200. It has an RS232 connection, which is fine because everything on DEIMOS is RS232 from via terminal. So I'll go ahead and order that and play around with it. ##Does this have any mech- anisms to turn itself on and off remotely? Far as I know, it's only activated when you tell it to take measurements. As long as you don't take a measurement, then there's nothing coming out. That's something I'll want to reverify. The other thing is, I talked last time about doing a noise test set-up, to see what kind of noise we could expect. ##There's a window of time we might have in the next few weeks to use the MOS CC systems down here to do some final work on. The boards are the only part of the system we can do noise tests on right now. But it should be the noisiest anyway We are sort of moving toward a decision to build a single beam. With this quarterly report will be a single beam schedule.My scheduling program has us ready to do this in February, but I think it will be April, due to the additional reviews that will have to get factored in. Nonetheless, it's a few months earlier than the dual beam schedule. When is the Lincoln completion schedule? ##I don't know, the last I heard was from e-mail the end of last year, they were in the middle of fabrication. The very first places we'll see them, you know these aren`t completed and packaged, their in wafer form, is the end of Feb or early March. ##How much time to complete all the other stuff?? ##Several months down the road. I would say middle of the year before we see a packaged end device. The packaged don't even exist. Lincoln isn't providing the package systems. ?? has to provide the package system. I don't think we've done anything about that. We've talked about helping him do that, but he's never initiated any- thing with us. ##Do you know what his scheme is for packages?? ##No. ##Would it meet with our approval? ##Well it has to, that's his plan, something acceptable to us. But I don't think he's done anything as far as I know. I'd say another year before we see the first Lincolns in place. I think the one year total schedule is realistic. ##Do you think we're on track to make a decision between them and your approach? Are those two efforts looking like they may come together? Probably about the same time. ##That's what I'm interested in obviously. ## One more question, is it true that your going to set up the new DEC Alpha? Have you started? ##I haven't started yet. ##What's your time schedule for that. ##By the middle of next week. ##Great!! How long do you think it will take. ##It will take some conferring with Ted, who has to be in. ## I want to get an account on that. Let me know when your ready to incorporate other books. Regarding the official announcement of a single beam system, does this still mean we`ll need books for double beam? yes. ##What's our plan for those side beam motors, I sent you an e-mail about. ##As far as I'm concerned, they were deleted. This is the motors that will drive the kinet- matic mounts of the 2nd. ##I think that's one of the things we learned from the CDR, is that we probably shouldn't leave those out. I'm having an anorthermeric fraction calculation done. We can reserved final judgement until then. But I think we may very well want that extra set of motors if we don't have any ??? ##Originally we thought we'd put those motors in with the ??? ##Yes that's the only time you`ll get a chance to test them. I think we took them out when we con- vinced ourselves that there was no thermal reason. But I think there is an atmospheric refraction reason. ##You'll let me know when you find out.