DEIMOS 10-17-95 Tape one Sandy: We're doing something yes. Like a lot of things there turns out to be more to than you thought. This is interesting for you Chris, this entered O?? directly with you. There are schemat- ics for the electronics, but the schematics as discovered by Bob, don't have all the hooks and bells and whistles to reflect some of the things that we said we were going to do with motions of ele- ments in the PDR document. I don't think anybody in designing the electronics ever read the PDR requirements and said, Oh, you want to sense this, Gosh that's going to take some sort of sensor here. #Ahh.. # Oh, you want to lock this, what does the lock look like, it is a pin, is it a clamp and we've got a sense of current. We have got to sense whether or not that thing is on or off. But Terry ?? and Rick actually had the PDR document. #I don't think... #If he read it he didn't pick up all of these things. #He didn't pick up all of these things, I think that for every one of our mechanisms, we've got to sit and think about exactly, what we need to know about this mechanism and we need to do with it at various times and make sure that we have included these hooks into the... #This is an area that is very different than HIRES. HIRES has nothing. you can take something out and lock it. #Ok. #Other than a user filter but you don`t count that. #Right. #You just do it quickly in case it takes off. You don't lose any figures [or did he say fingers!?] #Is Bob coming? #Yeah, he was on the phone. ... #Sandy: about doing software right at the very. #Sure, detectors, it has become our tradition. #Detectors. what's happened since the last meeting? We have decided to go ahead and hire a graduate student up at Davis. #Good. Did you get such a person. #Yes. So things are moving along much more quickly now than they have been in the past. We basically have somebody up there working on it all of the time. #So what has happened? #Well we are still trying to.. its just taking forever.. decide the best way to bong?? these wickers up. Ccd wickers, handler.. I am open to ...??? #You are spinning the glue and playing with viscosities and things like that. #Sandy: Where is that going? #We have been testing all sorts of different kinds of epoxies and polyamides and things, and they haven't had the right properties that will stand up to the chemicals and be thin enough so that we can spin it on, and we tried a couple that were silico- phobic and we would spin it on and it just heat up and would let the surfaces slide?? #Silicopho- bic! #Sandy: Where is that going on? Up there or down here? #We are going to be doing that doing that here, since we are being kicked out.. #Sandy: Marleen! What is going on with you? I didn't know you had this problem. When did this happen? #August 26. #Really and I didn't know anything about this. My God. Sorry. #Me too. #What is it? #I fractured my tibia riding my moun- tain bike, so it's an honorable break, that's why I'm laughing. #There is a certain cache?? I am still not used to that part. # I think that is probably about all that's really significantly different since the last ?? #Now you've done a whole wafer run that's was not good, and that hasn't changed I guess. #Yeah. Unfortunately we did four wafer runs, only one of which was for Dei- mos, which have been on a steady downhill slide. Each one is worse than the previous one. #There was one good one with the bad transistor or something on it. #Right. #Yes. #And we got most of that test. #I only know about two wafer runs, maybe you better go back and tell us about two more. #Yeah there's four wafer runs ..? One of them was the University of Texas run, that's the one where the implant levels were wrong, we could still operate the devices with different voltages. The next run was the first of the Deimos test runs, two runs, that one's got an even worse implant problems, which not only affect the way the amplifiers work but charge transport in the imaging area, we've got problems. The next two runs are a combination of various people's funding, some of our NSF funds and some our China project. #Also some Deimos or not? #No Deimos. #No Deimos, these are from the original mass set. They have even worse implant prob- lems, so that everything we saw in previous ones are there but much worse. One of those runs, there is not a single device that you can make work with any set of voltages between higher ??? and ??? #Ok. #So at this point we have stopped doing runs. #There's a fourth run. $Those are the fourth, there were two of those. #Combo run one and two. #They're still not together. #And those are quite recent? #Yeah. #Most recent? #Yeah. #Ok, so what do they have to say. #I sent them back wafers that we fully characterized from all of these runs plus one of the ones that we did a year and a half ago, which didn't have any of these problems and all of the current wafers are sup- posed to have the same characteristics, the same implant levels as the one from a year and a half ago. #They are supposed to, that is to say that was the spec? #That was the spec, that's right. #I said make it like this one, we like this one. It is clear that none of them ??? #So I have sent all of those back, they are now going to send them to a lab where they will slice them up and actually probe the implant concentrations and the doben? concentrations in the silica and attempt to deter- mine what really happened. Apparently they stamp these wafers out to have these implants done. #Remind us what the implant is? # There is an implant that is done over the entire wafer that pro- duces a potential that keeps the charge inside the silicon as opposed to right at the surface. The implant, or what we called the f?? channel. #Ok. #There's that one and we originally thought that was the problem and then there is also the MPP implant which defines a small potential under one of the three phases so that when you lower all of the clocks to a negative voltage, the charge remains under that phase, even though there is no vertical?? flox?? That's the MPP implant. Those are the two implants and they're done with basically a phosphorous ion beam that they expose the wafer to. #And these are both done by a subcontractor? #Apparently. #So the subcontractor has lost its memory, or its a new subcontractor? #I don't know the details. #But orbit is sending these things out I guess so that they have some ammunition to go back and say look this what they're doing. It seems to me like there is something degrating in their system that every run we get out is worse than the run before. #But it's not clear whether its at Orbit or at this other place. #Well I think it is at this other place. Geometrically, they look perfect. That is when you look at the poly- silicon layer you find the clock face and everything. There is no improper etching or anything like that, we looked at them very carefully. #Do you have an understanding with Orbit as to who bears responsibility when something like this happens. Did they choose the subcontractor, did they monitor the performance of the subcontractor? #Oh yeah. I did not even know until this all came up that they had a subcontractor, that's not part of our responsibility or they probably do not want people to know exactly. So I do not have any control over the subcontractor or even know who it is. #Right. #But obviously they are concerned because they are producing garbage here. #Does this effect their other ccd's? #I can't imagine that ??? #So it may be that their whole com- pany production has ground to a halt over this issue. #Yeah. #Let's hope so. #Yeah, right. #How about the money that has changed hands? #I don't know that any money has changed hands, yet. The Texas run has been paid for, but that's the run which there was a very good yield at the end, even though there were implant problems. #Right. #I don't remember if we paid for the Deimos run, but I told them I wasn't going to squabble over that one anyway. These are only these twenty thousand dollar runs. [!] #On the other hand, twenty thousand dollar runs do add up and you have two more here. #No, there are still two more Deimos runs that we haven't run yet. #No, there are two combo runs. #Those are fifty thousan dollar runs! #Oh, ouch! # Those have not been paid and I am not going to pay for those they're are going to redo those. They are starting they had a backup run which they started in just a preliminary phase and then stopped. And they have restarted that and there going to monitor what is happening to that run as they go through the pro- cess. and they're going to do some, what they call process ??? That is, they will do some the same as they did it before and then some with less of an implant. #So do you have any feeling for the time scale on which the next real try will take place? #I think it will be, it may be a month before we have all of these issues resolved, or at least before we get these test results back from this destructive test that they do to the wafers that I sent back to them. #And only then would I even consider redoing the next two Deimos runs. Plus there's a couple others. There's one other and one other potential that are waiting as well. ??? #Is there a way for them to test their subcontractor performance without actually committing themself to an expensive run? #Yes, and I don't under- stand the mechanics of what they are doing, because I know that you can implant a wafer and then just a single wafer and send it out to the company that will be analyze the profile of the.. #And that is clearly what we think we should do before, or they should do before we do any more runs. #I thought they should have done that months ago, when these problems first showed up. #I see. #I am a little concerned about Paul Sooney's attitude about all this. He is not with orbit any more, but he+'s the front man. He's set up his own company. #Do you deal with him or with Orbit? #Well, we are supposed to deal through him, he is supposed to be sort of an Orbit contact, but his attitude is just, oh well these random things happen, and the next run probably won't show this problem at all. #So rather than analyze a problem, we will just try again. #Just try again. Right. #That's not good. #No, no. Fortunately the head of their fab line was much more reasonable. #Who was that? #Shirley Young. #Can we help in any way? #Probably not, at this point, they are moving forward on this analysis ... #The one thing that I would be curious about if you could find it out easily, maybe from Shirley, is to what extent the company is mobilized as a whole to solve this problem or is our problem a tiny problem in the context of much more important things. We found the latter coupled with Sooney's sort of ho-hum attitude, then we might get really worried. #Yeah, in fact I was going to call up today and explore these very issues, to see what her feeling was because I don't think it is the same as Sooneys. #Ok. #Richard, does anyone there have any comment on their financial situation, or their recent drop in the price of their stock by half? I asked Sooney about that, he said their last quarter earnings weren't as good as expected but they were still Ok. #Ok. Next one is camera couplant and optical tools, I guess not too much has hap- pened on the camera except have we heard anything on the optical glass, other than what you heard from Ohara a month ago, I guess it was. #I got another email message from this women there and she had called back to the factory to see if there was a problem. She wasn't specific but she said there is nothing I know of or the factory knows about that would impede development more or less assumed schedule, which is mid November. #Mid to late November. #Yes, right. So things are on track at Ohara. I have not called Optibac to find out anything about the ???? #Ok. Couplant. Who knows the most about this? #Why don't I report on these things since I have been sort of looking after various aspects of this as you have been away. #Ok. #Could I ask a question? #Yeah. #The Ohara glass is going to be here in mid november, #Ah, mid to late November. #But we won't have the meltsheet data until.. #I assume, David, that it would come with it. It that something that we should need to look into, when the meltsheet data.. #Often it comes with it, sometimes it's delayed. #Ah, Ok, so I should... #If Harlan has requested up front that it be sent with it then it probably will be with it. #So we have a copy of the purchase order let`s first of all consult that. Can you look at that... Alright. And if it seems ambiguous, or if it makes a statement, tell me what the statement is and I will send her an email message and find out if that is going to be on track too. #So it's conceivable we could have a final design within a few weeks after reciept. Harlan had said by the end of year, he had given himself sort of four weeks after ... # So that is the state of the glass. The next question which has been central is the couplant question, and to give a brief review, you know that in August I believe it was we did a couplant test with the Silguard 527 as the preferred coupling material and it, when cooled, the material failed. It seemed to develop internal stress fractures and developed a reticulated network of gaps in the layer between the two lenses. This was a test done with roughly nine inch calcium flouride blank cou- pled to flint, and that is fairly representative of the kind of temperature coefficient differences we`ll have in the actual lense. So that failed, and that came as a surprise to most of us but in retro- spect, an analysis by Terry Mast and Gerry Nelson explained probably with 95% certainty why that failed. The reason is that as the gap changes shape as you change temperature, this material is required to flow from the cooling from the inner part to the outer part, but it is rather rigid and it can't flow because it is not a liquid and so in fact it tears apart. So we reconsidered what our options were at that point, there seemed to be two. One was to fill the gap which is rather narrow in the normal design, its a three thousands of an inch gap. One option is to fill that with a fluid and then that raises the isse of how to confine the fluid. What fluid to choose whether it will have an adverse effect on the lense surfaces themselves and the confinement mechanism, the fluid is going to be in contact with that so there is an issue of stability, and we don't want a fluid that dissolves whatever we are confining it with. And in general, confining fluids is a more complicated prob- lem, so,.. it's straightforward, perhaps the most straightforward solution from the optical stand- point because a fluid doesn't exert thermal stresses on the material as you cool it. But it has a complication in the sense that you have to confine it and also it exerts pressure head, pressure forces on the lense element. And we need to consider that as well. Have I sort of summarized the problems with the fluid? #Ok, Then the other, for all those reasons because fluid was unfamiliar and not looking so straightforward, we thought that we should put most of our effort in the short run into testing grease coupling, because grease has been used successfully in previous cameras that were like this and now has also been used in low res. Now the problem with grease probably increases as the size of the element increases and the challenge in our camera is that we have larger elements, up to eleven inches. We have never here tried to use grease to couple elements that were that large in diameter and furthermore we also knew that the low res grease coupling, at least one of them failed when low res was sent to Mona Kaia. You can see this by looking in at the camera. It was said that it failed because the camera was subject to thermal shock, it wasn't shielded slowly as the instructions said. But there was no evidence or any test made to show that it was the rapidity of that that lead to a problem. Maybe it would have happened inevitably any- way. Just because the gap changes shape, the grease must flow as the camera is cooled, so there is a large amount of ignorance attached to that. For all those reasons it seemed advisable to do grease tests and we have done a number of them, and let me see if I can summarize the results. The results grow that grease is in between, it's an iffy material. On the one hand, we never had outright failures when we put together two elements and chilled them down to -20, warmed them up, went down, back up again. We never had an ouright failure with the grease in the sense that it did not develop a network of gaps the way our silguard couplant test did. On the other hand it showed itself to be rather touchy in two ways, I would say. One way in which it is touchy is the fact that Girard was constantly monitoring the edge thickness of the gap. One thing we wanted to see was if we cooled down the gap changed shape, when we warmed up again, would it go exactly back to where it was before? It was suspected that grease would probably flow in this gap the way the silguard was supposed to and one of the questions when things are moving is do they go back exactly where they were before, because if they don't and you do this many times over years on the mountain, then there is going to be a steady evolution of the system in one direction and you may not like the results of that. So, we did these tests and we found that the gap sizes were repeat- able to the accuracy of our test and the number of cycles that were run, but we found that constant probing at the edges did cause edge defects to appear, little fingers and gaps, kind of analogous, I would say, reminiscent of the sort of tears that appeared in the Silguard material, would appear and then on warming they would disappear. There was a certain unpredictability to this material. And finally, the last test that Dave Hilyard ran, which I thought was probably, for me personally, the most conclusive test.. What Dave did was put little particles of glitter as tracers in sandwich of grease, cool it down and keep track in a rather elegant way - I must say I thought the way you did it was very nice -- He kept track of the motion of these little glitter pieces and they moved by two to three millimeters, which is exactly what I calculated that they should. But they didn't move in that predictable a way, they did not move on cooling down exactly radially the way the model said they should, and worse, when we warmed up, they didn't retrace their tracks, they didn't go exactly back to where they were. Some of them went off at odd angles, right? #So, this was evi- dence that this was not a predictable material, if glitter particles could go in random places after ten years, where would that glitter particle be? It might be at the edge because all of the grease had flowed out of this gap, in any case, this was a view of the nonanalyzability of grease, which is somewhere between a fluid and a cement, was pointed at as a big problem in the mind of Dan Fabricant, who was here last week and consulted with some of us on this coupling problem there's a similar problem in the camera that he is building - ??? MT spectrograph. So, these experiments though I don't think they have conclusively ruled out grease in my mind, they certainly have shown that there are risks associated with it and unpredictable that might become apparent only after years of use. In the meantime then finally, Fabricant came and explained yet another method which we might use to couple these lenses. It's the fluid method. And rather than using O ring seals which we had been thinking about here he suggested an elastomeric rubber RTV seal which simultaneously would support these lenses and would also provide an impermeable fluid gap and a chamber there in which the fluid might be supported. That looked very promising, we need to think much more carefully about this though, because when you actually run the numbers, it may be that this gap that we have to create here, the pads, the rubber enclosure that would actually have to support the weight of these lenses, could be awfully tiny, something on the order of one percent of the radius of the lense, which for some of these lenses works out to be something like five hundreths of an inch and we need to think as a group as to whether or not we can actually fab- ricate pads and seals that are that narrow. So, I think there are a number or questions that we are in the process of studying. #At least at this size, this is something we haven't done so it's something we will need to experiment with to prove to ourselve we can do it. #We would have to make actual tests on the properties of these pads and we also need to verify that the fluid that we would be containing here doesn't eat away or interact chemically with the rubber sealant, so that, you know, after five years all the sealant runs out onto the floor. So, we are still in process, I guess is the way to put it now, although the leading horse on the racetrack is the fluid approach. #Ok, and it'll be probably post-CDR before we actually get to doing part of the test, we'll start in early november working out what sort of tests and how we want to do them for the fluid that I think realistically we have to recognize at this point that its probably post CDR that we would do a full scale test and actually build up a elastimeric fluid cell, which I think we have to do, we are proba- bly going to have to do this with our calcium fluoride and some other lense. Some combination of lenses. #Uh, not brass? #Maybe, but we'll have to think it out. #I think we need to think real care- fully about what the ring material is. #Oh, definitely. # Because I have thought about the problem of CTE of that outer ring.. we could make our lives a lot easier if we could find a material that had a higher cte than aluminum. Is it fair to say that the geometry, that the thinness of these pads could be a problem, that occurs in supporting the calcium flouride element. Is that a fair way of saying it? Whereas the other glass elements, there is a bigger difference between glass and aluminum so that the thickness of this pad can be thicker and that is probably not so bad, is that...? #Generally as you get thinner, or your dimensions decrease, the tolerances are often a percentage of that. #Exactly. #You got to know everything, you got to know the cte of your glass, you got to know the cte of your aluminum and exactly how your pads go. #In dealing with a material with a very high cte the rtv itself, so then all sorts of things become critical at that point. #Well, it might be very important to look at a pad design which is maximally compliant, so that if we miss some of these things we put minimum stresses back into the lense. #Again, realistically, given this com- plexity, it's going to be post CDR that we ?? Dan Fabricant has stated that he doen't think that the couplant is a show stopper, he said it is simply an engineering solution to this, I feel a little bit bet- ter in going to get the CDR with this ??? #That's true, but it is also fair to say that he has never done this problem he has used this method, but first of all in larger pieces, but everything kind of scales with the size of the piece that means his pads were bigger and his tolerances were that much easier and he`s never tried to confine fluid with this solution either at the same time. So we are trying to do some new things that he's not seen before. Is there any easily machinable mate- rial that we could get that has a cte of around 30 rather than 22. #Delrin, but it's not a metal, Alu- minum is your best bet, that was my conclusion of the same issues. #Why not delrin, why can't we use that? Why do we need a metal? #The reason for the metal, according to Dan and I agree is that's where you do your O ring sealing is with your O ring in a metal groove. We understand that. We have cycled O ring groups in metal collectively millions of times on this planet. O rings in Delrin I am less sure about. #Actually we are not using O rings though. #Oh, yes you are. you couple the rings together with O rings. #You are just not touching the glass with O rings. Move the O rings problem with metal. #You have your lense that you pod? in with this rtd. #Yes, I understand. [Tape Ends] Tape two. #... about, you said you were worried about casting something that was a millimeter wide and holding the tolerance to a thousandth of an inch. I think that that is just a problem, it's not a ... # I wouldn't be worried about it if I was using brass or something other than a thing that I just cast and... Material that you buy rather than one that you fabricate on site is much more stable gener- ally with this rtv we are going to mix with a catalyst and set off righ here, which to me means that it is slightly variable in its characteristics. Having played with epoxy, I remember that there's a lot of things that have to do with chemistry and things that you have no knowledge of when this stuff goes off, and you can often get quite different material properties thinking you did exactly the same thing. I mean to the limit of your abilities you did, you did it at the same temperature you used the same weights, some of the time. So anyway that is just my caution, so that is why I worry about hitting fairly tight tolerances with a manufactured or home material. So it's some- thing to experiment with and we build our level of confidence. #By the way, this is a little tangen- tial, but if anyone ever wants to know about the stability of inbar and it's cte, Dan has a wealth of information on that question. #Yes, Jerry Nelson mentioned that it is variable. #Yes I know but it turns out to be much more interesting than that. #Much more interesting? #Yeah, and complicated and they actually monitored the making of invar pieces by monitoring the casting chemistry while the pieces were cast, and they had to learn to understand how the cte depended on annealing and the actual composition of the alloy itself, they know a lot. The bottom line is if you spend money and pay attention to the actual making of the invar, you can control cte to ten percent, but if you don't do something like that, then things wander around by a factor of two and in particular I got the feeling, that if you just buy invar stock, the way we have in our lab, we don't know, it could be wildly variable. #The cte could be off by a factor of two? #Easily. #Typically the delrin has that problem also and it's not ten percent it's factors of two hundred percent. We buy a long piece and measure it cte, that's kind of what you do. Design your athermal cell around that. Rtv will be tough, because you've got to cast it and you don't know until you cast it what it.. #We've ordered some, a test two gallon pail, of this rtv, and its free and it should be here soon. We'll experiment with it as much as we can and do the tests. #What did you find out from GE about the fluid prop- erties of Dow 200. #Not very much, this has got all the data, and it's the same thing as you see, it's got a coefficient of expansion, its got electrical resistivity and the statement that nothing attacks it. And I quizzed them again this morning, how about this other silicon fluid the Dow Corning 200, well it shouldn't be a problem. But there's nothing written down about any test they have done. #So I was thinking.. you've asked, maybe I got confused, you've asked GE I guess about the rtv and they've made the statement that nothing attacks it. How about asking Dow about 200? #That's a good question. #To find out if we can find out something about it. #Ok. #Complete. #I have never talked to the Dow people, so that could be something worth looking at. #Right, Ok. #Did you read Dan's paper? #Uh huh. #There's four of us who have that and it is very hard for me to wade through the mathematics. #I understand the mathematics, so I am happy to discuss that with you. #Perfect, because I would going to recommend that he design us a test. #Oh, no I think that Terry and I can design a test. #I think that Terry Mast understands the mathematics too. #Yeah. # He seems to .... #I've also noticed that they put 18,000 psi as a working stress in the in bar, that is way higher than I would have imagined they do it. #Ok, I think that there is a combina- tion of engineering nuances which only you would pick up and mathematical physics type nuances which Terry and I can pick up. I think we probably should sit and review this paper and everybody flag interesting aspects that, you know, Because I read it and I said, This is interest- ingk, I wouldn't have thought of that, apart from the basic mathematics which is simple. #So, maybe sometime after about November 2nd we should gather together if we have time before the CDR. #Well.. # It would be nice if we got to have this meeting before the CDR. #Yeah, see you and I decided we are going to propose here I think at this meeting that we have a few working sec- tions, one of them was on the camera mount and, oh sorry no. We didn't have one. Of course not. I am wrong. # We did schedule that by the way, that is next Wednesday and what we would like to do is go for an all day session and just go through all of them. #Right through all of these. #Well, I have two more to add to the agenda. #In this case it is easier not to. we have to check that Terry and Gerry Nelson are available. #And what day did you say, next Wednesday? #24th or 5th. # We have an optical meeting on Tuesday and Wednesday was the next available ?? day. #When is the optical meeting? #There is an optical alignment is it meeting? #Optical planning meeting. #That would turn out to be meeting number three because this afternoon is meeting number two. #Meet- ing number two this afternoon is to discuss the components of the error budget and make a list. And Brian, you are.. I just want to alert you that you need to come to this meeting that was men- tioned at our last optical meeting. But I want to make sure that you come. That's at one thirty, I think. #Yeah, and this room is scheduled, however, Deanne picked up an email traffic that some- body thinks they might be using this room. #We have it scheduled, but if the other group is bigger though, #No, I think there is a lunch meeting here, but they should evaporate about one thirty. #So it's official that the oil is the running horse in the lane. #The lead horse. # That`s what we are thinking about oil. #No I think we have just about completed our grease couplant experiment and Jerry did a fine job of writing down results and stuff. #So did Dave, Dave did an enormous amount of work this was a double effort. #Oh, sorry. #So we need to keep that we don`t want to lose that in case we come back. # I would appreciate hearing your instincts as we go along, and I think you did tell me last week sometime that on balance from your observation of what the grease was doing, that you were not so keen. #Right, I wasn`t holding anything back at that time. It's too unstable. #It it clear that the news?? absolutely aligned with a radial brush is this specifi- cally??? #It's hard to tell. #It's a flat plate. #It's square, probably two millimeters square, two and a half thousanths thick. #Any viscosity then zero would basically result in pressures on these things to rotate ??? #As long as its thoroughly trapped in the grease, its going to move with the grease, its a test particle. The issue is whether or not it could be dragging on one of the surfaces, right? But still I don't see that that would make it go in the wrong direction. It was the fact that it didn't.. if it had gone in the right direction but a lesser amount then I could have said, oh well, it got hung up, but when something actually winds up going off seventy degrees from the direction it went in the first place. #Did you double check that it locks in by having a finer set of tinier, dust particles ??? at least to make sure that those points are also locked, unless you can be guaranteed those glitter chunks are really locked to the grease they will in fact be ??? #You are saying that if you put in dust that that would be locked in ??? some glitter piece which is pretty big ??? relevant to ?? #It would be hard to measure the smoke particles, hard to track them, cause it takes like eight hours to do something. #I think that yours would be a good suggestion if it was emerging that this was the one sticking point that we were unsure about. #I'm ignorant of the others, so if you have others then.. #There are other issues and I think the totality of the picture is kind of negative. #Ok, I thought that if this was the only one I worry that the band of grease, that double check that thing doesn't move. #That's a good point it may end up coming back. #It could come back. #Is is pos- sible that we still might use grease on the less curved elements? #I don't know. #Every part of this camera that doesn't involve fluid will be a simpler part. #The question in my mind was the quality of the edge grease interface, we know that when we probed into the gap, that fingers developed, but on some of my fingers on my doubling, there wasn't any visible probe, there was still a little bit less grease there underneath. #You are raising the question of whether the properties of grease might be different if we had more parallel plates. Is that the gist of your comment? #Yeah, we know that our failure is dependent on radius of curvature of the element. #Well, some failures are, one class. #I would be reluctant to mix the two. #Fair enough, but we will see how we go, I mean if dealing with the rtv is simple, it would just be easy to carry on in the same vein, however.. #Yes, I suppose we could do some insurance experiments. Are you doing anything else right now that this would compete with or do you? #Very soon he going to start on the ??? #Yeah, but Gerard could do. #He's just starting out Johns Hopkins.. #Let's, maybe we'll just curb this for now and see where we go. I'd like to have a planning meeting or a meeting to go over our plans for the fluid. #The fluid might involve those pieces of glass. #Yeah. #No, I'm thinking of a flat plate, which we don't now have, it would be flat brass, would it? and a flat piece of glass. #Could be. #My assumption is nothing will happen, because nothing happened with the gel, which was very much, #Silguard. #Anyway let's postpone this for a few weeks until we get together and talk spe- cifically about what we want to do in planning for the fluid tests and then we will get some idea of the complications at that point. #Ok. #I think we've made a lot of progress actually. #Oh, we have! #Yeah. #We've gone from thinking gel would be the perfect solution to knowing that it is not and actually coming close to developing a solution. I think we're within striking range and I agree with Dan's assumption, or at least what I hear is Dan's assumption that this problem will be solved. He didn't all that stressed about it. It's really not a show stopper. #Well I might add that people are cheerfully using low res anyway,#They're not cognizant..# nobody has recorded any behavior in it which seems bad in light of the fact that it has this cosmetic problem. #I asked Tom ??? about this when I was there last week, and he looked at me and said, "Oh yeah, right! I remember that, nope, haven't thought about it." # So we may be totally hyper here in our.. #At one time we were considering using air as a couplant, and we had worried about the angle of inci- dence on those deep curvatures, but I talked to Ian Thomas about angle of incidence using Solgel AR codings. He says he doesn't have a lot of experience because most of their experience is nor- mal. #Who is Ian Thomas? #Ian Thomas works for Lawrence Livermore, and he is there Solgel guru there. But he says what evidence they do have says that solgel coatings are relatively unaf- fected by the angle of incidence up to about forty-five or fifty degrees. #They are very benign, the very fact that they are broad band also goes to the fact that they are quite benign. Nevertheless, we are at the hairy edge of having coating problems. That is to say general drop in throughput because of so many surfaces. So I think it would be wise to gather some more information here, try to make it quantitative and ... #Coupling with fluids still seems to be about the best one, but its not the only one. We might be able to use this as a backup if we have to. What is the current state of our calculations on the fluid deformation of the elements. #Rick ? #That's a to do, that hasn't been done. #Nobody's done that. #Nobody's been in Hawaii. #Somebody's been in Hawaii. #What was your question again? Fluid deformation? #There is a hydrostatic force against the ele- ments and being as how they are twelve inches there is half a psi difference across the element. It needs to have lots of area so... #And one element is rather thin. #and we have to assume that they are going to deform. #Ok. #With all that area that you have the volume and surface tension is going to draw off a lot of that pressure. #Surface tension uses a liquid surface, you actually get pulling as this wedded surface but there is no wedded surface. There is no surpilary?? action. #So it's strictly a pressure head. #Zero at the top and half a psi at the bottom. #Supposing you had a whole cavity. #The thickness doesn't matter in this case. #If you just have a totally full cavity then, unless we get down to the molecular level at which point... #Clearly there is going to be a thickness wherein its no longer behaving. #Like just a column. ??? #You are saying its much big- ger, much smaller than whatever we have. #I think that the first trial will be to assume that we have an infinite volume of this stuff a foot deep and try that and see if it causes an effect, if it does then we will have to back up from that assumption. #You will have to know if that effect is big enough to be a problem. We'll decide that ??? #Once you have a deformed shape you will have to do the next thing. #Yeah, and I won't be able to do that analysis before this afternoon, but once I do I will send it on to Brian and ... hopefully it will be the same order as the other deformations that we have seen. #Ok, I think that was couplant. Tools, Optical tools. #Carol still has..# an infi- nite number of #right. I'll need at least one rod I want to point at at the CDR. #You need what? #At least one drawing tool for an element. #She can do them pretty quick, maybe you should get together with her and pick your favorite couple, a few. #I'll also need, if it's a problem going to through all of them, I'll need to have some idea of what order I am going to attack these lenses in so that I have the tools available. I can make a judgement call and probably not be very wrong, but I need input from whoever is going to sit in the designers chair. #Now, with regard to you we sat yesterday trying to organize the chapter outline for the optics presentation at the CDR a little bit more clearly than it had been and this comment also relates to you Brian, so there's a matrix of tasks, such as design assembly test fabrication and a matrix in the other direction of elements, such as slitmask, collimator, tent mirror and so on. And we have highlighted certain boxes which are bigger than others, more important and assigned people to talk about those various elements. So there is a specific plan for you, there is a specific plan for you now if Harlan winds up giving this presentation as he conceivably might, then I think your assignment becomes his, but in the meantime I think that we should proceed on the assumption that you're going to give the presen- tation, I'd like to sit with you and talk with you in more detail about exactly what's in those boxes., Ok.? And we need to do the same to you and give you some help I think, either providing the view graphs for your part of the talk or talking with you about the ones that you yourself would want to present. #Obviously that part optical fabrication. #This is a little separate meeting which we need to have in the near future. Two of them, actually, one with Brian one with David?? #We might actually have both of our ??? ??? #Yeah. The collimator I hope will be on the machine or very close to it by the CDR. #Yeah. #Move on to the next topic: collimation and alignment plan. #Directed at Jack? #I know, I've thought about it??? I've forgot where we stand on that. I think we had two meetings? #No we had one meeting, no, that's true, we had two meetings on this and I think we have a broad outline of a plan, you heard the first meeting.. #Uh huh. #...and the responsibility has fallen to me, one of my two major things that I have to write out for the CDR, right? I am going to right out the description of this and do this soon in the next week and have it reviewed by the people who have heard about it and it consists of lots of little detailed elements, I don't think there's any point in sitting here and trying to discuss with us. It's a step by step walk through. There are some elements that are missing, notably the slitmask and I need to meet with Jack and Eric to get a plan for surveying in slitmasks kinematic mass?? that's still missing. #Is your view that this will become an appendix to the CDR? #No, my view that I am going to write a briefer more accessible summary at this stage, which I think is all we really need at this stage and the later more detailed report will be produced on a longer time scale. #Ok. #So I am thinking of writing at most a three or four page summary of this, something on that order. #Having forgotten when we had the last one of these meetings, one of the major things that dropped out of the first optical alignment meeting was the way in which we can actually test this instrument in Santa Cruz, which we didn't have before, and that it using the virtual image of the primaries formed by the secondaries. We can calculate precisely where that is and that gives us a way to actually phys- ically test this instrument here in Santa Cruz, which is a huge step forward.. #And the main thing that fell out of the second meeting which you weren't able to attend was the exactly the same kind of plan for putting the instrument on the telescope on our naysmyth platform, in fact putting a light source out there in virtual??? It was Jack who noted that there is a hole in the middle of the chopping ??? #So seven feet out there you can put a light bulb. #Good, and that's still inside the dome. #Barely. #That is going to allow us also to align the tertiery?? #Great. #So, good progress. #When would you like to do that Sandy, me and Eric. Survey the slitmask support, at least the notion for how to do it. #Ok, I start to be... somehow I have something in my mind that prevents me from carrying my calender. I do not know what this is. So I don't have my calender here. I will get in touch with you. I assume you are generally around. #Yes. #And I am too, so the end of this week. #You won't be here. #Oh, that's right, the three of us will be in Sunnyvale, going to an opti- cal alignment course. #Good! #So, at least we will hear how the experts think it should be done. #We will ask them how they do their slitmasks. #I forgot about that. #Ok, so the beginning of next week then. [Tape ends] Tape three. #...model? #Certainly we are adding bits and pieces to that. # Yes the Deimos model has now the collimator cell installed inside of it as a convenient way to rotate that around and test it's function- ality of a piece of concrete. It also has two pieces of plywood which represent the front end of Deimos, you know where that is and supported with some metal it gets around the slitmask han- dlers. We will now define the grating slides, they are bolted to that piece of plywood. You should take a look at it all of you. It is something we are able to also rotate around to see how things interfere with each other, and we are starting to come up with a shape that comes through the hole in the drive disk to the exit pupil where the grating slides will actually stop and then rotate a grat- ing. That is where the model is headed. #We also have the piso actuators that have arrived, and I think we are waiting for a couple of little bearings but other than that we should be able to put the tent mirror here very shortly, actually be able to test it. Are we planning to mount that assembly in the model? #Yeah. #Now do you think that is the final assembly? #Pieces of it are, pieces of it are not. #Because remember we had this discussion, I think it came up at our first alignment meeting in which it flexures and various thermal effects. I wondered if you had gone back and ... #I didn't do a complete analysis, I did do it a kind of rough analysis, and it looks like it shouldn't be a prob- lem, the amount of relative movement that we will see due to thermal changes is small enough I don't think it is going to be a problem. #Ok, so basically what you did is, you went back and you looked at your old picture, your former picture and you asked new questions about it in light of the discussion. #Yeah, these are the questions that Gerry Nelson had brought up about thermal changes, about overconstraining the tent mirror. And I just went back over it and said, how much is it going to move given the delta t of something and it doesn't look like it is a significant prob- lem. #Ok. Alright, well, is this... we alluded to it earlier, we thought we would have some working sessions on particular mechanisms, so I'd like to go over that and see exactly what the numbers are. #Ok. #So just a curiosity, not to home in particularly on you, or pick on you, but we are all writing things for this CDR, right? #Right. #So on the subject of tent mirrors and their mountings, there will be words? #There will be words. #And will this be discussed as part of those words? #For Wednesday? #No, for what you are writing down. #For the CDR, oh yes. #Ok. # I've got about six or eight such stacks of stuff to do, its not just the tent mirrors, but that's one of them. #Ok, as I actually don't have tent mirrors stuck in this agenda, we do have the piseoelectrics for that so they got delivered as a single channel unit, and Polytech says that they've never ever made this mistake and humbly apologizes and send us the parts to make it right, but please we could use the parts that we have now and have as much fun as we want with them and send them back later. #So we are borrowing parts we will return these parts later and they will send us the correct parts? #Yeah. Once they have sent us the correct parts and we are happy with them they would be pleased to get back the parts that they inadvertently put in ?? But there not... #What are these parts? #Its just the control rack. #There like fifteen thousand dollars worth of stuff. #Which they sent us by mistake? #No, they sent us the wrong configuration. #It takes three months to make. #What we have now is adequate for our testing. #We actually only have one piseoelectric driver, which is probably what confused them, because I ordered the controller as a three axis unit, which will subsequently be used in the real unit, but I only ordered one of the actuators, because they are about seven thousand dollars each. Rather than order everything. #Why would we need three rather than two? #You can order it in lots of one or three. #Oh. Ok. And they sent us one? #Yeah. So its got a serial interface, and at some point Bob and Chris will be able to work with it and we will actually be able to move the mirror, because we have actually made up the real supports for that part of it. One of the things we will be testing with the aluminum dummy mirror is the reac- tion of this whole assembly to the pieso. #Ok. Deimos structure. #Oh, I think that most of the progress that we have made on that we have already discussed which is at least that we have put the bulkhead in the model and the collimator mirror and we will be rotating that and seeing how that works. # Now that bulkhead's a lot softer than I guess the real bulkhead will be, right? It's got a natural frequency in the drum direction or fo?? direction. #Yeah, we are probably putting some little flanges in there--there's no place to attach them on the model-- that will be on the real instru- ment. #Make that point in the CDR. #Ok considering the strip?? go ahead and bong into some- body and say, look we've got a natural frequency of three, its not good ? and we knew that. #So we will talk about this too in our little working session. #Yeah. #We should have this session down there obviously where we can see things. #Yeah in the engineering lab. #No right there in the shop I think. #Whatever. #I just went to find Gerry and Terry but they weren't there to see you. #Railway. #Oh, yes we can talk about the railway. We need to move Deimos in and out of its where its deployed there and then be able to store it on the naysmyth platform, we always knew that. Recently we discovered that we are going to be sharing that with NIRSPEC which is near red spectrometer they are building at UCLA. NIRSPEC needs to be able to move from the left to the right platform which means we need a way to get that from the left to the right platform and what we are going to do is tie into a little rail system that they already have on the acasagrane?? deck and we will be able to roll NIRSPEC out onto the acasagrane deck and then rotate the telescope then back on the other naysmyth platform. This also means we need to have a switching system between when NIRSPEC goes in and out and when Deimos goes in and out. So we came up with the idea of having a little turntable, much like a railroad turntable that they have -roundhouses. Keck has volunteered to build this whole system for us which is very much appreciated. The only thing we actually have to build then is the rolling part of it. The part from the wheels up so that was a big load off our minds not having to do all of that. So that is the handling system part of it. #??monitoring ?? software wise or sensors and #That's one question, the other question that I have in the PDR where is talks about Deimos being moved out of the way for another instrument, it makes the statement, "We are not planning to disconnect anything and so Deimos will be fully functional in its storage location. The question is, what is the mechanism for dealing with the cabling that goes to the instrument while it is rolling in and out of the railroad tracks, in particular things like the cryogen lines, the liquid nitrogen. #They will not be disconnected. Deimos isn't going to be moving that far. It will only move from one part of the naysmyth deck to another part. #Right, but how do you get those lines to flex. #They're flexible. #Ok. #Why do they have to flex at all? #They can't be hard tubes. # Why can't the tanks be moved with Deimos, are they that big? Deimos has a big moving platform, why not put the... #Well that's a question, do the tanks go with Deimos or are they sitting in the vault? #This would be the 160 liter dewars? # None of that has been designed yet so we don't know. But the other wires, the fiber optic cables and such, we have to be careful not to run over them, like the low res, they have the same problem. Except I think they disconnect everything. #Well that's the question, there are some operational issues here. They raised this too, do we have any way of determining what is the thing that is installed in the telescope or not, do you disconnect things and power down while it is in transit and reconnect it once you get it off the roundhouse. We need to think about that. #I don't think we've totally sown?? out. In concept what we'd like to be able to do is move it and we are only moving it a cou- ple of meters onto a turntable, turning it and running it back towards the vault. #So at least in con- cept we'd like to keep everything attached. #Does it go through ninety degrees? #No, about seventy-five degrees. #Now there is another cable which will disconnect which is the actual drive system for moving it along the track. But that has nothing to do Deimos, it's entirely ?? it clearly a different system. And that will probably be shared with NIRSPEC. They'll have a single power source out on the deck and they will plug it in to either NIRSPEC or us. #Ok. I would at some point like to just go through the operational issues, cabling and what does and doesn`t go to the ??? #Yeah, with Chris and Mary, I guess. #When would be a good time to discuss that? #You have a lot, some other things to discuss... #I have pages of stuff. #.. the same so I think maybe you ought to reserve that question, don't forget it, but that's a big meeting. #Yeah, I think at this point its a desire that we disconnect as little as possible and I think this is a detail that will have to be dealt with a little later. #But ??? have to be notified?? because the controller wants to know if you are in position or not in position. #No we just need to decide if we need to know. #Well in partic- ular, I don't want an automatic dewar fill to start pouring as your trying to roll this thing onto the railway, for example. #Now, I don't think that would be a big deal, I think we can just some limit switches in there, as long as we remember to do that that tell us, a la Mos, which side we are on. #Ok, it's exactly that list of things that we need to write down and this and many other things. # And we know we are short on more than a few of them. #Chris had something to do ??? #Yeah. #But I think going back to the actual railway system, as designed I can't think of anything that is more benign than what we designed. Essentially, the minimum movement of the instrument and... #It's done manually, there are people standing around. # There's people standing, it`s done in the same in the same manner as forward Cass and anything going into the cass hole is done. There is a rack in the deck and a little motor of some sort that engages the rack and moves the whole assembly, it's captive at all times. So should you have an earthquake, the thing is not going to go sailing off the platform. #Are the other instruments that roll in and out, such as LRIS and Nert essentially powered up and live while they are in transport? #I don't believe, no, in fact I can state that they are not because they are on the deck they're on the mezzanine. #They are off the tele- scope. #Ok I know that LRIS can be run while on the deck.. #Yes. #They can and do that, the question is do they turn everything off, disconnect it while in transport and reconnect it once it is in the telescope? #Absolutely, they must. #Ok. So what we are proposing to do here is a little bit different from what the ??? practice is. #Easier, much easier. #NIRSPEC wants to do exactly the same thing that we are doing. they would like to stay cold and fully powered at all times. #Ok, #They can stand as little as a half an hour, of being depowered, and may be in trouble. #Sounds like the space telescope. #They've got a huge dewar that they have got to constantly keep working at ?? cold. It's got a radiant load of about 75 watts, so she's going to warm real quick. #That brings up one other thing that came as a bit of a surprise to me and that is NIRSPEC is planned to be an instrument on our platform for at least two years, they will not be going to AO until they actually upgrade that instrument. So it's not a case of just being there for a short time. They're going to be with us for a sizable period.. When they went through their PDR and funding talks, they never got funded the part of their instrument that would couple to AO, so their slit size and detector and things like that are all designed for conventional focus. I didn't realize that, so they are going to be with us for a couple years. The other thing is that AO NIRSPEC and Deimos all have June 98 as their commissioning date, so obviously something is going to change, and we will just have to see what happens as we move farther forward with these projects which one is going to take the lead. #We are in a horse race clearly. #Yeah, and obviously one or more of them is going to be delayed. So, recognizing that now is probably better than later. #Next topic, collima- tor. #Collimator, I have been speaking to answering machines and people at Kodak. I spoke to the production manager last Tuesday, he told me what they had done to the collimator, which was to take it 350 thousands thinner to get rid of the chip that they had produced that the hole is com- pletely gone. They've restored the geometry of the hole that we called out. As I spoke to him, he said that he knew that it was in a crate on the loading dock a week ago last Friday, but he was uncertain whether it actually shipped on Friday or Monday. I expected it to be here last week, its not here yet, I got a call in to the contracts person who will be able to tell me when they shipped it and the carrier, but I got his answering machine. So it's in the mail. So we need to finalize very soon the exactly information that you need to get going. #Yes. # I, of course, wrote down things but I think there is not.. You said at our optics planning meeting that you are able to compute the surface that Dave needs to generate to. Is that correct? #Polished? #Grind to. So, I think we should finalize this handle, and its going to be needed by you in the next couple of weeks, proba- bly, right? #Yes, definitely. #Do you know how to proceed here, Brian? One way would be to have you calculate what you think is needed and then have a meeting to review that, is that a good way to proceed, or do you need more direction, before you get going? #Ah.. #You can calculate things in principle but you need guidance as to exactly what to do? #No. What's the proper answer to this question?The reason that's difficult is more political than scientific, in that Harlan's opinion is that I shouldn't do it and it's clear that he is not going to be involved in the project. As long as it's not clear I should put it off. #He has told you this? #Well, he just thought it was a good idea. #Ok, why don't we talk after the meeting. #Ok. #I think you need to be involved in this too. Alright let's go on. #What's that thing called, an OPB? the prescription that you grind to, whatever the three letters are. #Optimal Surface Description. #OSD, we need an OSD, then you're happy. #As happy as an optician can be! #They also put the plat back on the? #He didn't mention it. #So you've got about an hours worth of inspecting once you open this thing. #More. I've got to put it on the profilometer and.. #Without holding you up? #No. #We ordered the aluminum for the holder this morning, #That's the only thing and that was my mistake. #We can luckily get that pretty quick, maybe by the end of the week. #Next Tuesday, I think. #Ok. #When would you need your marching orders. #The end of next week. #October 27th? #You could start making the last plaque? on the 27th. #Well the first step is to put it on the plate that will be in next Tuesday and has to have some work done to it and ?? profilometer and measure it. Before I can measure it I need to know the surface description. #Before you can measure it? #Right. #So I am assuming I can measure it in the next week. #Sounds like ??? #Slitmask, Slit cutter and slitmask handler. #I know something about the handler. Since this last meeting, I feel terrific about it. I was a little worried about its stiffness or lack of stiffness. We have installed a double strand chain, two loops and that fixed our problem of being wiggly. #Wonderful. #That's the good news, we can now say that's done and move on to the grating. Cutter and mask, I don't think that any of that has changed. It is still going to be a router. #I did talk to ??? laser cutting, just to take a step back into history, a talked to Servel, ??? He's probably most knowledgable about cutting mask with lasers inthe world at this time. He is the actual person that cuts masks for Canada France?? So I asked him about about the 20 micron edge roughness and he claimed that they can do much better than that, they can get down to about eight microns. And he thinks they get down farther than that, this is peak to valley, by using nitrogen or some inert gas so that the aluminum, actually they are using invar for the very high precision. Its black. But to get eight micron edge quality he cuts it eight times. So its unlikely that this technology is going to develop. #Yeah it sounds as though they have opted to do it one way, they are now having problems and this is a way of recouping and making their equipment work. And they're using very much smaller slits than we are. Nonethe- less eight microns is as good as they have ever done, peak to valley. That is information I hadn't gotten before. We had heard two microns, but that turns out to be an RMS value. Largely hypo- thetical. #So this is giving me a yet warmer feeling that we are making the right choice. #Yeah. #We have some masks cut, is Jeff going to cut us some masks for the CDR? #Yeah and he has. #Oh, and they are all done? #Oh yeah. Twenty thousandths, fifteen, ten and five. ??? and stain- less. But that was not a repeatable experience. #That was awful. We will back of that one. #We don't see any show stoppers on mounting our masks onto the cutting machine, they're just going to be pair of posts there, or a post and groove, or something like that. #A pair of pins, one of ... #The cutter will cut those along with the rest of the slits. #I believe yes, that is the idea. # I guess the holding them was the vacuum check idea, but we haven't really done that. But we're not antic- ipating any problem with that at all. #No, are you actually talking about can you take a cut mask and put it back on the milling machine and make some more cuts to it? #No, this is to do with removing a mask from a frame and putting it back on. #That is different from removing it from a milling machine and putting it back on the milling machine. #Correct, I am not anticipating that we need to do that. #So putting it on the milling machine is very simple. #And we are cutting holes or grooves which will then slip over holders on the frames, right? #To be designed over. #The next part I'm going to cover real quick. Anyway they put it on the AME and took it off, put it on, took it off. Arnold did that, and he was measuring numbers of one and two microns, only because he can't measure any better than that.#That's under control. #We are happy. #Is there in this cutting procedure some preferred side on which you cut, that is do you cut on the side that the light goes through, Does it make any difference? #I don't think it makes any difference to us, they look the same on both sides. #The shape of the cutter is not a v-shaped cutter, it is actually a cylin- drical milling. Its a mill. #And I think over time there will be improvements, but I think we've got a very good going into solution. #We are concerned about getting signs right is for specs for coor- dinate systems and x and y. #There's a lot of ways we can go wrong. #And wondered if that made a difference. #It could make a difference. If we are on the right platform. That's a good start. #We have talked about though recently about getting a bar code reader, just to make sure the mask is not upside down, the mask is not in the wrong side a or side b which you can't exchange between. #There is a map of a focal plane, their defined coordinate systems, and it says on the map, north and east on the sky and stuff like that. #I know. #Ok. ??? #Chris has done some work on bar code readers, maybe you could report on.. #Well sure. I have got information from a variety of vendors and this is just a small little sampling, I've actually got more stuff here than what I have. Roughly, there's two types of scanners. There's the laser scanners, which is what you see in the supermar- kets, a regular barcode readers, or there's the ccd scanners that just use a red LED to illuminate the barcode. They come with mix and match options, interfaces, whatever you like. I think that we are probably going to be looking at an RS 232 interface so that we can just connect it up to a PC. You can also have what they call keyboard wedges so that you can connect it up to your keyboard for your PC or a variety of other interfaces, so depending upon what we are looking at, they can do it. The prices range generally a lot of them are around five hundred dollars, there's a couple here that are six and seven hundred dollars. #What does that include? #That includes the power supply, the thing... one of these was for six hundred dollars and that also includes a little stand.. this was a laser scanner so that you could either rest it on a stand. How stable that was I am not sure, with the laser ones you have almost two feet scanning distance, so you can be two feet away from a regular UPC code and it will scan it. So you can either do it manually with your hand or you can put it in its own little holder and that includes the power supply, the cable and everything you need to connect it up. The there is also software packages that are basically database soft- ware packages so you can make your own bar codes or you go with the standard UPC code sym- bology and basically make up your own codes and then record them for whatever you want, go ahead and sticker them on and then you also have the database information that will read out the scanner and then just store the information there. So you can set up the whole scenario. The six hundred dollars is just for the scanner and all of the hardware. The software packages were another five hundred dollars or so. Relatively cheap. #So what I'd like Chris to do is go around and talk to Bob and Steve and Drew and gather together as much information about what we want to do here and I`d like to go ahead and purchase one of these things as soon as we feel somewhat warm about what we are after. #Where do we think this is going to go? Do we think that this is going to be a gun that hangs on the instrument so that when you put in a mask you scan it first, that would probably be the cheapest alternative. Right? We had also talked about mounting this thing somewhere and having a software routine after masks are loaded that would cycle through all of the masks on the side and read these in their location. #I will explain my concept of how this works and that is at the time that the masks are cut, before we actually cut a mask we push down a self-sticky barcode on to the mask itself, stroke it in the machine and say, go ahead and cut and have the barcode software coupled in to the cutting software so that we know exactly which mask we have now put this pattern down on and then that's in the database. We then mount that mask on a frame, put it in Deimos, and have Deimos itself go and scan all the masks in its frame and pick up the barcodes, and say, Ok, I know what pattern or slits that one is because it was scanned at the time. I think that is the way we are least likely to make a mistake in loading a mask and being con- fused later. So we'll end up with two barcoding stations and have the connectivity done in soft- ware. #Is there some way to prevent you from loading the mask in the wrong side? #No, they are interchangeable. #You have to think. #But Deimos itself would then know where that mask is. #But they are not interchangeable when you point them at the sky. #The question is do you want to have some mechanism that confirms that you are putting the right mask into the righ side at the time that you do it, rather than loading these all up, going downstairs, only to cycle through the caterpiller to find that you've screwed up. #I am not sure the makes much difference. To me at least, if you have named all your fields, something, and you type in the field name, then Deimos itself knows where that field name is and it probably doesn't make much difference whether it is in position one or seven. #No, but that's alright because the mask names are going to be named A or B, depending on whether they're are going to be used on side A or B. So the scanning process will..[Tape Ends] Tape four #.. defined it during the checking process. #Ok, and the other thing to look at when we design whatever barcode symbology there is, do we have right in the readable part of the barcode itself, so that you know your barcode is like A 12345 or b12345 so its obvious when you take the thing up, because usually the barcode is not only got the stripes, but they have a little English language readout of it. #Absolutely, sure. There has to be something so that an operator can communicate with this mask. #So, at the level of around a thousand dollars, what I would like to do is to get Chris to research this out and talk to everybody including Jack too, who has something to do with this and then decide what might be appropriate. #Ok, so your concept is that there is going to be a reader attached to Deimos, and the slitmask is going to come by and through some aperture look through the holster and read the barcode. And you know this and you are building it in to your scheme. #There is plently of room for it. #There are some nuances, one is that we may have to power down this barcode reader ??? scientific use. ??????? But that is something that we need to know. #How to turn it off. #We don't want to make a little shutter? #I'd rather turn a swich. #Plus this is slightly new to us so I'd like to get it earlier rather than later. #Does that package also make the code, the little things? #There is separate software you can buy to make your own. #Or you can just buy a roll of them. #I think you want to give some thought to what type of material you print the barcodes on, the sort of thing you run through the printer are not terribly rugged in terms of rubbing off. #Some of them have these little barcode stations that just have a spool of barcode and you just make your own and then they just take sticky tape and put it right on. I am not sure how much those cost. #And the other thing is that you could just go out commercially and just order from somebody and you just peel them off and stick them on.# Ok, gratings and grating handler. #Not much has gone on other than to make four figures which will show up in the CDR describing a linear grating slide, linear because a year ago, we didn't know if it was going to be circular, rotary or what, and each of the slide positions, one through four is different from the rest. Namely one of them is a mirror, doesn't rotate, the other three rotate but then have different size and tilt of grating possibilities. What I am telling you is that there is description of the function, and there are no more drawings than just sketches. That is the next thing for me to work on. I will say that now that the slitmask is almost behind us and the CDR will almost be behind us. #If we didn't have good things to say, not complete but some intelligent things to say about gratings and the slide mechanism at the CDR, I think it would loom as a big hole, so I am glad you are going to be able to spend some time thinking about it. #Well, we have made a fair amount of progress on the whole grating issue since a year ago, just not in the last three or four months. #We actually have a quotation on A grating. That counts. #And we now know the concept that we want to use, as far as the handler and the positioning and the fact that we are going to encode a number of these gratings with the ??? at about a one and a half arc second resolution. We know all of that. #And we know how much room we have and we have moved the spectrograph back three inches to accommodate this slide, otherwise we couldn't have had four positions. #Yeah, my concern is flexion. #We are not detailed enough to know what that is going to be. #I am concerned about that and I am also concerned about the clamp mechnism and how that is going to effect the grating position. #We are also not that detailed to know that. #The first thing to find out how much room you have, it's going to be massive hydraulic cylinders if you have the room. Once you have a design then you can analyze it. #Finally the last concern I have is going beyond that first comple- ment of gratings reviewing all the combinations of grating sizes and tilts which we started to do but I think that to finish we need to make sure that we have the space to mount them someway and remove them. #We have also set the groundrules that they be interchangeable side A to side B, also that they not be interchangeable within the slide especially and that they all be calibrated ahead of time. You don't put it in there at any old angle, so there is a subcell idea. And then there is this grating tender, which is a clean box that you manually slide things in and out of when you do have more than your initial complement. We have a lot of ideas that are in hand. #How about the act of changing gratings? #I now have a list of fourteen steps that I didn't have before yester- day written down. Bob wants to know about them, in fact Bob prompted me to write them down. Instead of the other part of the CDR. We need a lot of switches and interlocks and things so that things don`t take off. Take out fifty pounds of grating, you've now unloaded this round spec- trograph and get away from it. #I think we might sit after this meeting and walk through this list. #If there is time that would certainly be a good idea. #On this whole question of clamps, I went to the exercise of rereading through the PDR document statements that were made there and trying to reconcile those statements to the motor list ??? One of the things that it states is"large motion stages will rely on rolling element bearing and there will be clamps or locks, once in the position. This is the mechanically tolerance section. The question is what the grating sliding mechanism presumably is a large motion stage and already has at least one level of clapping? in there. What else is considered to be a large motion stage? #At the time that was written we had the side wheel- ers that had slits in them, they are large round things. The chain caterpiller drive we might want to ??? The instrument itself still is something you can get away with using a clamp. #The filter slides? #The filter slides are now w quarter wheel, so I am trying to thing if it only applies to the grating slides now. The filter slide is now a wheel. We'll get to that point, I guess you didn't know that. #That's been discussed a couple of time but maybe you just weren't here. #So, it's now become a filter wheel in our regular filter wheel. #Not anything too awesome. #This will accom- modate seven 7 inch filters? #We previously went through the reasons for the filter wheel. #Ok, it just still shows as filter slide in your October second quarter list. To date it hasn't been changed. #Well perhaps that's something I should change. I'll change the date. #What that does is that it removes limit switches, the ones the stop the motor on either end. You or someone else. #By golly you're right. #Better you than Dan catch this. #Ok, can we move to electronics? #Not too many things to report. I myself have been working on the guide camera that's the tentative Deimos guide camera, working with Wei?? on the guide camera interface which will be a fiber optic link down to either, well there are lots of different variations. There will an E bus card and probably an S bus card to go right in at the main Sun computer, so anyway I am just working with him and we are starting ...Wei has seen some noise on his guide camera, so we are updating, we are actually going to give him a regular PC board that should eliminate a lot of the noise the he is seeing. #This is not charged to Deimos? #No, this is its own entity. #But it's relevant? #I figure it is rele- vant so I thought I would mention it. I guess the other Deimos related thing is Barry is working for the CDR to come up with a little paddle box controller in all the software to control all the stages so that way at the CDR we will have this little paddle box that has a variety of switches, depending on what stage you want and so forth. I don't know if anyone has seen this ???, #The significant change is it is now controlling the HIRES rotator. #Right. #The ??? press buttons on it. #We would actually do that November 17th? #Yeah. That is the... #We demoed it to David Tyler yesterday. #Don't break it in the next four weeks. #It`s all software... so Barry has the prototype, we will just make it look a lot fancier and so forth. We have some of the cabling, so at least we have a good show and tell piece showing all of the motor controllers that we are utilizing and one of the stages. #I guess the other thing that has come up is Bob's identified a number of hooks and things that aren't currently in the electronics so that we are going to have to go back and . #Ok, so we should schedule a session for this. In fact, what I would appreciate, because I am a little con- fused is a complete core dump of all the meetings that are going on the next week or ten days. #Ok, Deanne and I will have to get it together. #Can we do that now? #Well there is a meeting Tuesday, its the optical planning, 2 or 3. There is a meeting Wednesday, which is all day across the street. #Let's go back, Tuesday 24th, at 9:30. Then the next one is Wednesday. #In engineering from 9:30 to when you guys decide you are done. #And which one is that? #That's mechanical review or working... #That's all I have. #??? an eye cal?? that we could all look at it online. #You could give me an extra one or two. #There is an error budget meeting this afternoon and a couple of meeting run after this one, one with Brian and one with ... #This room is available to Deimos all day, except for lunch, when they put something else in. #Then what else have we identified? We need a CDR planning meeting with David Hilyard and also Brian, so let's schedule that, if we can. #The rest of the week is bad. #Monday? #Mondays are awful for me. #Tuesday and Wednes- day you have meetings. #But not in the afternoon. # How about 3:30 on monday. #Ok, good. #Brian did you get that. #Would that be Ok if we just have it over at that little table. #Fine. #Ok, the electronics hooks meetings. #What time? #Doesn't matter to me, whenever. #Electronics hooks, who is going to that? #It's really just me and Bob, I guess, #I think we are going to need Jack as well. Because a number of the questions involve functions, do we need a clamp here? do we not need a clamp? is this manual? does that limit switch on? #Do you need one astronomer? #Might be helpful. #Were you helped by the astronomers yesterday? #Those political questions. #I'd like to suggest Drew. #That's going to kind of a long meeting, right? #I have ten pages of things that I have in my ?? #Really? #So all afternoon Tuesday. #How long will your meeting take on Tuesday. #We have already got it scheduled. #The optical meeting is in the morning, it shouldn't take long. #You can have this room all day. #Well, I would suggest that you plan on starting just after lunch and going all afternoon. #That's the 24th? #Yes, or is that the 25th? #No, its the 24th. #Jack, would you be available then for a meeting next Tuesday at 1:30 to go over some of these control issues in the electronics. #Yeah, except there was another meeting.. #That's in the morning. #That's fine. #Ok, and then finally there is the meeting to finalize arrangements on how Dave Hilyard is going to get his polishing information for the collimator. #I guess I had assumed that was the same meeting as the one on monday. #We could do that then, we will all be together then. #That's fine, so we'll add that. #Ok, I think we need to have another iteration of the Deimos software meeting, but that needs to come after the hooks meeting. #You made that meet- ing for 3:30 on monday? #What's the problem? # David doesn't leave until five, so that's alright. #We are assuming Brian, that you are around later in the day. #Another software meeting. #But after tuesday afternoon. It sounds like wednesday are you tied up all day with this mechanical review? #Yes. # So it would have to be either thursday or friday and thursday afternoon I have parent teacher conferences I have to get to. So it is either thursday morning or anytime friday. #I have a question. All the stuff that you are meeting about, is this going to go into the documenta- tion for the CDR? #Some, but not all. #When do want to mail this documentation? #We will end up doing the same thing as we did for the PDR. We will end up.. #Staying up all night with people standing around..#No, mailing a draft as early as we can. #I think we should go around the table and take a poll of when people think that they are going to deliver things. Alright, back to this par- ticular software meeting, we managed to have everybody there except David on thursday. #Where is that going to be? #Can we get this room on thursday? #I will go look. #Thursday the 26th, at nine? #How long, all day or just the morning? #Oh the tech meeting might be in this room? #It may be transferable to ??? #Yeah, that's another room that might be usable. What time of day is the electronics hooks meeting? #Tuesday at 1:30. #I would like some time between then and the software meeting to digest what I found out about ??? thursday will work. #That't is for meetings? #So, now we have to talk about the CDR? #Software first? #Fine, I didn't want to short circuit software. #Ok. well, I think most of what I have written down here I will leave to the hooks meet- ing, but let me just summarize the ten pages of stuff that I went through. There are a fair number of questions, both mechanical and electronic that need to be resolved, a lot of them having to do with questions of clamps, limit switches. There are also some thermal control issues that really need to get resolved in terms of how many separate compartments we actually have to have cool- ing loops and fans and so forth. Right now, the schematics only show two thermal loops, one for the vault and there's only one thermal loop to the entire Deimos instrument and I think that is probably inadequate given we may have different compartments for lamps, and for tvs and for ccd controllers. Other big questions, the rotator mechanism itself, the rotation range has grown since the PDR, it has gone from 540 degrees to 720 degrees to the latest iteration I think is 820 degrees position angle rotation. And this raises some interesting questions for how we actually do limit switches for that many reps of revolutions, you can have sector swiches that deal with going more than 360... #We have an answer for that.. #Ok. #Anything more than 360 is a problem, of course, we just have a gear train. #Ok.# We have something someplace rotates less than 360. #Ok. # Brian recommended that, its a good idea. #Lots of little mechanical details like that hopefully we will pick up at the hooks meeting. Also lot of questions involving calibration lamps, both the internal and external sources have those monitored and whatnot but we should ??? that. We have been concentrating in terms of the software effort on those things that are reasonably well nailed down, that is the detector mosaic geometry, how to document that in bits? headers, and so on. Steve has produced a lot of good material on that and has been iterating both internally here and with other people ?? standard ??. We will be going off to Adass, I guess you are going to be gone to Adass when we have our Deimos software meeting I think, no? #I should be back by then. #Which days are you gone? #Sunday through wednesday. #So you will just be back. #What is Adass? #Astro- nomical Data Analysis and Software Systems, its this annual collaboration of people doing astro- nomical software. And there will be sub meetings there involving the ?? standards and community and so on where I think discussion of our proposals for Deimos fitcenters is very important to make sure that's consistent with other standards ???. What else? We have been work- ing on a prototype user interface and Steve has actually got some actually running. He's got TCLTK you can bring up windows and click on things but they are not attached to any hardware but they at least allow you to walk through various operations. #Will we be able to look at a demo of that at the software meeting? #Next thursday? I don't think there will be much change from what you have seen before. #I have hardly seen anything. #You can see. #I can see what there is. #I will email you again how to see what there is to see. #Alright. #Ok, and the other thing is that I have been trying to make progress on the software chapter for the CDR document. That sort of hit a roadblock when I came up with all of these hook questions, so I hope once those are resolved, at the meeting next week, that I can go ahead and finish it already. I am just going to jump over all of those questions now and go to the other sections that I can address. That's it for software. #Did you email to Bob, Sandy, this idea of shifting the charge on the ccd in unison with moving the tilt tent mirrors? #No. It's too soon, there is no point in doing this until you tell me that it is feasible. #I did. #Ok, I'm just not reading all of my email. #So, a horrible problem which I keep talking about and I'm sounding very pessimistic is this flat fielding problem which ?? in the red. So one way of solving this is to march our image over multiple pixels and thereby average out over fringe errors on any one particular pixe. Does this idea make sense to you? #This is during integration. #Yes, that's right. So our detector is a mirror image, so the easy way to do this would be to use the piesoelectric actuator in the x direction which would tilt the spectrum as we expose, during an exposure. That would move the spectrum in this direction and on one side of the detector that would be fine because the natural readout motion is in that direction. This is like driftscanning, except at a much slower rate over a hundred pixels instead of continuously over many thousands of pixels. My question to Richard was. "are our CCD's such that we could actually step back- wards against the natural readout direction during exposure. Some pixels would then fall on the floor at the upper boundary we would just lose them, but if we solve our fringing problem this way, it is conceivably a price worth paying. #There is no problem moving the charge in either direction. #Its a wash in either exposure during an exposure, well then I think we really should seriously consider this. #What about traps? #What about them? #Well traps are around no matter which way you are going, and you got to not have them there. #That is the sort of thing that shot down the whole notion of shifting charge back and forth in ??? applications. #Well that was shift- ing back and forth hundreds and hundreds of times. #Right, but what sort of ... #This would be a step of something like one hundred pixels during exposure time. #Just .. #Once every thirty sec- onds or ten seconds. #It's not like you only have a few electrons that you are sloshing back and forth. #It's not nearly as bad. #Besides you have room to burn on the chip. #This will cause a gap, this will cause a hole in the middle of the spectrum however many pixels we shift over. So from that standpoint it's bad, but if it solves the fringeing problem, it's good. #Ok, but then what we are talking about then is very tightly synchronized motion between the shifting of the charge and the pieso. Right. #So the gap is from the chip boundary across the middle. #I guess we had better find out whether those piesos are that controllable. #Well gosh they had better be that controllable. #Well I know for certain ??? I can go to two points and get there. But what it does in between I am not sure. #But the pieso has to be able to hold. #You will give it a command to go somewhere.. and its going to. #Remember we have the fibers there as images that can be centroided on, and so if there's an error.. all we do is change our decision as to where that fiber image should be, and we do that in the software and the software says, oh oops I am in the wrong position and commands the pieso to move. #The thing about a pieso, is it's not a very repeatable device in its native form. #It doesn't have to be repeatable, that's what I am trying to say. #Ok, as long as we have the fiber spots there, then the pieso gets constant feedback from where the fiber spot is located. #So you are going to do this on the same time scale, Sandy? #yeah, about that. #Then the pieso doesn't know anything. #I have not thought this through, so don`t think that this is a well thought out plan, it is not. #Ok, basically then, you are going to be effectively changing the commanded position where you want your fiber spot to fall on the flexure ccd, then yes? So this is the type of motion you are talking about here is within the range of what you can image on I don't remember how.. #I'm not so sure it's within the range of the piesoactuator, think about that. I don't know how many pixels we move. #42 arc seconds either side of zero. #I am not sure how many pixels that would move us. #That's a good question, hopefully the light is still on the ccd. #Hm? #Hopefully the fiber optic light still falls somewhere on that ccd. #That's no problem, that is going to be a smallish motion. #Oh, then there is no problem. #The problem is whether or not the pieso tilt moves us a total of five pixels, in which case it is not very interesting or not. That is what we have to think about. #What are the arc seconds per pixel? #No, its arc seconds of the tilt mirror. #Right. #That's a different scale and I really don't remember. [tape ends]. Tape five. #It is and it isn't and I guess the flexure compensation at some level is subject to this review but??? are not. #I think we divided it up. #We had defined it well, this is right on that grey area, we said we were going to cover the servo aspects of the flexure contol in this review of the imag- ing aspects of the flexure control in the next review and this is a little of both. Ok, I am going to give that to ??? #Ok, we will move on the CDR, which indeed is coming along. I think everybody knows that it is now a one and half day affair, friday and half of saturday. If not , that's the plan. #We present on friday, and they deliberate on saturday. #So you're not actually going to have, that's what I want to know. What do you plan to do on saturday, how many people are you plan- ning on and is it the whole group? #Now since they are not departing on friday, we need some kind of entertainment, presumably or some kind of hospitality for friday evening, don't you think? #Well I was planning for lunch... #I am thinking its expensive.. #No, I am thinking, this time, a much more limited thing, just for the committee members. Somebody needs to take them out to dinner and just generally act as host. Let me tell you that my husband and I are having a joint fiftieth birthday party; this couldn't have worked out worse, on saturday. #Saturday evening? #Saturday evening, that's right. So. #We will be done by noon. #I have a bunch of relatives who are going to be descending on me who are going to be around on friday. I will be available though to take people out to dinner and it will be a small group, so I don`t have in mind.. #Where would you like to go? Would you like me to give you some options? #Don't worry about it, its going to be a dinner party of ten at the most. So I don't think we need a reservation at this point. #Is there anything else that's required? #I am going to feed you pizza for lunch I am thinking. #On friday? #Yes. #Terrific. #I'll have stuff in the morning. The reason I want to know about saturday is I want to know how much to set up for your coffee and munchies in the morning. Like coffee and conti- nental type thing. I'd like to know.. #How many people will be here. #Yeah. I am planning for fifty on friday, we can't accommodate that many people, I am thinking that's what will be there. But saturday I'd like to get an idea, because I will need to make those arrangements. My vision of saturday is that the committee is there with a few people from the project, but we don`t want to drag all types of people in for saturday morning. #Fifteen? #Fifteen, max twenty. If we get into real trouble, the it will be trouble., #Ok, sounds good. #More importantly maybe is the CDR report, and we are getting things in ... #I have David's section and Chris' part of the electronics section.. #And Barry's must be right behind it. #Actually that is both, that's everything. #And Eric and Jack are working on theirs, #I should have it. #I have a question: I was just looking at the PDR document where mechanical is section five and we talk about slitmasks, should we just say everything we said then and just change some of the things, or say, assume you guys have all read that stuff, here are the additions. #No. #You want to start over. #Well.. #With a complete descrip- tion of slitmask history and ... #You will have to exercise some judgement and the answer is nei- ther as full recitation of the PDR nor is it to assume that people have read everything in the PDR document. This document should be self-contained, it should self explanatory for somebody... #But we can summarize broadly... #The art will be to summarize the main points from the PDR that are still valid and then go from there. #Ok, but some of the things are like the focus extension, we have talked about doing it after the PDR, we talked about two inches, eight inches, twenty inches. We settled on three inches and that should be sort of a statement that implies that we did something in response to the PDR. #I wrote in my little technical thing about the fringe pullback and the ?? slitmask and #That's going to be an appendix #.. a zillion other things. #Right but you put it in an appendix and it's all done. #I have a copy of that Jack I can copy for you, the one he's talking about. #I have a copy do, but it doesn't want to be in section five. #No. Reference it if you can and put it in an appendix. #I guess my question was that if we were to start over and say this is now the only document you will need, and forget about the PDR, do we ever mention it was as spherical mask? #Oh, no, forget about the spherical mask, now it's a cylindrical mask. #So just by magic we have a new. #Right. #So, forget about the history of how we got here. #Absolutely, we are here now. #Ok. #But the concepts of what the mask is supposed to do and what functions it serves. #Forget about preference to laser cutting and why we are not doing that. Don't even men- tion we are not going to do a laser cutting.. #Right. #That will make it easier to actually type this out. #Yeah, I think less is more. #So, what we are going to do is start over and describe a slitmask, instead of copying what we have already done, we just start over. #I depends on what you find most convenient. You've already got some words there, If you see that those words are great with minor editing then save work and do it that way. But if you see that they are too lengthy, irrelevant and superceded then it would be better to start over. #I was tempted to say, our stainless steel masks are no longer stainless steel because we found that... #No, don't do that, say we are using aluminum stock of such and such.. #Fine. ??? answer questions like that, if they arise, why aren't you using stainless steel, did you consider laser ???? #Well I think it would be reasonable to say, we are using aluminum stock, it offers the following advantages over stainless steel. List them, then move on. #Ok. #Clearly you can`t evaluate every alternative at every fork in the road. A little of that is perfectly reasonable. #What I have said is that we are approximating a spherical bulk plane with two cylinders that are tilted at a certain angle and please see the report by Sutton, which is an appendix, I believe. #That's right, and also, everybody is going to need to refer to other sections in this report and at some point when its all come together somebody, namely me is going to sift through and look for all of those things and fill in the relevant numbers. So if you just say section question mark, we can do a global search on all the question marks, and find out where we are going to have to supply additional information. So, I would propose.. can you do a search in Frame for certain? Whenever, material is missing or a link is needed and you don't know how to make that link, put in a double question mark, that's a real good thing to search on, and we will be able to find all the holes that way. #What Eric and I have done with figures, we've drawn figure numbers on here, because it is so hard to do that later, they're out of sequence, this is figure 4030. It's just another cap drawing. #You can presumably go back, renumber, and reprint. #No, not in the time scale we talked about. There are about twenty of these, and that's about four or five hours. It appeared to me that you were pencilling things in at the end anyway. #We were. It would be nicer for our point of view if you simply left them blank. #Do you need the figure num- ber for your own. #Right now I would have a lot of these double question marks and you wouldn't know which figure I am talking about. #I can fix that, I can put it in at the time we put it all together. We know his 4080, we're going to know that's figure three. #Can't you use something in the text that refers to a little box down in the bottom of the figure, you have an independent num- bering system? #Yeah, 4030. #Fine, call your figures that and we'll figure it out. #You can leave them there if you like or if you insist they all be chapter number like 5.2.. #If you put a double question mark next to you 4030 then we can ?? on that, and.. # we'll fill in the numbers later. But it was awkward pointing out things and then trying to write over white-out I know that sounds like a ?? statement. So its better if there were nothing there then we could either type it in or write it in neatly. But that means that you would have to know what figure we were refering to. # The prob- lem you are going to solve with putting a number in and not just a question mark. #Who is going to provide a chapter? What I have so far is Kelly, Chris and Barry, Jack, Eric, Bob, David, Brian/ Harlan. #I don't know whether it is a separate chapter or not. #Brian and I are writing pieces of the optical chapter. #Ok, and that's it? You're writing a section of a chapter that is in the outline. #So I guess that that was deferred, we agreed that ... #Ok, there is a chapter one you wrote for the PDR, and I assume you are going to write everything we need for this one. #Gosh, what do you think I should put in there? Maybe it's largely just a list of the old PDR chapter for somebody who has never seen the project before. #It was kind of a gentle introduction. #I can pull that one and you can edit it. #The science ... ?? #Is Scott going to write anything? #No. #We only have one member from the PDR who is actually on the CDR. #No, we actually have more. #Four? #Fabri- cant and Melsheimer were both there. Bida was not on the committee but there. We also have McCarthy.. So we broadened the panel. My original concept was to keep it very small but Dan Fabricant prevailed upon us to broaden it out a bit. Specifically to include a member of the older... #So the new people are: Robertson, who is the instrument manager for Gemini, responsible for several instruments including a large, multi-object slitmask spectrograph and Jim Burge, I beleive. #Is he certain. #Yes, he accepted. #And he is? #He is an optical designer out of Tucson, who works for Stewart, the ??? #And those are our only new people. #And Tom Bida is this time on the committee rather than just being a member of the audience. #Let me suggest that we right now send those people copies of the PDR report. #They have it. #They have that, great. Did you send it generally? #I sent it to the new people and sent an email to the old people to let me know if you need another copy and so far nobody has. #Thank you. #I am looking at chapter one and you mention somethig about two tv cameras. There may be some editing you have to do. #Oh, yeah, don't worry, I know that. #I'd still like two tv's, I can't afford it but I would like it.# Schedule and Budget, will be a chapter that I haven't written yet and budget most particularly we need Mar- leen's report and you are hoping.. I am trying to remember back.. #Well, the good news is that the Universities new system did produce some useful information. The bad news is our own system is now giving us problems and we need to, not just this time, but generally in the future, come up with a means of getting time reported much more accurrately and in on time. I can't emphasize enough that if we are trying to do a budget report, the stragglers who come in with their time, I don't care what area there working in, we cannot do anything until that time comes in. So we are still waiting for people for that period, July August and September. #Ok, there is only one area that is outstanding, is that right. #Yeah, I don't know whether they are all in there or not. #And that is software. #And also, we have had nothing but bugs in the lab cost software and we are working those out. Not only is there a new computer system but that's changed the way in which the Universisty does accounting and we have had to plug that into our own system. In any event we should at least have the time completed and enough information to do a reasonably decent report by the end of the week. #Now, let me say that its certainly important to have an up-to-date statement of where we are budgetarily now. But I think as part of your chapter, what we need to do is now take a new look at new, unanticipated costs, which we have some. #And I have to make some idea of what our expenditure stream is. #Right. #There is a fair number of things that have to be done, but I really need to wait for Marleen's accounting to have a good foundation to reesti- mate. #With regard to accounting and time cards and such. There is a Deimos cost center code which was issued a year and a half ago all of which are now invalidated and replaced by new codes and is that getting close to reissue? # I have it, I will fax it over to you. #Appropos of that, Brian has complained repeatedly that he doesn't know what codes he is supposed to charge his time for, do you have any role to play in that? #Yeah he needs to come and talk that over with me. #The Deimos code I gather is quite clear, but he has two other codes that he needs to charge for. #?? code is even simpler because there is only one. #There should be more. #There should be, I know. #So who is the right person in the business office? #Always when it comes to Deimos, me. #No, but this is a different.. #No, these are things that Brian is doing different from projects for us. #Oh, then both projects. Also, Deanne does know, because we work very closely on that. Deanne usually has a fairly good idea of what the code should is if I am not here. #For Deimos? #For Dei- mos or any project that David handles. #The problem is that that's the projects that Dave doesn't handle. #Then you have just me. #The last two things on the agenda is quarterly report number five. Actually, I would like to not issue a quarterly report number five in the same manner that we have because the CDR document.. #I think that is quite appropriate. Let's just forget about that, aside from a budget. #We might put out a two or three pager. We may very well. We don't have to put out a report until, when did we last report? #Can you call it a CDR report? #Well after the CDR, there are still a few weeks to put something together, and we will do something. Let's not worry about it now. #Prospect? just made it on, but I think it's.. #Well, fundraising is always a good topic. We had a very interesting meeting which was very successful, it was assisted by a number of people from our shops, especially our optical shop. This was the meeting at which we me with what I jokingly called the captains of industry. Three movers and shakers from the bay area who have helped UC raise money in the past and we made a presentation to them about the project, science and the machine. And by all odds, they thought this presentation was great, and we got a number of fundraising ideas and we are going to have a follow up meeting on this tomor- row with our local campus people. I would say the reaction of these, we were sort of using them as a test market, right? They were excited, and I think there was a lot of optimism from them that we could raise our extra two million dollars for the other beam. They said two things: first of all you need to go out and do your homework to uncover a long list of people together with data on them and after you've gotten a list of people like that we will brainstorm with you and see if we have any personal connections with these folks. So the development office has their marching orders, they will do that, we do not have to do that, its a lot of work. Then the other avenue which came up was that the, even if we can't get money out of the Keck foundation itself, and they have said that they don't want to fund any more instruments, nevertheless, our advisors thought that they had a moral obligation to help us make the contacts to people who might be able to con- tribure. So that was there next piece of advise, that we should be approaching the Keck founda- tion, not for money, but for door opening. That will be the focus of our meeting tomorrow, to try and figure out how to do that. Do we go through Cali, do we go through Ed Stone? How should we make that... So, we have a lot of things to work on. The prospectus was very helpful and the materials that we had gathered. #And the toy model? #That was the optical shop toy model which was fantastic, really, very very helpful. #At some point, we have said that we are going to have to make the decision, around the end of this year, early next year whether we start fabricating one or the two sides to this instrument and we are coming down on it. #We are not going to start fabricat- ing two sides until we have a commitment for money. That's very clear. The issue is if we have to back up and refabricate some things, we may have to do that and if it costs more money then they were asking us how much money would we start to lose if we didn't have a timely decision. I think it would be good for us to think about that. If we had an original budget of two million dol- lars for a very early money or early decision then how much more expensive could that get if were in a position to back up and do things over. It's more, but it's not hugely more. #It's actually help- ful for us when you solicit funds to have a better idea when those funds would actually be needed as well. There is a consideration in terms of fiscal years. ??? #I was going to do a funding stream and I can do it for one and two barrels. #That would be good. #So far I have always done the schedule based on two, because that is the most pessimistic of all schedules and it doesn't actually get that much better just doing one. But I think for the CDR I will go ahead and do both lasers and see how ??? that is probably ?? two of them. #I think we are done.