Date: Thu, 14 Jun 2018 23:08:02 -0000 Subject: Re: [sloan-25mlog 4142] Re: 2.5m obslog 20180612 (SJD 58282) From: "Dmitry Bizyaev" Hi Audrey, > I don't think that we need to go this route (yet). It always depends on actual temperature, of course. Renbin's answer suggests that we can get not completely useless data if the scale must be kept off its best position. On the other hand, such not-best scale spectra cannot be counted towards the completion of the main MaNGA program, so we'll have to re-take the spectra under better conditions, again. This is an issue, but not a big one since we have only 5 nights of MaNGA left until the shutdown. We expect the improvement of the scale situation after the shutdown. Best wishes, Dmitry >Tuesday night, we had > problems with the 5C drilled plates when the temperatures were 16-18C. > Looking at the forecast for the next 10-days, we may (knock-on-wood) catch > a break with much needed rain storms. They will likely bring cooler > temperatures to the mountain. > > Tonight, the temperatures remained around 14-15C. We observed eBOSS plates > drilled at 5C close to the M2 limit, but luckily had a little room to > spare. It would be nice for the temperatures to behave another couple of > weeks before shutdown. > > eBOSS does not have any plates drilled warmer than 5C on the mountain, at > least as far as I can see. So, if we do run into problems with those eBOSS > plates later on (no exposures were acceptable on Tuesday night), it might > be a good idea to have a backup MaNGA plate ready to go. > > Also, with Dan's and Renbin's input, it makes sense to sacrifice a little > with scale to reach prime focus. We should be able to make this work. > > Thanks, > Audrey > > >> Hi Renbin, >> in this case would it be reasonable to swap time with >> some other survey (eBOSS, but may be APOGEE, also), >> if the night it too warm to handle some cold-temperature >> plates? >> Best wishes, >> Dmitry >>> Hi Audrey and Dan, >>> Thanks for these numbers. It seems the scale at the M2 limit last night > was >>> at 1.000431 . Compared to a scale of 1.00000, this corresponds to about > a >>> 140 um shift at the edge of the plate, which is about 1.16 fiber > diameter. >>> I think we can use the data in some way even if the scale is off by > half >>> of >>> that, say 2x10^(-4), if that could help put the plate in focus. i.e., > we >>> can put the scale to 1.00022 or 1.00065 to help with the focus. This > should >>> still keep all the galaxies and all the standard stars sufficiently > within >>> their fiber bundles. Though, for now, we don't want to count data taken > in >>> such condition in our complete sets or for completeness calculation as > they >>> could screw up the sampling uniformity. They could potentially do count > in >>> that but would require some careful vetting. >>> best, >>> -renbin >>> On Thu, Jun 14, 2018 at 6:59 AM, Daniel Oravetz > wrote: >>>> Hi John, >>>> We were really close to the M2 limit the other night when I was on > shift, >>>> too, so there are a few numbers still relatively fresh in my mind when > I >>>> was looking things up then. I'll include those here. >>>> The first thing is "a 10C temperature mismatch (between ambient and > drill/design) can be corrected by translations of 2.4 mm and 2.0 mm for >>>> the primary and secondary, respectively." (see >>>> http://iopscience.iop.org/article/10.1086/500975/pdf) >>>> Here are some useful numbers from looking at last night's example > where >>>> M2 >>>> was at it's limit for cart. 5, 8314: >>>> Time Inst Scale M1 M2 Focus Az Alt Temp Wind Dir > >>>> ------------------------------------------------------------ >>>> -------------------- >>>> 05:41 5-8314A 1.000431 -4255 -5657 150 169.2 83.3 18.0 25 130 > nan >>>> boss >>>> 05:58 5-8314A 1.000431 -4255 -5657 150 196.5 83.2 17.8 20 132 > nan >>>> boss >>>> 06:14 5-8314A 1.000431 -4255 -5655 150 217.4 81.5 17.6 21 130 > nan >>>> boss >>>> 06:31 5-8314A 1.000431 -4255 -5653 150 229.8 79.1 17.4 20 128 > nan >>>> boss >>>> 06:47 5-8314A 1.000431 -4255 -5651 150 237.1 76.4 17.2 22 148 > nan >>>> boss >>>> 07:03 5-8314A 1.000431 -4255 -5647 150 241.4 73.4 17.4 27 137 > nan >>>> boss >>>> 07:20 5-8314A 1.000431 -4255 -5642 150 244.0 70.3 17.3 25 134 > nan >>>> boss >>>> 07:36 5-8314A 1.000431 -4255 -5636 150 245.5 67.2 17.2 24 136 > nan >>>> boss >>>> 07:53 5-8314A 1.000431 -4255 -5630 150 246.4 64.0 17.4 22 136 > nan >>>> boss >>>> FYI, with M2 having been at its limit, then the M2 piston value of > about >>>> -5650 um reported in the above table corresponds to a Galil value of > about >>>> -7.2 million steps. >>>> The second thing is, how much more travel do we have available for the > M1 >>>> piston to begin with? I.e., if M2 were to move 4 mm, is there enough > piston range left for M1 to "utilize" that whole amount during scale changes? >>>> It looks like (also according to >>>> http://iopscience.iop.org/article/10.1086/500975/pdf) M1 has 12 mm of > total piston range available. Last night, when we were at the M2 > piston >>>> limit, the M1 piston was at -4255 um. If the primary moves between > -6000 >>>> um and +6000 um (i.e., equally centered around moves of scale factor = > 1.000000), then the primary only had -1745 um of piston range left before >>>> it hit a limit itself. That would mean that M2 should move at least as > far >>>> as is necessary to take advantage of that full remaining M1 piston > range, >>>> I think, right? Using the 2.4/2.0 ratio of M1-to-M2 translation from > above, that would mean moving M2 by (2.0/2.4)*1745 = ~1.5 mm. >>>> Hopefully I understand these numbers correctly, but I do tend to get > confused on units between what's in the night logs, what's in the TPM, and >>>> what STUI reports in certain places, so I'll have to double check for > my >>>> own peace of mind. >>>> Dan >>>> >> Notes: >>>> >> ====== >>>> >> * Cart 5 - We were at the M2 limit for the entire duration of the >>>> plate, >>>> >> so we >>>> >> could not account for additional scale and focus adjustments. At > the >>>> >> same time, >>>> >> we were also battling clouds, high seeing, and high altitude bad >>>> guiding >>>> >> for the >>>> >> first few exposures. >>>> >> * Cart 17 - Again, at the mirror limit, and severely out of focus. >>>> Still >>>> >> some >>>> >> clouds around. There are no other plates to observe. >>>> > We're working on a plan to fix the M2 position problem first thing >>>> during >>>> > the summer, the tricky part is building something stiff enough to >>>> take a >>>> > physical measurement so we can avoid recollimating and taking sky > measurements to see if we've improved things - I'm hoping this will >>>> speed >>>> > the process such that we will finish in a day or two rather than a >>>> week >>>> or >>>> > more. >>>> > One thing that would be a big help would be an actual target number >>>> for >>>> > how far we need to move the M2 support frame out from the primary. >>>> The >>>> > figure I remember from discussions last fall is 4mm, but since we >>>> don't >>>> > want to move too far it would be good to confirm this. >>>> > ____________________________________________________________ Logs of > Nightly APO Activities >>>> > https://trac.sdss.org/tracmailman/browser/private/ >>>> sloan-25mlog/index.html >>>> ____________________________________________________________ >>>> Logs of Nightly APO Activities >>>> https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html >>> ____________________________________________________________ >>> Logs of Nightly APO Activities >>> https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html >> ____________________________________________________________ >> Logs of Nightly APO Activities >> https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html > > > > > > >