Date: Mon, 19 Dec 2011 09:47:17 -0500 From: Fred Hearty Subject: Re: [sdss3-25mlog 399] Re: ADDENDUM to 2.5m obslog 20111215 (MJD Thanks Peter, Dan, and Kaike for taking the initiative to resolve APOGEE differences from practices established for BOSS and/or unique features of co-observing with MARVELS. I think this response to Mark's list of issues should be sufficient for today's meeting -- Mark, would you agree? If not, let us know where clarification is needed. Fred Frinchaboy, Peter wrote: > > > From: Mark Klaene > > Date: Fri, 16 Dec 2011 11:52:35 -0600 > To: Fred Hearty >, > apogee >, Bruce > Gillespie > > Cc: Dan Long >, > "obs-report-25m@apo.nmsu.edu " > >, > plug > > Subject: [sdss3-25mlog 399] Re: ADDENDUM to 2.5m obslog 20111215 (MJD > 55911)] > > > Fred > > Dan, Kaike and I talked about this and after a lot of research by > Dan --We need a fair amount of discussion on this at the next > tuesdays ops telecon > > APOGEE has some fundamental differences form the past 10 years of > SDSS that is causing this and not all of us were aware of the > fallout of these differences. > > here are some of the issues that need clarification and discussion > > 1. drill run nomenclature . how does drill run number relate to > observing run ? > > > Sorry, this is probably me, I think of run runs based on when they > arrive, rather than the design date in the platedb, I'll try to break > this habit. > > 2. autoscheduler plan observing does not match the long used > platedb plan observing page > > > The autoscheduler selected plate utilizing the SAME observing limits > as the platedb. The beginning and ending time of the slot must be > within the platedb observing limits (the scheduler pulls these from > the platedb directly). I'm not sure why it is still believe that this > is not the case. The COLORS of current HA and Alt on the scheduler > page don;t excactly follow these rules, but are just there as to > assist the observers. If this cause confusion they can be removed. > > 3. platedb is updated during the daytime (after plugging) but > Autoscheduler does not know this for that nights observing > > > It would be difficult to update this, while a specific cart is not > required, it is listed in the schedule that way. The two solutions > are to 1) have the plugger match plates to requested carts, or 2) > remove any cart designation from the observing plan and make the > observer do the check. > > Someone will have to make an extra check, either the pluggers or the > observers. > > 4. pluggings do not match plugging request. (this was never a > problem before but is now) because of item 3 > > > I don't think it should be difficult to require the pluggers to match > plates to the carts requested, but I'm not actually sure. > What do the pluggers think about this possibility? > > 5. Observers must go by actual pluggings not plugging request or > autoscheduler plan because they may differ. Thus there must be a > step to check platedb vs autoscheduler plan every afternoon. > > > The problem is the platedb listed the drilled surveys and we have ~ 60 > plates which were drilled as SHARED plates but will not be observed > that way (2011.07.e.apogee-marvels). It would be a waste to redrill > these plates as APOGEE only, so the MARVELS plugging is requested > through the autoscheduler through the plugging request. The > scheduler/plugging request is what should be followed for determining > if MARVELS fibers should be plugged and observed (as it is the only > thing which has the logic to identify the remaining approved MARVELS > visits/plates ). > > -- Peter > > > > Note the problem that occurred last night occurred twice this > month already but was caught. Last night it was not. So this is a > serious issue we must work through to everyones satisfaction. > Right now there is just too much manual and human interaction that > is needed and apparently no computer checks > > > > Mark > > > > > On Dec 16, 2011, at 8:23 AM, Fred Hearty wrote: > >> Hi Dan and Mark, >> The plates 5575/76 were part of the NOVEMBER drill run. >> Request that someone physically check that all co-observed plates >> from that run were properly marked. >> Also, the statement 'cart 1 plate 5568 (APOGEE+MARVELS) - we >> observe only APOGEE, 4*500 sec' from the night log also indicates >> that there was a SECOND missed co-observed plate, since this >> plate was supposed to be plugged for co-observing (and likely >> was), but the MARVELS integration was apparently not taken. >> >> Fred >> >> >> Dan Long wrote: >>> Hi Peter and Nathan, >>> >>> There is a difference between what the autoscheduler is >>> asking for and what we were told was to be marked for MARVELS. >>> Plates 5575 and 5576 are both from the September drill run, >>> which we were told not to mark for MARVELS. These plates are >>> being requested to be plugged for MARVELS by the autoscheduler. >>> Please advise if these plates need to be marked and plugged for >>> MARVELS or if the autoscheduler is mistakenly choosing them to >>> be plugged for MARVELS. >>> >>> Cheers, >>> >>> Dan >>> >>>> >(We have not received any signal for the MARVELS: PMT was less >>>> than 6.) >>>>> After APOGEE exposure we disconnect cart 3 and have found that >>>>> many drilled >>>>> holes are not connected with fibers. >>>> >>>> Yesterday's APOGEE autoschedule requested for cart 3: >>>> 3 5275 --> 5576 X None Shared plate, please plug MARVELS fibers >>>> || Please >>>> plug B plate fibers >>>> >>>> The platedb today shows that plate 5576 need A-side plugged. >>>> Same we have >>>> in the STUI platedb. >>>> We started observing the MARVELS with the plate at side A, then >>>> got zero >>>> signal, and tried to switch to side B on the fly for checking. >>>> Got the >>>> same zero signal, checked the plate visually after dismounting >>>> and found >>>> many unplugged holes. >>>> >>>> Dmitry & Viktor >>>> >>>> >>>> _______________________________________________ >>>> Message 395 of sdss3-25mlog - Logs of nightly APO activities. >>>> sdss3-25mlog@sdss3.org >>>> https://sdss3.org/mailman/listinfo/sdss3-25mlog >>> >>> _______________________________________________ >>> Message 397 of sdss3-25mlog - Logs of nightly APO activities. >>> sdss3-25mlog@sdss3.org >>> https://sdss3.org/mailman/listinfo/sdss3-25mlog > > Mark Klaene > Site Operations Manager > Apache Point Observatory voice: 575 437 6822 > PO Box 59 fax: 575 434 5555 > 2001 Apache Point Rd cell: 575 430 0210 > Sunspot NM 88349-0059 web: www.apo.nmsu.edu > > > > >