Subject: Re: [sdss3-25mlog 1186] Re: 2.5m obslog 20130401 (MJD 56384) From: Kaike Pan Date: Sat, 6 Apr 2013 06:26:13 -0600 --Apple-Mail-6-396917320 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii Thanks. Yes. Let's discuss the range that we would like to remove=20 the voice warning for the rotator fiducial in the script. -kaike On Apr 5, 2013, at 9:17 PM, Elena Malanushenko wrote: >=20 > I think Dmitry was worried about necessary actions but not=20 > bothering warnings.=20 > Warning informed him about problem, and he was looking for > how to deal with it.=20 >=20 > I do not like an idea to overwrite mcp actor settings.=20 > They stated that no correction if error > some error limit.=20 > I read that limit (600) and current values and make warning if no=20 > correction applied.=20 >=20 > I can delete voice warning if you think it is better.=20 >=20 > Regards,=20 > Elena >=20 >=20 > On 04 /05/ 13, at 8:21 PM, Kaike Pan wrote: >=20 >> Maybe we can think that the fiducial monitor script does not give = verbal warming on rotator fiducial errors smaller than >> 2000 tickets(?). -kaike >>=20 >>=20 >> On Apr 5, 2013, at 11:29 AM, Dan Long wrote: >>=20 >>> The rational for putting an upper limit on correction is to = prevent bogus readings (which do happen fairly frequently) from being = applied as a correction. 600 ticks was chosen as a value which was = considered about as large as a real error would get. Experience over = the years has shown that real errors do occur which are larger, though = they are not all that common. >>>=20 >>> The best indicator of a real error is if it is confirmed by = subsequent fiducial crossings. I think it would be quite reasonable to = consider adding an additional layer to the existing fiducial logic. The = first layer would be as things are now: a single fiducial crossing with = error >>=20 >>> I don't know how difficult it would be to implement such a change = (this is all in the mcp I believe). I also don't know how much sky time = is lost to dealing with corrections manually, but I wouldn't think it is = much. So that would have to be weighed against the effort required to = implement the change. >>>=20 >>>=20 >>> -Dan >>>=20 >>>=20 >>> On Apr 4, 2013, at 3:53 PM, Kaike Pan wrote: >>>=20 >>>> Not sure if there is another constraints on the system. =46rom = acquiring fields and guiding aspects, it would be nice to do so. >>>>=20 >>>> On Apr 4, 2013, at 3:49 PM, Mark Klaene wrote: >>>>=20 >>>>> can we/should we change the acceptable error margin for the = rotator so that a "too large to correct" won't happen unless it really = is too large that it will cause problems? >>>>>=20 >>>>> mk >>>>>=20 >>>>>=20 >>>>> On Apr 4, 2013, at 3:30 PM, Dan Long wrote: >>>>>=20 >>>>>> Hi Dmitry, >>>>>>=20 >>>>>> It takes 38 arcseconds of rotator error or 1900 encoder ticks = to generate 1 arcsecond of error on the sky at the edge of the plate. A = 15 arcsecond error in rotator position would result in less than half an = arcsecond of error on the sky at the edge of the plate, and even less = towards the center . It would probably not have hindered acquisition. = While it is a little troubling to have fiducial errors too large to = correct, in the case of the rotator 600 ticks is only 12 arcseconds, and = it would take something on the order of several arcminutes of error ( = ~10,000 ticks of encoder error) to cause trouble acquiring. >>>>>>=20 >>>>>> Cheers, >>>>>>=20 >>>>>> Dan >>>>>>=20 >>>>>> On Apr 2, 2013, at 3:44 PM, dmbiz@apo.nmsu.edu wrote: >>>>>>=20 >>>>>>> Hi Kaike, >>>>>>> thank you for the comments! >>>>>>>=20 >>>>>>>> If the rotator fiducial error when slew to a field is the = ONLY >>>>>>>> problem we have, I think that >>>>>>>> we should be okay to continue. Most likely, fiducial errors on = the >>>>>>>> rotator will not stop us from >>>>>>>> finding guiding stars. After we get stars, no fiducial = corrections will be >>>>>>>> applied, so we are okay >>>>>>>> for that field except for observers have to tolerance multiple = verbal >>>>>>>> warnings of "fiducials". >>>>>>>=20 >>>>>>> The rotators fiducial error suggested some 15" offset. Once it = happens >>>>>>> with rotator, it is hard to predict where guide stars go, so I = recommended >>>>>>> to >>>>>>> fix the fiducials first. >>>>>>>=20 >>>>>>>> There is a bug on the "stop" of "goto Field". Clicking = (applying) the >>>>>>>> "stop" >>>>>>>> button puts sop into a confusion status although"sop ping" and = "sop >>>>>>>> status" show >>>>>>>> SOP is okay. One can get SOP back to its normal status by issue = two >>>>>>>> commands >>>>>>>> from STUI: >>>>>>>>=20 >>>>>>>> sop restart (this may not unnecessary, so you can try without = it first) >>>>>>>> sop reinit >>>>>>>=20 >>>>>>> Can we put it to the observing notes that we cannot use the = "stop" button >>>>>>> in STUI? >>>>>>>=20 >>>>>>> Dmitry >>>>>>>=20 >>>>>>>=20 >>>>>>> _____________________________________________________ >>>>>>> Message 1176 of sdss3-25mlog - Logs of nightly APO activities. >>>>>>> = https://trac.sdss3.org/tracmailman/browser/private/sdss3-25mlog/all/1176.h= tml >>>>>>> sdss3-25mlog@sdss3.org >>>>>>> https://sdss3.org/mailman/listinfo/sdss3-25mlog >>>>>>=20 >>>>>> _____________________________________________________ >>>>>> Message 1180 of sdss3-25mlog - Logs of nightly APO activities. >>>>>> = https://trac.sdss3.org/tracmailman/browser/private/sdss3-25mlog/all/1180.h= tml >>>>>> sdss3-25mlog@sdss3.org >>>>>> https://sdss3.org/mailman/listinfo/sdss3-25mlog >>>>>=20 >>>>> Mark Klaene >>>>> Site Operations Manager=20 >>>>> 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=20 >>>>>=20 >>>>>=20 >>>>>=20 >>>>>=20 >>>>> _____________________________________________________ >>>>> Message 1181 of sdss3-25mlog - Logs of nightly APO activities. >>>>> = https://trac.sdss3.org/tracmailman/browser/private/sdss3-25mlog/all/1181.h= tml >>>>> sdss3-25mlog@sdss3.org >>>>> https://sdss3.org/mailman/listinfo/sdss3-25mlog >>>>=20 >>>=20 >>=20 >> _____________________________________________________ >> Message 1186 of sdss3-25mlog - Logs of nightly APO activities. >> = https://trac.sdss3.org/tracmailman/browser/private/sdss3-25mlog/all/1186.h= tml >> sdss3-25mlog@sdss3.org >> https://sdss3.org/mailman/listinfo/sdss3-25mlog >=20 --Apple-Mail-6-396917320 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=us-ascii Thanks. Yes. Let's discuss the range that we would like to = remove the voice warning for the rotator fiducial in the = script.-kaikeOn Apr 5, 2013, at 9:17 = PM, Elena Malanushenko wrote:I think = Dmitry was worried about necessary actions but = not bothering warnings. Warning = informed him about problem, and he was looking forhow to deal = with it. I do not like an idea to = overwrite mcp actor settings. They stated that no = correction if error > some error limit. I read that = limit (600) and current values and make warning if = no correction applied. I = can delete voice warning if you think it is = better. Regards, ElenaOn 04 /05/ 13, at 8:21 PM, Kaike Pan = wrote:Maybe we can think = that the fiducial monitor script does not give verbal warming on rotator = fiducial errors smaller than2000 tickets(?). = -kaikeOn Apr 5, 2013, at 11:29 = AM, Dan Long wrote: The rational for putting = an upper limit on correction is to prevent bogus readings (which do = happen fairly frequently) from being applied as a correction. 600 = ticks was chosen as a value which was considered about as large as a = real error would get. Experience over the years has shown that = real errors do occur which are larger, though they are not all that = common. The best indicator of a real = error is if it is confirmed by subsequent fiducial crossings. I = think it would be quite reasonable to consider adding an additional = layer to the existing fiducial logic. The first layer would be as = things are now: a single fiducial crossing with error <600 ticks gets = corrected during a slew. The second layer could be: if one = sees two subsequent fiducial crossing errors which are less than 6000 = ticks and agree to within 10%, apply the correction. This would = cover nearly every real error. I = don't know how difficult it would be to implement such a change (this is = all in the mcp I believe). I also don't know how much sky time is = lost to dealing with corrections manually, but I wouldn't think it is = much. So that would have to be weighed against the effort required = to implement the = change.-DanOn Apr 4, 2013, at 3:53 PM, Kaike Pan wrote:Not sure if there is another = constraints on the system. =46rom acquiring fields and guiding aspects, = it would be nice to do so.On Apr 4, 2013, at 3:49 PM, = Mark Klaene wrote:can we/should we change the = acceptable error margin for the rotator so that a "too large to = correct" won't happen unless it really is too large that it will cause = problems?mkOn = Apr 4, 2013, at 3:30 PM, Dan Long wrote:Hi = Dmitry, It takes 38 arcseconds of rotator = error or 1900 encoder ticks to generate 1 arcsecond of error on the sky = at the edge of the plate. A 15 arcsecond error in rotator position = would result in less than half an arcsecond of error on the sky at the = edge of the plate, and even less towards the center . It would = probably not have hindered acquisition. While it is a little = troubling to have fiducial errors too large to correct, in the case of = the rotator 600 ticks is only 12 arcseconds, and it would take something = on the order of several arcminutes of error ( ~10,000 ticks of encoder = error) to cause trouble acquiring. Cheers, = DanOn Apr 2, 2013, at 3:44 PM, dmbiz@apo.nmsu.edu = wrote:Hi = Kaike,thank you for the = comments! If the rotator fiducial error when slew = to a field is the ONLYproblem we have, I think = thatwe should be okay to continue. Most likely, fiducial = errors on therotator will not stop us = fromfinding guiding stars. After we get stars, no fiducial = corrections will beapplied, so we are = okayfor that field except for observers have to tolerance = multiple verbalwarnings of = "fiducials".The rotators = fiducial error suggested some 15" offset. Once it = happenswith rotator, it is = hard to predict where guide stars go, so I = recommendedtofix the = fiducials first. There is a bug on the "stop" of "goto Field". Clicking = (applying) the"stop"button puts sop into a confusion = status although"sop ping" and = "sopstatus" showSOP is okay. One can get SOP = back to its normal status by issue = twocommandsfrom = STUI:sop restart (this may not = unnecessary, so you can try without it = first)sop reinitCan we put it = to the observing notes that we cannot use the "stop" = buttonin = STUI?Dmitry_____________________________________________________Message 1176 of sdss3-25mlog - Logs = of nightly APO activities.https://trac.sdss3.org/tracmailman/browser/private/sdss3-25ml= og/all/1176.htmlsdss3-25mlog@sdss3.orghttps://sdss3.org= /mailman/listinfo/sdss3-25mlog___________________= __________________________________Message 1180 of sdss3-25mlog - = Logs of nightly APO activities.https://trac.sdss3.org/tracmailman/browser/private/sdss3-25ml= og/all/1180.htmlsdss3-25mlog@sdss3.orghttps://sdss3.org= /mailman/listinfo/sdss3-25mlog Mark KlaeneSite Operations Manager Apache Point Observatory = voice: 575 437 6822PO Box 59 = = fax: 575 = 434 55552001 Apache Point Rd = = cell: 575 430 0210Sunspot NM 88349-0059 = web: www.apo.nmsu.edu = ____________________________________________________= _Message 1181 of sdss3-25mlog - Logs of nightly APO = activities.https://trac.sdss3.org/tracmailman/browser/private/sdss3-25ml= og/all/1181.htmlsdss3-25mlog@sdss3.orghttps://sdss3.org= /mailman/listinfo/sdss3-25mlog_______________= ______________________________________Message 1186 of sdss3-25mlog - = Logs of nightly APO activities.https://trac.sdss3.org/tracmailman/browser/private/sdss3-25ml= og/all/1186.htmlsdss3-25mlog@sdss3.orghttps= ://sdss3.org/mailman/listinfo/sdss3-25mlog= --Apple-Mail-6-396917320--