Subject: Re: [sloan-25mlog 3277] 2.5m obslog 20161213 (SJD 57736) From: Kaike Pan Date: Wed, 14 Dec 2016 08:57:27 -0700 --Apple-Mail=_04BC3153-CD45-4C5C-A319-34D2D03E9E3E Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=iso-8859-1 will show observers so that everyone knows what to do if this happens = again. -kaike > On Dec 14, 2016, at 8:43 AM, John Downey wrote: >=20 > Looking at the plots, this does look similar to what we've seen = before, with one important exception: >=20 > >=20 > >=20 > The exception is that the rotator seems to have been physically stuck = rather than slipping, since what ultimately cured the problem was = manually moving the rotator. >=20 > At a guess, the amplifier got pretty unhappy too between 11:06 and = 11:09, presumably power cycling it corrected this but the rotator was = still binding. >=20 > Based on this event and previous ones, I'm going to take a guess that = the previous ones were slipping and this one was binding. It would be = good to have visual confirmation though, before we try to come up with a = solution. >=20 > I would suggest that when this problem occurs, the cold observer check = for slipping of the capstan, then try to move the rotator a couple = degrees off with the MCP from the podium. If it moves then all is well, = and if not, troubleshooting can proceed to power cycling amps and = physically jogging the rotator. Using the MCP is important here if we = are to understand the problem, because it is the lowest level means of = moving the axes, so using it eliminates other possible trouble sources. >=20 > We do realize that the observers' priority is getting back on sky = asap, but if we are to correct this problem, the delay of real = troubleshooting will be necessary. >=20 > If the problem is slipping alone, we think that will be an easy fix. = If its actually binding, we have a bit more work to do to come up with a = solution. >=20 > John >=20 >=20 >=20 >> On Dec 14, 2016, at 6:52 AM, Dmitry Bizyaev > wrote: >>=20 >>>>=20 >>>> 11:03Z when unmounted eng. camera, rotator stopped at 00:03:xx >>>> An attempt to move it off this position at zenith did not work: >>>> it shows the motion progress bar in SOP, all axes are clear in MCP, >>>> but it did not move. >>>> Tried at different ALT - no. >>>> Tried to power cycle rotator's amplified - no >>>> Cleared axes and pushed rotator manually. It came to zero >>>> and worked after that without problem. >>>>=20 >>>=20 >>> Two questions: >>>=20 >>> 1. Did slewing to ~2:00:00 and then back to 0:00:00 fail? >>=20 >> Yes. SOP showed the motion progress, but ROT coordinate did not = change. >>=20 >>> 2. Was the capstan observed to be slipping on the drive surface? >>=20 >> I don't know, sorry. >>=20 >>=20 >>=20 >>=20 >=20 > ____________________________________________________________ > Logs of Nightly APO Activities > = https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html --Apple-Mail=_04BC3153-CD45-4C5C-A319-34D2D03E9E3E Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=iso-8859-1 will show observers so that everyone knows what to do if this = happens again. -kaikeOn Dec 14, 2016, at 8:43 AM, = John Downey wrote:Looking at the = plots, this does look similar to what we've seen before, with one = important exception:The exception is that the rotator seems to have been = physically stuck rather than slipping, since what ultimately cured the = problem was manually moving the rotator.At a guess, the amplifier got pretty = unhappy too between 11:06 and 11:09, presumably power cycling it = corrected this but the rotator was still binding.Based on this event and previous ones, = I'm going to take a guess that the previous ones were slipping and this = one was binding. It would be good to have visual confirmation = though, before we try to come up with a solution.I would suggest that when this problem = occurs, the cold observer check for slipping of the capstan, then try to = move the rotator a couple degrees off with the MCP from the podium. = If it moves then all is well, and if not, troubleshooting can = proceed to power cycling amps and physically jogging the rotator. = Using the MCP is important here if we are to understand the = problem, because it is the lowest level means of moving the axes, so = using it eliminates other possible trouble sources.We do realize that the = observers' priority is getting back on sky asap, but if we are to = correct this problem, the delay of real troubleshooting will be = necessary.If = the problem is slipping alone, we think that will be an easy fix. = If its actually binding, we have a bit more work to do to come up = with a solution.JohnOn Dec 14, 2016, at 6:52 AM, Dmitry Bizyaev = wrote:11:03Z when unmounted eng. camera, rotator = stopped at 00:03:xxAn attempt to move it off this = position at zenith did not work:it shows the motion = progress bar in SOP, all axes are clear in MCP,but it did = not move.Tried at different ALT - no.Tried = to power cycle rotator's amplified - noCleared axes and = pushed rotator manually. It came to zeroand worked after = that without problem.Two questions:1. Did = slewing to ~2:00:00 and then back to 0:00:00 fail?Yes. SOP showed the motion = progress, but ROT coordinate did not change.2. Was the capstan = observed to be slipping on the drive surface?I don't know, sorry.___________________________________________________= _________Logs of Nightly APO Activities https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/= index.html= --Apple-Mail=_04BC3153-CD45-4C5C-A319-34D2D03E9E3E--