You are here: Foswiki>Commissioning Web>Com20090216 (16 Feb 2009, TorstenLeibold)Edit Attach

Lucifer February Com Night 10 February 15-16 2009

Observers: N.Ageorges, W.Seifert, V.Knierem (LUCIFER), DThompson (@LBT)
Telescope Operator: W.Wack
SW Support: NCushing on call.
Telescope Support: Brian


  • Opened near midnight. M1 ventilation off because of ongoing HP3 issue.
  • Had an error: BAD Poly:AZ = -1.552521e+00 rad. Why did the telescope not simply go to the other wrap? This sounds like a HW limit (-90) vs. a SW limit (less than that) and PCS calculates for the HW limit. See IT#2018.
  • TO station hung around 3:45am. Called Norm, rebooted machine. Preset was sent while lbtmu01 was hung and it appeared to work, but when MCSPU was brought back up the EL axis went into waitopr state.


Fully clouded over at sunset. wind is low and precipitation is not in the forecast, so we have the side vent doors open and mirror ventilation on.

21:15 Still cloudy. Tair=-2.6 RH=16.8% Dewpt=-23.4 Wind=3m/s avg. Tmirror=-3.0

21:20 Starting some flexure testing for LUCIFER

21:35 Noted on TO station (lbtmu01) top shows X taking 33% of CPU and 15% of the memory. Load average is ~2.5

21:50 Closing ventilation doors

23:30 clearing enough that we will open. Tair=-3.2 RH=35% and rising, Dewpt=-16.6 Wind~5m/s avg, 12m/s gusts. Right shutter door having trouble opening. Mirror ventilation is ON for the time being.

23:42:55.383 Preset Command Failed because Chase state not IDLE. That did not take long.

23:44:55.714 Preset TRACK/PARALLACTIC for pointing correction. No star visible. We reloaded the default pointing model but still no star. Could be clouds or extreme defocus, but it was a 7th mag star.

23:51:16.004 Preset to new source. Still nothing. We are resetting the mount encoders (done @00:18). Note that the last pointing model update from last night had us 131 arcsec off of the model in IE and 52 arcsec in CA. I believe this is evidence that the encoders are drifting, as they were last set ~2 days before and the temperature is not that different from the end of last night. Also, we had a brownout this afternoon, could that have messed up the encoders?

00:19:11.648 Preset TRACK/PARALLACTIC to correct pointing. Off by +20"IE and -27"CA. I recommend that we reset the encoders every night when on-sky observations are to be made. This should be added to the OSA startup wiki pages, along with sending the test preset in /home/telescope/PresetCheck.

00:28:15.059 ACTIVE mode preset to collimate, on axis. First image in decent focus (we had tweaked focus +0.4mm during the encoder reset), poor seeing 1.5" on the guider. WFS rms: 1189, 528, 441, 237. Switch to the "split" mode.

Note that the IIF GUI is reading that the IIF GUI reads "paused" for the WFS althgouh it is running and updating the mirror figure. This seemed to be working correctly just a couple days ago.

00:34:46.085 Preset same field but with off-axis guide star. Chase failed to make tolerance. M1 ventilation turned off at 00:39.

00:39:59.786 Preset to new source (GOODS-N field), successful.

Preset just sent gave this error:
Mon Feb 16 00:54:15.076 2009  Preset alert for left LUCIFER mode ACTIVE: PCS reports MCSPU error returned - BAD Poly:AZ = -1.552521e+00 rad. Too small. Slew not started. Polycount =480059
This looks like we were requesting to go somewhere too close to the AZ limit (about -89 degrees). Why did the system simply not go to the other wrap? IT#2018 (medium priority) submitted.

01:48 Forces on mirror rising. We are turning on ventilation now. Cleared c00 for good measure, but that might not do anything more than make the wfs take a couple more cycles to converge.

Now the WFS status on the IIF GUI is working again.

02:00 We moved away from the south because the enclosure was hitting torque limits. To the north (AZ~13) the building was OK but the seeing was awful. Mirror ventilation is still on.

02:28 And now the WFS status on the IIF GUI are no longer working.
02:41 Now the WFS status on the IIF GUI is working again.

02:43 Ventilation turned off. M1 and ambient within 0.2 degrees.

03:46 TO station appears to be hung. Calling Norm. I killed all of my connections into lbtmu01 from lbtmu05 and I still cannot ping the system. Power-cycling lbtmu01. A preset was sent from LUCIFER while lbtmu01 was hung and it appeared to work (telescope slewed to the new position) but when MCSPU was brought back up the EL axis went into waitopr state (not 100% sure of the timing).

There is a discrepancy between the acquisition image filename number and the number displayed on the GCSGUI acquisition image display.

UT12:51:47 The error message from an offset to a point off the patrol field was "couldn't move probe to position (SFP)... :AGW command status timed out waiting for probve to reach destination." Is this a real timeout? If it knows it cannot reach the position because it is out of the patrol field, then why would there be a timeout message? After this, an offset came in from LUCIFER which brought the star back onto the patrol field and the guider did pick it up and resume guiding, so that part worked fine. The timeout message was confusing.
Answer: The timeout is real. It means that the positioning call to the oacontrol library did not return within 60 seconds. Usually this indicates a motor axis failure but the real cause cannot be determined as the call just wouldn't return. This is logged as an error as the state of the AGw probe is unknown. The fact that a follow-up position call succeeded suggested that there was no problem with an axis. The message shouldn't be confusing but tell you that there was indeed a problem for the GCS controlling the stage. If a position would not be reachable that should result in a different error message and should happen immediately and this was not the problem in this case. At least not as far as the oacontrol interface reported. I think I've heard about a complaint about one of these positioning timeouts before so I assume this did not happen the first time. If this is true, I suggest you create an IssueTrak and assign it to Jesper and or Thomas Hahn to resolve the problem that the setxy function would not return within 60 seconds and not report an error. -- TorstenLeibold - 16 Feb 2009

GCSGUI: On acquisition image display, would it be possible to also draw a 3 arcsec circle around the location of the hoptspot?
Answer: Yes. I've sent you an email asking for more detailed specifications of what is wanted here. -- TorstenLeibold - 16 Feb 2009

Deep-ish imaging with LUCIFER for most of the remaining night. The weather was not optimal, with clouds blowing through all night.

End of night: 06:27 - no twilight flats because it is cloudy.

-- DavidThompson - 16 Feb 2009
Topic revision: r6 - 16 Feb 2009, TorstenLeibold
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback