LUCIFER Commissioning (1a) and AGw Commissioning for 16-17 September 2008 (Day 19)
Observers: N. Ageorges, W. Seifert, and numerous other LUCIFERians (Peter, Kai, Marcus, Volker, Andreas) on mtn, D. Thompson. J.Hill and D. Miller in Tucson
Software Support: P. Grenz, N. Cushing, C. Biddick, M. De La Pena in Tucson
Telescope Operators: A. Ceranski
After discussing the IRS hang with Jose it was learned that the IRS can only support a single client connection at a time. Please only connect a single client to the IRS.
The AGw WFS sensor/controller needs to be swapped out. Jesper is sending instructions how to do this from the mirror side. The spare components need to be delivered to the mountain.
Some data collected for a LUCIFER pointing model (~15 points).
Confirmed that the IIF/GCS modification to home the guider stages when a "TRACKING" mode preset is sent worked as expected.
Opened ay 9:55pm. Fairly thick and uniform clouds, but we can see some stars and there is no rain threatening.
Initial preset in PARALLACTIC mode (aligns detector with AZEL) with the telescope open-loop tracking worked. After ptmodify, John pointed out that there were some significant global offsets remaining from a few nights ago that should have been cleared. We cleared them and re-did the pointing model tweak (offset was about 45 arcsec).
Initial preset in ACTIVE mode to pull in the collimation succeeded, but the wavefront sensor had basically zero light on it. We tried a brighter star (5th mag!) but still nothing. The guider does close the guiding loop. Several points noted during this startup:
- Guider does not complain when the star is heavily saturated
- WFS does not complain when there are zero counts
- WFS does not complain when it is saturated (from last night)
12:24 Still trouble-shooting the WFS problem. It has persisted through a couple of complete power cycles and software restarts. Light gets to the guider, the WFS shows nothing but a bias signal. A physical inspection of the guider optics did not show any obvious problems. A flashlight illuminating the pinhole does not show any difference.
1:45 Contact was made with Jesper Storm to discuss WFS sensor problem. WFS images are 528 by 536 with no over scan. By 4:15 AM Jesper, Doug, and John have not been able to get the WFS to produce an image. The WFS/controller need to be swapped out. The backup components need to be delivered to the mtn.
Tonight we were completely unable to do wavefront sensing with the
AGw. Upon investigation, we find that there seems to be no light
being detected by the WFS camera. We see what appears to be a
bias frame from the WFS camera.
We power cycled the AGw (stopAGW, startAGW) to restart everything in the AzCam
We sent Dave and Walter upstairs to look into the front side of the AGw. Even with full dome lights on, and with a flashlight shining on the pinhole no light is detected on the WFS CCD.
With help from Thomas Fechner and Jesper Storm, we modified
c:\soguiders\guider.tcl so that the AzCam
GUI could read the
images with overscan. We were hoping to see overscan to
confirm that the bias frame we are getting was connected to an
actual CCD. A 10 minute eposure in this mode didn't look any
different in the real versus overscan pixels.
At 4:15AM we handed over the AzCam
vncviewer to Jesper and Thomas to see if they can figure out anything. They are also preparing instructions for how to change / disconnect the controller from the primary mirror side. They will send these to AGw mailing list.
Our present idea is that somebody should take the spare cameras and spare controllers from Torsten's office to the mountain so that we can connect them temporarily to figure out if the camera head or the controller has failed. Try not to touch the cables at the WFS camera as this can mess up the optical alignment.
Note that what we see could be consistent with a failure of the CCD
clock drivers, so a test of the power supply might also be a good first step.
1:55 IRS/IIF hung. Called Jose and he is reviewing the problem. It hung two other times between 1:55 and 2:40.
2:10 Jet has crashed for he 3rd time since 1:30
2:40 Requested Lucifer team to recompile and relink with the last dirt.ice file so the updated version of the IRS could be put into service. The decision is still being discussed and it is now Lucifer's decision on when the upgrade will take place.
4:15 Manually attempting to collect images for a pointing model. Telescope is out of focus when changing elevation. Adjusting collimation model and it is still out of focus at high and low elevations. Collected ~15 data points. See /home/telescope/20080917.ptlog.
- 17 Sep 2008