Telescope Performance Verification
with IRTC at LFBG - 14-15 May 2009 MST

Observers:: D.Thompson, J.Heidt (mtn)
Telescope Operator: A. Ceranski
Software Support: C. Biddick (phone)
Telescope Support: T. Shih, J. Little

Summary

Open shutter time 35.5% (see end of this page for details)

Current transform apparently taken with LEFTZEROPOINT = -186.29, this is clearly better than -187.703.

Discovered that there is still a systematic residual in the transform: f(PA)
Later discovered that this was due to poor coordinates or uncorrected proper motions. The catalog coordinates of BS9148 that we use differ from the GSC coordinates in Simbad by ~0.77 arcsec and the radius of the "arc" we saw was 0.71 arcsec. This is precisely the behavior you would expect to see if the coordinates of either the target or the guide star were incorrect. This does not mean that the transform was bad.

Something still causing problems on writes (in iraf) to home directory, made a local LBTtools to compensate. After this fix the rest of the night went very smoothly.

Re-measured AGw coordinate at center of rotation, agreed with config values to ~2 pixels
Took some 20-minute guided observations at high and low elevations
Took another set of AGW calibrations, 4 subsets at different PAs plus two sets with one guide star at 5 PAs.

Preparations:

Previous night's problems/solutions:
  • TCS/OSS from centos5 killed, centos4 restarted and it worked thereafter last night
  • M1 heated: PMC patched to fix this
  • Transform correct, angle corrected, still need to test
  • Slowdown: LDAP server died, NIC ports replaced and it seems to work again

Preset closed dome (before sunset) ACQUIRE mode to test
  • hung because OSS stuck moving M3, OSS restarted OSS
  • GCS hung due to timeout waiting for image, restarted GCS
  • crashed GCS (needed to redo selectAGW), restarted, reselected
  • worked! We should be ready to open/start
  • "stop" button on GCSGUI for guider will interrup ACQUIRE mode

Details

Tout=10.9 T(M1GLASS)=9.4 RH=21.7 Wind 2-6m/s

19:20:30.830 pointto M5_0444 TRACK mode to see if we can see it on the IRTC. Successful, with the star quite well collimated and just above center on the IRTC. IE changed from -72 to -67, CA from -10 to -7. Now we need to wait until it gets dark. Took an image sequence to show the initial collimation: 023221 0002_cube.

19:36:21.427 pointto BS9148 ACTIVE to collimate. Reached sub-200 nm rms with 0.7 arcsec FWHM on guider.

19:47:34.398 pointto BS9148 gs=4 ACTIVE.

LDAP, or something else, still seems to be causing problems (manifesting as trouble writing to pipes in the uparm directory). So, I installed a local version of the current LBTtools in /home/LBTO/iraf and set that in the login.cl file to overwrite the system definition.

19:58:20.380 pointto BS9148 gs=4 ACTIVE to reset for this test (PA defaulted to 180.18). Takeseq 2.0 30 - 025943 0003_cube. Star at 107.5 132.5

20:02:57.279 pointto BS9148 gs=4 ACTIVE PA=120. Takeseq 2.0 30 - 030514 0004_cube. Star at 130.5 85.0. So the transform (or something) is still not working. Confirmed in the PCS log that the last change to LEFTZEROPOINT set it at -187.703 on May 14 18:08:22.

Changes LEFTZEROPOINT for IRTCBGF to -186.29 in PCSInstrument.conf and PCS restarted to take effect. Confirmed in syslog that it changed to this value on May 14 20:10:31.

20:11:31.974 pointto BS9148 gs=4 ACTIVE PA=120. Takeseq 2.0 30 - 031359 0005_cube. Star at 154.5 123.5.
20:15:43.057 pointto BS9148 gs=4 ACTIVE PA=180.18. Takeseq 2.0 30 - 031703 0006_cube. Star at 155 132
20:19:33.532 pointto BS9148 gs=4 ACTIVE PA=240. Takeseq 2.0 30 - 032033 0007_cube. Star at 161 136
20:22:56.366 pointto BS9148 gs=4 ACTIVE PA=210. Takeseq 2.0 30 - 032425 0008_cube. Star at 158 135
20:26:33.147 pointto BS9148 gs=4 ACTIVE PA=150. Takeseq 2.0 30 - 032837 0009_cube. Star at 158 128

offset 5.0 5.0 DETXY and takeseq 2.0 30 for sky for above.

NOTE The transform is clearly better using LEFTZEROPOINT = -186.29. However, there is still ~1 arcsec scatter in these points. They are correlated with PA. It looks like there is an ~0.5 arcsec offset between the AGW coordinates of the rotator center in AGW space and the actual mechanical center of the rotator. Called Jesper

20:31:21.618 dice5 on BS9148 gs=4 10x10 30x2.0sec
Starts with 033703 0011_cube and ends with an abort at 035502 because we hit a Y limit at 87 degrees

OK, we are off the Y limit and cleared back to the default collimation, plus a bit of focus offset. We had observed this field yesterday through zenith passage, but we conjecture (i.e. wild speculation) that the hot mirror compensated for the collimation lookup. 09:28 we are back to collimated at 83 degrees EL. IE adjusted to -79, CA to -10.

We will run a dice5 pattern, single pass, here with 20-minute integrations. Seeing is good at ~0.45 on the guider. This will give us time to discuss options with Jesper.

21:33:10.161 pointto BS9148 gs=4 ACTIVE, PA will be 180.18. Star is 10.82 mag and close to saturating the guider. Switching to GS#3 at R2=15.68 mag for the 1200s guider tests

21:38:05.674 pointto BS9148 gs=4 ACTIVE PA=305.37
21:42:54.412 dice5 10x10 500x2.0s aborted because of the write problem to uparm, but an integration is in progress so I will let it continue 044350(see save.irtc.20090515.044350) and 0025_cube?

Changed LBTO account to write to /scratch/uparm (set to be world writable). Note that this will only work if you set up a uparm directory on the scratch disk. This seems to make the iraf operation more robust.

IE updated to -92.3 CA to 2.0

22:18:29.489 pointto BS9148 ACTIVE gs=3 - no guide star found! Pointing checked out OK (off ~2.5 arcsec). Pointto same source, gs=4 also not found. Checking on an M5 star.

pointto M5_1332 TRACK mode - IE now -58, CA now -10. A ~20 arcsec offset from a prior valid pointing correction.

pointto BS9153 gs=0 ACTIVE to recollimate. AGW stages at: -0.132 612.416

Re-measure AGW rotator center:
  • collimate on-axis
  • stop guiding and WFSing
  • read out guide cam ~60ms until it saves to disk
  • offset 3.0 0.0 arcsec to make the circle bigger
  • move AGW stage to 0.0,612.5
  • start readout of guider for 60000 ms
  • send rotator to +/-360 from current position

GCS froze when we tried to stop guiding and WFSing. Restarted GCS.

guider000002 0.029 612.501 center at 283.75 278.50 (AIP_IRTC_L.cfg has the hotspot at 282.9 279.6). EL=69

0.63um in preset
guider000003 0.029 612.501 center 281 285 , out of focus (my fault)
guider000005 0.015 612.501 saturated
guider000007 0.015 612.501 center 285 279
guider000009 0.029 612.501 center 285 280
*1.4 um in preset*
hang on collimating M3. Restart OSS. guider000010 0.029 612.501 center 283 280
guider000011 0.015 612.501 center 283 281

While calling Jesper...
pointto BS9153 gs=7 ACTIVE to recollimate. Guider sees 0.4-0.6 arcsec FWHM.
dice5 10x10 600x2.0sec loopN=1
Start: 064520 0026_cube
End: 075056 0029_cube (aborted before the last cube to go back to transforms)

Jesper made a new transform using a better determined rotator center on the IRTC (now 159.84 128.45).

pointto BS9153 gs=7 ACTIVE just to reset things. PA=208.13 by default.

Check centroid vs PA as before (it is still moving):

Single# Cube# PA
081811 0030_cube 208.13
082109 0031_cube 238.00
082519 0032_cube 268.00
082801 0033_cube 178.00
083017 0034_cube 148.00

Still offset with changing PA. Jesper suggest another set of AGw calibration data. This time, we take 4 different PAs on the Stone M field, plus two sets of data using a fixed guide star but varying the PA (this is how we found the problem tonight), so...

01:33:01.056 pointto StoneMagw gs=0 ACQUIRE mode. IE now -47, CA now -20

01:40:22.016 pointto StoneMagw gs=41 (allow to set PA 190.46)

Note that takepic and takeseq do not seem to setr the exposure time on the IRTC. Perhaps the restarts of iraf to clear the uparm errors did this, but communication still works (data is taken).

Stone M AGw Calibrations (all cubes are 30x0.5sec):
Time GS# Single# Cube# PA
01:42:07.531 41 084931 0036 190
01:51:28.011 67 085238 0037 190 no getdat
01:54:18.940 60 085553 0038 190
01:56:37.256 67 085744 0039 190
01:58:41.750 36 090140 0040 190
02:03:01.022 10 090425 0041 190
02:05:27.113 16 090807 0042 190 may be star 11
02:11:06.162 45 091218 0043 190
02:13:09.243 74 091434 0044 190

02:15:09.107 10 091723 0045 268
02:19:10.990 34 092107 0046 268
02:21:58.246 45 092300 0047 268
02:23:47.257 31 092508 0048 268
02:25:55.126 40 092732 0049 268
02:28:15.821 20 092949 0050 268
02:30:22.616 37 093138 0051 268

02:32:59.276 53 093429 0052 075
02:35:50.004 52 093714 0053 075
02:37:57.646 72 093912 0054 075
02:39:56.470 69 094113 0055 075
02:41:41.870 83 094240 0056 075
02:43:21.322 67 094448 0057 075
02:45:37.860 74 094725 0058 075
02:48:23.920 60 094953 0059 075
for sky sub 095118 0060 075 offset 5.0 5.0
02:52:05.468 55 095334 0061 075
02:54:15.229 51 095519 0062 075
02:55:53.645 68 095655 0063 075
02:57:28.420 44 095836 0064 075
02:59:10.847 58 100026 0065 075
03:01:15.037 62 100223 0066 075

03:03:07.708 39 100521 0067 342
03:05:53.714 20 100658 0068 342
03:07:25.621 10 100907 0069 342
03:09:34.408 44 101107 0070 342
03:11:50.411 51 101253 0071 342
03:14:12.606 72 101542 0072 342
03:16:35.036 37 101805 0073 342
03:19:37.960 31 102011 0074 342
03:21:28.451 58 102239 0075 342

Range of PAs with star at high field angle...
03:24:01.165 68 102525 0076 043 248 arcsec radius
03:26:15.674 68 102747 0077 113
03:28:22.443 68 102923 0078 078
03:30:02.307 68 103106 0079 008
03:31:59.711 68 103313 0080 -27

And one at small field angle...
03:34:19.838 53 103528 0081 030
03:36:03.574 53 103659 0082 065
03:37:31.006 53 103853 0083 100
03:39:40.539 53 104053 0084 135
03:41:41.641 53 104237 0085 170

1200s guiding stats at low EL
Will let this run until we can no longer wfs. Field is setting.
pointto BS9160 gs=11 ACTIVE EL@start = 30 degrees
Start: 105918 0086_cube, end: 114310 and 0088_cube. Script aborted because of advancing twilight, last 150 exposures of the cube were in a closing dome and unguided.

Closing at 04:58. The WFS is showing very bright background.

For an overall efficiency number I note the following:
  • 11947 seconds (3.32 hours) of all IRTC data collected tonight
  • 9.36 hours between the sending the first collimation preset to when we closed the dome
  • 3.32/9.36 = 35.5% "open shutter time"

Guider Statistics
  • 6212 guider images taken over the entire night
  • Excluding only 11 outliers, all remaining points fall within a 1.62 arcsec box
  • RMS of centroids on 6201/6212 points: X=2.25 pix (122 mas), Y=1.92 pix (103 mas) * Plot: GuiderXY.20090515.pdf

-- DavidThompson - 15 May 2009

Topic revision: r9 - 16 May 2009, DavidThompson
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