LBBG Commissioning with AGW1: 01 November 2011 UT

Observer: JHill, JHeidt, RPogge (LBTO) with Doug, Mike, Dan, Jesper, Paul assisting remotely
Telescope Operator: SAllanson (LBTO)
Telescope Support: JUrban (LBTO)
Instrument Support: TShih (LBTO)

Summary

We had lots of problems, but we made alot of progress. We set the rotator/positon angle by adjusting LEFTZEROPOINT. We verified guiding off and on-axis. We aligned the guide probe hotspot to the rotator center.

The big questions

Why are 30% of the WFS camera images rubbish?

Do we need to change the DSP rotator limits since we changed LEFTZEROPOINT?

Is the SX M3 alignment we did on-sky to match the MODS pointing OK?

Is the WFS Camera alignment OK?

Why does this AGW at this focus have an X-flip in the probe position compared to LUCI?

Plan

See http://wiki.lbto.org/twiki/bin/view/Commissioning/Com201110LBBGPlan

Preparations

All kinds of chaos during the day caused by water main and network failures. Network is back at 00:20 UT, but with some DNS wierdness still.

New patches of IIF and PCS installed.

Tom Sargent wrote:
If you need to use the left rear rotator, since it doesn't have a cable
chain, you have to first override the cable chain.  You can use the
"roverride lrbg on" command in the engineering interface or, using the MCS GUIs.

Details

LBBG

00:24 Rebooted agw-control to get it to talk to GCSL with LBBG, then restarted AGW and GCSL.

00:45 Test preset crashes PCS - not sure why. Another preset succeeded, but didn't find a star.

Even a brighter star doesn't help. Jeff confirms that M3 looks to be pointing at the correct focal station.

01:07 ptspiral. No hint of a star. We see cosmic rays on the guider, and Steve saw twilight.

RFBG

01:20 Switch to RFBG to verify that we can find a star there, but our idea of a quick check was foiled. (more details in Steve's log)

RFBG rotator had a SERDES error.

01:29 Install the 20111029 pointing model and collimation models on all of the SX Gregorian focal stations (not LBC).

Doug says DX M2 won't turn on because of a (perceived or real) temperature problem.

LBBG

02:00 Meanwhile, a 40-sec long sky exposure on the WFS camera shows that the lenslet array only has half of the secondary on it. See /OldRepositoryAGW_Data/20111101/left_wfsc000002.fits. We take this as a small confirmation of our suspicion that M3 is in the wrong position (-30 deg compared to -26 that we expect).

These were the SX M3 values in oss.conf against which we applied our global offsets during the night.:
oss.side[0].terc.focal_stations.bentGregorianBack string "-615.0 -20.0 0.0 -30.6877"
oss.side[0].terc.focal_stations.bentGregorianFront string "-632.0 -95.0 0.0 +26.0123"
oss.side[0].terc.focal_stations.bentGregorianCenter string "-431.0 -120.0 0.0 +0.0000"
oss.side[0].terc.focal_stations.bentGregorianRearFiberFeed string "-632.0 -95.0 0.0 -45.0000"

Restart OSS to be able to home SX M3. Still had problems moving to LBBG, so we cycle power on M3 UMAC.

02:38 readWFSCam failed (image invalid).

02:40 selectAGW left AGW1_IRTC.cfg

02:41 left_wfsc000004.fits - pupil hasn't moved.

02:42 Steve uses OSS to move SX M3 to -26.0123.

02:44 left_wfsc000005.fits - pupil is now 80% on the grin array

02:47 Steve uses OSS to move SX M3 to -25.000

02:48 left_wfsc000006.fits - image is all gray ????

02:50 left_wfsc000007.fits

02:53 Steve uses OSS to move SX M3 to -25.600

02:54 left_wfsc000008.fits - now the pupil is centered left-right on the WFS image, but 2 rows to low

02:59 move Tilt from -20 to +980 arcsec

03:00 left_wfsc000009.fits - wrong axis (+1000 arcsec Tilt was 1 row left)

03:03 undo Tilt, move Tip to +315

03:04 left_wfsc000011.fits - wrong direction in Tip (+1000 arcsec moved down 1 row of lenslets)

03:07 move Tip to -1515 arcsec

x=0, y=612.4

Why is every 2nd or 3rd WFS image blank/gray?

Why are most of the WFS images blank/gray?

03:14 left_wfsc000016.fits - Pupil is almost all on the lenslet array now.

SX M3 for LBBG determined for WFS: Tip=-1515, Tilt=-20 Select=-25.6

Steve applies the equivalent global offsets to PSF - dTip=-900 arcsec dRZ=5.088 deg

03:26 ACQUIRE preset (with RZ collimation enabled) - ptspiral - no star - a confusion about RZ

03:38 ACQUIRE preset - ptspiral - we find star in focus at IE=+44 CA=-15 (default IE=-65 CA=+36)

CCIT operations says that they'll be able to track down our DNS problems more thoroughly in the morning.

Adjusting the Rotator Zeropoint

We are adjusting LEFTZEROPOINT in PCS - we haven't done anything to the DSP zeropoints.

03:48 GUIDE preset to Persson star - acquisition offset goes the wrong way

LEFTZEROPOINT from 0.0 to +100 deg - restart PCS

04:00 ACQUIRE preset to Persson star

guider000001 CA=-10 56.87 276.20 guider000002 CA=-30 416.43 181.03

LEFTZEROPOINT from 100.0 to +114.8 deg - restart PCS

This is as good as we can do until we get the image better focussed and collimated.

Testing the Acquisition Offset

04:14 GUIDE preset on-axis - guiding is now working on-axis

04:18 left_wfsc000019 - no star is making it into the WFS

04:21 GUIDE preset BS9188 gs=1 (on Y-axis) - works OK

04:26 GUIDE preset BS9188 gs=9 PA=0 - works OK (But this might have been a random star - see troubles at 11:40)

Use the Default Pointing Model to refine the M3 position

04:30 ACQUIRE BS9188 gs=0 IE=46 CA=-21

Move IE to 36, then compensate with M3 selector -0.08 degrees with some crosstalk in negative CA

04:55 M3 motion hangs - restart OSS - power cycle

We messed around for an hour fighting UMAC and OSS. It turns out that there is no error message of any kind if you ask M3 for a position out of range. It doesn't move, and OSS reports success. Issue 3771.

IE=-65 CA=+36 Tip+330 Tilt-920 RZ=4.25 - are the values to match the default pointing model from October 29.

06:04 left_wfsc000022 shows alignment on the lenslets not as bad as we expected.

gs=9 PA=0 x=94 y=90

06:10 left_wfsc000024 shows that the pupil moved up (the good direction)

gs=9 PA=INDEF=46.2 x=0 y=131

06:10 left_wfsc000025 isn't too bad. Ideally the pupil would be one row higher and one column to the right.

Measuring the rotator center

06:20 Preset back on-axis parallactic, check focus

06:26 readGuideCam left -e 80000 (while slewing rotator 360 -> 0) - circle incomplete

06:28 readGuideCam left -e 100000 (while slewing rotator 0 -> 360) - left_guider00004 - Rick measures 255.7,283.0 as the center of a slightly elliptical circle. See ~MODSeng/agw1cen.png Why is it elliptical? Flexure?

second image imager guider000005 - see ~MODSeng/agw1cen2.png

06:45 GUIDE preset on-axis - timed out waiting for probe - why? - restart AGw1

06:51 GUIDE preset on-axis

Measuring hotspot position

06:56 GUIDE preset to a brighter pointing star

guidecam_hotspot_x 310.0 guidecam_hotspot_y 239.0

LBTO@obs2:104 % setHotspot left -x 240 -y 239
Hotspot set to (x,y): 240, 239
LBTO@obs2:105 % readWFSCam left -e 1000 ; ls

Oops it looks like there might have been a typo in the hot spot 210,239 is very close while 310,239 is far away.

07:19 GUIDE preset to BS9107 on-axis

000040 = bad readout 000041 = good readout

07:27 preset to fainter star off-axis so we can see sky in lenslet

LBTO@obs2:122 % getHotspot left Hotspot is (x,y): 220.00, 229.00 corresponds to left_wfsc000046

Update hotspot coords in AGW1_IRTC.cfg 220,229

07:49 Stop and restart GCS - resend preset - This puts light on WFS, but lower row of pupil missed the lenslet array when on-axis

getxy -o 16 -u 1 Guide probe position x: 37.000000 y: 1020953.000000

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: 0.558000 y: 612.522000

0.054 arcsec/pixel
0.032 mm/pixel

08:26 resend GUIDE preset (but that's not what we wanted)

setxy -u 1 -x 1.104 -y 612.522

08:29 TRACK preset

readGuideCam left -e 100000 (while turning rotator) guide000006 (went wrong direction in X)

08:37 Acquire Preset

setxy -u 1 -x -1.164 -y 610.826

stopAcquire left

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: -1.171000 y: 610.826000

readGuideCam left -e 100000 (while turning rotator) left_guider000007 209, 235

setxy -u 1 -x -0.85 -y 610.6

readGuideCam left -e 100000 (while turning rotator) left_guider000008 (collimating)

readGuideCam left -e 100000 (while turning rotator) left_guider000009 227,228

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: -0.864000 y: 610.599000

[telescope@agw1 etc]$ getxy -o 16 -u 1 Guide probe position x: -59.000000 y: 1010955.000000

These are the encoder numbers we want to have.

sudo vi oacontrol.conf

stopGCS, stopAGW, restart oacontrol, startAGW, start GCS, resend preset

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: -0.059000 y: 612.517000

[telescope@agw1 etc]$ getxy -o 16 -u 1 Guide probe position x: -4.000000 y: 968925.000000

Increased the homeoffset increment from +96 to +218 (I'm confused about those units), and adjusted limits accordingly.
Reduced the ooffset increment from 10000 to 2000 (now understanding the comment about units are microns), with no change in limits.

stopGCS, stopAGW, restart oacontrol, startAGW, start GCS, resend preset

09:52 ACQUIRE preset

[telescope@agw1 etc]$ getxy -o 16 -u 1 Guide probe position x: -7.000000 y: 1010484.000000

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: -0.118000 y: 612.509000

I think Y is correct, but I screwed up X since you can't tell if you changed the homeoffset in the right direction.

10:05 readGuideCam left -e 100000 - approx 144,226 so X is far overcorrected

Correction to AXIS_ROT homeoffset (and limits) +72 (history was +96, +218, +72)
Correction to RAD oofset +2000 microns

stopGCS, stopAGW, restart oacontrol, startAGW, start GCS, resend preset

10:24 readGuideCam left -e 100000 - guider000011 218,226

OK, it worked with only a bit of coaching from Jesper!

Rotate position angle to investigate whether chief ray is OK

10:27 GUIDE preset on-axis PA=0

10:32 readWFSCam left -e 20000 ; ls - wfsc000048

offset 0 0 90

10:34 readWFSCam left -e 20000 ; ls - wfsc000051

offset 0 0 90 (relative) - offset screwed up GCS was looking for star while rotator still unwrapping

10:39 GUIDE preset on-axis PA=0

offset 0 0 90 - returns to GCS before the rotator slew finishes - rotator never goes off-source as seen by IIF. Issue # 3774

10:45 GUIDE preset on-axis PA=0

offset 0 0 -90

10:48 readWFSCam left -e 20000 ; ls - wfsc000052

offset 0 0 -90 (rotator ends up at 159.1)

10:50 readWFSCam left -e 20000 ; ls - wfsc000053

The wobble of these 4 images is about two lenslet spacings (probably that is bad).

Unwind the +1200 arcsec Tip on M3 while allowing the pointing to vary.

10:55 ACQUIRE IE=-45 CA=-5 Tip = -300 Tilt = -920 RZ=4.25

11:02 GUIDE preset on-axis PA=0

11:04 readWFSCam left -e 20000 ; ls - wfsc000055

offset 0 0 90

11:06 readWFSCam left -e 20000 ; ls - wfsc000056

offset 0 0 90

11:08 readWFSCam left -e 20000 ; ls - wfsc000057

offset 0 0 90

11:09 readWFSCam left -e 20000 ; ls - wfsc000058

This was better, it reduced the pupil wobble by at least a factor of two. It's hard to say when the wobble is completely gone as there is some residual aberration.

Initial Check of Active Optics Corrections

(Trying this without any new pupil positioning parameters - while Doug is working on those.)

11:22 ACTIVE preset on-axis - got the goofy readout again on WFS

This is badly slowed by bad readouts.

11:33 -900 Z8 to M2

11:43 ACTIVE preset off-axis to check IRTC image PA=354.61 - star is visible on IRTC although not well centered.

11:46 Resend preset with PA=0.0 - but we don't see a star

11:50 ACTIVE preset off-axis to check IRTC image PA=354.61 - works

11:51 Resend preset with PA=359.99 - no star ????????

GAP IN NOTES

12:14 ACQUIRE off the y-axis

X-coordinates of the probe are reversed. Why? Are the AGWs different. PCS configurations should be the same as LUCI1.

Changed leftxyflip from false false to true false in PCSInstrument.conf

Now PCS sent the probe where John sent the probe manually.

Active Optics Tests with IDL

12:30 GUIDE on-axis - now we can't find on-axis star

12:35 put the flip back to false false, Restart PCS, restore IE/CA ??

12:41 Doug running active optics with IDL, plagued by goofy/rubbish images. Corrections starting at wfsc 000062, 63, 64

12:52 0.8 arcsec FWHM on guider at EL=33.5 degrees.

wfsc 000073 is a good image to determine lenslet edge values (as rising twilight illuminates the lenslets).

Rotation is fairly close, but maybe not exactly right. new Pupil positions are in the ~LBTO/idl/wfsc/

More on Flips in Twilight

13:05 ACQUIRE preset on-Y-axis

13:08 ACQUIRE preset off-Y-axis

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: -0.086000 y: 513.200000

[telescope@agw1 etc]$ getxy -u 1 Guide probe position x: -9.456000 y: 513.649000

[telescope@agw1 etc]$ setxy -u 1 -x 9.456 -y 513.649 Command successfully transmitted!

X-flip is confirmed, but not solved.

13:11 Close for impending dawn T=5.2 degC, D=-9.9 degC, Sky clear, wind 5-10 m/sec.from SW

-- JohnHill - 28 Oct 2011
Topic revision: r9 - 03 Nov 2011, JohnHill
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