NAOMI summary commissioning log
NAOMI technical information ·
WHT ·
Astronomy ·
ING
The entries below summarise results from each NAOMI run
(defined as any period when NAOMI was scheduled to be available
for commissioning or observing) and from some of Durham's Rayleigh-laser
tests:
- 2004 Apr 10
- 2004 Feb 2
- 2003 Oct 6-9
- 2003 Sep 6-16
- 2003 Jul 9-23, Aug 10-16
- 2003 Jun 13-14
- 2003 May 12-13
- 2003 Apr 13-21
- 2002 Nov 13
- 2002 Sep 22-26
- 2002 Aug 16-28
- 2002 May 22-31, Jun 21-22
- 2002 Mar 31
- 2001 Sep 1-4, 27-30
- 2001 May 29-31, Jun 4-15
- 2000 Dec 20-23, 2001 Jan 4-7
- 2000 Aug 14-15, 21-25, Sep 9-11, 14-17
- 1999 May 26-27, 29-31, Jun 21-27
- 1998 Jul 4-7
- 1997 Jun 17-20
- 1996 Aug 26-28
Total NAOMI commissioning time 2000 Jan 1 - 2004 Feb 2 = 47.4
clear bright nights (the earlier runs were with ELECTRA,
total time on-sky ~ 10 nights?).
Total OASIS commissioning time 5.0 nights.
More detailed reports can be found in
~crb/ao/yymmdd.report.
Laser tests by Durham team.
Nights used: 2 for RLGS tests
Conditions: seeing ?
Observers: Tim Morris, Richard Myers, Colin Dunlop, Paul Berry,
Roy Ostensen Tim ?
Successes
- Laser worked fine, first 2 half nights used for alignment and focus.
Second two half nights wiped out by bad weather, no chance to close loop
on WFS.
Problems
NAOMI in GRACE with OASIS/MIT3 as science camera, to test on-sky
performance after discovery of problems with FSM.
Nights used: 0.4 for NAOMI tests
Conditions: seeing 0.8 arcsec
Observers: Chris Benn
Successes
- Confirmed that NAOMI still corrects the FWHM in the optical
from 0.8 to 0.5 arcsec, i.e. it is performing as well as it was
last year, so the FSM problems probably aren't new.
- Closed the AO loop on the nucleus of galaxy N4151 (B = 11), and
took a test 600-sec spectrum with OASIS through the HR667 filter,
with the 22-mm enlarger.
- Measured throughput to WFS, and counts from moonlit sky.
Problems
- SDSU crashed ~ 15 times, usually on changing WFS integration time,
and nearly always requiring a power-cycle of the controller rack in
GRACE. DR 16127.
- Table doesn't float properly, so difficult to laser-flatten.
NAOMI in GRACE with INGRID and OASIS/MIT3 as science cameras.
Nights used: 3.1 of the 4 nights were lost to bad weather (rain and clouds),
0.5 to technical problems, 0.4 were used for NAOMI/OASIS tests.
No science programmes observed.
Conditions: cloud and rain most nights, seeing varied 0.7 - 1.3 arcsec
on the night we opened.
Observers: Gilles Adam, Roland Bacon, Chris Benn, Richard McDermid
(from Oct 7), Ilona Soechting
Successes
- First stage of Craige's acquisition gui tested on-sky.
- Excellent simplexes (FWHM = 2 pixels) obtained on INGRID, SBIG and MIT3.
- LR160 grism installed in OASIS;
lenslet array realigned with its mask (to stop vignetting in one corner
of CCD);
wedges in beam-steerer rotated; FITS headers now correct for OASIS;
XOASIS successfully tested with data from MIT3 in its new orientation.
- Initial tests of TT mode (using doublet on WFS).
Problems
- Mirror (DM) lost shape on the one night we opened, most of night lost
to this (possibility related to sudden temperature change in GRACE)
and to second loss of shape in the morning.
- DM segment 45 died.
- Y movement of pickoff probe stuck twice 'soft limit'.
- Pre-OASIS focus corrector out of range, not possible to focus
pinhole on MIT3 CCD fo all filters.
NAOMI in GRACE with INGRID and OASIS/MIT3 as science cameras.
Nights used:
4.2 for NAOMI science, 1.9 for MCAO commissioning,
1.2 for NAOMI commissioning,
3.5 for ISIS science,
0.2 lost to tecnical problems,
0.0 to bad weather
= 11 nights in all.
Conditions: 4 nights clear, 5 nights with some cloud, 2 nights dusty, seeing
ranged 0.5 - 0.8 arcsec on 4 of these nights.
Observers:
Chris Benn, Sebastian Els, Roy Ostensen, Ilona Soechting
Successes
- 4 nights of NAOMI/INGRID science observing.
- Observing efficiency measured on the night Sep 10,
= total integration time on
sky / length of night = 6.0 hours (840 closed-loop exposures)
/ 10.0 hours = 60%,
comparable with that achieved for typical observing with spectrographs.
- WFS mode g11 (8*8 pixels on WFS binned to 4*4) tested,
delivers better correction than default mode.
All-time record achieved with this mode on-sky Sep 10, FWHM = 2.8 pixels
in H band (0.11 arcsec).
- Initial tests of tip-tilt-only mode (using doublet in WFS).
- RLGS/MCAO tests in GHRIL by the Durham/MPIA team. RLGS launched
successfully on-sky at expected power.
Problems
- 1 hour lost to low-level communications problems causing dither script
to end unexpectedly.
- Two instances of pickoff mirror failing to move.
NAOMI in GRACE with INGRID and OASIS/MIT3 as science cameras.
Nights used: 2.3 for NAOMI commissioning, 4.8 for OASIS commissioning,
5.2 for NAOMI science (plus 4.7 for ISIS science, 3.2 lost to
bad weather, 0.8 to technical problems).
Conditions: most nights clear, seeing < 0.7 arcsec on 8 out of 21 nights.
Observers: Gilles Adam, Roland Bacon,
Chris Benn, Sebastian Els, Roy Ostensen, Ilona Soechting
Successes
- New NCU and acquisition camera AG3 commissioned.
OASIS flat, simplex flat and focus drive commissioned.
- OASIS first light Jul 11, most of tasks on commissioning list completed.
Many PR / test targets observed.
- Very good simplexes routinely achieved.
Craige's new simplex (August) running at 10 Hz.
- Best FWHM on sky 0.14 arcsec. FWHE and strehls being measured up now.
- 5.2 nights spent observing 7 NAOMI/INGRID science programmes.
- Craige's acquisition tool (acquire guide star, offset to target)
successfully commissioned.
- Further tests of Savart-plate interferometer, now nearly ready to use.
- New GRACE pointing model gives 1.6 arcsec rms.
Problems
- Gains on some modes found to be defaulting to incorrect values,
which might account for some of the lack of performance
improvement on switching between WFS modes. Gains corrected.
- Repeated WFS mechanism failures (usually foreoptics stage), not always
easy to spot at night, difficult to reproduce during the day.
- Acquisition gui needs improving for October. Control guis need
rationalising (we typically keep ~ 40 windows open on 5 displays).
- ND filters needed in WFS for use with very bright stars.
NAOMI in GRACE with INGRID as detector.
Nights used: 0.8 for NAOMI service (Jun 14) + 0.3
lost to technical problem (Jun 13).
Conditions: Photometric, bright-of-moon, seeing 0.6 - 0.8 arcsec Jun 14
(dipping as low as 0.4 arcsec according to RoboDIMM).
Observers: Chris Benn, Sebastian Els
Successes
- Best simplex on 5-micron pinhole was Jun 13, 2 diffraction rings in
H and K, strehl = 0.8 in both bands.
Less good Jun 14: FWHM = 2.5 pixels in H but strehl only 0.2.
- Closed-loop performance (V = 8 star) on-sky in different modes:
With ELECTRA reconstructor, modes 1, 2 and 3 yield FWHM 2.9, 15
and 3.2 pixels. Strehl in mode 1 is 0.11.
With NAOMI reconstructor, modes 1, 2 and 3 yield FWHM 3.2, 5 and 2.9.
Strehls are 0.08, 0.03, 0.08.
Summary: modes 1 and 3 give similar PSF, mode 2 gives poorer PSF,
NAOMI and ELECTRA rconstructors give very similar performance
in modes 1 and 3.
- Observed 14 9-point dithers for Bejar, with
WFS in mode 1, 2.5-msec integration time, corresponding
to ~ 100 Hz loop speed. The first target appears to be a
clearly-resolved 0.15-arcsec double.
Targets (= guide stars) ranged V = 9 - 12. AO-corrected FWHM typically
0.2 arcsec for all targets. Strehls ~ few %.
System very stable all night, no problems apart from one crash of
dither script, cause unknown.
- Target acquisition done with INGRID, in movie mode.
- Analysis of sequences of images shows no evidence for settle time;
first image is as good as later images.
- RoboDIMM was measuring seeing 0.6 - 0.8 arcsec all night Jun 14,
and the open-loop seeing on INGRID was consistent with this.
The seeing values reported by RoboDIMM dipped as low as 0.39 arcsec.
- Scale on INGRID = 0.0390 arcsec/pixel, +- 2%.
With sky PA = 0, N-S direction is rotated from vertical by 0.67 deg
anticlockwise (on image as conventionally viewed).
Problems
- Poor on-sky performance Jun 13, traced to high noise on WFS,
due to Peltier coolers switching off when chiller failed.
- The delivered PSFs are severely degraded when the guide star
is positioned near the edge of the INGRID chip.
Mean fwhm was 6 pixels (0.23 arcsec) within 10 arcsec of the chip
centre, but 10 pixels (0.39 arcsec) at radius 15 arcsec (based on
17 images, each with guide star in a different position on chip).
- INGRID naomi-obs pupil was misaligned ~ 1 segment on DM, default
position of wheel should be changed to 273.8 (this position was
used for most of the observations).
- Pinhole image found at pixel 770, 672 on INGRID,
not as well centred on
Rotator centre not on chip. Probably some misalignment.
NAOMI in GRACE with MIT3 CCD as detector (this CCD will be used for OASIS).
Used with with OASIS dichroics, OASIS feed flat + doublets,
MITLL3 CCD at position of OASIS (and SBIG simplexing camera fed by
deployable flat just in front of MITLL3). Field ~ 1 arcmin.
The new NCU, ADC, dichroic changer and OASIS were not present.
All objects were observed at low airmass.
Total time used: 0.8 nights (the remainder was for INGRID tests + service,
in poorer seeing).
Time used:
Conditions: 0.6 - 0.8 arcsec seeing during the tests.
Observers: Chris Benn, Ilona Soechting at night;
Sebastian Els, Tom Gregory for daytime setup.
Successes
Problems
NAOMI in GRACE with INGRID as detector (first time on-sky
after move to GRACE).
Total nights used: 2.2 for NAOMI commissioning, 0.1 for OSCA,
1.7 for ISIS service, 4.7 lost to bad weather, 1.7 lost to technical
problems.
Conditions: seeing 0.7 - 1.2 arcsec during the time used for NAOMI tests,
worse (up to 3 arcsec) on the other nights.
Observers: Chris Benn, Sebastian Els, Tom Gregory, Roy Ostensen,
Paco Prada, Ilona Soechting
Successes
- Excellent simplexes again obtained with the 5-micron pinhole, up to
4 diffraction rings in H.
- Fringes obtained with Tom's Savart-plate interferometer (FELIX) view
of DM. This is expected eventually to supercede the white-light
flattening.
- With the bright light source in the NCU, (nominal) integration times
as short as 0.02 sec can be used for the simplexing, and it runs
at a brisk 4.0 Hz.
- OSCA - Pete Doel's team were at ING to align OSCA and install
3 new stops. Sebastian will be circulating a full report on
the OSCA work.
- There was no change in peak height on changing from Hbarr to
the old H filter, so the latter doesn't degrade the wavefront.
- Good working environment in GRACE, no unexpected problems.
- Tom has bought a thin plane mirror which can be mounted over the DM.
A comparison of best PSFs obtainable with and without this mirror
will measure any degradation of PSF due to DM being segmented.
- We closed the loop in modes 1,2,3,8,9 with a range of exposures
for stars with m = 8, 10, 12, 14, 15, and also tested different
reconstructors. No strehls measured yet, but we can draw a few
conclusions on the basis of FWHM measurements in sub-arcsec seeing:
- For a bright star, m = 10 or 12, mode 1 2.5 msec gave the best
PSF, fwhm = 3.4 pixels, modes 2 and 3 were almost the same,
mode 8 worse (fwhm 5 - 7 pixels), mode 9 worse (fwhm 7 pixels).
Strehls in mode 1 not measured yet, but the PSFs look very good.
Ilona suggests that speed-gain for modes 2 and 3 might be offset
by loss of photons from the ignored pixels. At what value of
seeing is it predicted we need to switch from the good-seeing
to poor-seing lenslets?
- Comparison of reconstructors:
Mode Reconstructor FWHM/pixels
open loop 20
9 electra 6
9 sor4 13
9 naomi 6
8 sor4 8
8 naomi 5
8 electra 5
open loop 17
- For m = 14 star, there was significant correction 16 to 10 pixels
with WFS exp times as short as 10 msec in mode 1.
Approx the same in mode 8, i.e. this may be the crossover mag at
which mode 8 does as well as mode 1 (consistent with Ilona's models).
We were impressed by how well the AO system could correct even
when the spots on the WFS were almost invisible to the eye.
- OSCA had ~ 1 hour of good-seeing tests on-sky, Sebastian will
circulate a report.
- Dither scripts and autoguiding tested at the new focus, required
changes of sign made in software.
N now up, E to right on INGRID as usually displayed.
- SLODAR (12*12 WFS) commissioned in GHRIL by Richard Wilson,
35-50 Hz frame rate, Cn^2 turbulence profiles measured on
4 nights (ground layer >95%, 93%, 55%, 78% on these nights,
with other contributions at 2.5, 5, 10, 13 km, wind speeds,
up to 20 m/s).
SLODAR now moves to Mercator Cass focus, will measure Cn^2 profiles
nightly Jun - Dec 2003.
- DIMM, NAOMI, WHT Cass autoguider, INT and JKT seeing were compared
on several nights. This was not a systematic experiment, but there
seemed to be general agreement, except that the DIMM numbers were
typically a few tenths of an arcsec optimistic (although they agreed
with the measurements from GHRIL SLODAR on one night).
Problems
- Persistent problems with white-light flat, typically making only
~ 50 measurements rather than 150. Sometimes the simplex was
excellent in H, but the images were poor in J and K, a clear sign
that the wlflat hadn't phased the mirror.
- Mirror event one night when trying to go to mode 8
with 200 msec exposure time, naomi reconstructor. The mirror
recovered ~ 15 mins later.
- INGRID controller SDSU2 has lower noise than SDSU1, but has a fan,
which generates vibration and airflow. SDSU2 can't be fitted
with the chiller plate because the case is slightly the wrong size.
- No problems with vibration or oscillations, but we did suffer
from elongated images for a while on the night Apr 17.
- Readnoise on WFS seems high, ~ 7 e- (assuming gain 0.6), probably
this always dominates sky, may be limiting the mag we can reach.
- Frequent topgui crashes, but recovery straightforward.
- Failure of pickoff probe movement during dither. Recovery using init
and index not at all straightforward, the omc was rebooted the next
day to get rid of remaining problems.
- In AutoOffsets (particularly in mode 9), the AutoCentre can get
stuck in an oscillation, such that in each iteration, it overcorrects
by a factor 2. Can be fixed only by manually typing in the required
pickoff x,y during this oscillation.
- Occulter in NCU is stuck, probably near max value, but we used
an independent light source anyway.
- Gates into GRACE can only be opened with telescope az set to certain
positions.
- Significant diffraction from cross hairs on OAP1, they may need to be
removed for the OSCA observations in July.
Total of 0.3 nights used tests, but seeing 1.5 - 2.5 arcsec,
nothing useful obtained on-sky. Optical detector (AG2).
Useful runs 550751-777.
Observers: Chris Benn, Tom Gregory (+ Roy for afternoon setup)
Commissioning successes
- AG2 used successfully for simplexing + on-sky tests.
- Established feasibility of measuring static + tip-tilt performance
by gradually reducing SG gain to very low value.
- Sky background (moonlit) measured on WFS.
- Background light contamination from laser measured through R filter
on CCD.
Problems
- Spots on WFS double in x direction, traced to misalignment
of DM and WFS.
- With OSCA on the bench, the CCD mount could no longer be placed in
its nominal position.
Relay lens installed to transfer the focus further along the
bench.
- Failure of various mechanisms on startup, slow to recover.
NAOMI in GHRIL with INGRID as detector.
Useful runs 537481-538178.
Total time used:
1 night was used for NAOMI service, 3.5
for NAOMI tests and commissioning, 0.5 lost to bad weather.
Observers: Chris Benn + ?
Conditions: seeing 0.7 - 1.0 arcsec on all nights except Sep 24.
Commissioning successes
- Performance characterisation in seeing ~ 0.7 arcsec, particularly
of degradation of performance with radius, via images of the
outer parts of the globular cluster M15.
- Efficient 1-hour dithered integrations on Kuijken's targets
(allocated NL time), overheads very low.
- Simplexed mirror shapes achieved which are excellent in all 3 bands
J, H and Ks, suggesting that problems in Aug probably caused by
deficiencies in white-light flattening.
Airy rings seen in pinhole images in all 3 bands, at predicted radii.
- Sky background on WFS measured.
Problems
On both of the nights Sep 22 and 23, we got blobby or banana-shaped
images on-sky (~ left-right), after getting perfect images with the pinhole.
On putting the pinhole back in, lamp images were agin perfect.
The blobs did not rotate with position angle, suggesting something
on the NAOMI side of the derotator, but on the telescope side of
the pinhole! RM suggested that the table might be oscillating
left-right, perhaps caused by removal last week of UES from the
other platform (although it was balanced with weights).
On Sep 22, the blobbiness seemed to vary slowly ~ mins, and was present
on more than one star image, then vanished at midnight.
- We explored the decrease in peak intensity of the pinhole image
on closing the AO loop. Flux is conserved, but the FWHM increases
slightly, and peak height drops, on closing the loop.
Effect of errors in the offsets? Needs exploring.
- Mode 9 (4*4 lenslet, WFS binned 2*2) tried on sky. Usually, but not always,
the FSM drives to a limit, and the WFS loses the star.
- The sky baffle is offset ~ 3 pixels (nearly 50% of a cell) in
both x and y on the wfs, which (a) increases sky background over that
when it is centred, (b) limits the range over which spot movements
can be measured.
- Lots of TopGui crashes, but not much time lost.
- ~ 3 hours lost to crashes of whtdr machine, which strangled all the
other cpus.
NAOMI set up INGRID as detector.
Total of 1.9 nights used for commissioning, + ~ 2 nights lost to
technical problems associated with NAOMI.
Useful runs 525181-536369.
Observers: Chris Benn, Sebastian Els,
Paco Prada, Roy Ostensen, Almudena Zurita
Commissioning successes
- Performance characterised in 0.6-arcsec seeing.
- Nearly 3 nights of NAOMI service observing, including
observations of near-earth asteroid which resulted
in a press release and BBC coverage.
- OSCA throughput and suppression performance measured.
Problems
- Poor initial alignment of INGRID with NAOMI and of OSCA with
both.
- INGRID grabs slowed from 2.2 Hz to 0.7 Hz plus long pauses,
severely hindering afternoon setup.
- Simplexes in one band no good in other bands, probably due to
failure of white-light interferometry in x-direction.
- Independently (?) of the above, the x-shift lenslet
is not well-aligned, shifts being made manually.
- NAOMIOMC and NAOMISDSU no longer pingable.
NAOMI set up INGRID as detector.
Total of 5.6 nights used for commissioning NAOMI, 2.1 for OSCA.
Useful runs 503131-503500.
Observers: Chris Benn, Nigel Dipper, Sebastian Els, Tom Gregory,
Richard Myers, Roy Ostensen.
Commissioning successes
- OSCA coronagraph commissioned on-sky by UCL team.
- AO performance measured for many combinations of band, seeing,
guide-star mag and radius from guide star.
- First scheduled observer to go on-sky (Rebolo), good data
obtained.
- Service observations, ~ 1 night (in addition to the above 5.6).
- PR images of solar-system, PN and globular-cluster targets.
- Setup scripts and new gui commissioned. Afternoon setup now
much faster, night-time operation close to push-button.
- Performance with faint guide stars improved by installation
of sky baffle near WFS pickoff mirror.
- 4*4 WFS mode (faint stars) used on-sky.
- SLODAR replacement for JOSE commissioned on-sky, measures
r0, and Cn^2 distribution with altitude.
- Dither script tested.
- Sporadic comparisons made between seeing measured
with NAOMI/INGRID,
SLODAR, WHT Cass aux-port, INT, JKT, Mercator, TNG and the
IAC DIMM (ridge).
Problems
- Cloud and/or bad seeing (up to 4 arcsec!) nearly every night.
- Occasional top-gui crashes.
- Occasional 'deformable-mirror events', degrading PSF.
- PSF sometimes very sensitive to telescope focus, implying DM
elements being driven to end of range.
- One instance of sticking fore-optics in the WFS, dirt on encoder
track.
The aim was to test Tom's sky baffle, fitted on the WFS pickoff mirror,
close to the focal plane. Bad weather, but reduction in intensity
of light recahing wfs from the illuminated dome was a factor ~ 10,
as expected for a 2.5-arscec aperture (vs ~ 9 arcsec previously).
The size of the aperture was also confirmed by imaging it with the WFS
doublet.
NAOMI set up with TEK1 CCD as detector, fed via Laser Components R or I
dichroics, to assess performance in the optical (for future use
of NAOMI with OASIS).
Total of 4.8 nights used for commissioning.
Observers: Roland Bacon, Chris Benn, Tom Gregory,
Richard Myers, Roy Ostensen.
Commissioning successes
- Closed-loop correction with guide stars to V = 15.
- Significant correction in R and I bands e.g. FWHM 0.8 to 0.5
arcsec in R band for guide star V = 11.
- Loop closed on several galaxy nuclei, e.g. M31, N1068.
- May/Jun misalignment probably traced to rotation of off-axis
paraboloid 2 in its mount.
- Much more stable now that fold flat and OAP2 have been
clamped to bench, and now that strain-gauge loop left closed
overnight.
- Throughput for WHT/NAOMI/CCD measured in R band, and predicted (with
MIT-LL/BIV CCD, and ADC and relay lens in path) to be 0.53,
compared with CFHT/PUEO/CCD throughput of 0.46, satisfying one of
the requirements for OASIS to come to the WHT. WHT in addition
has 1.4 times more collecting area than CFHT.
Problems
- Closed loop in 4*4 mode, but still not working for faint stars
(for which it provides the main gain), probably because of a flaw
in the background subtraction.
Fast-guiding mode not ready for testing.
- DM needs more and more shimming to get light through correct
lenslets. Shift of ~ 0.5 element discovered after commissioning,
corrected by moving DM carriage.
This may have degraded correction slightly.
- WHT later to found to have been severely comatic during Sep 27-30
(1-mm tilt on secondary, after focus drive collided with a cable
connector)
but NAOMI removed the coma so well we weren't even aware of it,
and the closed-loop FWHM gains Sep 1-4 and 27-30 were similar.
- FWHM of pinhole should have been ~ 1 pixel on the detector
(0.07 "/pixel), but was typically 1.5-2 pixels, and sometimes
wouldn't reduce below 3 pixels, i.e. a significant part of the
minimum FWHM of 5 pixels seen on sky may have been due to the
mirror flat.
NAOMI set up with INGRID as detector.
Total of 7 nights used for commissioning, 3 for service observing.
Observers: Chris Benn, Tom Gregory, Johan Knapen, Andy Longmore,
Richard Myers, Roy Ostensen, Almudena Zurita.
Commissioning successes
- Closed-loop correction with guide stars to V = 13.1.
- ING staff ran NAOMI without external help, during the last 2 nights.
- Dithering from INGRID script controlling TCS and pickoff mirror.
- COHU TV now reaches V ~ 12.
- Top-level engineering gui in routine use for first time.
- New calibration lamp.
Problems
- Navis hacked, Durham team lost ~ 5 working days, pre-commissioning.
- DM X-stage failed.
- Misalignment in NAOMI, images cross-shaped before simplexing.
- DM strain-gauge event Jun 7, mirror suddenly lost shape, after
being stable ~ 36 hours.
Control of mirror elements regained slowly, over ~ 1 hour.
- 4*4 and quad-cell modes giving worse performance than full-frame,
perhaps because of a background-subtraction problem.
(See AJL report 12/6/01.)
Total: 8 nights scheduled observing.
The Dec run was wiped out by bad weather, the Jan run by poor
seeing and technical problems.
Commissioning successes
- First scheduled-science exposure r386265.fit, 4 Jan 2001.
Problems
- 5 nights lost to bad weather.
Total of 11 nights used for commissioning (12 were allocated).
Commissioning successes
- Near-diffraction-limited images in Ks, good PR pictures of PN.
- Throughput measurements to INGRID and to WFS.
- Pickoff-mirror coordinates calibrated.
- Initial tests on ADC positioning.
Problems
- Telescope oscillations during Aug.
- INGRID foreoptics designed to match old NAOMI layout,
INGRID (and dichroic?) had to be moved.
- High background in Ks, emissivity ~ 100%.
- Significantly worse seeing in GHRIL than at Cass after change
to GHRIL, on several
occasions, perhaps because NCU lamp not cooled.
- Acquisition TV reaches only V ~ 9.5.
Total of ? nights used for commissioning (ELECTRA).
Commissioning successes
- Near diffraction-limited images in J, K.
Problems
- Bad seeing local to GHRIL, traced to warm air flowing through
the derotator (now sealed with windows at each end).
- Throughput ~ 1 mag lower than predicted, probably due to dirty
surfaces in the derotator, again due to the flow of air through it.
- Telescope azimuth oscillations (now eliminated).
- RF electrical noise (now eliminated).
- Using WHIRCAM as a detector instead of INGRID (whose commissioning
was delayed).
- GHRIL derotator misaligned (no serious impact on run).
Total of ? nights used for commissioning (ELECTRA).
Commissioning successes
- Strain gauges working well in GHRIL environment.
- Co-phasing of segments (i.e. piston as well as tilts) to give near
diff-limited images in K in one dimension.
- Mirror shape stable for periods ~ 24 to 48 hours in Ks.
- Diffraction limit in Ks, FWHM ~ 0.15" reached for short periods.
Problems
- Vibrations from WHT oil pumps, UES cooler, WHIRCAM, new (damped)
optical bench needed.
- Poor quality optical bench, transmits vibrations.
- Poorer seeing in GHRIL than at Cass.
- Limit on integration time is duration of stable seeing
before burst >> 1"
seeing which causes poor segments in DM to lose lock.
Total: 4 nights commissioning (ELECTRA).
Commissioning successes
- Reached guide-star mag ~ 8.
Total: ? nights commissioning (MARTINI/WHIRCAM).
MARTINI was first used at the WHT ~ 1989.
NAOMI technical information ·
WHT ·
Astronomy ·
ING