GRACE = phone extension 637, intercom 78
-----------------------------------------------------------------------------
Tel focus on WFS = 98.40 (13/6/03), 98.45 (3/9/04),
on INGRID 98.32, on AG3 98.24
OASIS focus 4.05 for SBIG, 6.1 for science.
INGRID focus -540.
-----------------------------------------------------------------------------
Acq TV AG3: 7/03 N left, E down (within 0.5 deg), at sky PA = 0,
0.407"/pix, 200" field
A star at 580 452 (6/04) (130 102 with the usual window 450:750, 350:600)
will be on the WFS.
-----------------------------------------------------------------------------
INGRID 1024^2 pixels as viewed by iraf, with simplex quadrant numbers
---------------
| 2 | 3| W <---- for sky PA = 0 in GRACE
| | | |
| | . | | Dot indicates position of normal
--------------- | acquisition, ~ pixel 583,643 => x,y=51,78 q3
| 1| 4| V 560,700 4/03
| | | ~ 0.0388, 0.0393 "/pixel in x and y (14/6/03)
| | | S and N-S dirn rotated from vertical at PA=0
--------------- by 0.67 deg anticlockwise (scale 0.038 5/02)
(3/9/04 - pinhole is at 599, 534)
Pupil-stop wheel naomi-clr: 236.35 (mfb 25/6/03) (237.4 4/03)
naomi-obs: 274.15 ( " " ) (273.8 13/6/03)
Order of 9-point dither on INGRID: 5-point dither is probably:
6 5 4 4 3
7 1 3 1
8 9 2 5 2
Rot centre 10/7/03 732 576
260 microns of INGRID focus corresponds to 0.1 mm of telescope focus
INGRID focus -1200 usually used with NAOMI
-----------------------------------------------------------------------------
MIT3 on OASIS, N up, E right, for PA = 0, scale = 0.02 "/pixel (checked 2/04)
Acq at 1052 2078
Scale at OASIS focal plane (lenslets) is 2.4 arcsec/mm
AG7 0.033 "/pixel
SBIG 240*320, 10-mic pixels, 0.025 "/pixel at f/16
TEK (at IR port) with mag*3 lens gave scale 0.029 "/pixel (SE 6/2/03)
-----------------------------------------------------------------------------
----
| | WFS cell 8*8 pixels mode 1, ~ 1500 units of tip/tilt on DM element
| | (spot typically visible over range ~ 2100 units, i.e. 2.5")
----
WFS pickoff position: 10.2,-0.8 7/03 mode 1 (7.0,-1.8 4/03), 10.8,1.4 17/7/03
If tel moved 1" E, need to move pickoff by +1" in X (2.93"/mm) to recover GS
1" N +1" in Y
With loop closed, to move image right on INGRID, decrease pickoff x.
To move up, decrease pickoff y.
Pickoff mirror as viewed from behind it (i.e. through SLODAR port in WFS):
+ve movement in x moves pickoff mirror right
+ve y down
Moving pickoff in x or y also moves both WFS foreoptics and camera.
When individual segments are moved from topgui 600 DAC in x and y,
the spots move on INGRID dx = -1 and dy = +1 arcsec respectively
(i.e. left and up).
-----------------------------------------------------------------------------
Band Seeing Sky cts/ Diff-limited Best perf (s = strehl, f = fwhm)
factor* pix/sec FWHM" pix
R 0.039
Z 1.13
J 1.17 ~ 1.5 0.07 1.8
H 1.24 ~ 5 0.08 2.1
K 1.32 ~ 35 0.12 3.1 s = 0.9 lab Durham
* factor by which seeing should be multiplied to get that in V band
2-mic pinhole = 0.009", 5-mic = 0.023" (need FWHM / pinhole >~ 5 or
strehl degrades)
J H K are for INGRID (K sky counts with nobs pupil)
-----------------------------------------------------------------------------
MECHANISM NOMINAL UNITS RANGE CONVERSION
--------- ------- ----- ----- ----------
NAOMI simplex lamp -50 -480 - 0
(for 5 msec wfs)
Old NCU cal lamp 90? 0-230 0 is minimum
FSM (= TT = OAP1) 4096, 4096 DAC 0-8191 Divide by 1640 for "
DM range corr to ~ +-700
Up to 20 kHz full stroke
Range corresponds to +-2" or +- 8V
DM (deformable mirror) 4096 4096 4096 DAC 0-8191 DAC = 7.5 A stroke
Tilt 0 0 (RM 2/03)
The range in tip/tilt is 6 microns, i.e. each element can move
+- 3 microns in piston, i.e. 0.73 nm = 7.3 A/DAC
Total TT range is about +-4"
WHT pupil covers 7.3 DM segments
Non-linearities < 40 A when DM on, rms error ~ 120 A, acc to Myers et al
DM X stage 0 mm
~3 for offset
Pickoff mirror 5.9, -1.9 " Div by 2.93 for mm
typical Mult by 0.5, -0.5
to give
change in AVE X, Y
16.4, 2.5 for ~ centre of INGRID chip 14/6/03
If telescope is moved 1" E, need to move pickoff -1" in X to recover star
1" N -1" in Y
If pickoff moves faster than 1.5"/mm, NAOMI can't keep up in closed-loop
operation.
x,y = 2.16, 2.15 corresponds to -29.5, -35.7 on epics page
WFS foreoptics focus 0 - 2 mm Mult by -0.1 for
focus shift on ALIGN gui
1.0 mm = 3.57 units on
epics page
(to expand array of spots on WFS display, increase telescope focus)
DM gain 0.3 good seeing
0.15 bad seeing
DECIMATE 10 10 (full frame, 20 or 30 for others)
(2 numbers, should be same)
Segment tip-tilt limit 0.1 (approx equal to apertures on WFS i.e. 2" usually)
FSM gain 20
(if > 50, fights DM)
FSM offload to TCS takes about 3 sec, with AUTOGUIDE on
SG gain 0.8
(0.001 to hold overnight)
SG loop operates at up to 3.3 kHz (doubling of this is possible)
WFS exposure time microsec 1250-50000
e.g. 2500 for bright star
White-light flat slope DACs/mirror
<~200
Zernike tip/tilt (x,y) < 100 Divide WL flat slope
by 3.2, -4.1 in x,y
to get Zernike with
same effect.
Zernike tip of 100 units results in -170 dx on DAC (100 tilt to +170 dy),
corresponding to ~ 2 black and 2 white fringes per element in x direction,
which is same direction as for a twist of the fold flat about a vertical
axis.
AddZernke tip 1 at command line corresponds to 40 units of Zernike in gui.
Simplex MF aim for < -0.03
De-staircasing 20 - 200 typical Adding X, Y removes
staircase to right, and
above, core of image
Reconstructor gains 109, 105
" convergence 0.001
SBIG pixels ~ 31 pixels/"
Filter wheel: GG495 (1), v_wide (2), red (3), blanks (4 - 6)
LENSLET WHEEL:
posn "/pixel pixels
1 x-shift (formerly reticule) 3
2 f195_1.1 good seeing 63 0.26 8 63.35 on epics page
3 doublet 5" field
4 f95_1.1 poor seeing 243 0.52 8
5 f195_1.1 y-shift
6 f95_2.2 4*4 lenslets 0.26 16
(driving lenslet wheel also causes camera stage to move)
WFS camera stage 'focussed' 210.44 on epics page
'near_pupil'
'pupil'
WFS MODES:
WFS pixels fast-mode
lenslets read image diam" readout/msec
1 8*8 8*8 8*8 2 full-frame 5.3
2 8*8 4*4 2*2 1 quad-cell 1.4
3 8*8 4*4 4*4 1 3.0
4 (modes 4 5 6 are for 2-CCD mode)
5
6
7 test mode
8 4*4 4*4 4*4 1
9 4*4 4*4 2*2 1
10 tip-tilt mode
11 8*8 8*8 4*4 2 2.5
12 8*8 8*8 2*2 2
In-hand:
13 4*4 8*8 4*4 2
14 4*4 8*8 2*2 2
15 4*4 8*8 8*8 2 lower priority
Image = size of array delivered. E.g. Mode 2, sometimes referred to
as 8*8_2*2 is for 8*8 lenslets, reading 4*4 pixels binned 2*2,
delivering a 2*2 array of values.
Modes 1 2 3 8 9 10 are not synched, 4 5 6 are synched.
The readout overheads were obtained from the top-gui WFS page, and
appear to be independent of exposure time.
WFS noise in fast mode should be 4.5 e-, 0.56 e-/ADU, i.e. 8.0 ADU
(from old ATC document, I think).
Last measurement was by SMT, 9 ADU.
Slow mode is too slow for AO.
WFS processor handles frame rates up to 500 Hz.
----------------------------------------------------------------------------
lambda/4 rms corresponds to strehl 0.84
NAOMI technical information ·
WHT ·
Astronomy ·
ING
