AOBIR cookbook
Last updated: 18/06/2001
Send comments to:
lai@cfht.hawaii.edu, forveill@cfht.hawaii.edu
AOBIR runs July 2nd to July 12, 2001
From / To |
PI Name |
Run Id |
02 / 03 |
|
E/D |
03 / 04 |
Stockton |
H8B |
04 / 07 |
G. Soucail |
F33 |
07 / 08 |
|
D |
08 / 12 |
C. Marois |
C41 |
2 FOR OPTIMIZED OBSERVING CONDITIONS
-
Morning / afternoon calibrations: Flats in
IR in different filters require time, especially with narrow band filters.
Please do organize everything ahead of time with the SA and OA,
especially if you need flats on the sky. In normal circumstances the dome
has to be accessible to the daycrew from 08:00am to 05:00pm, so please
be aware that, according to the operation schedule, it may not always be
possible to fulfil your request. The SA will let you know what calibration
data are already available.
2 AOBIR in numbers
pixel size |
18.5 microns |
pixel scale |
0.0348 arcsec |
imaging field : 1024x1024 pixels |
36x36 arcsec |
Subraster: any size is allowed, as long as
the number of pixels is a power of two, anywhere within one of the
4 quadrants.
Do not use the center of the focal plane array as it is the intersection
of the 4 chips composing the whole detector.
Orientation: North up, E left within 2
degrees.
Bonnette angle: do
not move!
Distortion: < 0.5% for the whole field.
AOBIR internal flexure: TBD.
Image quality: good (SR > 0.9 in K band).
Quantum Efficiency: > 50% from 1
to 2.5 microns (peak 65% in K).
Readout time (full frame): ~9 seconds,
including writing to disk.
Readout noise: 20 to 25 electrons (some correlated noise).
Dark Current: 0.15 electrons per second.
Filters: 16 positions, including one blank
for dark
4 AOBIR BASIC OBSERVING PROCEDURE
Beginning of night
-
The Real Time Computer (RTC) should be reset at the beginning of the night
by the SA or OA, before you log on Neptune as "aobir".
-
You should have two AOB windows permanently available on screen:
-
"Adaptive Optics Bonnette ", open this window by clicking on AOB
in Pegasus menu (Pegasus/AOB for similar later references),
-
"APD" will appear automatically whenever the AOB is restarted.
Standard Procedure
-
Ask the OA to point the Telescope to your target.
-
If you need to change the hot spot, see the Offsets paragraph below.
KIR
-
Click on Pegasus/Filter to choose the filter (this requires a few
seconds and you must wait for the end of the operation before doing
anything else).
AOB
Density filter
Magnitude Range ® |
Density Filter |
R < 8 to 9 |
open |
6 < R < 8 |
1.0 |
R > 6 |
2.0 |
Those are indicative values, and will vary with sky absorption.
The flux on the APD should be < 8000 (linearity limit) and must
never be > 15000 (permanent damage limit). Adjust the density accordingly.
The flux should ideally be (> 80) to provide an accurate wavefront
correction, but the AOB will still provide some improvement with counts ~20.
The flux mentioned here is the number of electrons per millisecond on the
whole aperture (19 sub-apertures). Some displays refer instead to the average
flux per subaperture (19 times lower).
Gain
The default value for the optical gain is 128 for a point object.
You might have to adjust it for an extended object, the optical gain modifies
the amplitude of the WFS membrane and should be reduced to observe an extended
object, typically 50 to 80, with values ranging from minimum 50 to maximum
128.
To observe a very extended or very faint object, it may be necessary
to guide on a close-by star. This star has to be within the physical
limits of displacement for the WFS (40arcsec) and, the closest the
star is to the target, the more accurate are the corrections, due to anisoplanatism
effects.
Please check with your SA before using any other option in "Start
Adaptive Corrections" as they may require a reset of the RTC.
-
AFTER the OA confirmed the Telescope is pointing at the target ,
-
check the flux is increasing on the APD,
-
click on "Start AO Correction" in "Adaptive Optics Bonnette"
menu,
-
select the apropriate neutral density filter,
-
it is in most cases best to try first with a gain value of 128,
-
click "accept" in "Start Adaptive Corrections" menu.
You must check the values of the voltages applied for tip-tilt corrections
while the AOB is trying to close the loop. Those values should gradually
diminish as the object is getting centered on the WFS, and then vary around
zero once the loop is closed, in good conditions. If conditions worsen,
the tip-tilt will saturate at 4 volts, won't correct turbulence anymore
at 3 volts, is reliable up to 2 volts, although the correction is then
questionable.
The "What now?" window will appear, giving you two options, stop
or pause the loop.
Once the loop closed, you may realize you still have too many counts
on the APD (see above) or you may want to optimize the corrections by adjusting
the gain:
-
click "Stop Corrections" in the "What now?" window,
-
select the proper density filter and/or adjust the gain,
-
start the correction again.
Focus
It is important not to waste part of the dynamical range of the deformable
mirror to compensate the defocus of the Telescope. The offset that should
be applied to the Telescope focus (ask the OA if you don't know how to do
that) appears in the "APD" window some 2 minutes after the loop is closed.
It should be adjusted whenever it exceeds +/- 5. There is no need
to pause the corrections or stop the observations for this adjustment. It
will take about 2 minutes before the focus correction needed is updated
in the "APD" window.
So, in most cases, there will be three values you have to adjust to optimize
the corrections, first the density filter, second the optical gain if
necessary,
and the focus whenever necessary.
DetI
-
You can work either entering commands at the prompt (>) in the
DetI window, or using the DetI form (easier to enter comments, for
example), or using scripts executed in a unix shell window (NOT the DetI
window).
-
Evaluate roughly the integration time and take a snap to check the field
(>snap), the temporary snap file will be written to the disk, then
overwritten with the next one. Please note that the snap command does not
take any parameters; you have to previously enter the exposure time with
the etime command: >etime #, with # in seconds, between 0.1sec
minimum and 4hrs maximum.
Three modes of observation are available, cube, coadd and multiple sampling
readout (MSR). The MSR mode is currently unreliable and therefore disabled.
-
cube 0: mode cube off, go n, save n images in n independent files.
-
cube 1: mode cube on, go n, save n frames in one file (cube).
-
coadd 0: mode coadd off.
-
coadd n go: mode coadd on, add n frames and take the mean value,
n has to be a power of 2 (2 to 64).
-
msr 0: mode MSR off,
-
msr n go: mode MSR on, reset the detector, calculate the mean
value of n readouts, take an exposure, calculate the mean value of n readouts,
substract the first mean value from the last one and save the result, n
has to be a power of 2 (2 to 64).
The overhead being 9 seconds, the coadd mode is more adapted to very short
exposure time.
Cubes of coadds and MSR will be possible too, cubes of cubes will be
tested soon.
Offsets
If you want to perform an offset, click on "Offset" in "AOB Special
Functions", the window "AOB Offsets" will appear with five options:
-
"Setup (telescope only) Fill in From pixels only": The last hot
spot position is kept in memory. To bring the object image to this
last hot spot position from wherever it is on the array, just enter the
current object position in pixels in From x, From y,and click
on this option.
-
"Normal (Telescope and WFS) Correcting offset in pixels": Enter
the current position of the object image in pixels in From x, From
y, and the position where you want the image on the array in To
x, To y.
If you need to define a new hot spot, offset to the desired position
on the array, then click on "Setup Functions..." in the "Adaptive
Optics Bonnette" window, and then on "Record Instrument HotSpot"
in the "AOB Setup Functions" window.
Ask the OA to record the new hot spot position on the guiding camera
screen.
-
"RA/Dec offset with correction": offset without pausing the loop,
up to 40arcsec (physical limit of displacement for the WFS). This option
is use for mosaics or dithering.
-
"Sky offset correction paused": the system pauses the correction
before to move the Telescope to a larger offset on the sky. This option
is used for sky background determination in case of extended objects.
-
"Back from sky offset, correction continued (don't change numbers)":
after the Telescope moved back, will continue the corrections in the same
conditions.
Please note that you have to click one of these options to perform an offset,
"Done" will only leave the "AOB Offsets" menu.
Orientation: North is up and East left, to within about 2 degrees.
The exact orientation varies slightly from one setup of the instrument
to the next one, so for a more accurate measurement ask your SA who may
have determined this value during setup, or can otherwise suggest an
astrometric calibration field that you can briefly observe.
The WFS might be sliding slightly for large offsets (> 25arcsec), and
won't be coming back in the exact same position, this will be tested further,
up to 10arcsec seems perfectly safe for now.
Scripts
You can write your own scripts in CShell. Small modifications of the
provided scripts will probably be enough for most applications, but it's
also possible to write highly customized scripts to support most observing
procedures. Please talk to your SA/OA for examples, command syntax,
and general advice.
For mosaics, we recommand that you avoid placing your object at the
center of the array (unless of course it is extended enough that it fills
a significant fraction of the detector). This is the intersection of the
4 chips that make up the full detector, and the cosmetic there is slightly
degraded. The detector also has a few bad pixel clusters, and over time the
camera has unfortunately accumulated some dirt on internal surfaces (a
cleanup is scheduled during semester 2001-A). You will also want to avoid
the affected parts of the detector, which are easily seen on any flat-field
image. On a totally different matter, please make sure that your script
does move the object back to the center before exiting: this will leave the
WFS in the proper central configuration to close the loop on the next object.
Warning messages will appear in yellow in DetI window, error messages
in red, including the warming up of the detector and the lack of disk space.
Ctrl-C to interrupt a script (xterm or hpterm window).
Then, in DetI:
>break: will break a sequence of exposures in cube mode or in
coadd mode;
in cube mode, the current frame will be completed and the FITS Header
will be updated accordingly, all frames being saved including this last
one;
in coadd mode, the whole sequence will be lost.
>stop : will interrupt the exposure, save the data and update
the FITS header accordingly.
>abort: will interrupt the exposure and that exposure will be
lost.
Please note that the break/stop/abort command may not be applied succesfully
at first try,
enter the command again if you get no feedback from DetI.
>? or >help for a list of commands, typing the command
with no parameters will give you the proper syntax.
Be careful to stop the correction before asking the OA
to move to the next target.
PSF: Permanent recording during the observations, using the WFS
data, available to use in scripts only.
Telescope
Click on "....check...." in the main Pegasus menu bar, you will open a
new window, displaying "Status", "Catalogs", "Safety"
and "Dismiss", to get rid of that window.
Click on "Status" scrolls down a menu offering informations about the
Telescope Control System Status, the first one is the most important one,
click on "Telescope information" will open a window displaying UT,
UT Date, HST, Telescope and target coordinates, Equinox, Hour Angle, Airmass,
...
"Tiny information" opens a smaller window with the same informations.
"Dome Slit display" gives you the relative position of the Dome
and the Telescope, the small circle indicating where the Telescope is pointing.
"Weather strip" is helpful to monitor the weather (evolution
of the humidity, wind intensity and direction, pressure,...).
The other options allows you to obtain TCSIV status informations as
complete as the ones available to the observing assistant, but the Telescope
Status is the most important one for the observer.
Click on "Catalogs" scrolls down a menu with the available catalogs,
SAO / USNO, User object lists (any list you may have given to your observing
assistant), ObsLog object list (objects already observed during your run)
and New GSC TCSIV (map of the field with available guide stars, this is
the tool used by your observing assistant to search for guide stars, ask
him/her for details).
"Safety" should give you the current limits according to the
instrument and general conditions.
End of the night
-
Quit the APD window.
-
Click on "System Sleep" in the "Adaptive Optics Bonnette"
window.
-
>detach in DetI window.
Click on Pegasus/Aloha to end the Neptune session.
A few notes
-
Note that you are not allowed to change the filename ######x.fits (constraint
due to archiving), an odometer will automatically implement the number.
After an integration, the name appearing in the DetI form is the next filename.
-
Make sure you use the blank filter position to do your darks.
-
There is a danger of saturation in flats due to remanence after long
exposures.
If you want to clean the detector before moving to the next filter, a cube
of 8 exposures of 0.1 second is usually sufficient. Script
detclean will do that for you
-
Flats in IR in different filters require time. Please organize everything
in advance with the SA and OA, especially if you need flats on the
sky. Be aware that this might not always be possible, according to the
operation schedule.
-
Be very careful not to click on the wrong option when using an AOB window,
as engineering menus are still available.
-
Always WAIT for the end of any AOB action. Asking for simultaneous
actions will almost inevitably cause trouble! For example, trying to close
the loop while the filter wheel is still moving may require a reset of
the RTC (several minutes) to bring the system back.
Copyright © 1998, CFHT Corporation. All
rights reserved.