Home · Search ·
About ING ·
Astronomy ·
Public
Information · Engineering
WORKING WITH AUTOFIB/WYFFOS
Autofib is a robotic fibre positioner at prime, feeding light to WYFFOS
in GHRIL. The optical configuration of the telescope is FOLDED (BROKEN)
CASS, in order to direct the light of the WYFFOS calibration
unit via the nasmyth flat up to prime.
A new LINUX machine is setup to run the AF2 robot and a new GUI to display
the AF2 status.
First Night Tasks, (determining Rotator Centre, Gripper Center)
More information like recovery procedures can be found on the AF2 main index.
Gripper centre
(viewsky 0,0)
|
172,213
181,221
151,219
|
21/11/05
17/08/05
20/04/06
|
Rot. Center Af2
viewsky -600,500
viewsky 0,0
|
148,116
202,208
192,174
|
21/11/05
17/08/05
20/04/06
08/06/07
|
Telescope aperture '0'
|
0.6,8.9
|
21/11/05
|
Telescope focus
|
82.50
|
09/06/05
|
AG2 window: [160:550,30:500]
|
Nominal Rotator offset:
-0.36 to -0.45 (20/04/06)
|
SOME LINUX COMMANDS:
|
TO>AF2 VIEWSKY 0 0
TO >AF2 WAKE
TO >AF2 SLEEP
TO >AF2 PARK
TO >AF2 RECOVER
TO >AF2 VIEWOBJECT 112
TO >AF2 MOVEFIBRE 16 0 0
TO >AF2 OFFSETTEL
.......more in the AUTOFIB web page
|
|
TELESCOPE SETUP -TCS-:
- USER> STATION PRIME
- USER> INSTRUMENT AUTOFIB
- USER> AGSELECT AF2_TV
- Zeroset AZ and ALT
to zeroset on the targets switch to computer control:
- USER> AZ 298
- USER> ALT 89
- USER> ENG
- USER> ZERO AZ TAR
- USER> ZERO ALT TAR
Drive AZ and ALT slowly in + direction over the zeroset target
points, then switch to computer control again
to zeroset absolute:
- USER> ZERO AZ ABS
- USER> ZERO ALT ABS
Zeroset absolute will go wrong in case an absolute encoder bulb
is broken! When slewing to the first object check azimuth and altitude
incremental and absolute encoders are in close agreement and updating.
- USER> ZERO ROT ABS
- USER> WAVELENGTH ...
- USER> TEMP ...
- USER> PRESS ...
- USER> HUMI ...
- USER> CAL LAST
- USER> FOCUS 82.45
Focus (depends where you are looking!):
The best focus both on the WYFFOS detector and on the TV fibre-view
is 82.25 mm. However, due to aberrations intrinsic to the WHT (mostly
spherical), the best focus in the outer parts (>=20 arc minutes off
axis) of the field may be as high as 82.65 mm. Hence a compromise value
of e.g. 82.45 mm can be used. Fiducial stars right at the edge of the 29
arcmin field will be elongated by as much as 3-4 arcsec!
On the mobile probe the best focus is at a position of about
83.20 mm. This may be important to know when acquiring faint sources
with help of the uDAS TV.
The uDAS Autoguider:
General Information;
The camera looks at the output end of the guide fibre bundles (also called
"fiducial fibres").
The new Small Fibres unit has 10 fiducial bundles:
- Each consists of about 10,000 (yes!) individual fibres.
- These make up a coherent bundle about 100 microns wide.
- Each give a view of about 8 arcseconds on the sky.
- The 10 bundles are arranged in a rough circle with numbers in this
order:
(clockwise from bottom) 144, 128, 112, 16, 32, 48, 64, 80, 96 with 160
in the middle.
|
The Autoguider commands and setup.
- Running on LPSS88
- Setting the SAOimage DS9 Zoom menu as: Invert Y and 0 deg.
- Display the region file /home/whtobs/af2_070312.reg, marking the fiducial
fibres.
uDAS "Acquisition" TV:
The camera looks at the output end
of the mobile probe fibre bundle.
The FOV is approximately 20" by 30". The mobile probe is used for acquiring,
especially standard stars on individual fibres and science fields if the
telescope pointing isn't very good. With good pointing you may try to acquire
science fields directly with the uDAS Autoguider.
On the mobile probe the best focus is at a position of about 83.20 mm. |
|
The TV commands and setup.
- Running on LPSS91.
- The suggested AG2 window to view the mobile probe fibre only is as follows:
"[160:550, 30:500]"
- Display the region file /home/whtobs/af2_ag2_20050609.reg,
where the gripper and rotator center are marked. Rotator centre (big circle)
and the gripper centre (small circle).
- Set the following in DS9 display. Zoom menu: ROTATE 90 and
FLIP Y.
Atmospheric Dispersion Corrector:
Check the PF atmospheric dispersion compensator is Enable. Looking the Instrument
Control Console, the ADC update: ENABLE.
Other screens and consoles:
- The AF2 TV screen is used to view the back illuminated fibres
when a new field is set up. You see the white spots defocussing as the
robot picks the fibre up. Please record every new setup with the VHS video.
- Configuration screen (on the side of AF2 TV) with the coloured
map of the fibres positions.
FIRST NIGHT.
Defining Rotator centre, Telesope Pointing and Gripper centre. |
| ROTATOR
CENTRE |
- Move to a 12 mag.star.
- Using the TV in continuos mode.
- Set a window in the TV.
- TO>AF2 VIEWSKY 0 0 (move the robot to the centre of the field.)
- ROT SKY 0.
- Measure the position x and y.
- Rotate 180º, then measure again.
- Calculate the rotator centre.
- Mark the rotator centre to further references.
|
| TELESCOPE POINTING. |
|
| APPERTURE OFFSET |
- Pointing to 12 star magnitude.
- AF2 VIEWSKY 0 0
- Center the star using the HANDSET in mode APOF.
- Store the position. USER>store ap 0.
|
| GRIPPER CENTRE |
- Field configuration setup done.
- Rotator in tracking with the demanded SKY PA.
- Goto a faint star (or some standard).
- Select a fiducial and offset the telescope to look the fiducial.
- Look the star position using the TV.
- TO>af2 offsettel SKYPA STARNAME FIDUCIALFIBERNUMBER (answear question;
'no' and then 'yes').
- TO> af2 park. (to see the star in the TV).
- TO>af2 viewobject FIDUCIALNUMBER (same fiducial as above).
- See where the star ends up and mark the position as GRIPPER CENTRE.
- Check using a science fiber.
- TO>af2 offsettel SKYPA STARNAME SCIENCEFIBRENUMBER. (answear question;
'no' and then 'yes').
|
Acquiring a Configured Field using the Fiducial Fibres
- Observer must have run AF2 SETUP for the field being acquired.
- The astrometry (star position information) used by the observer to
generate the configuration, must be accurate in order that several fiducial
stars coincide with fiducial fibre positions AT THE SAME TIME.
- The telescope pointing must be reasonably accurate (Calibrate Sky
r.m.s. <6 arcsec approx.). You can test the pointing on a bright
pointing grid star by putting the "moving probe" (camera on the fibre
positioning robot) in the center of the field using the command AF2
VIEWSKY 0 0. Moving this star onto the GRIPPER CENTER using handset
will tell you the pointing correction. If the pointing
star is close to the target field, you can apply this correction at
the field position and the fiducial stars should fall into the fields
of view of the fiducial fibre.
- Arotator offset must be applied. If the pointing is accurate (or the
above correction is applied) then fiducial stars close to the centre
of the field will show up in the field of view of the corresponding
fibres. If there are no fiducial stars close to the centre then you
may need to move the rotator a small amount (usually 0.2 to 0.5 degrees
in the negative direction) using the HANDSET. Wait for 3 readouts of
the acquisition camera to see the updated position. Adjust the rotator
position and XY position to bring the fiducial stars as close to the
fibre centres as possible.
|
|
If the above procedure fails to acquire fiducial stars, try the "long acquisition procedure", which uses the
larger field of the moving probe to first get a fiducial star close to the correct
position (the gripper center) and helps to find the rotator offset from scratch.
Acquisition (Long version, when pointing is not good enough to put fiducial
stars directly in fibres):
This assumes the sequence 'rot centre' -- 'calibrate' -- 'aperture offset' has
been executed successfully. You must have a field setup (AFSETUP) ready, the
telescope must track the field and the rotator must track the demanded sky PA
of the field.
- To acquire move the robot to a fiducial object close to the instrument centre
e.g.
TO> AF2 VIEWOBJECT 96
and centre the star with the XY HANDSET on the GRIPPER centre (how to define
- see next section) (last gripper centre found was x= 181 y= 221 with AG2
window set as mentioned above on 17/08/2005.
- Now move the mobile probe to another fiducial star at the edge of the field;
this time move the rotator to centre the star as good as possible, then park
the robot
TO > AF2 PARK and switch the power off
TO > AF2 SLEEP as now we proceed with the final tweak
using the Autoguider.
Acquiring Standard Star onto Science Fibre
Once the GRIPPER centre is defined the procedure is reduced to:
- TO> af2 offsettel SKYPA STARNAME SCIENCEFIBRENR
- Answer the questions (first 'yes' then 'yes')
- centre the star (TV) on what you now know is the GRIPPER centre
- TO> af2 park (or af2 sleep)
The detour via a fiducial fibre is to determine the gripper centre as the fiducial
fibres have the same optical centre as the science fibres; this way you cannot
miss the standard star.
Arcs:
When you take an arc, remember to offset the telescope away from the object
or close the primary mirror cover as otherwise you get the two signals superimposed.
Telescope @ acces park:
Move first to AZ 300, EL 15; Switch to ENG; Press AP2 on platform; Open gate;
Press AP3 on platform
NEW LINUX INTERFACE GUI
How to reset the micro in GHRIL: (Obsolete - there is no Miro
with the new linux system)
On the AF2 mimic, there is a status of the network connection. If this is
showing 'micro off the network', reboot the AF2 VME crate in GHRIL by turning
the plug for the crate off and then on after 10 seconds. Also check that the
network connection is properly plugged into the front of the crate.
If the ICL is giving problems after this consider killing and restarting the
AF2 task.
-------- © 2007 Isaac Newton Group of Telescopes --------
Last Updated: 2007 Juan Carlos Guerra