Simple Instructions for Running the PSF Subsystem for Adjusting Collimation

General PSF Description (from observer's point of view)

These instructions have been adapted from an email from D. Thompson. See the PSF User Manual (T481s00062) by C. Biddick for precise details.

On the PSF sub-GUIs for each mirror there are two relevant fields to look at. The "global offsets" in the lower left are where you can add manual offsets to the physical mirror position. They are an offset to all of the other inputs, but not relative to the current position in the traditional sense. If you enter, for example, -0.5mm on Z and click "update", the mirror will move down by 0.5mm. Clicking the "update" again will not change anything because the half-millimeter offset is already included. If you change the Z offset to -0.6 and click "update" then the mirror will move down by only 0.1 mm. If you clear the global offsets and update, then the mirror would move up 0.6mm.

The other relevant field is the "total collimation" at the top right. Those numbers represent the sum of all of the other contributions (global offsets, collimation model, offsets sent by GCS active optics, etc.). These are the numbers that are recorded in the headers of the images. They represent the actual position of the mirror relative to the (arbitrarily assigned midpoint of available range called "magic") defined coordinate zero points. So, for example, the Z position of the primary mirror can range from about +2.6 to -2.6 (millimeters). The "total collimation" numbers turn red on the PSF GUI if you try to exceed the available range. They turn yellow if the mirror is not in the present "total collimation" position.

If you clear the global offsets and then update, you are not reverting to the collimation lookup table. For that you need to also "Clear Active Optics" about center left on the Primary Mirror sub-GUI and upper left on the SecondayMirror sub-GUI. If there were any global offsets in effect when you made the collimation model then they should be cleared from the global offsets and you should revert to the collimation model ("Clear Active Optics", "clear global offsets" ("Clear Inputs" and "Update"), and "initialize" the new collimation model) before taking more data. Note that some collimation models do include predefined global offsets.

Launching a PSF GUI

• ssh obs2 (or whichever mountain machine can launch GUIs)

• PSFGUI left & (or right)

• click on "Primary" or "Secondary" or "Tertiiary" or "Control" to get the various sub-GUIs of PSF.

Basic Training to move Primary Mirror around with PSF (from TMS point of view)

Here's the basic training for how to operate the PSF subsystem of TCS to move the primary mirror. (in a style for TMS folks)

Login to mountain control room workstation (ssh to obs1, obs2, robs, etc).

Launch a terminal window: pull down root menu from GNOME or MATE blue background and select "open in terminal".

Launch a PSFGUI: in the terminal that pops up type: PSFGUI right &

Note: This PSFGUI has the live control of the telescope optics, so don't practice with it during the night.

On the PSFGUI, click the button "Primary" to bring up the sub-GUI for M1 named "Right PSF Primary Mirror Active Optics and Collimation Control"

On the upper right of the Primary sub-GUI is the current commanded "Position" vector (X,Y,Z,RX,RY) named "Total Collimation".

Before you start any measurement sequence, click the "Collimate" button to position the primary at the current lookup table position (function of elevation and temperature).

Just below "Total Collimation" is the line "Global Offsets" with white entry boxes to type in. To move -1 mm in X, type -1 in the X box, and hit carriage return to make it turn from red to black (only black numbers are accepted). To apply the current vector of Global Offset click the "Update" button. You will see Total Collimation turn yellow, and then back to green when the mirror is at the new position.

Global Offsets are "relative absolute". They are RELATIVE in the sense that they are added into the "Total Position" command, but they are ABSOLUTE in the sense that doing X=-1 twice will leave you at X=-1 (not X=-2).

Controlling Active Optics Behaviour

See this detailed description of how to control ActiveOpticsModes with the Control sub-GUI.

PSFCLIENT command line interface - new with Build BP3

(based on a Chris Biddick email on 21-SEP-2010)

You can set global offsets, instrument_offsets, pointing tip/tilts, active optics offsets, Zernike offsets, shell offload, and show position from the command line. Entering an incomplete command will give a short help message. The command string may be abbreviated. 'rel' will add to the existing values, 'abs' replaces the existing values. (Note that adjust_pointing is always 'abs'.) Case is ignored. Whitespace separates tokens. Units are mm and arc-seconds.

Usage: psfclient side command optional_parameters
       side is left, right, or both
       command is global_offsets, instrument_offsets, pointing_offsets, manual_pointing_offsets, active_optics_offsets, shell_offload
                  wave_global_offsets, wave_instrument_offsets, wave_active_optics_offsets, wave_opd_offsets, wave_secondary_offload
                  show_position

global_offsets/instrument_offsets usage:
psfclient side [global_offsets|instrument_offsets] [M1|M2|M1M2] [rel|abs] x y z rx ry rz
psfclient side [global_offsets|instrument_offsets] M3 [rel|abs] tip tilt piston rz

pointing_offsets usage:
psfclient side [manual_]pointing_offsets [M1|M2|M3|M1M2|DEFAULT] tip tilt coll_flag
coll_flag defaults to true; set to 'false' to not collimate

active_optics_offsets/shell_offload usage:
psfclient side [active_optics_offsets|shell_offload] [rel|abs] x y z rx ry rz

wave_global_offsets/wave_instrument_offsets/wave_active_optics_offsets/wave_opd_offsets/wave_secondary_offload usage:
psfclient side [wave_global_offsets|wave_instrument_offsets|wave_active_optics_offsets|wave_opd_offsets|wave_secondary_offload] [M1|M2|M1M2] [rel|abs] Z1...

show_position usage:
psfclient side show_position [M1|M2|M3]

By-Eye Active Optics

See this guide to ByEyeAO if you want to make manual active collimation or focus adjustments based on the appearance of the images.

Directions of Motion for all primary and secondary mirrors

As understood by TCS subsystems PSF, OSS and PMC, the following are the sign conventions for the motions of the LBT primary and secondary mirrors.

+Z is up the optical axis to the sky when telescope is zenith pointing
+Y is toward zenith when telescope is horizon pointing
+X is toward the left wall (left/right as defined when you are standing at the back wall looking at the telescope)

If you want to know more than that, here is an excerpt of email exchange between J. Hill and T. O'Brien.

I confirm that what you report is true. I can also report that a +Z move raises the secondary up toward the sky.

The useful thing that you guys could do is to look at the sketches Ross made while we were tilting the secondary, and confirm that the alpha, beta angles combine with X,Y,Z to make a proper right-handed coordinate system.

Now I'll risk confusing you by telling you that the above/below reported hexapod coordinates are rotated 180 degrees (around the optical axis) with respect to our standard coordinate system for the primary mirrors. So last night we modified the OSS (Optical Support Sub-system) software to reverse the signs of X,Y,alpha,beta (nothing changed in UMAC land). Now +Y is to the sky when telescope is horizon-pointing, and +X (on left mirror) is away from the C-ring. So from the astronomer's point-of-view, the position with the secondary centered above the direct Gregorian rotator is:

X=-5.5 mm Y=-3.5 mm Z=0 mm RX=+75 arcsec RY=280 arcsec

We ran out of Y position travel on M1 to collimate the telescope with M2 at the above position. So we have decentered the secondary in Y so its permanent coordinates for the next two weeks are:

X=-5.5 mm Y=-2.5 mm Z=0 RX=+75 arcsec RY=+280 arcsec

I'm building a full primary collimation model for the secondary at that position. When Andrew gets back from down under, we'll measure the F/15 aberrations and adjust the secondary to correct those.

The hexapod operation has been very smooth. Once or twice a night, the OSS software gets confused talking to the UMAC and we have to restart OSS.

Ciao, John

On Tue, Apr 15, 2008 at 11:55:14AM -0400 or thereabouts, Thomas P. O'Brien wrote:
> I'll attempt to communicate to you the sign conventions on the Hexapod
> motions.  These should be verified.
>
> Telescope Zenith Pointing
> Viewing SX  Secondary Mirror from behind primary mirror at Direct Gregorian
> Observer has right shoulder toward "C" ring and is looking up at secondary
>
> Initial postion of target annulus on Secondary mirror is at about 10'o'clock
>
> Correction moves of Hexapod required to center target annulus on Secondary
> mirror on alignment telescope crosshairs are:
> +X = 5.5mm  Hexapod command moves secondary toward "C" ring.  (toward the right)
> +Y = 3.5mm  Hexapod command moves secondary toward observer's feet ( down)
>
> These two moves center the secondary mirror with telescope zenith pointing.

Directions of Coma-Free Pointing Tip-Tilt

M1M2 (mode 1)

Here's a little table of which way things move when you apply coma-free pointing corrections to M1M2 together (which are also binodal-astigmatism-free). (from IT 5030) For so-called "Mode 1 Pointing" the primary and secondary mirrors rotate together as a telescope, with no relative collimation changes between them.

TIP=0 TILT=+10 arcsec (which would be compensated by deltaCA=-10) moves:
M1 dX=-0.148 mm dRY=-10 arcsec
M2 dX=-0.665 mm dRY=-10 arcsec

TIP=+10 arcsec TILT=0 (which would be compensated by deltaIE=-10) moves:
M1 dY=-0.148 dRX=+10
M2 dY=-0.665 dRX=+10

M2 (alone)

Here's a little table of which way things move when you apply coma-free pointing corrections to M2 alone (which are NOT binodal-astigmatism-free).

TIP=0 TILT=+10 arcsec (which would be compensated by deltaCA=-10) moves:
M2 dX=-0.511 mm dRY=-9.82 arcsec

TIP=+10 arcsec TILT=0 (which would be compensated by deltaIE=-10) moves:
M2 dY=-0.511 dmm RX=+9.82 arcsec

Directions of Motion for Pointing-Free Coma

When sending pointing-free coma to the M1 hardpoints:
-1000 nm Z7 moves dY=-0.8536 mm dRX=-9.17 arcsec
-1000 nm Z8 moves dX=-0.8536 mm dRY=+9.17 arcsec

When sending pointing-free coma to the M2 hexapod:
-1000 nm Z7 moves dY=+0.932 mm dRX=-97.1 arcsec
-1000 nm Z8 moves dX=+0.932 mm dRY=+97.1 arcsec

Directions of Motion for Tertiary Mirrors

For both tertiary mirrors, a more positive Selector Angle rotates the tertiary CCW as seen from above.

Negative Tip makes the tertiary mirror more horizontal (at zenith).

For the SX tertiary mirror, positive Tilt is the opposite direction as increasing selector angle.

Pointing Sensitivity for Optics Motions

The following pointing sensitivity analysis was done by Dave Ashby in March 2013.

The pointing sensitivity to various mirror displacements are as follows:

  M1-X/Y     :  21.0 as/mm
  M1-Rx/Ry :    1.98 as/as
  M2-X/Y     :  22.6 as/mm
  M2-Rx/Ry :    0.219 as/as
  M3-Z        :    2.36 as/mm
  M3-Rx      :    0.0605 as/as
  M3-Ry      :    0.0856 as/as

-- JohnHill - 14 Apr 2008