Procedures To Record Interaction Matrices
Prior To An AOB Run :

Here I present the procedures to set-up the AOB to take a set of interaction matrices. This is to be done prior to any AOB run.

Basic steps

1. Center the TAS with respect to the WFS
2. Adjust the WFS membrane mirror frequency to be at the resonance frequency (important)
3. Tune the phase of the lock-in detection to maximize the signal (very important)
4. Fine centering of the WFS and Taking of a set of interaction matrices (important)

Detailled procedures

1. Centering the TAS/WFS : Use the more recent estimate of the center field coordinates for the TAS and the WFS and fill the according fields in the form "Define set-up configuration" (or use the one already in this form, they are most probably good enough for the purpose of this first step). The TAS intensity should be set to 12. At this point, you may try to determine offset voltages. If it fails (it may fail because the alignment of the WFS and TAS is not good enough), you may want to apply old offset voltages if you have any.
2. and
3. : WFS Membrane mirror frequency adjustement and phase of the lock in detection : The most convenient way is to use the ISIT camera in pupil mode visualization.
   1. Turn the ISIT camera on.
   2. Bring up the maintenance user interface
   3. Open an hpterm and type
      aopch -pass 'tvm 2'
      to move the tv mirror in the pupil view configuration. At this point you should see the pupil/extra focal image on the camera display. If you don't, then probably :
      • The astar intensity is not high enough (should be 12)
      • The ISIT gain is not high enough
      • The TAS image does not fall in the WFS 4 arcsec field of view -> restart at step 1 up above.
   4. If the ISIT camera focus has not been done yet, do it. To focus the ISIT at this point :
      • Set the optical gain to 0 (no modulation on the membrane mirror) using the appropriate command in the menu "WFS" of the maintenance user interface.
      • Send a pal at the fifth floor to slide the ISIT back and forth while you are monitoring the pupil image on the ISIT display, in the warm and cozy environment of the control room (you should see the spiders when in focus)
      Once the ISIT is focused, you can proceed (don't forget to call you pal back) :
   5. Set the optical gain to 128 using the appropriate command in the menu "WFS" of the maintenance user interface. The image of the pupil is now an extra-intra focal image. It should be fuzzy and/or double/elongated.
   6. Using the "WFS tool" in the CFHT user interface, move the WFS in X or Y to have a clearly elongated extra focal image. There should be two fuzzy "rings" of light, not necessarily of the same diameter. The WFS translation should be large enough to have the image span approximately 15 centimeters from end to end on the ISIT display.
   7. At this point, adjust the focus of the WFS using the "WFS tool". When moving the focus, the rings will change size. A good focus is obtained when the two "rings" have the same diameter.
   8. We are now ready to adjust the WFS membrane mirror frequency. Turn the frequency knob (open the left panel in the WFS electronics) around until you have the largest span of the image on the ISIT display. You should be able to see a significant difference between the knob all the way down (or all the way up) and the optimum position you just determined. You are all set. Don't touch this knob before the end of the run except if there is a significant temperature change (>5 degrees C?), in which case you may have to do the whole procedure all over again.
   9. Now it's time to align the WFS / TAS. Use the WFS tool and move the WFS to have the two "rings" aligned as best as you can. You are all done with the ISIT camera. Remove the tv mirror by typing
      aopch -pass 'tvm 0'
      in a hpterm window.
   10. Next step is phase adjustement of the lock in detection. Go to the computer room and open the left panel of the WFS electronics. Left switch should be on "ON". Right switch should be down. The analog display is a measure of zero. The goal is to make it zero by changing the phase delay value with the wheels below the analog display. The phase value SHOULD be between 10-0 and 12-B. If not, there is a problem (but we never had any problem with this).
4. Fine centering and acquisition of the interaction matrices. You are now ready to start the procedure of acquiring a new set of interaction matrices. This is going to be done using the maintenance user interface. But first you need to fine center the alignment TAS/WFS and do accurate offset voltages. Why ? Because the goal here is to calibrate the relationship between the mirror deformations and the WFS measurements. To obtain a good calibration, you have to insure that the WFS is working in its linear domain. The way the calibration is done is that a positive and a negative voltage is applied to each electrode consecutively : For each electrode, the difference of the measurements corresponding to the positive mirror voltage and to the negative mirror voltage is done, and divided by the mirror voltage to get the WFS response to a unit voltage command for this electrode. This calibration is done for each electrodes + for the tip-tilt mirror, and for 12 characteristics optical gain than span over almost the entire range of optical gain (16 to 256). You have to insure that when the positive/negative voltage will be applied on the mirror, the measurements are not going off the linear range of the sensor. The voltages to be applied to both mirror have been determined in the lab, for each optical gain, so you won't have to worry about this. However, it is indispensable to insure that when the mirror is at rest (meaning flat, or in other word with the offset voltages applied), the WFS measurements are as close to zero as possible. This will insure that application of positive and negative voltages will not result in off-linear-range measurements. This is particularly sensitive for large values of the optical gain. Indeed, when at the largest optical gain value, the WFS is VERY sensitive to the slightest aberration. So that accurate offset voltages and accurate centering is necessary. To perform this, follow the steps below :
   1. I assume you have passed thru the previous steps of 1,2 and 3 described above, so that the TAS/WFS should be pretty well aligned.
   2. You should have a maintenance user interface up. If not, bring one up.
   3. You should have a "WFS tool" form up. If not, bring one up.
   4. Before determining a new set of offset voltages, you need to set the optical gain at which these offset voltages will be computed :
      • In the maintenance User interface, menu "Control", click on "Closing the loop"
      • In the subwindow "membrane mirror", click on "Use default" Another window pops up. This window allows you to select the optical gain in CLOSE LOOP. (not the same process to set the open loop optical gain). Set it to 128. Click on "Accept". This window disapears.
      • Click on "cancel" in the big close-loop set-up form.
   5. In the maintenance user interface menu "calibration", click on "Offset voltages determination". A window pops up. default parameters are ok. Click on accept. The offset voltages are determined. A big red button on the maintenance user interface will stay on till it's done (30 seconds max).
   6. The new offset voltages are determined and applied. You need now to center the WFS/TAS with a better accuracy. First set the OPEN LOOP optical gain to 128 as described in (2.d). Then use the 'WFS tool' with fine X/Y step values (max of 5) to fine tune the alignment. You can use both the realtime display in "modes" and try to cancel the mode number 2 and 3, or you can use the X-Y display in the WFS tool (center the black dot on the cross).
   7. You have done it for an optical gain of 128. Now let's go to a larger optical gain. Re-do (b) and select a optical gain value of 200. Pass thru (e) and (f) again. Once you have done this carefully enough, the mirror is flat even at the largest optical gain, and the TAS/WFS are aligned well enough not to induce a significant tilt aberration on the WFS. You are all set and can start the interaction matrix measurements.
   8. Click on the maintenance user interface "calibration" menu. In this menu, select "Interaction matrices ..." (first line). In this submenu, select "Interaction matrix measurements". A window pops up. Make sure "all" matrix are selected. Click on "accept". The process has started. It should last 40 minutes. You can check the behavior of the WFS/mirror using the real-time display. Make sure "temporal filter" is at 1. You can reduce the "time delay" (left scale) a little (not to its minimum value tough because in this case it monopolizes the CPU for the display).
   9. Once it is finished, just answer the question. You WANT to save the new set of interaction matrices. The suggested name is our standard, don't change it.
   10. If you are happy about these matrices, go to the directory where they have been saved and make a copy in to aosave.ao (this last file is the one that is read and loaded by default into the RTC when the ao servers are started).