MMTAO April 2005 Run Douglas Miller The MMTAO Spring run from April 22nd to May 1st was quite a success. We had some poor weather at the beginning of the run as well as some technical difficulties but during last five nights of the run the AO system, ARIES and the MMT all worked very well and we gathered a lot of astronomy and engineering data. I believe that the last two nights of observing were an indication that the AO system and ARIES are turning the corner from instruments being commissioned to being a reliable system for acquiring high quality scientific data in a very efficient manner. Overview of the Run ------------------- Friday - Sunday 22/4/05 - 24/4/05 As scheduled, Friday night was a secondary change night so we did not open. We tested the new electronics built by MMT personnel to latch the hexapod brakes if a collision was detected by the tape switch. We found it worked perfectly and are now confident that any collisions will be detected and the hexapod properly stopped. We also spilled a small amount of cooling liquid on the DM (to Hexapod) interface plate. We estimated that only a few cc's of water/methonal made its way into the DM, pass an actuator and into the gap between the reference body and the thin shell. This water in the gap was seen because about a third of the shell would not release from the reference body. Saturday and Sunday nights were lost to bad weather. Also on Saturday, with DM pointed downward in the clean room, we removed all forces and slowly the shell moved down onto the four retaining clips. After a couple hours of drying, the mirror was tested and operated normally. We had one capacitive sensor that was flaky for the first hour or so of operation, but it settled down and we have not seen any lasting effects of this water in the gap. Monday 25/4/05 In the first part of the evening we were able to open the dome, collimate and close the loop. In the last run we were not able to collimation the telescope by moving our f15 secondary because the range of motion of the hexapod was limited by our secondary so we had to tilt the primary mirror in order to collimate. Since the last run we found that one of the electronics boxes of our DM was not positioned properly due to interference with the other two electronics crates. We minimized this interference and were able to get all the crates positioned properly to make the size of the DM at the position of the hub stiffening ring (the position of collision) 5+ mm smaller. This was the cause of our collimation problems last run and had more than a mm of extra space for collimation this run. At 10:00 the humidity increased to 90%+ and we closed the dome. At 2:00 the atmosphere dried enough that we opened again. We found the our closed loop was not stable and we would get high forces on our DM after about 30 seconds of closed loop operation. After a couple hours of debugging we determined that a software error was causing the feed-forward matrix not to be loaded. This matrix, and the related calculation, is needed to allow for round-off errors in our reconstruction as well as to correct for modes that our reconstructor is not able to correct (eg modes about zernieke number 56). This problem was fixed Tuesday during the day and we were ready to close the loop again on Wednesday evening. Tuesday - Wednesday 26/4/05 - 27/4/05 All systems worked and both night we closed the loop soon after sunset. On Wednesday we spent some time working on AO/ARIES communication in order to automate positioning of the science object on ARIES and to allow automated dither patterns. Most problems were solved. The observers, Mike Meyer and Xaiohui Fan, were able to able to take data in a fairly efficient manner with some interaction with AO and ARIES personnel. Thursday 28/4/05 The first hour we spent setting up a finder telescope on the side of the MMT telescope so we could search for satellites. After about 20 minutes of searching we were about to locate our satellite and close the loop. However, wind made running the AO system difficult and at about 11:30 we had to close the dome due to high humidity. Friday 29/4/05 The weather finally cleared up and in the first half of the evening we were able to lock onto and image two satellites. (see the attached jpg image **NOTE** do not distribute this image. If you want to show it to people other than CAAO people, please talk to Phil Hinz) The second half of the night was spent performing engineering tasks. Many of these task have been on our engineering list for three runs and have not been completed because of poor weather. We were able to take a full set of reconstuctor on the sky data for Guido and Vidhya. The plan is that a new and improved reconstructor matrix will be used during our next run in June. We also fixed a couple items in the AO/ARIES communication, align the Indigo engineering camera and take some 60 Hz data for strehl measurement, and we tested new routines to offload tip/tilt/focus building up on our DM. This was our first full night of engineering in three runs, and we accomplished may several of the high priority item. We now have a lot of data to analyze in order to improve our system, and prepare for our engineering nights in June. Saturday - Sunday 30/4/05 - 1/5/05 These two night were, by far, our most productive science nights with the AO system. The AO/ARIES communication worked great so that after an initial setup of about 20 minutes when a new object is acquired, the astronomers, Beth Biller and Eric Nielsen, were able to position the object on the ARIES chip, in closed AO loop, from the ARIES user interface and execute a dither pattern, in closed loop, to acquire data. No verbal interaction or intervention from AO, ARIES or MMT personnel was need for 1-2 hour stretches when data was being taken on an object. During this time I would estimate 90% of the time was spent collecting useful science data. In addition, Beth and Eric let us do a bit of engineering during several 15 minute slots between science object. We were able to create speckles at specific positions next to the star being observed by putting a sine wave on the DM via slope offsets in the reconstructor. (see attached jpg image). Craig was also able to test the tip/tilt sensor in ARIES. ----------------- With poor weather the last two AO runs, we finally had five nights of good weather, an operating AO system and some good luck. This last run in April was by far our most successful in moving our system toward an efficient, user friendly system for taking high quality science data. Doug