This file contains a summary of my progress at the IOTA site during my visit April 6 to April 14. In addition, a list of external memos, to be hung off the website, is described. Memos for website: ------------------ 1. DAQ_Documentation. I have written a first draft of a rudimentary users manual for the IDL-based Data acquisition. JP Berger currently is working on the second draft and I hope to post this soon. 2. Details on the configuration of the second analog output board. 3. Memo on how to convert stroke to OPD for the data taken so far. There have been some simple mistakes which will lead to confusion if one naively accepts header/log information. 4. Summary of my trip to mt hopkins Summary of my Trip to Mt Hopkins: --------------------------------- --------------------------------- A. PICNIC Camera 1. For some reason, using the PICNIC ADC card causes the ADC to be out of range when the pixel is fully charged (0 ADU after reset). Swapped with NIMCOS ADC card and it works ok now. 2. We note that the light still gets in the camera even when the blocking filter is in place. Is this good/normal? 3. We also note the background is very high on timing patterns for large number of reads/loops. This DOES limit our capability to go faint under stable seeing conditions. ** We need a cold baffle matched to the IONIC optics. ** 4. We still need somebody to do basic camera tests -- tint, background, read noise, and gain for all potentially usefeul timing patterns. ------------------------------------ B. Delay System/PZ Scanners 1. The internal delay for DL1 changed by 1.7cm because the dichroic was removed and replaced in a different location for the FLUOR group. We should decide as a group if the best way to remove the dichroics is to individually unscrew the mounts, or to remove the whole plate. Its not clear which is best. 2. SD2 is limited to move between 0 and 100 cm. This is less than 1/2 the nominal range, due to problems with track, airpads, or both(?). I hope we can get the replacement pads soon since this makes observing difficult. 3. In addition SD2 has been oscillating fairly often. From what Marc says, we probably need at least one fresh pad for SD2. SD1 behaved pretty well when I was there. 4. When operating with 1 loop 4 reads (0.64 ms integration time / pixel), the piezo scanner in Delay 2 oscillates/rings a bit at the beginning of the scanning ramp -- even for pretty generous time for 'flyback.' For instance -- for 256 samples in the ramp and 100 samples in the 'flyback', there is still detectable ringing in the interferograms. We had to up this to 140 samples to eliminate it. There is no sign of this in the scanner for SD1 which is scanning doulbe the stroke of LD2. Hence, there is seomthing different/wrong with SD2. We should try to either stiffen the mount more, or put some thought into a more clever sawtooth pattern that eliminates this problem. 5. I wrote memo on how to convert stroke to OPD given the recorded information. There have been some mistakes and I tried to correct them in the code (something was being divided by cos AOI instead of multipled). However -- we should improve DAQ code to have a different conversion factor for each piezo scanner since the AOI are slightly different for each beam (for IONIC). -------------------------------- C. Auto-Alignment Progress 1. We have a shmem interface to move the picomotors. I recommend that somebody write an OT interface -- it should be trivial. All this runs on iota-ppc0 and is part of the startup code. The directory is ~iota/iotamc/vxworks/picomotors 2. I have an IDL routine, accessible from the main menu of the IDL DAQ code, which allow us to automatically align the fibers with the startracker pointing centers by using picomotors on the relay flats between the dichroics and OAPs. See DAQ_Documentation. 3. Marc wired the interface for the new analog output board (8 voltage outputs). JPB and I wrote a short document describing the configuration (see web). I calibrated the outputs, and wrote the vme code which a) sets the voltages (copying brewer's code for the first board), and b) a vme daemon which allows us to change the voltages on any of the outputs using only shmem. Again, somebody should write a little OT tool for this since I've done the hard work already with the vme daemon. This code is in ~iota/iotamc/vxworks/analog_outputs2 and is already automatically loaded on iota-ppc0 upon vxworks-start. ----------------------------------- D. Star Tracker 1. I added an autothresholding feature. When you take a set of dark frames, it automatically sets the threshold. Very nice feature I think. 2. I changed the default search step from 0 to 128 for the piezo search option. I tried this out for the first time (successfully) this trip, and it works really well! -- a few small improvements could be made, like automatically moving the piezo offsets to the roll/tilt offsets once a star is detected (pete?) ----------------------------------- E. Data Acquisition -- VME side 1. Major Major improvement in data headers. See DAQ Documentation for 'full' description of all fields. I want to add a few more but later, like voltages on the fiber piezos. 2. Changed code to operate in CMD_SCAN mode until commanded to CMD_REST. This means DAQ is always reading out frames, but not recording them. In the past, one had to set a variable 'nscans' and then command to CMD_SCAN, thus readout out 'nscans' scans and then rest. This was not convenient since we want to always be monitoring the fringes even if we aren't taking data -- same model as startracker. This makes the system more responsive. ------------------------------------- F. Data Acquisition -- IDL side 1. I've grouped _all_ the IDL helper routines under a single main menu; see DAQ_Documentation. I list here only important changes or major upgrades. Here is a screenshot of the main menu to give you some idea: ------- ------- -- W E L C O M E T O T H E O B S E R V I N G T O O L -- 1. Align pixels on Picnic Camera (** OPTIMIZE FLUX ON CAMERA **) 2. Setup Observation Parameters (** SET REAOUT AND SCAN PARAMETERS **) 3. Display raw scans (** SEE RAW DATA **) 4. Find fringes (** TAKE DATA **) 5. Fiber Explorer (** ALIGN FIBER TO OPTIMIZE COUPLING ON STAR**) 6. Align Fibers on Startrackers using Picomotors 7. IOTA3GUI (** USE TO DETERMINE OPTIMAL LONG DELAY POSITIONS **) ------ ------ 2. Starting combine some of the vxcom*.so code. Now using vxcom_all.so instead of vxcom.so which combines analog output2 control iwth DAQ. 3. Ran into memory limitations of the bus-to-bus adaptor when using IDL to grab more than 1 shmem block at a time. This is REALLY slowing things down on the IDL interface, and even causes IDL crashes due to a out-of-memory fault by the bus-to-bus adaptor. I am working with Gary/Kamal to find a solution -- such as using the 'unmap' function to free resources. Crashes ONLY seem to happen in the 'fiber explorer' function, not during normal data acquisition (thankfully!). 4. Tons of small improvements, like help menus (hit ), and auto-defaults (for instance-- when prompted you can always hit return to use a default answer, etc.). Plots have all been improved and more diagnostic messages are displayed. Automatic locking and unlocking of fringe-tracker, etc. Change observing parameters on the fly, etc. 5. matrix. This great key lets you automatically take a set of shutter data. Tel a Only. Tel B Only. Tel C Only. Background. See DAQ Documentation. 6. Fiber Explorer. This program lets you move the fiber around and map out the coupling efficiency in a 2-dimensional image display. Very useful -- although currently slow and crashes (due tothe bus-to-bus adaptor issues!). 7. IDL will not only create the appropriate sub-directory based on UT date (as before), but now will look inside and number the first IOTA# data file high enough so that we never overwrite existing data. This is nice. 8. Read DAQ_Documentation before coming to IOTA, since it goes through lots more details and important new features.