MODS Controller Upgrade – Instrument Availability Delayed
update: Mar 1st
The MODS instruments underwent controller upgrades this summer, replacing the legacy head electronics and a set of aging computers. These upgrades were carried out to ensure the long-term reliability, robustness, and stability of the instruments. Additionally, with the new controllers, readout overheads have been significantly reduced.
Both MODS are mounted on the telescope; they are not yet released for science although much progress has been made in the past month.
The focal station checkout was completed during ECD time on Feb 1 UT. On this night, we also tested both MODS in binocular operation, executing MOS, longslit and spectrophotometric standard star alignments, a couple of spectroscopic observations and a dithered imaging series. The system ran well, although it uncovered issues in timing and in the performance of the MODS2 Red IMCS and these, in addition to the known readout problems, held back release for shared-risk observing.
Since then, the timing issues have been addressed and the problem with the MODS2 Red IMCS was traced to a wiring error which is being temporarily addressed in software until summer shutdown, when a hardware solution can be implemented. The only remaining issues are with the controller configurations: for all 4 detectors, the last column read out in each quadrant has systematically high or low counts and, consequently, the central two columns in the resultant merged image are too high or too low; there are some charge transfer effects; and requested exposure times longer than ~17 minutes are not possible, although since the overheads on full-frame unbinned data are about 5 times lower than before, the recommendation to take multiple 15-minute exposures will be a viable workaround until this problem can be resolved.
Some changes which will affect the planning and execution of observing programs are noted here, while updates to the sciops webpages are being made. There have been changes to the procedures for starting up MODS and running scripts, but the user interface looks the same and, after a short time, operation of the upgraded MODS will seem familiar to users.
It is important to note that the raw data are stored as multi-extension FITS files, where extensions 1-4 hold the data from each quadrant, extension 5 is a FITS Bin Table containing the Archon controller status at the start of exposure, and extension 6 holds the merged image. Operational software (modsDisp, modsAlign) has been updated to use extension 6. A new header keyword CCDCTRLR was introduced to indicate that the data have been taken with the upgraded system:
CCDCTRLR= ‘Archon ‘ / STA Archon CCD controller
Two changes which affect observation planning are that:
(1) All acquisition images will be taken using the 3Kx3K ROI, which covers the full 6′ x 6’ field of view. Neither OT programs nor generated scripts need to be changed, however, since the OT will force the 3Kx3K ROI when generating acquisition scripts and the “acqMODS” script engine will overwrite the ROI in the script with 3Kx3K. This change was made because the detector overheads for 1Kx1K readouts are short enough, 7 sec, that there is a tendency to run into race conditions, and it is safer to use the 3Kx3K ROI, which has only a slightly longer overhead, 12 sec.
(2) Individual exposures must be <= 15 minutes, as mentioned earlier.
