Minutes of GBT Commissioning Meeting Friday, 13 December 2002, 10 AM EST 1. Spectral Baseline work From Dana: 12/9/02 DCR gain stability tests show that short term gain instability introduced between the ODM2 detector and the OR 2 outputs, but the 8-minute gain cycle is coming mostly from converter rack modules'response to equipment room temperature cycling. The analog filter rack modules' gain responds to the equipment room temperature to a lesser extent. The Equipment room temperature cycles by 2F p-p, and causes total power cycling of about 0.5% p-p. 12/10-11/02 A new test noise source with "cal" switching capability was used to inject noise into the input of the analog filter rack to test stability of the spectrometer system from the AFR inputs through to the spectrometer data outputs. Several 200MHz and 800MHz configurations were tested. 200-MHz BW tests showed very good stability and (on-off)/off spectrum flatness to better than about 0.12% RMS for off-to-on time differences up to 20 minutes using SF1-8 to HSS0-7. 800MHz mode tests exhibited a narrow birdie at 340MHz, and good (on-off)/off flatness (<0.05% RMS) for off-to-on time differences up to 6 minutes. Longer differences exhibited growing baseline structure. The source of the 340MHz birdie has not yet been identified, but may be in the test noise source. 12/10/02 DCR monitoring of total power from the IF noise source at the outputs of the IF rack and the analog filter rack showed that fiber links 2 and 5 have considerably poorer short term gain stability than fiber links 1, 3, 6, and 8. The direct modulated laser in fiber 8 appeared roughly as stable as the four good links. No definite evidence for azimuth or elevation dependence of fiber link insertion loss was found to the level of a few tenths of a percent. Total power variations of this order was seen on time scales of minutes, but part of this is probably due to the converter modules. 12/11/02 Survey of 2.4 MHz ripple in optical fiber modules 1, 2, 3, 5, 6, and 8 at an IF frequency of 1500, 3000, and 6000 MHz using the Spectrometer with an 800 MHz BW. Longer (30 min) integrations at 3000 MHz were made to characterize an intermittent ripple (~40-70 MHz) thought to reside in the IF system. Results will be forthcoming. 12/12/02 Inspection of new optical modulator which was installed during the day (12/12/02). Preliminary results indicate that the ripple is down by 20 db. 2. PTCS project work From Richard: Little specific to report this week. Kim and myself continue to discuss the best way to present the high-level system design. Amy, Jason and J.D. have started work on the Active Surface Improvements Project Charter, and I will be reviewing the first draft with them this afternoon. Kim, Ray and Nicole met to discuss the EMS, but unfortunately Ray is now out for shoulder surgery. I have reserved a sequence of commissioning slots and am working on a detailed plan for these. Jim is refining scientific requirements. The metrology group are continuing to work on the LRF calibrations, but are having some problems measuring the mirror and back prism offset from the mirror axis to better than about 50 microns. This will be facilitated by some additional fixtures which will be available mid January. They are also looking into our options for replacing the replicated mirrors while fine tuning the calibration of all instruments, experimentally checking the glass offset calculation vs a hollow retroreflector. Work at Triad on the laser shelters is proceeding week, the first article should be ready by early February. 3. Az Track work Bob reported: - bolt loading test pieces in fabrication - anti fretting coatings being researched; possible tests in January. - Zinc test shim material ordered; installation planned in January. - Draft plan for AZ track work should be ready next week - calculations of a "beam on an elastic foundation", independent of FE modelling, have produced some interesting results. - Flex plate FE work will probably be contracted out. 4. K-band and other commissioning From Ron: Commissioning activities this week were almost exclusively on training Harvey Liszt and Barry Turner on the GBT. Training during the days consisted of discussions about the GBT hardware, software, and observing strategies. Three nights were spent with the GBT on learning how to use the system and in checking out their observing proposals. Harvey's observing used the lower section of the K-band receiver while Barry's used the upper section. They seem to be very happy with the quality of their data but have numerous suggestions and criticisms concerning the GBT and Aips++ software. I expect that Barry and Harvey will provide more details when they write up their experiences. On Tuesday, we had a flurry of K-band activities which included: o recabling the IF Rack so as to allow multi-beam, dual polarization observing even when an one or more IF Rack optical drivers have been removed. o establishing that the K-band system requires 3 to 6 dB more gain in order to optimally drive the IF Rack optical drivers. o altering the M&C system so that the software would reflect the removal last summer of the receiver's L.O. multiplier. In other commissioning activities, Anish spent two long observing sessions on polarization tests. And, during a lull in training on Monday, I was able to determine that the beam offsets for the upper K-band receiver are, to within the measurement error, as predicted from the lower K-band receiver measurements. Activities for the coming week will be few since most of the telescope's time is scheduled for non-commissioning activities. We have scheduled two short sessions for spectrometer mode checkouts by Frank and two six-hour sessions for polarization tests by Anish. Frank will also be getting Glen involved in supporting radar observations. At the end of next week, Al Wootten is scheduled to join us for another training session. 5. Project and observing program scheduling: From John: The main progress this week was that the new consolidated calendar was released for general use, and it replaced the old calendar on the Web pages. We are proceeding with the baseline investigations, with that project receiving all the time the participants request. Spectrometer checkout time likewise is available. PTCS time is being scheduled later in the month and in January. K Band commissioning and astronomy is taking place when the weather and hardware cooperates. Some RFI suppression work is being introduced into the schedule on Monday in support of pulsar observations being corrupted by 28.75 and 60*N Hz interference. From Carl: Last week: ======= Completed GBT02C-03 (Pulsar observations at 800MHz & L band) Continued observations on GBT02A-062, GBT02B-019, GBT02C-034 (long term pulsar monitoring) GBT02A-049 (short term pulsar monitoring) Next Week (Sat-Fri): =============== GBT02C-035 daily radar Observations of Titan to the end of the month GBT02A-052, GBT02A-012 (long term pulsar monitoring) GBT02B-010 NH3 obs at Ku band BT064 - VLBA observations December ======= GBT shutdown from Tue Dec 24 at 7pm to Thu Dec 26 at 7am January ===== Astronomy scheduled = 35 hours Maintenance scheduled = 142 hours Tests scheduled = 300 hours Un scheduled commissioning = 265 hours February ====== Will begin scheudling astronomy next week. 6. Software status: see SDD weekly report at: http://software.gb.nrao.edu 7. Spectrometer status From Rich: This week work on the spectrometer concentrated on the verification of M&C version 3.9. Approximately two full days were devoted to this task. On the positive side, the multibank enhancements for robustness seem to work fine. We experienced no crashes. We did note that, on occasion, data for Banks B, C and D is incomplete. There is still some vestige of the Manager OFF/ON problem. The initial version of the test program did not treat the 9-level partial products correctly. This was corrected quickly by the M&C group and verified by us the following day. Found that, for 200 MHz bandwidth, two-phase data is not properly handled. Noise did not scale with root-time in our tests. This works properly for all other bandwidths, as it did in the previous incarnation of software. Found that mode 2N4-4A-12-9 (12.5 MHz, 9-level) has very small scale structure in the acf. The corresponding 50 MHz mode does not. We may need some help from Ray on this one! The spare System Monitor was upgraded to the latest version of firmware. The spare VME interface board was upgraded to the latest version of PAL. Both were tested in the system. Near term plans include: 1 - perform tests to see if the acf structure seen in mode 2N4-4A-12-9 is present in other 12.5 MHz modes. 2 - support one more iteration of spigot tests. We plan to take advantage of Ray's visit to get some help on the 12.5 MHz problem. 8. AIPS++/Dish status Jim reported: - next release of aips++ will coincide with next M&C release. - Jim has been working with Gaussian fitter to allow ranges to be specified. - Significant support provided to Harvey and Barry this week; Jim will travel to CV next week. 9. Any other business None Richard Prestage 13th December 2002