GBT Operations & Commissioning Meeting 3 October 2003 AGENDA 1. Az Track Status & other telescope news -- Bob A. 2. Observing news -- Ron 3. Spectrometer status -- Rich 4. Spectral Baseline, Front-end, and IF work -- Roger 5. Software status -- Nicole 6. Schedule -- Carl 7. Project scheduling -- John 8. Any other business REPORTS 1. Az Track Status & other telescope news Azimuth track: No developments this week. Materials evaluations and analysis continues. Structural inspections and repairs: Work in progress on the walkway repairs. Disposition of elevation shaft defects still not finalized. 2. Observing news A significant fraction of the observing time over the last week went to proposal checkouts, spigot card testing, and system tests. Nevertheless, we serviced about seven observing proposals. Highlights include a substantially successful run with the Spigot card which Karen and Frank will summarize. Most of the problems during the observing time were caused by human errors. The Spectrometer did not have a single failure even though it was used for long periods for many of the tests. The Spectral Processor was not in much use but still failed on occasion. I still cannot say whether the failures of the SP are more or less with the new cables. The antenna went into "Safe Hold Mode" four times, much more than it typically does. Some of the operator's reports suggest an intermittent brake or other hardware issue may be the cause. 'Human errors' include: The noise diode was left firing for most of a pulsar run, the spigot card test source was left in place during Karen's spectral-line with the Spectrometer. The hardware "RAID" failure on the spigot card will be avoided in future runs since the RAID's configuration was not optimal. We are following the same K-band commissioning plan as last year. The data we have collected so far will be used to derive Tcal(freq), Tsys(freq), and feed "Suck outs" for all polarizations for all four receivers. We have data for a gain curve at 25 GHz, a substantially higher frequency than our old 20 GHz curve. We have collected stability data for the lower-band receivers. We still need stability tests of the upper band receivers. We plan on taking 'typical' spectral-line data using NOD, OffOn, frequency-switching, etc. These observations will allow us to test various calibration methods, to determine how well the system integrates down, and to provide a basis for the observing methods we recommend to observers. -- RJM 3. Spectrometer status Rich and Holly were at the ALMA Correlator CDR this week and so there is no Spectrometer activity to report. 4. Spectral Baseline, Front-end, and IF work We have now had problems with two of the Q-band cryo LNAs. Gary has found wiring errors on some of the PCBs that provide transient protection to the LNAs. The problems are being corrected. The Ku-band "waffle-iron" thermal gap assemblies were tested with the network analyzer and the data looks good. The receiver is being reassembled to be cooled. Next week, I plan to repeat tests done earlier with the spectrometer and the old thermal gap design. All of the old optical fiber modulators with 2.4MHz ripple are now out of the system. Two more ripple-compensated optical modulators were installed this week by Steve and Bill (in channels 2 and 4), but final adjustments and engineering checks are not quite completed. When these are released for use there will be seven of the new modulators in place; channel 8 still has a direct-modulated laser installed. Bench testing and alignment of two of the new modulators is yet to be completed. -- RDN 10/2/2003 5. Software status Single Dish Development IPT #53 - Friday, October 3, 2003 This week we started the new 6 week development cycle, which is the 8th cycle in 2003. Please note that during this week's project planning and resource allocation meetings, the decision was made that work will be limited to the PTCS project, Ease of Use Initiative, and Data Handling this cycle. Work on the spectrometer balancing issue, the only scheduled maintenance item, has already been started. The Plan of Record for the current development cycle is now available from the wiki at Software.PlanOfRecordC82003. This cycle our software status will be kept online on the Plan of Record page, making it easier to keep up to date. This will replace the status board currently in the 2nd floor hallway. M&Cv3.17 was successfully released on Tuesday, September 30 after one round of astronomical regression tests and three rounds of integration tests. Details can be found on the wiki at Software.ReleaseNotes. Work on Cycle 8 deliverables for PTCS started early in the week. The MR for the Quadrant Detector has been completed and is online at Software.ModificationRequest11C803 for review. Antenna work is progressing; out of 6 major items, 4 have been coded and unit tested, ready for systems testing and review by the PTCS project. Substantial additional work is required to scope out the remaining antenna issues. One of the 8 items on the list to support EMS, interfacing to M&C samplers so that the data is accessible in Matlab, has been completed. Remaining Tier 2 items for the PTCS Python scripts were sponsor tested successfully. Configuration testing has continued on schedule and additional work, particularly to add high-frequency receivers and prime focus receivers is scheduled for the upcoming week. Test usage of the configuration utility was successful during Jim's program checkout last week. This is the second of three development cycles dedicated to the configuration task; during the current cycle our focus is on making changes based on feedback from program checkout and observer support. The transition to this next stage will happen in mid October. Regarding Ku-Band, Jim may have found a problem with the calibration algorithm, which he is currently investigating with Ron. Operational support this week included adding new K-Band calibrations to the receiver calibrations database. This was a milestone, since it was the first time Mark's scripts were used for the task by Electronics, rather than the tasks being performed manually. Another operational support task on Thursday was making changes in date and time formats used by GO, which was required in the move to Red Hat 9 to keep GO operational. -NMR 6. Schedule Last Week ======= Observations for: GBT03B-026, GBT03B-022, GBT02A-031, GBT03A-014 Completed: GBT03B-022 October ====== Scheduled hours [backup] Astronomy ~ 247 [117] Maintenance ~ 123 Tests & Comm ~ 361 [42] Un assigned ~ 14 November ======= Scheduled hours [backup] Astronomy ~ 241 [50] Maintenance ~ 102 Tests & Comm ~ 209 [8] Shutdown ~ 26 Un assigned ~ 131 -- RCB 7. Project scheduling September 29th Planning Meeting Minutes Rm 241 10:00 A.M. [0] Observer comments None We discussed the problem of spectrometer balancing taking way too long. This seems to have been introduced by a recent version of M&C. SDD will look into it. It is a show-stopper according to Karen. [1] This week's schedule HVAC pump repair floating Tuesday-Wednesday This was nailed down to Tuesday. [2] October Observing Schedule discussions We discussed the impact that an overnight shutdown for welding would have. The next 3 weeks should be OK, that is, only local observers will be inconvenienced. We will work around pulsar monitoring and good K band weather in scheduling the welding. We must have at least 1 day notice to begin the weld preparations before the overnight shutdown. About 1 week of preparation is needed for each side. [3] November Observing Schedule discussions Carl is starting to fill in the November schedule. [4] GBT development planning Planning Cycle 8 planning meeting The meeting (really, a set of 2 meetings) will be this afternoon and tomorrow. [5] AOB Proposal for pre-observation tests There are proposals floating about to define an end-of-the-day routine test that can be run to verify that the telescope is ready for the night's observing. Karen, Nicole, and Ron are thinking about the form these should take. 3 levels of test have been proposed. A quick test, taking about 5 minutes, would let the signal paths be verified through observations using cal-on-cal-off, but not necessarliy on a source (antenna parked) A longer test, taking perhaps 45 minutes, would do more in-depth checkout. A full on-sky test would run the telescope to a known source, and compare the output with a control output. Some variation of this based on avaiable effort and cost-benefit in terms of time will be implemented in the future. [8.] Any other business