GBT Operations and Commissioning Meeting 10 December 2004 AGENDA 1. Az Track, Antennae Report -- Bob 2. Observing News -- Ron 3. Scheduling Status -- Carl 4. Microwave Electronics Status -- Roger 5. Digital Electronics Status -- Rich 6. Software Status -- Nicole 7. Project Planning Status -- John 8. Any other Business REPORTS 1. Az Track, Antennae Report Track: * The all-consuming effort this week was the panel review, which went very well. We had some good discussion and some good recommendations, which we will begin working on. Our thanks to everyone who supported us. Telescope: * We tried to replicate the problem with locking the turret with S-Band in position, but it didn't happen. 43m Telescope: * The oil has been changed, and is circulating through filters to clean up. We will look at remaining activities the first of the week to decide on a moving test, possibly later in the week. -- RAA 2. Observing News In the last week, we supported ten projects in nineteen observing sessions for a total of 130 hours. Maintenance (17 hrs), Ka-band commissioning (9 hrs) and hardware and software tests (12 hrs) took up the remaining time. Lost time this week was low (4.4 hrs or about 3%). The average time to setup an experiment was 45 minutes, a typical value for the last year or so. The notable problems were: Inexperienced observers/extended remote setups: 1.75 hrs Config_tool/Grail/M&C Tests: 1 hrs Spectrometer/Spectral Processor: 0.5 hrs Winds: 0.5 hrs We can expect the same level of support for the next week. Note: Since multiple IF's and NOD'ing are the norms for high frequency observing, all operators and support people should be aware of the recent hardware problems with ODM1 in the IF Rack and channel R4 in the K-band receiver. -- RJM 3. Scheduling Status Last Week ========= Observations for proposals GBT02C-050, GBT04C-021, GBT04C-056, GBT04C-057, GBT04B-026, GBT04C-036, GBT04C-025, GBT02C-025, GBT04B-029, GBT04C-008, GBT01A-029 Completed proposals GBT02C-050, GBT04C-057, GBT01A-029 Next Week ========= Observations scheduled for GBT04C-040 PI Jean-Luc Margot The interior of Mercury revealed by its spin dynamics GBT03A-011 [P] PI Fabian Walter The Molecular Gas Content of Quasars at z~4.5 GBT04C-023 [B] PI Haydon Knight High Time Resolution Giant Pulse Searches GBT04C-056 [P] PI Paul Demorest Precision Timing of Binary and Millisecond Pulsars GBT04C-025 [PB] PI Maura McLaughlin Investigating a New Class of Transient Radio Sources GBT04C-036 [B] PI Rajagopalan Ramachandran Searching for young pulsars in the Cygnus Super Bubble region GBT04C-008 [B] PI Yurii Pidopryhora Mapping the Galactic Halo HI: Evidence of Outflow from the Galactic Plane? Scheduled hours [backup] ======================== Category/Month-> December January February March Astronomy ~ 453 [133] 433 [303] 0 0 Maintenance ~ 45 [034] 0 [136] 0 0 Test & Comm ~ 174 [030] 253 [058] 0 0 Shutdown ~ 36 0 0 0 Un-assigned ~ 37 58 0 0 Proposal Checkouts ================== Awaiting checkout - 0 Active checkout - 9 On Hold - 5 Schedule - 51 Completed - 251 Current backlog [hours prior to 04C*] = 897 Total time to discharge [hours] = 3265 * Includes projects that are on hold for trimester 04C -- RCB 4. Microwave Electronics Status Operational issues: There was a report of an instability in one K-high channel, and that is being investigated. Also, an intermittant instability was found in IF 1; the config tool device health file was changed to default to IF 2. A 6hp compressor was moved to the GBT and put into service. This is significant because the extra gas capacity of these units are needed to run 1020 and 350 refrigerators together; that capability will be necessary when the Penn Array system arrives (it requires a dedicated compressor), and if we replace the model 22's with 350's. There may be some operational issues with the 6hp higher start-up currents when transferring to the 50kW emergency generator. R. Norrod 12/08/2004 5. Digital Electronics Status Spectrometer Report 3 December - 10 December 2004 *****Operations, Problem summary***** Dec. 6: The spectrometer turned itself off after experiencing serial errors. This occurred right after a spigot run, so the manager was probably off. Clearly the spectrometer manager needs to be turned on before trying to configure the spectrometer. Nevertheless, the spectrometer should not have powered itself down. However, it is designed to protect itself against out-of-range physical parameters including system voltages and temperatures. If it reads a bad value, it shuts the system down to protect itself. The value which caused the shut-down is kept in the System Monitor^Òs memory and can be accessed via the engineering port. Almost always, when a shut-down has occurred, the out-of-range value was clearly an erroneous reading. In this particular instance, Rich decided not to spend the time investigating the cause of the failure and asked Donna to simply restart the system. If this type of fault starts to occur annoyingly often, it should be possible to improve both the firmware and hardware to avoid erroneous reading. *****Development***** *LTA replacement: Continue working on LTA firmware. more than 50% done. *Cross-Correlation Test Fixture: A rough chassis layout was completed. A power estimate for the system was completed and suitable power supplies were located in-house. The boards for the cross-correlation-and-output module have arrived. The electrical design of the front panel was completed. A detailed system block diagram drawing was completed. The schematic for the front panel interface board was completed. The printed circuit for the front panel interface board is 70% complete. Construction of the second filter module is 50% complete. *****Plans***** Continue LTA replacement and Cross-Correlation Text Fixture development. -- Rich and Holly 6. Software Status Single Dish Development IPT #108 - Friday, December 10, 2004 Today ends week 3 in C8, the last development cycle of 2004. The Plan of Record is available from the wiki at Software.PlanOfRecordC82004, and focuses primarily on maintenance activities and other items which require less scientific support. Ron, Jim, and Paul spent a portion of this week starting beta tests for the IDL work, using the low level I/O functions as well as the GUIDE layer which is the part that provides the unipops-like functionality. I/O is largely complete, and work is now focusing on fleshing out the contents of the toolbox and fixing bugs identified during early beta testing. Documentation is also being expanded and updated as code changes are made. Melinda is halfway through writing the group of scheduling blocks that will become our regression test suite, and spent some time with Jim yesterday working out issues like which sources to use, and how long scans should be. We will meet with the operators next week, who will run the scheduling blocks on our behalf, in preparation for running regression tests in an automated dynamic way sometime in 2006. The SDD used available backup time to start integration testing early, primarily to work out any issues that may have been introduced by the addition of the full telescope software simulation capability this cycle. IF Rack was ported to Linux this week, and is now being tested. PF balancing was tested through config tool and turtle, and is now ready for sponsor testing. BCPM can now be turned off through turtle, and work is underway to determine how best to balance in the middle of a scan. The bug that produces empty strings to be sent to the Archivist has been identified, reproduced in the new simulator, and a solution has been put together which is also being tested. The DATEBLD keywords in the FITS files have been made FITS standard. Operational support over the past week included: * Fixed the 1 window, ACS, Rcvr26_40 IFPath problem reported by Toney in the config tool; patched in on 12/7. -- NMR 7. Project Planning Status December 6, 2004 Planning Meeting Minutes Rm 137 10:45 A.M. [0] Observer comments & operator log items resulting in lost time: Observer's comments received from: None. Operator's log comments Significant failures the past week stemmed from: Computing: Titania needed rebooted. Software: GO froze up, config tool failed to set up for a peak (restarted everything), MM converter manager, config tool failed to find path (see Minter's report), antenna Manager The first config tool failure was due to the example script on the Web being wrong, and causing the error. It has been fixed. The MM converter manager just failed for some reason and needed restarted. The second config tool problem is real and is being worked on. The antenna Manager problem is also being looked at. Electronics: Intermittent spectrometer problem The problem disappeared before anyone could look at it. If it happens again, callouts are recommended. Telescope/Servo: Wind, Az transducer fault (software problem) Other: Several long setups There was a bit of discussion about this. 3 separate incidents caused most of the lost time. One was the GASP setup, which took longer than allowed. This was chalked up to the novelty of the backend. Second was a computing problem with an observer's account. M. McLaughlin's default shell was tcsh, but the pulsar spigot scripts only work with the user's shell set to bash. The command to change the user's shell did not work on the telescope network, either. It was recommended by Ron that the scripts be modified to work with either shell. The third problem was a run in which a remote observer had no previous experience with the GBT. This is clearly a violation of the conditions whereby we agreed to allow limited remote observing. Other Problems from Observing Reports: Frank Ghigo's Ka band commissioning report had three minor problems: GFM slowness, config tool oddity, and MM converter extra prepare. Nicole reports that the second 2 problems are known, and will be addressed in time, while the first is likely due to network problems. [1] Resource calendar schedule conflicts and discussions None It turned out that Bob and John missed the fact that there was antenna motion needed for Quadrant detector work on the second maintenance day. Kim pointed this out to us. After consulting the task list, that activity was approved with a slight change of time, but not duration. Thanks to Kim for alerting us to the problem. [2] Observing Schedule discussions None. Carl and Phil were absent [3] GBT development planning Development planning with Project Managers Richard gave a brief overview of the ongoing attempt to get a handle on our development projects, and to re-plan based on our new estimates and our strategic plan. Everyone should re-read (or read, as the case may be!) the 2005 program plan. IF Amplifiers John presented Roger's plan to build 2 batches of IF amplifiers with improved gain stability and reliability. Richard and Ron questioned if these were proven to fix the non-linearity problems. Consulting with Roger, they are a necessary, but possibly not sufficient step. There are also zero-point offsets and detector issues to deal with, which are being studied right now. The plan to build the first batch of 10 amplifiers is underway. 20 more are needed later, but we have to find money to build them, as the M&S budget cannot support that activity. [4] AOB None -- JMF 8. Any other Business