Notes from Commissioning Meeting 30 August 2002 1. Observing program checkout Richard reported that Carl has placed a complete list of observing programs on the web, with links at www.gb.nrao.edu/gbt/gbtstatus There are five categories -- Awaiting Checkout, Active Checkout, Schedule, On Hold, and Completed. There are 20 programs on the Active Checkout list. These programs are for frequencies lower than K-band, and are for trimesters prior to 02C, for the most part. They appear to be observable, in principle, and are thus the priority for checkout in the next month. Commissioning scientists have been assigned to each program. Actual testing has started on about 5 of these, but the remainder await attention. A specific action for next week is Action 30aug02-1: Commissioning scientists: Identify specific observing requirements of assigned programs. Prior to next Friday's commissioning meeting (or as soon as possible if you are out of town), please review all proposals on this list in detail for specific and detailed observing requirements. Please send this information to Carl Bignell so that he can update the web pages. For spectral line programs, this should include identification of the exact Spectrometer mode string(s) required. In general, it is important that we specify the _exact_ mode required so that we can set priorities on Spectrometer engineering and astronomical checkouts. There are 15 programs on the Schedule list, 8 are on the On Hold list (meaning they have been tested at least once and require further development). There are 46 programs on the Awaiting Checkout list; ==> we have a lot of work to do. At next week's commissioning meeting, we will go through the programs on the Active Checkout list one by one and review status. 2. Software status. Mark reported that M&C v3.6 was successfully released this week. This is the August M&C update. Major features and enhancements are saved for the formal version release, but bug fixes are done on an intermediate basis throughout the month, as required. Earlier in the week, Amy sent out the release notes for this version. Major features include: Improvements in LO1 Reliability - Feature 1: Elimination of synchronization loss between the LO1 and the backend sig/ref binning process - Feature 2: Communication with synthesizer improved - Feature 3: Proper disabling of interrupts at the end of a scan - Feature 4: Ported LO Counter to Linux (Phase 2 task completed early to aid in sig/ref testing) - Feature 5: Reboot Improvement to LO Coordinator Spectrometer Improvements (Phase 0 - Enable More Efficient Hardware Testing) - Feature 6: Enhanced Message Generation - Feature 7: Identify/constrain Integration Times - Feature 8: Support On/Off Recovery - Feature 9: Fixed problem with data collection for wide band mode Other - Feature 10: Ability to handle increased number of switching signal masters - Feature 11: DCR Output Reduction - Feature 12: Receivers handle default configurations correctly. - Feature 13: Fixed problem with reading past the ephemeris end in the GO User Table. Mark and Melinda have made good progress on the PF2 manager, and expect to complete this by next Wednesday. Work has proceeded on an Interface Control Document needed for the Caltech continuum backend, which has its preliminary design review meeting on next Friday morning, September 6. Work is proceeding on the new users interface. It is still expected that a UI for the Spectrometer will be available from a button on the GO screen following the September M&C release. The cable swap and accompanying changes to the FITS files and Filler necessary to fix the S/R flip in the first sample of a frequency switched scan using the Spectral Processor will be coordinated for the September M&C release. Mark and Bob are working together on this. 3. Spectrometer status Rich has updated his spreadsheet of Spectrometer modes. This is now on the web from the /gbt/gbtstatus page listed above. Rich has made good progress with the checkouts, and added 20 new modes that have had engineering checkout (some of these require follow-up to resolve some problems). Rich found that in several of the high resolution modes, very narrow birdies are present. He intends to eliminate the possibility that these are coming from the noise source next week. There are 109 modes on the Spectrometer spreadsheet, and 86 of these have now had initial engineering checkout. No modes have fully passed the astronomical checkout prescribed by Phil's checkout memo. 4. Baseline investigations Roger reported that efforts to pin down the source of reflections in the optics at L-band have not been successful yet. Earlier this week, some reflecting panels were placed on top of the Receiver Room to see if any standing waves could be deflected onto the sky, but this did not produce any improvement. Work continues. Dana has written a draft memo summarizing baseline investigations to date. Following his return at the end of next week, this memo will be finalized and released. 5. Scheduling status. Carl reported that September is now mostly scheduled. He set out to reserve ~50% of time for commissioning, but more than this is presently on the schedule. Many observing programs need Galactic Center time, and not all can be fit in. Carl is beginning to work on the October schedule. 6. AIPS++ status Bob reported that he believes that the present intention is that AIPS++ will try to resolve the problem of handling names with minus signs and other embedded symbols that could also be interpreted as mathematical operators. 7. Other business. Glen reported that the comb of RFI lines in the 300 MHz band (with 1.966 MHz spacing) has, fortuitously, disappeared. No one knows of any specific action that may explain this. The 50 MHz bandpass than Glen showed was pretty clean. There is some ripple in one polarization that might originate in the IF converter rack. This will be investigated. Glen also showed a curious effect in which the signal strength in the two polarizations diverged over the course of a long observation of on/off pairs. Rich reported that the digital group was making good progress in installing filters in the feedarm servo cables to eliminate that source of RFI. 10 of 56 cables are now done. The group may need to have an overnight shutdown to complete the last phase of work. They will coordinate this with Carl. Rich also noted that J.D. and Jason are finding that the short maintenance days that end at 3pm for the track inspection are making it hard to put in a good day's work on the actuators. A long maintenance day is much more effective for this work. They will work with Pete and Carl to see if a solution can be found. Rich noted that all 2209 actuators are now working (the first time ever). Roger reported on the status of the RFI window screen replacement in the upstairs control and equipment rooms. The first set of 4 screens have been replaced, and produced a 15-20 dB improvement in attenuation through those windows. The remainder of the screens need to be removed and sent off for refitting, and will be away for some weeks. This will increase the RFI levels at low frequency while they are away. September already has a lot of low frequency observing scheduled, so we should wait until after that. The IPG should propose a time that they would like to do this work, and then coordinate this with Carl. Roger noted that overall, the shielded areas are still only attenuating to the 40-50 dB level, which is below the original spec. This may be the result of a variety of penetrations in the room that are no longer fully shielded. SUMMARY OF ACTION ITEMS Action 30aug02-1: Commissioning scientists: Identify specific observing requirements of assigned programs. Prior to next Friday's commissioning meeting (or as soon as possible if you are out of town), please review all proposals on this list in detail for specific and detailed observing requirements. Please send this information to Carl Bignell so that he can update the web pages. For spectral line programs, this should include identification of the exact Spectrometer mode string(s) required. In general, it is important that we specify the _exact_ mode required so that we can set priorities on Spectrometer engineering and astronomical checkouts. PRJ // 31 August 2002