GBT Operations Meeting 20 June 2003 AGENDA 1. Az Track Status -- Bob A. 2. Spectrometer status -- Rich 3. PTCS update -- Richard 4. Observing news -- Ron 5. Spectral Baseline, Front-end, and IF work -- Roger 6. Software status -- Nicole 7. Project scheduling -- John 8. Observing schedule -- Carl 9. Any other business REPORTS & DISCUSSION 1. Az Track Status a. The new wear plate over splice 45 was removed for inspection of the shim material and baseplate.  The shim material had compressed by 0.012".  The shim was trimmed to fit better and 0.020" of additional material were added.  An inspection of the weld showed that nothing further could be done to improve it's contour.  The wear plate was reinstalled, and the wheels appeared to have the same tilt when they passed over it as last week.  Further analysis is needed. b. Additional samples of the old and new wear plates were sent to the metallurgical lab for testing. c. Simpson, Gumpertz and Heger are making progress with the finite element analysis.  A conference call was conducted to evaluate the results. -- RA Discussion. Bob added that the FEA analysis is showing that at the bolt preload used, the wear plate is not being locked to the base plate. However, it also shows that the whole plate should shift, which is not what we observe. More work is needed. 2. Spectrometer status The spectrometer continues to operate fairly reliably. Of the two major faults that have been occurring, one has been worked around in software and the other persists. The software fault involved the "CCC Xilinx load from ROM" error. The latest software clears this error after a successful load from file. The other error is the "DMA error" where the computer does not get all the data it expects. The software group has improved the logging; we hope to glean some clues about this error from these improved logs. Holly is also working on observing the failed data transfers with the logic analyzer; this may also help. Work on the microprocessor clock speed upgrade to 32 MHz continues. Ray has came up to help on the LTA board on 6/19. Although several stumbling blocks were overcome the board is still not quite working correctly at the higher clock rate. There is still hope however! Progress is being made in the construction of filter boards for the high speed samplers. Ten of ten are built, and eight are tested. These are part of a solution for cleaning up the sampling clock. Documentation for the system monitor board and pulsar spigot has been updated. There are still several items that need to be updated. Plans for next week - Continued work on microprocessor clock speed upgrade - System tests of the hardware reset - Work on the improved test fixture - Pulsar spigot testing (mostly by Kaplan) - Upgrade documentation - Equipment room clean up. -- RL Discussion. Phil asked if an LTA upgrade is still envisioned. Rich and John replied yes, that the LTAs are still slot dependent, which is not desirable. Rich added that the pulsar spigot card is still not available for general use, although it appears to be working from a hardware prespective, until results show otherwise. David Kaplan is working on software. 3. PTCS update Recent PTCS work has focussed on upgrades to the antenna manager for improved trajectory control, and continued EMS development. In the last development cycle, Joe Brandt has been addressing some problems with the calculations of antenna trajectories. There were three known issues in this area: 1. The "trajectory/scan mismatch" problem. This was a bug which caused a slight misalignment between the ending point of the slew trajectory and the beginning point of a scan. This may cause vibrations to be induced when the servo attempts to jump onto the target position. 2. The "small-step" problem. This occurs when the antenna is slewed to a point very nearby to the starting position, or the same position with a slight change in velocity. Again, the trajectory preprocessor can produce an impulse in acceleration, instead of a smooth profile. 3. The "fails to generate a trajectory" problem. In some cases, the preprocessor can fail to find a route, generating an antenna fault. Fixes to issues 1 and 3 were included as part of M&C V3.14, which was released on June 18th. We plan to tackle issue 2 in a later development cycle. V3.14 also includes additions to the antenna FITS file which will be used for more direct "peak" data analysis procedures, and some general refactoring of the antenna software. We had hoped to perform rather extensive astronomical tests to demonstrate the improvements in the antenna trajectory calculations. Unfortunately, the recent spell of bad weather both delayed reinstallation of the X-band receiver, and limited us to observing at C-band or below. We have demonstrated via inspection of the commanded and actual encoder values, and accelerometer data that the antenna behaviour is certainly improved. A short C-band all-sky pointing run was performed to ensure no systematic errors had been introduced by the modifications. V3.14 has been released for use, and more extensive tests will be performed when telescope time and weather allow. EMS Developments ---------------- The first experiment with the Engineering Measurement System will be to perform a "feed-arm tip trilateration experiment" - trilaterating retros on the tip of the feed-arm as it is tipped up and down in elevation. We had hoped to run this on 19th June, but bad weather delayed the required realignment of the rangefinders. This work is ongoing, and the new target date is Tuesday 1st July. We expect this to be the start of significantly increased use of the rangefinders, which will occur either during maintenance time, or during scheduled PTCS test time. -- RMP Discussion. Dave Parker added that the next iteration on the laser cover prototype was expected from the contractor in the coming week. If it is accepted, the production contract for the remaining covers will begin. Ron noted that the holography receiver has returned. As an initial strategy to deal with the saturation problem, Ron plans to look for satellites with weaker signals. Don continues to work with the Cambridge group on the OOF holography analysis. 4. Observing news We serviced about seven proposals this week. The BCPM problems at the start of the week required a bit of reassignment of observing proposals that we were able to handle. The significant problems for proposal 3B7 were with the Spectrometer when, on most night we had no problems but on others we lost up to 30% of the observing time. There was also during 3B7 an E-stop for an unknown reason that cost us about 1 hour. The pulsar BCPM proposal 3B35 got off to a late start, losing about an hour, due to a combination of issues that are summarized in the operator's logs and Glen's e-mail. -- RJM Discussion. We discussed the appropriate amount of setup time to be listed on the observing schedule prior to the start of a session. Presently, one hour is alotted, although the typical time for setup is now down to about a half-hour or less. We decided to leave the hour setup slots for now, but will continue to review this over the coming weeks and months. We should reduce the time once the large majority of programs are being set up in a half-hour or less. 5. Spectral Baseline, Front-end, and IF work The X-band receiver was installed on Tuesday, but bad weather has prevented further observations for the baseline investigation. We have scheduled time next Wednesday evening, and will, if the weather is good, concentrate on the X-band baseline structure on strong continuum sources. Yet another mechanism that can introduce baseline ripples in the 1-8GHz Converter Modules has been identified - phase change in the LO2 (but not LO3) connections. LO2 feeds a triple-balanced mixer used to upconvert the 1-8GHz first IF to 9-10.35 GHz. A double-balanced mixer we substituted showed much less sensitivity to the LO phase. (We cannot use this particular DBM in the real system because it does not have sufficient bandwidth, but it worked for the test.) We hope to discuss our findings with industry mixer experts to better understand what is happening, and perhaps find a solution. Also, we will follow-up with additional measurements in the system to better judge the significance of this mechanism. (It appears to be a strong effect, but we found it while doing lab tests on the Converter Modules). And, we continue to pursue the ripple mechanism related to crosstalk through the Optical Receivers, but have no new findings to report. The PO for new optical fiber modulators has been placed and confirmed. RDN 6/19/03 Discussion. Roger was away so there was not a great deal of discussion. The cleanup and modifications of the X-band feed were discussed. Dennis noted that the inspection of the feed corrugations for metal flakes and other debris was extremely laborious and that he would like to find a way to automate this. 6. Software status Single Dish Development IPT #40 - Friday, June 20, 2003 A standard, semi-quarterly development cycle will be put in place at the beginning of July. In the meantime, we are in the middle of our last nonconforming development cycle, and the release date for M&C v3.15 has been reset to 7/9/03 in the late afternoon. M&C v3.14 is in place as of this week. The plan of record which shows goals for M&C v3.14 is available at http://tryllium.gb.nrao.edu/images/POR_May03.pdf, and all of the MRs are viewable through http://projectoffice.gb.nrao.edu.Tests have been conducted to characterize the antenna trajectory preprocessor fix through early next week; results will be described in a comprehensive memo within the upcoming weeks. Work is in progress on the new Plan of Record for the month of June 2003, which is available at http://tryllium.gb.nrao.edu/docs/POR/POR_June03.pdf. Out of 7 MRs for this cycle, all have been written and all but one has been approved. The activity for improvements to the laser rangefinders has been completed, and 13 out of 14 items in this activity have been successfully sponsor tested. PF2 software has been completed and is ready for the telescope. The SDD has successfully prototyped a Python script for analyzing pointing data, and work continues on a similar prototype for focus. Scientists and software engineers are now all involved in evaluating this approach for its utility and scientific applicability. Several analysis activities are underway, including the Project Charters for the DCR Improvements exercise, and the Data Export/Data Analysis Project Charters which Ron and Nicole are working to complete by the end of the cycle. It is intended that these will provide the strategic roadmap for efforts spanning the next six months. Please visit our updated web site at http://projectoffice.gb.nrao.edu, where we are now actively integrating telescope control and data analysis activities. Also, our status board describing the latest news about items on our current Plan of Record is now outside Nicole's office. If you have any questions about the status of your development, feel free to refer to "The Board". -- NMR Discussion. As noted above, the software release cycles are going to a 6-weekly (semi-quarterly) schedule. These will be aligned with the quarters of the year. We are planning that all GBT projects will work on this release schedule. By aligning with the development cycles with the calendar quarters, it will be much easier to prepare Quarterly Reports and do annual program and budget planning. 7. Project scheduling June 16th Planning Meeting Minutes Rm 241 10:00 A.M. 0) Observer comments A few spectrometer failures per night over the past week. 1) This week's schedule Spectrometer Spigot card tests are scheduled for Thursday 2) Next week's schedule Holography is scheduled next week. It may or may not be done, depending on the receiver status and Ron's workload. 3) June Observing Schedule discussions None 4) July Observing Schedule discussions The PF1 recevier will be off the telescope for 3 weeks. The first weekend in July is available for observing. 5) GBT development process and program proposal review A meeting is scheduled for Thursday to review and finalize 2 new documents: GBT Project Planning Process A document to discuss how we plan and execute our development projects for the GBT. GBT Current Development Program The current development program that we are managing. 6) HVAC RFI spare boards/modules -- We will buy a PCM out of ops funds priority -- This is a high priority project funding -- We can cover the filters out of 43450.XXXX for a subset of the HVAC units. This is ending initial design efforts for better cost numbers. 7) AOB BCPM status -- BCPM was down since the power outage last weekend. (It has since been fixed by Ford, Shelton, Shank, Lacasse, Bennett, and Don Backer) A plan is needed for ongoing support of the BCPM. August test/commissioning time schedule discussions There is lots of time left at the end of August -- JF 8. Observing schedule Last Week ======== Observations for: GBT03B-034, GBT03B-007, GBT03A-021, GBT02A-015 Completed: GBT03A-021, GBT03B-007, GBT03B-009, GBT03B-034 One project cancelled (moved to July 2) because of BCPM failure: GBT03B-015 June === Remaining: Astronomy ~ 160 hours Maintenance ~ 42 hours Test & comm ~ 68 hours July === Astronomy ~ 329 hours Maintenance ~ 218 hours Tests & comm ~ 198 hours August ===== Astronomy ~ 329 hours Maintenance ~ 192 hours Tests & comm ~ 223 hours -- RCB 9. Any other business The Users Committee meeting appeared to go well as regards the GBT. At least one UC member was very eager to see the pulsar timing mode of the Spectrometer delivered as soon as possible. We will examine this as part of the Spectrometer development program we are working on. PRJ 24 June 2003