USNO VLBA Support Meeting

Date: November 13th 2014

Time: 11 am Mountain, 1pm Eastern

Phone: 434-817-6519 (NOTE CHANGE IN PHONE FOR NOVEMBER ONLY)

Attending:

USNO: Dave Hall, Dan Veillette, Chris Dieck, John Spitzak

NRAO: Ephraim Ford, Jim Ogle, Walter Brisken

Agenda

Status
  • Correlator: James was unable to attend and requests to do a stand-alone meeting with USNO in early December to walk through the punchlist.
  • VLBA Status
    • NRAO Mauna Kea Lightning Hit Update
      • MK is up and running, mostly electrical damage but some mechanical work done due to realignment of FRM.
      • Dan: Regarding SSDs, n4307b needs LA data. NRAO will check to see if disk has been shipped. Nov 3rd.
      • The runs from SC have been correlated well. It has been analyzed fine and there have been different values seen in the analysis, but that is expected to be due to the different baselines.
    • Filter removal at SC & LA.
      • See C. Walker 10/29 email
      • Filter removal is not recommended
      • Longterm, USNO is evaluating.
      • NRAO: No plans for installing filters
      • Walter: We could move from the polybank filters to the DDC filters. You would only have 8 channels, and could observe at half the bandwidth for twice the time.
          • USNO: How tunable are the sub-bands? NRAO: Very - can be tuned to non-integer intervals with a few areas that can't go. Maximum width are between 1 & 128 MHz, power of two, only 8 available, not 16.
        • A test could scheduled if USNO provides a rough tuning plan. USNO: A benefit of this would be better determination of where the ambiguities are between each band. The fewer number of channels it would be tougher to space them well enough.
      • A new frequency set is still needed for the fortnightlies.
        • Walter would need to know more about what the use cases are.
        • USNO is not sure whether it is mandatory to use the same frequency set on all observations
        • Chris to contact Walter regarding what the proposed frequency set should be.
  • 4 Station Observation Plans (F4)
    • USNO Update
      • Need to test the new 13cm frequency set suggested by CWalker (Test 1). USNO to evaluate data. Walter will develop schedule based on info from Chris (see above)
      • Move forward based on those results.
      • To be scheduled dynamically like daily observations. Scheduled in a template manner like the DQ runs at first and see how it performs.
  • Data Quality Improvement
    • USNO Update on effect of 5MHz pulse cals:
      • Dave: Has instructions from Haystack on how to pull out HOPS PCal data. This hasn't been done this week. Goal is to compare HOPS Pcal and DiFX Pcal data.
      • Walter: We operate at 32MHz, most geodesy is at 8Mhz. This can affect amplitude, which HOPS looks for, so be aware of this. Dan: Has been looking at the amplitude parameter with little luck, possibly due to a different build he is using. He is still getting data, although it is flagged. What is strange is that he can get a good amplitude measurement of phase cal in S-band, but loses fringes if he uses it.
    • USNO Update to increasing cable cal measurement cadence
      • all comparison experiments successfully processed with SOLVE, but no relevant output yet.
      • Experiments have been run with no, low-freq, and high-freq cable cal data. In order to run through post-interactive analysis USNO needs to run a test scenario that has not been done before. This can be done after catching up on observation analysis. ETA on completing this analysis is in 2 weeks.
  • Daily Observation Report for October
    • Things are still looking good, the transition to SC went really well from the analysis side.
    • There are more bogus points identified. These are a flag only, no clear understanding of source for them but will follow up at a lower priority.
    • It is good to see that the change of the baselines with SC shows up in the fit data. MK-PT is definitely better baseline than PT-SC.
      • Walter: Still wants to look at what contributes to the error budget. Can USNO give him access to the delay rms across the scans. Single scan errors would be independent of the baselines in use. The USNO fits incorporates the offsets between the models, so it is baseline dependent.
    • Walter: Have you noticed a difference in Tsys? USNO doesn't look at this and Jim says there isn't much to be expected.
    • On October 10th we had 3 of 4 antennas out, so no usable data.
  • Open issues
    • Walter: Met with John at DiFX meeting
      • Roger C. is going to be going to half-time very soon. (He maintains HOPS, Mark4) Not sure who picks this maintenance up
      • A new DiFX release will be coming soon, validation testing is pending. Goal is to show 5454 and 2.4 perform comparably
      • Vex2 standard will be coming out soon. For most of DiFX Vex2DiFX will require the effort. Short term we will probably have to generate a Vex2 and Vex format file. Dan says they also worry about OVex (HOPS). Walter: we need to lean on Haystack to support the Vex2 format quickly, probably Art Niell
      • Mark4 telecon could discuss this next week at their telecon. Walter will add the Vex2 to the agenda through an email to Art.
    • Dan: Knows there will be a separate correlator meeting coming in and we can discuss this, are there known issues with control commands
      • Multicast message dropouts after delays? This is only seen when our GUI was running on a separate network. This has been seen if receive buffers were set too small, but that is unlikely to be USNO's current setup. Use TestDifXMessageReceive or Send.
      • Daemon not running or not proper environment? Dan says it was running.
      • Could also adjust Time To Live parameter higher to allow more attempts of sending the message.
    • Next meeting on December 11th, 2014. Correlator meeting a week prior.

-- EphraimFord - 2014-11-03
Topic revision: r5 - 2014-11-13, ChristopherDieck
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback