Digital Group Status 01-JUL-06 THRU 07-JUL-06

***Operations, Problem Summary***

Significant lost time (9.95 hours total) was reported with the Spectrometer. Of this lost time...

7.25 hours were due to "Configuration issues"... Ron Maddalena seems to have a handle on this as an operational issue.

2.00 hours were due to "DMA errors"... John Ford seems to think that these errors may be indirectly related to the LTA's; but specifically relate to an infrequent problem wherein the system-wide "Start" command is not being recognized by one or more of the uProcessors in the system.

0.70 hours were due to "Standard Spectrometer errors"... Most likely related to the LTA's...

QUESTION: Since the Spectrometer (and in particular the LTA's) seems to be increasingly problematic, should we focus less attention on the long-term solution (i.e., re-design of the LTA's) and explore some sort of near-term, band-aid fixes/improvements?

Summer maintenance continued with lots of work being done on the active surface.

***Development***

Cal-Tech Continuum Backend (CCB)...

The 2nd CCB has now been fully re-assembled; has successfully undergone bench testing; and awaits further testing with the Ka Band Receiver.

A large percentage of the circuit boards and sub-assemblies required for the 3rd CCB are complete and await bench testing. At this point, we feel that the 3rd CCB could be completed and readied for final testing in the course of two calendar weeks.

Spectrometer LTA Re-design...

Work has resumed on the LTA re-design project... A meeting with John Ford to discuss current and proposed new architectures, design approaches, and the implications thereof is planned for next week. Following this, a project schedule will be prepared and distributed.

Lab Spectrometer...

The Lab Spectrometer appears to have some remaining issues which result in 100MHz comb artifacts appearing in the spectra along with occasional ACF spikes and resultant ringing. Roger seems to think it's a manageable situation, so we'll most likely wait until John Ford returns after the 4th of July to tackle this issue (no change from last week's report).

ALMA...

John Ford's summer student, Manasseh Obi, continued work on a detailed, predictive noise analysis of the bias board with help and input from Randy and Jason.

***Plans***

- Begin construction of CCB #3...
- Evaluate Tektronix scope...
- Continue Active Surface maintenance...
- Continue ALMA support as needed...

Randy

We have resumed normal tracking after the shutdown for upgrade of the MIT/LL electronics. Plan normal tracking next week as well.

We will start testing low frequency mapping with the 43m over the next month, using the up/down conversion unit built by Steve White. Right now we have a single polarization unit, but will have a second in a few weeks.

Still planning on further MIT/LL electronics upgrades in September.

Statement of work for 2007 and 2008 is being reviewed by George Clark in C'ville.

This ends Week 1 of 6 in the fifth development cycle of 2006 which can be viewed at PlanOfRecordC52006.

Operational support included items in Ron's report above.

There is not a lot to report as it was a short week and we spent a lot of time this week reviewing and interviewing candidates for our open software position. Because the wiki is public, I will include no further details of our interview process in this report but we will be happy to answer any questions in person.

We supported 14 projects in 27 observing sessions, including one radar run and some polarization experiments. Average setup time was 13 min. The telescope was scheduled for 13.5 hrs of maintenance and 2 hrs of receiver checkouts.

Lost time was significant this week for many reasons that will need a bit more explaining than I've usually supplied over the past months. I'm not going to depress people with the percentage of time lost.

• Project 6B49 was an unsuccessful radar run probably due to a problem with the Goldstone transmitter. 2.5 hrs.
• 2A53, a Robishaw polarization experiment, lost 9.5 hrs due to various Spectrometer failures.
  • 7.25 hrs lost when the Spectrometer did not configure itself properly and the observer did not catch the problem until after the session was over. Dana might be able to provide more information.
  • 2.0 hrs lost due to a rash of DMA errors. Mark Clark was involved in helping with this problem. The operator's logs don't give information on how the problem was fixed.
  • 0.2 hrs lost to 'standard' Spectrometer hang ups.
• CGSR2 hardware/network problems : 0.5 hrs
• Other, more 'standard' Spectrometer problems (3 instances) : 0.5 hrs
• Observer didn't show up : 0.4 hrs
• Astrid hangups (2 instances) : 0.3 hrs
• SDFITS Filler : 0.3 hrs
• Others (Spectral Processor, Scan Coordinator) : 0.3 hrs
• No-Lost-Time problems:
  • RFI reported @ 1420.31 MHz -- Any conclusions as to what this might be?
  • Astrid/Titania Lockup

Of the 14 operator shifts, five had no problems and 2 had no-lost-time failures.

This rash of problems brings to mind a couple of suggestions:

• Operators should attempt to contact Carl, then me, when an observer doesn't show. In this case, the operator made their own decisions as to what to run. Since Frank and I were the derelict 'observers' -- we actually thought we'd had given this time to a backup project -- we could have made a different choice as to how the time should be used.
• The 7 hrs Tim R. lost to the spectrometer configuration problem was noted in the operator's log about 12 hrs after the operator's shift was finished. Those of us who read the last day's logs each morning would have seen an error-free shift. For significant loses like this, and especially when they are logged after-the-fact, we should use the observing email exploders to inform folks. The DMA problem could also be considered as being under-reported and might have been another item to mention using the exploder.

Operations

The K-band receiver has an intermittant problem with the dewar vacuum readout that caused some warning messages and e-mail, but no lost time. We spent some time searching for a 1421.xxx tone-like interference signal, but it appears to have turned itself off.

Development

High Frequency Receivers:

Q-band Receiver: We continue to test the Q-band receiver with the Lab Spectrometer. Detailed progress reports are on the wiki at:

http://wiki.gb.nrao.edu/bin/view/Projects/SpectralBaselinesKaQBand

Ka Receiver: The receiver is currently being dismantled to install the Zpectrometer signal taps and the upgraded phase switch control system. While that's being done the physical configuration is being changed so the differencing loops no longer have to be broken to install the receiver on the telescope. The primary support components are in the shop queue and reassembly is expected to start later next week.

In microcircuit assembly, started work on the following MMIC units:

Qty 3, 5040 amplifiers for Ka-band rx, approx 20% complete.

Qty 3, 5618/5040 LO freq doublers for Ka-band/zpect, approx 20% complete.

Qty 2, 2-28 Ghz amp with 44Ghz LPF for Ka-band, approx 90% complete.

%META:TOPICINFO{author="RichardPrestage" date="1152195420" format="1.0" version="1.1"}% %META:TOPICPARENT{name="CommissioningMeetings"}%

GBT Operations & Commissioning Meeting

7 July 2006

AGENDA

• Az Track, Antennae Report -- Bob
• Observing News -- Ron
• Scheduling Status -- Carl
• Microwave Electronics Status -- Roger
• Digital Electronics Status -- Randy
• Software Status -- Amy
• Pulsar Related -- Scott
• MIT/LL 43m Project -- Glen
• Any other Business

Az Track, Antennae Report -- Bob

Observing News -- Ron

Scheduling Status -- Carl

Microwave Electronics Status -- Roger

Digital Electronics Status -- Randy

Software Status -- Amy

Pulsar Related -- Scott

MIT/LL 43m Project -- Glen

Any other Business

-- RichardPrestage - 06 Jul 2006