Marty, Randy, Joe, Richard, Jason, Adam, John, Ray, Dan
Agenda:
Disposition of Action items from last week
Review of urgent action items
Review of timely action items
Maintenance plans for this week / next week.
Review of astronomical commissioning tasks.
Other
Meeting notes:
Disposition of Action items from last week
Nonlinearity problem
Action: Adam, to circulate brief note to gbsapp mailing list with conclusions of Ron's analysis as of last Friday
Ron saw the problem was not too bad, but a small gain instability - possibly not significant IF artifact
Divots in spectrum
Action: Adam, to provide plots of On and Off separately, refilled so that the center spike is not interpolated over.
Completed
Action: Adam, to confirm whether width of divot is constant in channel space, or frequency space, or neither, between modes
Circulated an email on topic with discussion. Issue is frequency space. Exact cause is not clear. Some loose correlation with total power. Next step is to investigate ADC snap blocks for consistent ADC statistics.Evaluate any possible BOF version dependencies. link Dan noted this looks like a possible interleaving issues - can that be replicated in simulation [Dan student] (needs power (1%?), spectrum (broadband noise), and integration suggestions (1 min.))? Also a test to deliberately set offset gains off 1-2% and see effect. [NRAO].
Action: Adam, to draft an Observers Alert based on this Ron's email, and send it to Toney.
needs to be done.
Action: Adam, to FFT the ADC snapblocks corresponding to one of the total power spectra that shows a pronounced dimple, and see if it is there also
Action: Adam, create a wiki page with our knowledge to date on the symptoms, and arrange a telecon with UCB
Wiki page built and UCB on weekly meeting telecon
Action: Adam, to inspect ADCSNAP data as proposed by Dave MacMahon
Mode 10-19 Hardware implementation
Action: Adam, to produce plots of bandpass shapes for mode 10 (s/w fix), mode 10 (hw version) mode 23 (8-subband version), linear and log space, for comparison.
Re-assigned and completed. [link] Old BOF 10 and 23 look different but stable, 10 is now in some bi-stable state. Unpacked simulated packets (white noise) and result was a flat spectrum. Is ADC processing suspect. Joe to inject a tone+noise at different frequencies in multiple subbands. Record some data packets and 'playback' in both modes for comparison.
L1 BOF (Mode 4-9) switch to "new" re-quantization gain and digital filter
Action: Richard, to confirm old and new calculations.
Answers match - close this instance - release after other tests
L8 BOF (Modes 10-29) Upgrades
Action: Randy and Jason to acquire snapblock data during maintenance on Friday; Richard to assist.
Complete. [action]
Missing 'Prepare'
Action: Richard to re-run Toney's scripts to confirm it is a Cleo bug or not
Scripts were not supplied. IDed by Ray as being in VEGAS Cleo app. Pass bug report to Ron.