Fiber Tracker Meeting Minutes
July 1, 2003 - 10am, Farside
- Calibration/data quality : Warchol
- talk slides
- problems with CFT X8
- what is in the PDAQ "mono" plot? It checks that channel numbers are read out
in order (monotonically increasing)
- SDAQ run - plot of pulse height distribution for 6A5 with zero threshold on
- 6A4 also has problems
- if look at pulse height distribution in the data, problem is there too
- when this board doesn't work, problems also show up in discriminator plot
- also shows up in vertex examine
- problem comes and goes
- 6A5 board was swapped, but new one is even worse sometimes
- the next day the sequencer was swapped, but it is still not fixed
- FEB: crashing SBC (loses connection to routing master)
- problem shows up in FEB utility as all boards being red
- has to be fixed by DAQ shifter - CFT shifter doesn't need to do anything
- temperature readback craziness
- manifests as intermittant major alarm
- fix is to pull out bayonnet, wiggle board, and replace bayonnet (experts only)
- Fred: more on 6A5:
- looks like all errors are HDI bus bit errors
- on 8-bit bus, looks like four lowest bits are all acting up
- in the past the problem has been on the AFE board (bit of solder, etc.); this time
it's not on the AFE
- might be a pinched cable
- requesting a ~4 hour access to plug in a spare cable
- during today's short access, will try a virtual backplan swap
- CFT Electronics Status : Borcherding
- spare boards have all been modified
- green dot: modification to unused channels
- one board has been tested - fixed 8/9 channels (last channel is channel 0,
so it does not go through the MCM; bigger job to fix)
- red dot: an almost backwards compatible modification to allow future running at
higher speed
- Alan: Any progress on split/moving pedestals? no; don't know how much time is
being spent on it
- Testing of the TriP chip at DAB : Estrada
- talk slides
- typo on page - uses a packaged chip, not 8 MCMII's
- first VLPC spectrum
- used the highest gain rebuilt cassette for this test
- each color is one channel, displayed are just a few channels on one chip
- fits to the spectra
- mysterious deformation around 128 counts
- something to do with the conversion? don't really know
- present in multiple channels, but not all
- measure a gain of around 21 counts/p.e.
- other typical parameters:
- pedestal width = 0.18 p.e.
- gain dispersion = 0.17 p.e.
- pedestal spread (RMS of 16 channels) = 0.05 p.e.
- note: histogram is not smoothed
- Z measurement - considering possible modification to take difference between
pulse arrival time and crossing clock to measure Z
- can only get timing measurement when the total (direct+reflected pulse) number of
photo electrons is above ~5
- some tunable parameters were changed - things got a little better
- need feedback from tracking people now to see if this would be useful
- conclusions and plans
- TriP still looks good
- for the moment , get sigma = 58cm for the z measurement
- when Paul returns, start 396nsec timing studies
- "undershoot" (our hobby)
- a big pulse will make the VLPC have a bias current lower than the bias current
with no pulse for something like 40ms
- even more confusing, the characteristic time is not constant for different pulse amplitudes
- Greg: if you integrate under the pulse and across the shoot, is the sum positive or negative?
very negative
- CTT status : Drew
- one of the links from downstairs to crate x13 is losing sync, needs SCLinit to recover
- happens about every 20 minutes
- does not affect triggering, just trigger studies
- will need to be solved before next trigger list is installed
- Status and plans : Borcherding
- trying to make short access to swap cables, longer access to add cable
- more boards with modifications need to be tested
- several things being planned for August access, including:
- major modifications to LV power supplies: remote resetter and quick-connects
- modification of all remaining boards to fix unuased readout chatter
- And Now for Something Completely Different - SciFi @ MICE : Bross
- talk slides
- MICE = Muon Ionization Colling Experiment
- international workshops towards a neutrino factory: NUFACT
- cooling needed to reduce emittance of muon beam
- current time scale : data in 2006?
- for VLPC readout, they hope to use a TriPII chip