T-474 phone conference

Friday 21 july 2006, 11:00 PDT, (19:00 GMT+1)

discussion points

• Usefull runs (24 hours of stability) :
  http://efd-mo.win.slac.stanford.edu/SHIFT+LOG/447 http://efd-mo.win.slac.stanford.edu/SHIFT+LOG/451 http://efd-mo.win.slac.stanford.edu/SHIFT+LOG/454
• Make a list of good runs, document conditions, etc.
• Add VSAM data to the ntuples
• Define some "beam quality" cuts -- e.g. klystron trips, feedback hiccups, compressor-induced jitter etc. should be marked
• Another the round of algorithm tuning (two-pole filter seems to look good, so would need to tweak bandwidth a bit)
• Process "real" calibration runs (normal calibrations as opposed to BPMCALI)
• Reprocess the full dataset again.
• Updated analysis: resolution, stability, vibration, temperature dependence, etc.
• Handling interferometer data

minutes

Minutes of the 21 June 2007 T474 Phone conference
-------------------------------------------------

HARDWARE DISCUSSION

- During the meeting some initial discussion between Mike, Yury and
  Stewart started about hardware needs for the future running. They
  agreed to have the weekly T474 phonemeetings alternate between
  hardware and software (analysis). We will start next week with
  hardware issues. By then a list will be put together with more
  specific requirements. The main items on this list will be :

  o electronics for Alex's BPM prototype (prob. use SIS ADCs again,
    but need to investigate whether we e.g. would want to go for
    new 200 MHz sampling ADCs )
  o electronics for a phaselock loop for the local oscillator (LO)
  o magnets, contact Dubna to organise concrete involvement in terms
    of people and magnetic measurements
  o expansions to the Zygo interferometer ? Need to contact Mike
    Hildreth on this issue.
  o ...

- We are currently using more than 50% of CPU power for readout, which is
  according to Zen I/O dominated. We need to think of ways to improve
  this when we'll have more hardware channels to read out.
  Maybe contact Zen to have the data being stored in binary format
  instead of ASCII. This will require some effort in LabVIEW
  programming.

- Some discussion took place on having a the local oscillator phase
  locked. It was agreed that this would be the preferred method to
  take care of the phase systematics compared to doing it by
  corrections in the analysis software.
  To control the phaselock loop, one could think of purchasing a NI
  DAC card which can easily be setup and controlled under LabVIEW and
  doesn't cost much money. This needs to be investigated, need to come
  up with a list what is needed by next weeks phonemeeting.

- Mike brought up the issue of temperature stability, need to think
  wether it makes sens to move all the electronics to the same place
  where the temperature is stable (as e.g. downstairs in the end
  station ?)

- We also want to think of extending the ground vibration measurements
  that were done by Filimon and Mike A. last summer. Especially
  measuringing vibrations on the SPEAR girders would be
  usefull. Stewart brought up the idea that we need to define a
  default set of positions where to take measurements. It's important
  that these positions are still easily accessible after the magnets
  have been installed so we can compare vibration before and after
  installation of the magnets. It would be good to have a student work
  on this, e.g. Mahsa, Yury's new student would be a good
  candidate. These measurements can take place during 2 weeks
  somewhere in August.
  The installation of the magnets can be simulated and measured by
  hooking up some cooling water as this will probably be the biggest
  source of vibration. Once this is measured we can decide whether we
  eventually need making changes to the cooling water system in terms
  of mechanical stability.

- Stewart brought up the issue of the large vibration on BPM4 and was
  wondering whether there might be a holding current on the stepper
  motors. It might be possible to lock it better into position. Or
  possibly move it and then shut of the power.

- Mike Woods and Michele met with people that do magnetic measurements
  at SLAC. We need to dicide on where/how/how many NMR probes. And
  which techniques we will use for the magnets. Need to take into
  account that because of the length of the magnets, fringe fields
  might be very important.
  The idea is to run the 4 magnets in series from the same power
  supply. We need to think what we need in terms of

  o controllers for the magnets (+ what kind of controller do we want?)
  o cable plant

  Should add these items to the list of the hardware meeting next
  week.

- Need to monitor the stray fields inbetween the magnets as there is
  quite some iron present in the girders. So we'll have to install
  some flux-gate magnetometers to monitor the effect of the field from
  the dipoles on the stray fields.


ANALYSIS DISCUSSION

- Need to come up with a list of good runs and discriptions (Filimon)

- Need to define a standard set of dataquality cuts (Stewart)
  o cuts on the PMT spil counter signals
  o klystron trips (eg. using energy BPM)
  o cuts on performance of the feedback system
  o charge fluctuations (compare signals of Q cavities)

  Would be good to have some kind of data quality bitpattern with all
  the different cuts represented by indidivual bits, one can cut on or
  use in a bitmask.

- Need to figure out whether the calibrations can be improved by means
  of some kind of regression analysis. (Mark ?) Mike brought up the
  issue that it would be nice to have someone check temperature
  variations during calibrations as well together with a "historical"
  calibration plot to see how much the scales change over time and
  thus figure out how frequent we would need calibration.

- The plan is to have some runs reprocessed soon and try to get rid of
  the phase systematics by applying an second gaussian filter. With
  the reprocessing this would also include the VSAM data.

- We need a webpage or some form of documentation with exactly which
  is what in the datafile.

- We need a timetag for the datafiles. Currently only the absolute
  time is written at the beginning of the runfile with minute
  precision. Mike suggested maybe to read out something called the
  "PMON" which is some kind of pulse-id counter.

- In terms of optics, we need to come up with a plan on what to
  do. Running with a waist at the center of the chicane proved to
  difficult in the current optics setup. The quadrupole magnets seem
  to be steering too much. So we need to figure out whether we have to
  align them or decide upon a different optics set ?

- Someone needs to have a look to the raw signals from Chris' BPMs
  that were taken with Zen's scope-vi.

- Next meeting will be on thursday, July 27th at 17:00 UK time, 09:00
  PST. Details will be sent later to the list.