Minutes of the Feb 03, 2005 US ROD meeting
==========================================

Follow up on pending action items from last week's meeting:
-----------------------------------------------------------

1. UCSB: can we improve humidity to go to lower T in multirod?

   Not we present hardware.  Air dryer not good enough.
   Feeds off building dry air which is not dry enough.
   Investigating other possibilities, but low priority.
   BTW: UCSB runs coolant at -20 C, FNAL at -25 C.


2. CalProfile at UCSB too slow.  Jim to try to remove sleep
   statement, also try local X.

   Done.  Now as fast as FNAL.  Stable, too.


3. Want to read all DCU out, put in root file.  Wim to do in V24.

   Software done.  Wim will try remotely at FNAL.


4. Check all rod monitoring that goes in DCU.  Ryan to figure out 
   and write up, part of single rod procedure.

   Not yet, but Ricardo has some ideas.


5. Standalone redundancy ring check --> Wim, Frederic will provide,
   Puneeth to follow up

   Contacted Wim, needs to check on some HW, will try remotely at FNAL.


6. Defect analyzer crashing, Puneeth to follow up with Valeri, Lionel.

   Nothing news since last week.  Puneeth to follow up


7. Jim to start talking to JT about scripts for ARC/ROD comparison

   Started talking.  JT could do it, but may be saturated with 
   module testing.  Not clear.

Discussion of default settings (led by Yuri):
---------------------------------------------
Settings at UCSB and FNAL different.  Long discussion.
XML file: Jim says most important difference is that 
continous triggers are off at UCSB.  This was after
Wim suggestion to improve stability.  
    ---> FNAL will do the same
Pedestal offset needs to be standardized (same as ARCS)
    ---> VPSP=36

Many of the parameters have to do with online fault finding.
This is not used, so not important (but having them consistent wont hurt)

Then, broadly speaking, there are two kinds of settings
(a) Those that affect the DAQ, e.g., UPDATEWAIT, MAXAQUISITIONERROR, ETC
(b) Those that affect the data processing, e.g., what algorithm to
    use for pedestals, etc.
Wim is the expert on (a), Valeri on (b)

ACTION ITEMS:

8. Yuri to talk to Wim, figure out what to do with settings of type (a)

9. Jim to talk to Valeri, figure out what to do with settings of type (b)


News from FNAL
-------------
MR: Yuri saw crash on V23 at end of scenario.  Jim has also seen this.
    Stable at room T.  Errors at low T: FEC, I2C, Headers.
    Yuri to continue debugging with Wim.

SR: (Ryan) problems with V23 install, solved by doing install from scratch
    from Wim's web page (except for SY127 driver).

News from UCSB
-------------
SR: Fixed by going from V23_new to V23_old.  Wim found bug.  Should be
    fixed if install from his website.

MR: Good news and bad news.  
    Good news: stable at both high and low T (but remember low T at UCSB 
    is not as low as FNAL, and FNAL is unstable at low T only)
    Bad news: Major chiller leak.  Under investigation.

Discussion of Noise
-------------------
UCSB started to look at noise issues.  On both SR and MR common
mode subtracted noise is smaller than row noise, by of order 10% 
in both cases.  In SR stand can still see opens as lower than 
normal noise, but in MR opens have higher noise.
Tried a few grounding schemes, connection of rod yellow-green
wire to building ground.  Will continue these tests.  Also 
improve grounding.

One issue came up: how to compare to ARCs noise since overall
scale affected by unknown and varying laser gain.  Apparently
normalizing off tick marks height did not work for TEC people
(non linear regime?).  Alternative normalize off charge
injection pulse height.  No firm decision taken, we'll certainly
come back to it.

ACTION ITEMS:

10.  UCSB continue grounding tests, will report again next week

11.  FNAL will check if noise from opens is higher or lower than normal.