From lenny@fnal.gov Mon Jun 21 10:16:06 2004
Date: Mon, 21 Jun 2004 12:07:45 -0500
From: Lenny Spiegel <lenny@fnal.gov>
To: Anthony Allen Affolder <affolder@neutrino.physics.ucsb.edu>
Cc: Lance Simms <simms@neutrino.physics.ucsb.edu>,
     Lisa Chabalina <elis@fnal.gov>, Marcel Demarteau <demarteau@fnal.gov>
Subject: Re: Hybrid Tester Questions


> Could you please send your list of hybrid tester questions
> around?   I'd like to be able to answer or find solutions for them prior
> to tommorrow's meeting.
> 

Tony,

   I have a few concerns regarding the XML files produced by the hybrid test
stands now in operation at CERN, FNAL, and UCSB. Most of these are
philosophical or might even reflect a misunderstanding on my part, so please
bear with me. The concerns are:

  1.  It's not clear to me if there is a single person associated with
      the hybrid XML files. That is, Radicci:Gantry XML, Costa:Bonding XML,
      ?:Hybrid Test XML.

      For the ARC and LT systems, even if they don't have a single contact
      in the sense that that person would write the generating code,
      document the XML file, and interface with the DB group, someone
      should at least give a green light for uploading files to the DB.

  2.  Similarly, documentation on the hybrid test XML file does not appear
      to exist on a single, standard site.

  3.  The criteria* for passing a hybrid appears to be different in
      the U.S. from that at CERN. Lance has indicated that this is
      an oversight and will be corrected soon, but I still think there
      is a link between this problem and points 1 and 2.

  4.  TESTWITHPA_val is set to :0: for failed hybrids. This means that
      the FAULTY status of the hybrids is valid even for failed hybrids
      and the DB offers no protection against their inadvertent
      assembly into real modules.

  5.  There has been a known issue with PLL's failing at -20. At the
      moment the hybrid test program is not capable of positively
      identifying the problem - Lance is working on this - but there
      also does not appear to be a flag in the XML file that would
      indicate the PLL condition.

I would have a similar set of questions and concerns for all other test
system XML files. 

Lenny

 

>A hybrid will fail if
>
>It has one or more shorted channels
>
>Or
>
>It has five or more noisy or open channels
>
>------------------------------------------
>
>Upon checking out the vis called teststation_report.vi and
>teststation_final_report.vi in Bruno's latest release of the code, I have
>found that he is now using a different set of rules.
>
>For "TIB_P_4D", "TOB_P_4U".."TOB_S_4D", "TEC_4U"
>
>One or more shorts will fail the hybrid
>
>0-2 noisy or open channels means an "A" hybrid
>3 or 4 noisy or open channels means a "B" hybrid
>5 or more means failed
>
>For  "TIB_S_6U".."TID_S_6D", "TOB_P_6U", "TEC_6U", "TEC_6D"
>
>One or more shorts will fail the hybrid
>
>0-3 noisy or open channels means an "A" hybrid
>4-6 noisy or open channels means a "B" hybrid
>7 or more means failed