From jang@physik.rwth-aachen.de Tue Jul 11 02:09:11 2000


 As a have stated several times, also in a mail to Ettore, we have at 
 Aachen 1 a complete chain for bonding and testing. For convenience 
 I repeat it here :

 The following list covers what we have/foresee for the
 module tests; it does not cover eqipment for the "burn-in"

 *   2 microscopes for optical inspection
 *   computer controlled machine for alignment tests
     (sensor-sensor-pitch-adapter-hybrid-frame)
 *   computer controlled equipment for measuring I-V 
     characteristics (including switching matrix 
     for up to 12 modules)
 *   readout system (detailed specs follow) 
 *   laser test stand for functionality tests
 *   planned : radioactive source test stand for
               gain measurement
 *   infrastructure : unclassified clean room, N2 or Ar floated
                      Al boxes for storage and/or transport
 For the bonding equipment I refer to Alan Honma's list in the web.


 Concerning the qualification procedures please find below  our 
 comments/suggestions to your draft1, where I assume that it concerns
 the M200 milestone modules and may be altered substantially according
 to experience.
 

 ad 5 : Module testing
 =====================

 5.1  Module test before bonding :

      add :  optical inspection; cleaning
      
      add :  connect/bond the necessary parts allowing for an I-V curve
             (needs only a very few bonds), which should be performed
             preferentially separately for both sensors of two-sensor-
             modules. Cross checks with sensors data base for variation.
             This is necessary at least in the beginning until experience 
             has been gained; it allows for tracing back failures
             introduced by e.g. handling, also during the glueing
             procedures, transport, storage, and the automatic bonding  
             (by comparison of the I-V curves before and after).
             It should be discussed wether at this stage some temperature
             cycling should be performed : it has been observed that
             the mechanical stresses lead to enlarged currents, the effect
             neither being understood nor having been studied yet 
             systematically, so it is not clear whether repeated 
             applying stresses and relaxing leads to microcracks   
             which finally may reduce the maximum bias voltage 
             considerably.

 5.2.1       o.k.

 5.2.2       the radioactive source should be of the same kind 
             in each module testing lab, preferentially Ruthenium106
             (because of its high electron energy of 3.5 MeV :
             it eases the calibration to mip considerably, as 
             compared to Strontium90). 

             otherwise o.k.

 ad 6 : Module Burn in
 =====================

             We will need from time to time during burn in (i.e. 
             during temperature cycling: between +20 degree and -10 
             degree ?) an I-V curve in order to see whether eg. the 
             maximum bias voltage has changed or significant changes 
             occured in the shape of the I-V curve.
             A somewhat  unsolved problem is the following : We are 
             going to bias (for two-sensor-modules) both sensors in 
             parallel and it might well happen that some systematic 
             failures  concern preferentially one of the two sensors, 
             e.g. the one far from the hybrid. How are we going to 
             realize that in order to be able to investigate the 
             cause of the failure in a determined way?
             


 ad 7 : System test
 ==================

             It seems premature to talk already now about any details
             of system tests, which depend, of course, in addition to
             what you mention also on the availability of the support 
             substructures, for the FWD petals including cooling and 
             interconnection boards, which we are going to have 
             probably by mid of next year.

 ad 9,10 : HW setup and SW tools
 ===============================

            In the last but one meeting it was agreed that for the
            beginning each module test center uses the system which
            is available at that center. 
            Do we really need the standard system at each center 
            already for the M200 ? Comparability (especially on
            the signal/noise ratios) could be achieved by referring
            to the radioactive-source "mip"-measurement.
            


            Greetings

            Wolfgang