Minutes of CMSUK-ECAL meeting, 2-Nov-1995

     Minutes of UK CMS/ECAL Meeting; 2 November 1995, RAL

  ------------------------------------------------------------------------


  Present: R Brown (RAL), W Cameron (ICSTM), D Cockerill (RAL),
           J Connolly (RAL), G Heath (Bristol), B Kennedy (RAL),
           K MacKay (Brunel), D Miller (ICSTM), T Virdee (ICSTM)
           R McClatchey (University of the West of England) for item 1.

  Apologies for absence: P Hobson (Brunel), D Imrie (Brunel),
                         S Watts (Brunel)

  ------------------------------------------------------------------------
  Minutes: B W Kennedy / RAL
  ------------------------------------------------------------------------


  1) Possible CMS involvement by the University of the West of England,
     Bristol.

  GH introduced this idea in the UK CMS meeting held in the morning of 2
  Nov.  Richard McClatchey came to the afternoon ECAL meeting to present
  his views on UWE's possible contribution to CMS.

  RMcC was a CERN fellow in the Aleph daq group until 1987.  He has
  industrial experience in database technologies, and is now a lecturer in
  computing at UWE.  The UWE group has expertise in databases, object
  oriented software and distributed systems, and comprises 8 academic staff
  and 2 PhD students, of whom one is currently based at CERN.  In addition
  they have one MSc student, also based at CERN.  The group has been
  involved in the RD38 project for 1.5 years.  They expect the RD38 work to
  tail off after next year, and are looking for further work in the HEP
  field.  The CMS engineering information system seems to match the skills
  available at UWE; other members of RD38 (Le Goff ...) are also active in
  the database project.  The group also works with industry; RMcC gave the
  example of a project with Motorola to develop network management systems. 
  He expects that the group could devote the equivalent of two full-time
  academics to CMS, with a post-doc and a student if funding can be found. 
  They are preparing an ESPRIT bid, and also hope to get EC or PIPPS funds
  for an RA.

  DJAC asked how the UWE plans fit in with CMS timescales: we expect
  crystals to be arriving in the UK from 1998 onwards.  RMcC is keen on a
  long-term involvement.  He foresaw a development project in 1996-98,
  producing a prototype system in 1997, with a "first cut" management
  system early in 1998.

  RMB asked about UWE physics interest.  RMcC's group is entirely  within
  the computing department, so they would not carry out physics analysis;
  however, they are interested in publications on the computing aspects of
  the project.  RMB felt that this would not be a problem as long as UWE
  were closely coupled to Bristol, but the UWE group could not become a
  separate institute in CMS without physics involvement. He will take
  informal soundings of the views of the PPC and the selection panel; our
  future plans must depend on their reaction.

  WC asked about the relationship between Le Goff's group and the regional
  centres.  TV said that Paul LeCoq has asked Le Goff to set up a database
  for the regional centres.  RMcC emphasised that this is separate from the
  CICERO project, but is likely to use similar software techniques.

  It was agreed that the Bristol and UWE groups should develop a proposal.
  Helen Heath will be the principal contact for UWE.

  RMcC then left the meeting, and further discussion took place.  Several
  points came up: (a) it is important that the software group should have a
  stake in the successful outcome of the project, to minimise the risk of
  ending up with a useless system; (b) the software designers must work
  with physicists to specify the system - they cannot expect to receive a
  complete specification at the start of the design process; (c) it is far
  from obvious that extra funding would be available to cover the costs of
  the UWE group - if not, is it cost-effective to use existing funds to pay
  for them ?

  ------------------------------------------------------------------------

  2)  ECAL Engineering & Design, and the long term UK prototyping
      activities.

  DJAC gave a summary of the recent ECAL engineering week at CERN.  The
  three design proposals were shown and discussed at great length, leaving
  all three design groups disheartened.  There was some feeling in favour
  of a decision by the end of the year.

  TV felt that CMS must decide on an approach (ie with or without material
  between the crystals) very soon, but it is not necessary to choose a
  particular design immediately.

  It is clear that both the RAL and Ecole Polytechnique groups regard the
  basket 600 scheme as unworkable, but the groups could work together on
  either of the other designs.  Some physicists were disturbed by material
  between the crystals.  TV is confident that 300microns of carbon fibre
  (or glass fibre), or 170microns of Ti, is no problem.  The
  crystal-to-crystal gap is more important; the nominal design gap should
  be no more than 600microns, and preferably less.

  DJAC then asked about the consequences for the UK groups if the RAL
  design is selected.

  TV felt that the UK has the lead responsibility for the engineering
  design whichever scheme is chosen.  The lead group would have to form a
  team to complete the ECAL design.

  JC felt that RAL lacked the resources to act as the lead group; RAL could
  contribute at most 2.5 sy/y to the mechanical design, while (for example)
  Ecole Poly had approximately 10 design engineers working on the ECAL.
  DJAC reminded the meeting that CMS had been allocated only half of the
  manpower requested, and that this shortfall had a serious impact on the
  role of RAL .  TV pointed out that RAL is perceived by the rest of CMS as
  the lead group in the ECAL design.

  RMB suggested that RAL and Ecole Poly together could form the basis for a
  design team.  Each would develop its own ideas until a rational decision
  could be made between them, at which point the groups would join forces
  to work on the chosen scheme.  If one of the groups (or both together)
  were formally recognised by CMS as the lead institute, they would have
  the authority to determine how the design should progress.  JC is clearly
  respected within CMS, and so would be a good choice to lead the project.
  TV was unwilling to accept Ecole Poly as the lead group, as they have no
  track record in delivering large systems.

  TV was concerned that the RAL and Ecole Poly manpower together still
  seemed insufficient - he estimated 40 person-years to complete the
  design.  RMB replied that RAL had diverted considerable effort into test
  beam preparations and work on the basket scheme.  Some manpower could be
  released by withdrawing from test beam work.

  RMB proposed that progress could be made in the short term at a technical
  board decision meeting - held every 4-6 weeks,  It should be possible to
  decide between the basket design and the RAL/Ecole Poly lattice designs
  quickly.  Assuming that the basket were rejected, DJAC would have to
  coordinate RAL and Ecole Poly, and decide on the extra resources needed
  to finish the job.  TV felt that if DJAC, Jean Badier, Paul LeCoq, and
  Jean-Louis Faure presented a united front, their views were sure to
  prevail.

  3) UK Regional Centre presentation for CMS ECAL week, and the link with
  the engineering & design.

  WC showed drafts of the transparencies for his presentation.  Detailed
  comments were made and noted by WC, and will be incorporated into the
  transparencies to be shown to CMS.

  3)  New questions from our PPESP referee, John Fry.

  4)  AOB

  5)  Next meeting