Minutes of the CALICE-UK Steering Board meeting, UCL, 10/11/04 ============================================================== Present: Jon Butterworth, Paul Dauncey, Chris Hawkes, Mark Lancaster, David Ward, Nigel Watson, Matthew Wing Phone: Roger Barlow, Fabrizio Salvatore Minutes: Paul News: Paul contacted Janet Seed, who (not very helpfully) gave no indication of the allocated level of funds PPARC might have available, but said we should apply for what we needed and could defend. Jon will contact Jenny Thomas to see whether he can get any more information. [Note added after the meeting: Jenny says there is only 500k total of new money currently allocated in FY05/06 for ILC detectors, although this goes up, to over 1M, from FY06/07. This is the planned line and is not binding so it could go up or down. This line is to be split between CALICE and LCFI (and any other projects which might arise). LCFI have been awarded around 200k for the first two quarters of FY05/06 but have no funding approved (yet) beyond that. They originally asked for 400k total in FY05/06. Jon hopes we can argue that the continuation of the CALICE programme is still "seedcorn" and so would then not come out of the ILC detector line. Even then, being restricted to 100k of new money in FY05/06 would be difficult.] The PPRP meeting at which we will be considered will be on 1 and 2 Feb. We will only be needed for one of the two days; which is not yet fixed. We will then be considered for actual approval by Science Committee; the date at which this will happen was not known but Jon will try to find out. [Note added after the meeting: The Science Committee meeting following the PPRP will be on 10 and 11 Mar.] Ken Peach has not yet signed the MoA; this really should be done before the Dec CALICE meeting in DESY. Nigel will chase this up. The situation with Edinburgh is still somewhat uncertain. Paul will meet with Pete Clarke and Steve Playfer when he is there next Wednesday. Paul has been contacted by Mike Tyndel at RAL/PPD who may be interested in joining. Mike has a very strong background in silicon detectors and is a member of the MAPS collaboration. He is meeting with Ken Peach and Nigel tomorrow to discuss PPD's future role. The first two production CERC boards were taken to Imperial last week and tests have started. There are no disasters so far and early results will be presented this afternoon. The rest of the meeting discussed each of the workpackages in turn. DAQ: Before the meeting, Matthew had sent round a draft document combining the DAQ conceptual design and a proposal draft (see web page). He presented the current status of ideas for this workpackage (slides also on web page). One thing which had been mentioned in previous discussions but had been left out here was a study of failsafe FPGA loading. The front-end FPGAs on the wafer PCBs will be inaccessible for the lifetime of the experiment. It is clear the firmware on them will not be ideal on day one and so will need to be upgraded with time. Hence, a way of loading new firmware via the fibre is needed, where any bugs or corruption can be guaranteed to never disable the FPGA permanently. One issue is whether we should bid for a full chain of the DAQ or only to study the parts which seem to be difficult or bottlenecks. It might be hard to argue the full chain would be needed in the next three years. It also needs to be clear what has been achieved by the LHC experiments already; e.g. Paul thought CMS had already demonstrated a network with a throughput of 500 Gbit/s ~ 60 GBytes/s which should be sufficient for our needs. Note, this value was given in Patrick Le Du's talk (page 4) at Durham; http://conference.ippp.dur.ac.uk/cdsagenda//askArchive.php?base=agenda&categ=a041&id=a041s3t1/moreinfo Also, Atlas have several fibre-PCI cards and if one will do something similar to the job we need, then we should simply aim to buy one of theirs for now. Another issue is to what extent we can couple the proposal to reading out future versions of the VFE ASIC. The information from Orsay on the schedule and scope of this work has been minimal so it is still not clear what we could bid for in this area. Overall, the feeling was that the proposal should include: o VFE ASIC readout o Network and/or passive router studies o Failsafe remote FPGA loading studies o Fibre-PCI card studies o Clustering simulation studies Another meeting is needed for a technical discussion on the scope and this needs to be within 2-3 weeks. The date will be set this afternoon when more of the relevant people are present. [Note added after the meeting: this was fixed for Tue 16 Nov at 11am at UCL.] Mechanical: Roger reported that there is interest in Manchester from him, Steve Snow and Ray Thompson. These last two people are currently heavily involved in the Atlas SCT construction but would become freer over the next year. Their Atlas experience is in the mechanical silicon mounting, which requires tolerences of microns. (The equivalent numbers for the ECAL are not known and need simulation to answer; they probably also depend on whether MAPS or diode pads are used.) They have also been involved in the silicon QA and have built up a semi-automatic system for measuring and recording all the silicon module measurements. One suggestion would be to use "second quality" silicon as this could result in a factor of two cost reduction. In this case, knowing where the dead channels are (or at least that they are not correlated) would be even more important. Some preliminary studies by Jean-Claude Brient indicate the energy resolution does not degraded quickly as the fraction of dead channels increases, but this needs further study. The Manchester group also have experience in mechanical installation but this is probably more applicable when we get to real prototyping, which would be beyond the next three years of this proposal. Another area is thermal studies and hence whether cooling will be needed or not. Manchester have a licence for the FLEXPDE heat flow modeling package and experience of using it. This is what Ian Duerdoth used for his CALICE studies a few years ago; he would probably not rejoin the collaboration but would be available for consultation. The talk by Steve Snow in the afternoon would give more detail on these points. One issue is coordinating with the existing work going on in France. A meeting has tentatively been arranged on Mon 22 Nov in Paris for Paul and some of the Manchester people to meet with Marc Anduze to discuss this further. Paul will try to get introductions done via email beforehand so a discussion can then take place to clear the basics before the meeting. The scope of the proposal is therefore still quite undefined. However, it is clear the cost would be almost totally for rolling grant effort. There may be a little equipment money but only at the level of ~5k/year; otherwise there is no new money foreseen. MAPS: Paul had sent round a very rough (V0.1) draft document of the MAPS conceptual design before the meeting (see web page). He presented the current status of ideas for this workpackage (slides also on web page). A major issue for this workpackage is the lack of simulation to defend the concept. It is probably now a higher priority to switch to this work than continue with clustering algorithm studies over the next few months. Simulation: Nigel also distributed a document and showed the latest status of the simulation workpackage (see web page for both). Although it is somewhat arbitrary, we will need to divide the list of tasks between the "CALICE" workpackage and the future simulation and physics workpackage. At a basic level, labelling the ECAL and HCAL data analysis (and supporting simulation) work as CALICE is the obvious division. Also, the two leaders for these two workpackages are not yet defined; it is assumed to be Nigel and David but they are currently sharing the tasks. (It would not be impossible for these two workpackages to have co-leaders if this makes more sense.) For the CALICE workpackage, the HCALs set the schedule and this is by no means firm; the upcoming DESY meeting should give some idea of the prospects for a summer 2005 AHCAL run and a 2006 run with the DHCAL. Paul also said the effort for upgrading the DAQ from the ECAL-only system to be used at DESY to a combined ECAL and HCAL system will not be negligible and needs to be included in the CALICE workpackage. AOB: Matthew, David, George Mavromanolakis and Nick Malden will be able to attend the CALICE meeting at DESY on 7/8 Dec. This is followed by an ILC (i.e. not CALICE-specific) simulation workshop on 9/10 Dec, also at DESY. Paul asked for people to check if their group has a 10 bit, 8 channel VME TDC which we could borrow for the next few months. This is to read out tracking drift chambers (lent to us by Japanese ILC groups) which will be installed at the DESY beam line. Next meeting: This will be in the afternoon of Wed 15 Dec at Birmingham. At this meeting, we need a first draft of the proposal text, together with milestones and deliverables. There are various appendices which are required for the proposal; Jon helped write the PPARC document specifying what is needed and will look for a concise summary of this. The latex files and eps figures for the first draft should be sent to to Paul a few days before the meeting. The primary authors are: o Current CALICE status - David/Paul o DAQ - Matthew o MAPS - Paul o Mechanical - Roger o Future CALICE programme - David/Nigel o Other simulation and physics - David/Nigel We should aim for around 30 pages of text and figures, with appendices in addition. This averages to 5 pages per section, although the more expensive (i.e. MAPS) might take one or two pages more and the less expensive (i.e. Mechanical) less. We will have a second round of text preparation following this meeting and produce a second draft for discussion at the LC-UK meeting in Imperial on Tue 21 Dec. We will arrange a CALICE SB session within this day.