Minutes of FT Upgrade meeting held at on Liverpool on 17/9/98 ============================================================= Present: Steve Burke, Rob Henderson, Terry Sloan (Lancaster) Erwin Gabathuler, Tim Greenshaw, Girish Patel (Liverpool) Derrick Hill, John Morris (Rutherford) Jim Kay, Graham Stokes (Daresbury) 1) News on magnet support structure. ------------------------------------ Tim reviewed the current rather confused situation w.r.t. the support of QO, the foremost of the new magnets to be place within H1 during the HERA upgrade. The magnet designs acceptable to HERA all place more stress on the LAr cryostat than is tolerable, regardless of where the support is place within H1 (i.e. regardless of whether the FT is split or the support goes between FT and CT). Action ------ a) Tim to maintain contact with DESY as regards magnet/support design and ensure FT boundary conditions are satisfied; express strong preference for keeping FT as single unit. b) Possible trip to DESY with Jim if necessary. 2) Review of planar chamber design (mechanical). ------------------------------------------------ Jim discussed progress with the planar chambers, in particular the situation with regard to: PCBs Work is concentrating on completing the PCB artwork as this is probably the component with the longest order times. Graham's goal is to have this ready for the end of September. There are no manufacturers capable of producing the full 1.5m diam. PCBs in one piece. One producer (Spemco, Dartford) who can produce them in three pieces has been visited. The delivery time for the three piece solution is 12 weeks at a cost of ~7kpounds. The producer could tool up to cope with 1.5m PCBs at an additional cost of 10kpounds, too much for us and perhaps associated with longer delivery times. Tests must still be done to check that the 0.5mm PCB can take the required HV. The possibility of testing all PCBs prior to etching was discussed. Graham is testing the possibility of transferring a CAD file to the producer, thus speeding up the PCB production process. Action ------ a) Graham to complete artwork. b) Graham to test CAD file transfer. c) Tim to organise PS for HV tests, PCB sample to be made available by Jim. d) Jim to discuss HV test of all PCBs prior to etching with PCB producer. Aim --- Order PCBs before Christmas. Nomex walls Still not certain whether 10mm wall needed (rather than 3mm) to increase stiffness of the gas tank, though Jim's feeling is that this is not necessary. Action ------ e) Jim to contact Geoff Tappern on tank structure. f) Derrick to contact Claire about Forward Proportional Chamber design, reason for thick supporting wall. Infrastructure Jim has agreed with John Fry that the clean room tent currently at SLAC can be transferred to Daresbury in the near future. (Exact date depends on time needed to solve BaBar endcap problems!) The tent is plastic coated canvas on a steel frame of height about 3m and area about 8x8 m^2. This is adequate for our purposes and can be erected in the Daresbury experimental area as soon as it is available, at the latest at the beginning of 1999. We have yet to locate a surface table that we know we can transfer to Daresbury. The equipment needed for testing the planars as they are built is: Gas tank (radial test tank is at Daresbury) Bottled gas (CO2/Argon CO2:Ar = ?) HV power supply, cables (to come from RAL) Feedthroughs for gas tank (unclear what is needed) Scintillator trigger (trigger to cover single cell to come from RAL) Simple readout system, i.e. discrimnators/coincidence units etc. (to come from RAL) Dark current measurement (to come from RAL) Action ------ g) Jim to organise transfer and construction of clean room. h) Jim to find surface table. i) Tim to find out what gas required. j) Jim to check availability of gas at Daresbury. k) Derrick to provide HV supply, cables, scintillator trigger, simple readout and dark current measurement. l) Tim to install test set-up at Daresbury. Tools The planar crimp tool design and old radial crimp tool are available. A new crimp tool must be built and tested (is one enough?). We must find out more about the necessary wire tensioning, how was this done during the construction of the original planars? m) Jim to organise construction and testing of one (three?) crimp tool(s). n) Jim to contact Ray Thompson (Manchester) on details of tensioning. 3) Review of planar chamber design (mainly electronics). -------------------------------------------------------- Derrick showed us some pre-amps, the old planar cell noryl end pieces and the old planar feedthroughs. The pre-amp PCBs have been ordered. Suppliers for all the components have been located. The testing of the pre-amps must be organised soon due to the impending retirement of the expert. The noryl pieces can be made at RAL. Note that three man-weeks are needed for this once the design is complete. Some of the feedthrough PCB artwork must be produced again as the original drawings are no longer available. Derrick mentioned that there is a possible problem with the pins. These are produced by Medelec in Lausanne, but the company is not responding to faxes and has not sent the promised samples, despite having assured Derrick over the 'phone that it would do so. It was agreed that, if necessary, Jim and Derrick should pay the company a visit. Derick has provisionally booked some RAL manpower for the period of the HERA shutdown. Action ------ a) Derrick to keep on at Medelec, visit them with Jim if necessary. b) Don/Derrick to organise pre-amp tests. c) Jim to organise completion of noryl end-piece design. 4) Overall FTD design. ---------------------- It is still not clear whether we will have to shorten the FTD (by removing one radial) or not. However, removing a radial will simplify the upgrade in several respects. Will this compromise the performance of the FTD? Steve Burke reported on studies of the effect of changing the inner cylinder of the FT Gas tank to lead (equivalent to about 5mm of steel in terms of radiaton lengths). This led to about 20% of high Q^2 events containing additinal tracks. There were typically three or so of these. It was agreed it would be useful to have similar figures for additional material placed closer to the beam line, allowing as to look roughly at the effects of material at small radius. This could be done by changing sections of the beampipe to lead. The orientations of the planars were confirmed to be, in order of increasing z, i.e. looking from the interaction point: 1st has vertical wires 2nd has wires 60^o clockwise from vertical 3rd has wires 60^o anti-clockwise from vertical 4th should have wires 45^o clockwise from vertical(?) [Minutes of meeting in Liverpool on 24/6/98 state that this last orientation should have wires at 90^o to the previous chamber, i.e. at 30^o to the vertical, measured clockwise.] The current situation is described in detail in: http://hep.ph.liv.ac.uk:80/~sjm/atlas/planar.html It was noted that the 5th and 7th planar chambers, counting that closest to the interaction point as one, are upside down (i.e. rotated through 180^o, this means the wire stagger is in the opposite direction to that expected). This should not be changed during the upgrade. Action ------ a) Girish and Mike Jines to simulate some high Q^2 events with additional material at small radius. b) Girish (with Steve B's help) to simulate some events and then run through FTREC with one radial turned off, with 4 wires in each radial (4 nearest window). Compare reconstruction results, i.e. efficiency for all tracks, momentum resolution for high momentum tracks (> 10 GeV). c) Rob to recheck desired orientation for new planar chambers. 5) Repairs to radials. ---------------------- Action ------ a) Tim to organise design of necessary PCB. 6) Forward trigger. ------------------- Derrick described progress with the forward trigger. This is to be built of scintillator (already purchased) and wavelength shifter (WLS, already purchased). The trigger will be outside the FT Tank. It is proposed that the track trigger have one layer in front of the tank with 32 fold segmentation (2-fold in r 16-fold in phi) and one layer behind the tank (between CT and FT, the segmentation of which has yet to be decided). There are as yet no results from the FIT, needed in order to help determine what segmentation is necessary. It is proposed that the readout be via Avalanche Photo-Diodes (APDs). Two Hamamatsu APDs with gains of the order 10000 and area 8x8 mm^2 have been ordered (cost about 500 pounds each). Lab tests using scintillator and WLS with APD readout will start soon with a possible beam test at DESY towards the end of the year. Note that work is also needed on pre-amps, cabling, bias and low voltage distribution, environmental monitoring, slow controls and trigger logic (latter to be realised in VME using a board being designed at DESY?). Action ------ a) Don to talk to Manfred about FIT results. b) Don/Derrick to perform tests. 7) New tank end-wall. --------------------- Not clear what is necessary here until situation with magnet support clarified. Action ------ a) See action 1a) 8) Schedule for regular meetings/organisation of group. ------------------------------------------------------- Jim will coordinate mechanical aspects of the project. Don is responsible for the electrical side of things and for the forward trigger. Dave Sankey will look after the DAQ. John will coordinate the software development. Tim is responsible for making sure the whole thing works as promised. It was agreed that this series of meetings be used to coordinate the FT upgrade effort, minutes being produced containing actions on which reports will be given at the next meeting. The meetings will take place monthly. The next meeting will be held at Daresbury. Action ------ a) Tim to organise next meeting. 9) Progress with simulation. ---------------------------- The new planar chambers have now been included in the GEANT H1SIM framework by Girish and Mike Jones. The code to produce FT hits from the GEANT output has been added by Steve Burke. This now needs testing and de-bugging. Action ------ a) Girish and Mike Jones to de-bug new FT simulation. Goal ---- Produce simulated data for testing new reconstruction code before Christmas. 10) Software, in particular ideas on segment finding and linking. ----------------------------------------------------------------- John mentioned some ideas he has worked on in the past with Rob on segment linking. These represent a considerable improvement over the current approach. John stressed his concern over the lack of manpower for software development. At least one additional person is needed to complete the project on time. That person should produce chamber monitoring code and also work on the event display. For more details see: http://dice2.desy.de/~jvmorris/upsoft.html Action ------ a) Tim to talk to Erwin about manpower situation. 11) Last things. ---------------- This is (probably!) Steve Burke's last FT upgrade meeting. I would like to thank him for all he has done for the upgrade project in the last year or so and wish him well in his future career, whether in or outside HEP! (Please inform me of mistakes and omissions in the above as soon as possible.) Tim.