From settles@mppmu.mpg.de Mon Jun 25 10:54:25 2007 Date: Mon, 25 Jun 2007 10:53:40 +0200 (CEST) From: Ronald Dean Settles To: lctpc@desy.de Subject: Summary of face-to-face meeting@lcws07 04062007 Dear lctpc friends, The WP phonemeeting this coming Wednesday will continue the discussion started at our face-to-face meeting in Desy 3 weeks ago. Here is a summary of that meeting. You have to study the slides for details (see the website below). Let me know if there are any additions or corrections... Cheers, Ron WP#31 Face-to-face LP1 endplate meeting 4 June 2007, 10:00-ca.16:30 Sem Room 4 DesyHH --Agenda of the open meeting ca.11:00-16:30 LP1 face-to-face meeting -LP1 mechanics 1. Dan 2. Peter, Ralf 3. Klaus 4. Ties 5. Leif -LP electroncis 6. Paul 7. Madhu 8. Sascha -LP DAQ 9. Ulf All talks are avialable at the LCWS07 site: http://ilcagenda.linearcollider.org/materialDisplay.py?sessionId=51&materialId=0&confId=1296 Summary -------------------- 1. Dan showed the earlier endplate sideview, then the latest version on slide4. The earlier endview layout and shape of the panels is seen on slide5. After the request to have the panels with the same center of curvature, the result is seen on slides 6,7,9-15; R=1600 above and 1430 below; the drawings have been sent to Akira and Paul. There was a discussion about fieldcage diameter (see below), and it turns out that the stay-clear radius of 357mm is unchanged so that the position of the panels is also unchanged. Dan then described the stress-relief tests (slides16-21). The process consists of 3 stages of stress relief after which a piece is machined closer to final dimensions. After the 2nd stage e.g. the mullions had about 50mum distortion (ultimate goal is 25mum) and have been strengthened. The gas tightness test (slides22,23) has improved after a change in the shape of the 0-ring grove. (Note added by RS: 2mbar overpressure has been used successfully by Star, so we should use the same for LP1.) Still to be specified: -gas inlet/outlet, -temperature stability/cooling, -cable supports, -what else? 2. Peter started (slide2) on the fieldcage inner-diameter which had been reduced to 73.5cm to give the SiLC detectors more space. Since the fieldcage outer surface is not smooth, more leeway is needed. Discussion: This led to considerations about what the final dimensions should be. They were written on the white-board which NN (?) photographed. Here is what I (RS) noted down and would NN and everybody please check: --i.d. of fieldcage 720mm --wall of fieldcage 25mm on the radius=50mm on the diameter --SiLC detectors 35mm on the radius=70mm on the diameter --clearance 10mm ------- --total 850mm Dan's stay-clear diameter of 714mm is therefore still valid. The discussion continued on the number of bolts and pins for fixing the endplate to the fieldcage. It was provisionally proposed to attach with 40 M6 bolts and 4 pin positions (Note added by RS: more than 2 pins at once is overdetermined, and with 2, one pin should be opposite a precision hole and the other opposite a "Siemen's Langloch"). Finally it was proposed that the endplate or the whole chamber should be rotatable up to about 45 degrees (see Klaus Dehmelt's talk below). Peter then showed the strip design (slide3) and endplate sideview (slide4) which was already discussed in Dan's talk above. Calucations of the field homogeneity in the corner region (slides5-8) are in progress. 3. Klaus Dehmelt covered several issues wrt infrastructure. The LP1 support structure (slides2,3), SiLc support structure (slides 4,5), how to move (slide6) and how to rotate for cosmics (45 degrees? see Peter's talk above) are under development. PCMAG (slides7,8,9) will be mapped starting around July 23 with the device shown in slides10,11. Details of the beam T24 are shown in slides12-19. The rates (slide17) seem to be such that there will be less than 1 track/TPC "picture" (~15musec per "picture"). The beam will not be available until mid 2008 (slide 19), so that cosmic running will be essential, and discussion is on-going as to how. Takeshi proposed using existing trigger scintialltors read out with MPPCs. Details of the layout, rotation, trigger counters, etc, must be worked out during the next couple of weeks. 4. Ties showed two schedules under development, one for LP1 and the other for the fieldcage. Items are being filled in, so it is too early to discuss details. For the fieldcage, the plan is to finish the design in June and construction by end September so that it is ready for use by end October. For LP1 we should try to have all bits and pieces together by end of the year, and the details should be scheduled by phonemeetings during the next few weeks. 5. Leif discussed aspects of the pad layout and connectors. The gas-tightness (slide 1) of the multilayer board for the pad plane is an issue to follow closely. In principle (slide2) there can be about 9000 pads and 300 connectors per panel (for 1mmx4mm pads). The layout (slides3,4) is not yet adaped to a curve pad plane; this has to be done as soon as the pad/pad-plane/panel dimensions are frozen. Takeshi pointed out that CDC prototype is with 1.17mmx5.0mm pads and a 6-layer board; here we need more communication to make sure things fit together in the end. 6. Paul covered a) the layout of the LP1 panel with bulk-micromegas+resistive-anode and b) the use of T2K electronics for LP1. a) Now that the panel size has converged, detailed design work (slides3,9,11) will be started. b) An alternative to the new_preamp-ALTRO line is to use the T2K-AFTER approach (slides4-8) which is thought to be cheaper, because ALTRO is not officially specified by EUDET. The proposal is that this should be kept as a back-up, and that (slide 11) panels with ALTRO be prepared at Carleton and ones with AFTER at Saclay. Leif commented that the new_preamp-ALTRO was discussed at the EUDET meeting in Amsterdam January 2006 where it was decided to produce a few hundred channels for EUDET, that 2000 are promised for the end 2007 and that it is too late to change this direction which has proven to be successful. Also it is part of the promising low-noise analog+digital development towards the LCTPC which is being designed by CERN. The ALTRO<->AFTER alternative will be discussed at the LCTPC electronics meeting on 14 June. Paul also described some features needed for the cathode (slide10), an Fe55 source and Dean's UV-produced drift-electron patterns. 7. Madhu described the electronics development at Carleton/Montreal. Details are give in slides2,3,4 and summarzed in slide5: 48 channel VME card with 10 bit FADC derived from new generation development of 12-bit 50MS/s digitizers. A 16-channel prototype is being fabricated which could be made plug-compatible with the ALTRO. Power pulsing will be included in the next generation which could be tried out along with the resistive-anode-TPC high energy SP tests being considered for Fermilab; this SP test will not compromise Carleton's contribution to LP1. 8. Sasha reviewed the Rostock work on time-to-charge conversion TDC approach for LCTPC for which a few hundred channels are part of the EUDET programm. They will be made compatible with the connectors described by Leif. A large number of channels (25000) could be produced for the next phase after EUDET if the present test are successful and this approach is approved by the collaboration. 9. Ulf described the DAQ being developed for the LP, which is based on the ALICE TPC DAQ. One issue is the timing, since we have asked Eudet (JRA1/David Cousins) for 1ns accuracy to demonstrate the time-stamping capability of the TPC/SiLC system. Ties and Sasha propose to measure the trigger time with scintillators and correct jitter arising from the TPC-clock time-bucket quantum. See slides 2,3 for definitions of the symbols. Ulf summarized that it will be necessary to modify the FEC (called mFEC) for the new amplifier and additonal data. The trigger must be distributed to the RCU and the trigger number to the mFEC.The RCUs need modification for the clock/trigger. Standard DRORC will be used and ALICE API/drivers for the DRORC. We must build our own DAQ and interface (slide 6). Brussels is interested in joining this effort. To summarize the tasks (slide 7): Electronics - hardware: Lund Electronics - firmware: Brussels(?) DAQ - readout: Lund DAQ - LCIO data format: Bonn DAQ - common DAQ interface: Bonn DAQ - local control/user interface: Lund(?) Finally, the question must be decided as to which event-display to use.