From paola.giannetti@pi.infn.it Mon Sep 18 02:17:56 2006 Date: Wed, 09 Aug 2006 23:37:18 +0200 From: Paola Giannetti To: Mauro Villa , Livio Lanceri , Luciano Bosisio Cc: cardoso@fnal.gov, Mauro Dell'Orso , Francesco Forti , Stefano Bettarini , Giuliana.Rizzo@pi.infn.it Subject: Re: informations about test team boards Dear Mauro, Livio and Luciano I met Guilherme Cardoso at Fermilab and it was very useful. I try to summarize below answers to our questions. If I misunderstood something Guilherme can correct me: Paola Giannetti wrote: > Hello, > I am Paola Giannetti from INFN Pisa. I am part of the SLIM5 project > that will need soon a test beam setup similar to the one used by > BTEV. I know that Dario Menasce spoke with you about our future > needs. I am in Fermilab now and I would like to meet you to > understand if we can buy with your help each piece of the DAQ system > you used on your Btev test beam (the old version that already worked > and the italian group is using on their CMS test beam). If possible we > would like to use your DAQ setup as it is now for Lab tests, to > practice as soon as possible. The DAQ setup they have ready for strip silicon detectors has not been used at test beam (at test beam they just used pixels) but it has been used for lab tests (as Trieste would like to do). We can buy a complete system. Guilherme will provide us the offer. > We also need a modified version of these pieces for the future test > beam, so we wonder if we can commit some work to Fermilab or to a > company you already know since it worked on the old version. In the > case it is possible we would like to have a rough estimate of the cost > (we would ask a more precise quotation when the work is better defined). Guilherme group can work to produce a modified system, even firmware changes (firmware is not in their hands, but they can take the code from the italian student, Milan group, and make changes). The cost will be very dependent on which piece has to be produced (see below a more specific description). > I attach a file with specific questions on all we need. We would like > to know for the set-up as it is: > >1) price and availability of each piece of the old DAQ chain: hibrid module with pitch adapter, the board patching the connectors, the mezzanine, the PTA. > > Guilherme will provide us the offer for all the pieces.. >2) FPGA firmware source code: for the mezzanine and the PTA. > Any description/manual of the code (if not on web). > > This is available, but it is the work of an italian student going to Cern in a short time, it is not under Guilherme control. The student cannot help us but he can for sure provide the old code, working in the past. >3) Is it possible to access very specific board documentation/information? > Schematics, layout, cad used to design each piece. Any documentation which > might be usefull in addition to what is on web. > (On web: descriptions, pdf schematics and layout) > This would help a lot also to define the modifications we > need to design new pieces for the test beam. > > Since Fermilab is available to work on the system we did not discuss this possibility. We should just define what we need in a document and they will do the work. Guilherme suggests to design in Fermilab a new system optimized for our application. The old system cannot be easily used because it has strong constraints (see below). >Extension/developments: > >4) To our knowlege the mezzanine has been tested connected with 2 hybrid > modules only, but on the manual it is said that it can be connected up to 4 > modules (on board there are 4 connectors usable). > It will be important for us if we can connect 4 modules per mezzanine. > > The limit comes for the available number of connections. If we plan to use all the six available output signals per chip (12 differential lines) the connector on the hibrid itself is too small (12 lines x 6 chips = 72 lines not available). Only if we use a single output (or few) signal/chip we have enough lines, but the system becomes too slow. Guilherme suggests to design in Fermilab a new hibrid module with new pitch adapters that will solve also mechanical problems and will be provided of a good connector going directly with a flat cable to a PTA for initial tests and to the EDRO for final work. The EDRO FPGA should include the firmware developed and tested with the PTA multiplied for the number of cables it has to handle. >5) Can Fermilab work on possible minor modifications of the boards > they designed and if yes how do they evaluate the costs? >[possible modifications: more powerfull FPGA for the mezzanine (if that >constitutes the limit to read 4 modules, see question 4) or update of the firmware] > > They can work on the new system. For the cost we really need to define what we need. If I understand correctly the hibrid/pitch adapter are much less expensive than a modified mezzanine or PTA. >6) Can Fermilab work on possible major modifications of the boards >they designed and if yes how do they evaluate the costs? >[possible modifications: adapt the mezzanine to the MAPS read-out, which > likely will mean at least a change in the firmware, if not more] > > Yes they can do it. >7) If Fermilab cannot work for the modifications, and if we >are CAD compatible, can we have the projects to implement >efficiently in Italy the modifications we need? > >8) Can Fermilab help us to get in contact with companies that already developed >some old pieces for possible changes? In particular for specific parts like the >pitch adapters. > > > This is not necessary, because they can work on the project. Guilherme will be at 12th Workshop on Electronics for LHC and Future Experiments. 25-29 September 2006, Valencia SPAIN. If some of you is planning to go it is a good occasion to plan the modifications. We can also communicate via e-mail and telelink. We should move fast. The system is complex and will need a lot of work. Ciao Paola