June 24, 2010 FDC meeting
- fADC125 status (Gerard, Fernando)
- ground modifications
- Cathode redesign: finalizing (Roger, Bill)
- Full-scale prototype tests
- tests with fADC FDC Log Book, pages 494 - 500 (Lubomir)
- cosmic tests (Beni)
Participants: Bill, Glenn, Casey, Roger, Mark, Fernando, Beni, Simon, and Lubomir.
- Bill: the PR for the clean room is in place but not for the ACs. We will need the AC, the one for the whole bay area, to start moving there some stuff and do some dirty work that doesn't require clean environment. Teresa Danforth is the person to be contacted and Glenn will help us in that.
- Bill: the cathode production plan is in a good shape: Fast Track procedure will be posted soon. The wire production planing will require more information since we have not done the stringing. Therefore, first we need to start setting up the fixtures from UVA and exercising and then clarify the procedure.
- We will set up the stringing initially in the tent in 126. The other option, F117, is no longer an option since there's not enough space. We decided to use the granite table in 126 for stringing. We are going to put wheels on that table. For the cathode production we will need another two granite tables. Lubomir: According to Eugene there are money for equipment so that we can buy two granite table right now. Glenn confirmed that and said that he could sign for Eugene, who is away from JLab till July 7. The plan is to put the tables directly in 727 B.C.
- fADC125: Lubomir did further studies of the cutoff seen in the spectra from the wires. As suggested by Gerard on the last meeting now a quadratic interpolation is used to find the maximum from the fADC samples. At page 498 in the FDC logbook, shown are the maximum signal distributions separately for all the channels. The cutoff position varies from channel to channel somehow and probably this is the reason we don't see a good peak at the saturation region. Fernando asked if there's a peak in the spectra for the individual channels. Lubomir: there's not enough statistics to see clearly the peaks channel by channel. The spectra from the strips look OK, the signals extend up to 16000 fADC units, the 14bit limit of the fADC. Lubomir will use the new scope and the differential probe to histogram the signals from the wire cards directly and see if there's a saturation.
- Bill proposed a modification of the cathode production procedure: cutting the foil from the transfer ring will be done not at the transfer ring I.D. position but at a larger (by several inches) radius. This will allow to flap the copper foil over the whole periphery of the cathode ring to improve the grounding connections. Thus, no modifications of the cathode g10 ring are needed. Everybody is happy with the new design. The back of the cards will be glued on the copper foil with conductive epoxy. The aluminum ground foil will be glued at one side to the cathode copper foil and Bill proposed to use conductive ink to connect the other side of the aluminum foil to the cathode. Fernando: conductive ink may not be the right choice, we will think about other options.
- Fernando will be in vacations next week; the extra ground cards we need to put the cathodes of the second layer on ground, will be ready on Friday.
- Roger has prepared the drawings of the cathode foils according to the Bill's modifications from the last week, but with the change of the procedure the flaps at the periphery are not needed. Instead, Bill asked for marks to be placed at the periphery to cut the foil. Some of the cutting positions are fixed by the cards, the other positions are not so important. Roger also showed his modifications at the connector region to make sure they are OK. He showed also the modified daughter card with a ground on the back of the card so that it can be glued to the cathode copper foil. Roger will work on the modifications that were proposed today by Bill.
Full-scale prototype test results
- Lubomir is analyzing the FADC data from the last week. Ther results are posted at the FDC logbook, pages 494-500. Shown are the areas that were covered by the test, by overlapping the three cards (for the wires, top strips, and bottom strips) at different positions. While the resolutions from the strips are OK, there are several problems with the results. The wire position as reconstructed from the ratio of the total top to bottom charge is shown at page 499. The wires are closer to the top at the middle of the chamber and closer to the middle between to two cathodes, at the shortest wire side. The wire position variations are up to 1mm. There are no visible variations along the wires within the +/-12cm covered by these tests. People asked if we can turn the chamber other way around to see if there are any gravitational effects. Lubomir: the chamber will not work after such a procedure because of the dust inside; can try this only at the end of the tests. Other possibilities, like frame deformation during the wire winding, were also discussed. The easiest test to be done is probably to scan the wires over their entire length, but requires many different positions of the cathode cards. At page 500, shown is the mean maximum signal from the wires, as a function of the position on the chamber. There's some inhomogeneity in the distribution, representing the gas gain, that has to be further studied. The speculation is that it is caused by the gas leakage. Bill asked if we can study this with an oxygen sensor. Beni: we got one but we need a pump at the output of the chamber because of the high impedance of that device.
- Beni: working now with two layers and two wire cards; need more ground cards to put the cathodes of the two layers on ground. The plan is to test also an old wire card that will go to CMU for their studies. Previous results, posted at page 492 of the logbook, show that the signals on the top strips are bigger than on the bottom for the middle layer too. Beni swapped the top/bottom cards to eliminate the gain differences in the electronics. Beni did also gain calibrations channel by channel using test pulses at the front of the cards. He needed 10V (actually ~3V because of signal splitting) test pulses to see reasonable signals (~50mV) out of the cards. Fernando will look into that after the meeting. Including the gain calibration in the data analysis doesn't improve the results so far.