From GlueXWiki
Jump to: navigation, search

May 27, 2010 FDC meeting

Tentative Agenda

  1. Production
  2. Electronics
  3. Cathode redesign Bill's changes proposed on the last meeting (Bill, Roger)
  4. Full-scale prototype test results
  5. Other


Participants: Bill, Eugene, Mark, Brian, Casey, Beni, Simon, Lubomir, Gerard (on the phone), and Glenn and Fernando (at the end)0


- Bill informed us that Ron Bartek had finished the design and the specifications for the infrastructure of the off-site space were given to the procurement. The information is that a DOE approval might be required. Eugene will contact Teresa Danforth to speed up the process.

- Bill will take care about creating the production flowchart based on the above document.

- We discussed a plan to build a one-layer prototype based on the new designs of the frames, PCBs, and cathodes, and also using the UVA equipment for wire stringing. We will use the first article PCBs and cathodes to make the prototype. Brian will help with the transportation of the wire equipment from UVA; Lubomir will contact Nilanga in this respect, as well. The idea is to set up first the wire stringing at room 117.


Lubomir showed two pictures linked above: with the raw fADC data, showing strong oscillations (coming from the CAEN HV) and after subtracting the oscillations using the unaffected channels (second picture left panel). Right panel shows the ratio of the strip to wire signal as it changes with time. We discussed what is the origin of the delay: top strips coming before, and bottom after the wire signals by about 8ns. By swapping cables, cards, fADC connector we should be able to identify this. The noise is reduced significantly after the correction. This allows to obtain very high resolution from the cathodes. Page 465 of the FDC logbook shows the wire position reconstructed from the cathodes minus the actual wire position. 5 strips were used on each side and the wire position resolution is about 400microns, which corresponds to about 70microns position resolution of the cathode coordinate. If three strips are used (page 464), the resolution is a factor of two worse. Gerard (on the phone) asked what is the algorithm for the strip coordinate reconstruction, and what information is needed from each channel. The best resolution is obtained by integrating the charge from each strip for the time of the first peak as defined by the wire. This probably will not be possible to implement in the FPGAs, but Gerard will work on that and also on extracting the timing information from the fADC. First of all, Gerard will work on the fADC so that the firmware is loaded from the permanent memory when the module is powered up. Now we need to load it from a laptop.

Cathode redesign

Roger is working on the Bill's modifications, hopefully will have them done soon.

Full-scale prototype test results

- Simon showed results of his simulations using Garfield and Heed packages (FDC logbook page 466). For the present gas mixture of 40/60 Ar/CO2 and present field configuration (2200V on sense and -500V on field wires) he showed several events with 90Sr source and also with cosmic muons. After convoluting the signals with 11ns time constant, the signals look very similar to those we see with a scope or with the fADC. The drift time spectra are also very similar to the ones obtained with the full-scale prototype. The long tails above 160ns are coming probably from tracks close to the field wire, but Simon will look into that. Eugene suggested to create an ntuple in Garfield so that one can easily correlate different parts of the spectra with the different trajectories.

- fADC data analyses were discussed above

- Beni was able to reconstruct the wire positions from the cathodes (FDC logbook pages 458 to 463) using conventional ADCs. The resolution is 700 microns. The systematics in the wire position reconstruction seen before is visible now as well: the reconstructed position varies with the coordinate along the wire. Beni still needs to do some work in his code related to the coordinate system. Bill was concerned that the testing the prototype took a lot of time and that the same can happen with the testing of chambers during the production. He and other people proposed possible ways to automate the testing and shared experience from other experiments. We will keep on discussing this at the next meetings.

- Lubomir re-visited the considerations discussed before (GlueX-doc-757-v1) for the field configuration of the FDC. The charge induced on the strips is not affected by the field configuration. The best, from resolution point of view, is to have a configuration where all the field lines from the sense wire are connected to the field wires. This, however, requires to increase significantly the present voltage on the field wires, while the sense wire high voltage can be reduced only slightly to insure maximum efficiency. As Eugene explained, the influence of the field wire voltage on the amplification is much weaker than the sense wire voltage. So, the problem is, based on our previous experience, the chamber will trip at higher field voltage. Still after the chamber cleaning we might be able to work with somewhat higher field voltage.


- Fernando: some minor corrections are still needed and Kim will be ready with the PCBs hopefully by tomorrow. Fernando did some simulations for the heating of the fADC125 and confirmed that the heating is not an issue, so we can power the module as long as we want. He explained that some of the positions in the VXS crate, just above the middle of the fans, have reduced air flow. Fernando was concerned that the software for the fADC125 to be used in CODA (first simplified version) is not ready, while that was promised by the DAQ people two weeks ago.