From GlueXWiki
Jump to: navigation, search

March 12, 2015 FDC+CDC meeting


  1. Instructions for Bluejeans meeting connection
  2. FDC meeting ID: 290664653
  3. To join via a Web Browser, go to the page [1] https://bluejeans.com/290664653.


  1. Gas issues
    • CO2 monitoring [2],[3]
    • FDC gain measurements
  2. CDC update (Mike)
  3. FDC update [4], [5] (Lubomir)
  4. Electronics: logbook entry 3325323, 3325070 (Fernando, Cody, Chris, Nick)
  5. Engineering (Dave)
  6. Other


Participants: Curtis, Naomi (CMU), Sean (NU), Fernando, Dave, Chris, Nick, Simon, Beni, and Lubomir (JLab).

Gas issues

- CO2 was monitored for another 4 days: the first plot is with a correction function with a power of 2.4, second with 1.4, as explained in the previous meeting. First, the sensor was connected to the CDC for one day(Thursday-Friday), then moved to EEl126. There the bottle was almost empty and as it turned out the CO2 was lower than 60%. On Sunday the bottle was replaced and by Tuesday the values were close to 60%. At that time the fan failed - as the instructions say it has 1000h life and we were using it one month already. We didn't learn a lot from these measurements, except that the change of the gas factor, done a week ago, showed almost no effect (within 0.5-1%, as expected) on the gas mixture. If we want to continue monitoring CO2 we need to buy another pump ($200) that will work for another month, or try to find a better pump. Beni suggested using the pump that Bill made for the O2 measurements.

- Beni did different changes in the settings of the mass flow controllers used for the CDC mixing and was looking at the maximum drift time. Changing the gas factor didn't show any difference (within 1%). Then he changed the controller mode from "mixing" to independent flow control, also made the flow rates in and out of the mixing tank very similar. With the latest modification Beni found a slight change in the drift time but in the opposite direction, i.e. shorter. Next steps: change the gas factors to 1 to see if it will make any difference. Beni suspects something is wrong with the controller boxes.

- Lubomir did measurements in the Hall with an iron source attached at the downstream side of the FDC. There were two sets of measurements (Friday and Monday) at different atmospheric pressures, both sets at three HV values 2100-2150-2200V. Results will be presented on the next meeting, but we do see indeed higher amplitudes in the Hall, compared to the measurements with the spare package, consistent with the lower CO2 percentage.

FDC update

- The chiller is working fine so far. There was a problem with one HV channel (FDC2, Cell1) that tripped on Wednesday night and then it took three days to condition it and set it back. In the plot attached you see that on Wednesday the humidity was >55%, which combined with the water on the stairs from the melting snow and the fact that the AC was not working at that time, may have caused the problem. Fortunately, next three days the humidity was as low as 20%.

- Sean was interested in the progress with the FDC alignment. Lubomir: the procedure is first, to align all the elements withing the package: 8 parameters per cell or 48 per package. Three packages have been aligned already and it will take one more week to align the last one. It turned out that the procedure converges only if there's enough statistics for each iteration, at least a quoter of the total, that's why it takes time. The last step will be to align the packages between each other, but there are only three parameters per package, so it will work faster.

- Today we did (together with Dave and Casey) a small modification in the spare package. We used the opportunity to look at the cathode: as seen from the picture above, there are no hints for corrosion in the place where the O-ring goes over the connector traces. The spare package was sitting in EEL126 for two years.


- Beni checked the fADC crates with pulser (Mike's code) and found ~10 bad channels in the FDC. 6 of them had some connector issues and were fixed by Chris and Nick, 4 are still bad (first log entry).

- Lubomir looked at the data (first runs after the chiller was installed) and made a list of bad channels (second log entry linked above). There's nothing new or unexpected, these are 0.6% of the total channel number. Noticeably, there are 8 new bad wire channels sitting on one ASIC in the pre-amp. Fernando wanted to have a code, similar to the one for the fADCs, to check the TDCs with a pulser. Bryan has implemented already this feature in his libraries.

- Beni: one FDC crate (rocfdc9) is crushing from time to time, he suspects broken SD.