March 1, 2018 Drift Chamber meeting
- Instructions for Bluejeans meeting connection
- Meeting ID: 290664653
- Current run analysis/issues:
- CDC thresholds
- CDC gas issues
- FDC average charge vs drift time
Participants: Naomi (CMU) Eugene, Beni, Alex B., Simon, Sergey, Alex A., and Lubomir (JLab).
Discussions about the current run
- Naomi proposed several options: setting the CDC thresholds from the widths of the pedestal, scaling the current thresholds, or doing dedicated threshold scan with 4 different thresholds same for all straws. After lengthy discussions that included all kind of topics, we agreed that it is worth doing the threshold scan but after we figure out what is going on with the CDC gas leakage. Initially, the observable to be optimized was signal/noise as taken from the amplitude spectrum (if you can fit the noise and the signal reliably). At the end of the discussion we concluded that it is better to have enough statistics to get the rho yields (like 10M triggers per setting). Sergey pointed out that we should not really be calling the initial "noise" peak "noise". With the cosmics runs we see a real noise peak from the electronics - the earlier threshold scan would have set the thresholds to make this noise level acceptable. What we are seeing now is (presumably non-hadronic) background. Later addition from Naomi: it would be best to do the threshold scan when the atmospheric pressure is stable, so that the changing gain does not confuse the analysis.
- Beni: excluding the hits with times before the expected signal may improve the tracking (the wire-based one) and then the effect of the threshold may not be so important. Simon will re-analyze the two runs that Justin used to compare the rho yields with different thresholds, using a more appropriate software time window on the CDCdigihits.
CDC gas issues
- Naomi showed amplitude spectra for different regions of the drift time spectrum and compare them for runs in December and now. Her conclusion is that we are losing some charge, possible due to oxygen contamination in the gas as Beni suggested this morning. On Monday during the access Lubomir will measure the O2 contamination. We will also sniff around the CDC. If needed we may increase the flow and re-measure the O2.
- Lubomir showed the average charge vs drift time for the FDC that is indicative of electron losses due to electronegative contamination. The relatively small slope was always there and is related to the way the charges are collected by the electronics, also because only one peak is readout. Important is to compare such plot for the CDC for runs in December and now.
- Cody is working on implementing multi-peak mode. The way it works now, without comparing to the threshold, is not very useful. Naomi: only the first peak is calculated precisely. We will still have to be very careful with the thresholds. Lubomir: Still, having second is better than nothing. Eugene; the best is to use raw mode and evaluate all these effects. It should be done with the nominal high current even with a dead time close to 100%. We plan to have such run with ~10M trigger.