Difference between revisions of "Minutes-7-30-2015"

From GlueXWiki
Jump to: navigation, search
 
Line 18: Line 18:
 
#* gas system [http://www.jlab.org/Hall-D/detector/fdc/gas_system/CO2_common_gas.pdf]
 
#* gas system [http://www.jlab.org/Hall-D/detector/fdc/gas_system/CO2_common_gas.pdf]
 
# Other
 
# Other
 
<!--
 
  
 
= Minutes =
 
= Minutes =
  
Participants: Curtis, Mike (CMU), Eugene, Luke, Dave, Chris, Simon, Beni, and Lubomir (JLab).
+
Participants: Curtis, Mike (CMU), Eugene, Luke, Beni, Dave, Chris, Nick, Simon, and Lubomir (JLab).
  
 
== CDC update ==
 
== CDC update ==
  
- Beni found the CO2 MFC for the CDC was not working. As a result there was only Ar in the CDC, explaining the max HV of 1kV. The MFC will be sent for fixing and two new MFC will be ordered. Scot is working on that but the time frame is not clear yet. In the mean time we decided to connect the CDC and FDC mixing tanks together, and fill both with the FDC MFCs set to 50/50 Ar/CO2 gas mixture.  
+
- Beni: grounding with a thick cable of the HVBs doesn't change anything in the noise: a typical noise event is shown above. Chris and Nick tried both connecting through 50 Ohms or without 50 Ohms (by removing it on some of the boards). Still the idea to be tested is to connect the grounds on many cards together.
  
- Mike showed more plots demonstrating the azimuthal asymmetry in the drift times for one "bad" straw. The time-to-distance is a little different in different directions. The main difference is in the range and, if time is converted into distance, the wire appears to be shifted by ~1mm from the middle in horizontal direction. Beni suggested to divide the tube into (at least three) parts along z to see if this is a real offset of the wire.
+
- Nick: One HV channel trips. For now Nick checked that it doesn't come from the CAEN, or HV cable. Next check is to swap the HVB cards.
  
 
== FDC update ==
 
== FDC update ==
  
- Lubomir showed latest cathode resolution results: now we have reached the design resolution of 200 microns. The problem was in the centroid calculations - fitting the signals improves the resolution significantly. The resolution extracted from the wire position improved from 150 to 80 microns, while the directly measured resolution using tracking improved from 250 to 180-200 microns (first plot). There is still systematics in the residuals as function of the distance to the middle of the strips. For the wire-estimated resolution (2nd and 3rd plot) it depends on the wire position (as expected) and varies as sine function with an amplitude of up to 130 microns. For the tracking-estimated resolution (last two plots) this amplitude is 4 times smaller, as expected. The problem is that in both cases there's a left-right asymmetry w.r.t. distance to strip middle - a slope of ~100 microns per strip width. One certainly can correct for this and further improve the resolution.
+
- Lubomir showed cathode resolution as a function of the charge (first plot). Normally, it gets better at higher charges. This is the case for the resolution in x (wire-reconstruction resolution) which changes from 300 down to 220 microns. The resolution in y is almost constant with the charge. This indicates, there's a contribution to it at a level of <180 microns. Part of it may come from the tracking, but significant part is coming also from another source. Lat two plots show the cathode resolution (residuals with an offset for better colors). One clearly see the different foils on the cathodes. Also one can see  the channel-by-channel gain calibration (done by Vlad) may help to improve the resolution.
 
+
 
- Luke showed more results with 5FFe source (linked above), now varying also the field HV. One can see that the distance between the peaks in direction perpendicular to the wire plane increases for smaller charges - no explanation at all.  
+
- Garfield simulations showed that screening from the wire creates the asymmetry in the signals induced on the two cathodes, depending on the side w.r.t. the wire at which the avalanche is created. This is demonstrated in the plots above.
  
- FDC is at almost the nominal HV, 2150V. For some of the channels it took a week to condition them after the power outage.
+
- Spare package: we took data with a random (pulser) trigger with the standard DAQ prepared by Sergey. We see iron source signal an every 200 events, still good enough to have some statistics now over much bigger area. After taking some data we changed the gas from 60/40 to 40/60 Ar/CO2.
  
== Electronics ==
+
== Engineering ==
  
- Beni looked at the FDC noise - it's systematically smaller than the noise in the CDC. Beni wants to try again to improve the grounding by connecting the grounds on the HV board together with a thick cable. Will discuss this with Fernando.
+
- When we decided to connect the CDC and FDC mixing tanks, wrong connection was used resulting in filling the FDC only with Ar and putting all of the CO2 to the CDC. Dave fixed that and now the two chambers are approaching slowly 50/50. All this was checked with the CO2 sensor - see the plot.
  
-->
+
- Dave is working on the interlock that shuts off the MFCs at the input of the FDC and CDC in case of pressure problems. For now, after such event it can be cleared only manually.

Latest revision as of 18:51, 31 July 2015

July 30, 2015 Drift Chamber meeting

Connection

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

Agenda

  1. CDC update (Mike, Simon, Beni)
    • CDC wave form
  2. FDC update (Lubomir, Luke)
  3. Electronics (Fernando, Chris, Nick)
  4. Engineering
    • gas system [8]
  5. Other

Minutes

Participants: Curtis, Mike (CMU), Eugene, Luke, Beni, Dave, Chris, Nick, Simon, and Lubomir (JLab).

CDC update

- Beni: grounding with a thick cable of the HVBs doesn't change anything in the noise: a typical noise event is shown above. Chris and Nick tried both connecting through 50 Ohms or without 50 Ohms (by removing it on some of the boards). Still the idea to be tested is to connect the grounds on many cards together.

- Nick: One HV channel trips. For now Nick checked that it doesn't come from the CAEN, or HV cable. Next check is to swap the HVB cards.

FDC update

- Lubomir showed cathode resolution as a function of the charge (first plot). Normally, it gets better at higher charges. This is the case for the resolution in x (wire-reconstruction resolution) which changes from 300 down to 220 microns. The resolution in y is almost constant with the charge. This indicates, there's a contribution to it at a level of <180 microns. Part of it may come from the tracking, but significant part is coming also from another source. Lat two plots show the cathode resolution (residuals with an offset for better colors). One clearly see the different foils on the cathodes. Also one can see the channel-by-channel gain calibration (done by Vlad) may help to improve the resolution.

- Garfield simulations showed that screening from the wire creates the asymmetry in the signals induced on the two cathodes, depending on the side w.r.t. the wire at which the avalanche is created. This is demonstrated in the plots above.

- Spare package: we took data with a random (pulser) trigger with the standard DAQ prepared by Sergey. We see iron source signal an every 200 events, still good enough to have some statistics now over much bigger area. After taking some data we changed the gas from 60/40 to 40/60 Ar/CO2.

Engineering

- When we decided to connect the CDC and FDC mixing tanks, wrong connection was used resulting in filling the FDC only with Ar and putting all of the CO2 to the CDC. Dave fixed that and now the two chambers are approaching slowly 50/50. All this was checked with the CO2 sensor - see the plot.

- Dave is working on the interlock that shuts off the MFCs at the input of the FDC and CDC in case of pressure problems. For now, after such event it can be cleared only manually.