Difference between revisions of "Minutes-7-2-2015"

From GlueXWiki
Jump to: navigation, search
(FDC update)
 
(5 intermediate revisions by 2 users not shown)
Line 9: Line 9:
 
= Agenda =
 
= Agenda =
  
# CDC analysis update (Mike, Simon)
+
# CDC analysis update (Mike, Simon)[[File:MillepedeResults_2July2015.pdf]]
 
# FDC analysis update (Lubomir)  
 
# FDC analysis update (Lubomir)  
 
#* Fe source results [http://www.jlab.org/Hall-D/detector/fdc/55Festudies/XvsZ_Fe.pdf],[http://www.jlab.org/Hall-D/detector/fdc/55Festudies/XvsZprojX_Fe.pdf],[http://www.jlab.org/Hall-D/detector/fdc/55Festudies/XvsZprojZ_Fe.pdf]
 
#* Fe source results [http://www.jlab.org/Hall-D/detector/fdc/55Festudies/XvsZ_Fe.pdf],[http://www.jlab.org/Hall-D/detector/fdc/55Festudies/XvsZprojX_Fe.pdf],[http://www.jlab.org/Hall-D/detector/fdc/55Festudies/XvsZprojZ_Fe.pdf]
#* S&T review pictures [http://www.jlab.org/Hall-D/detector/fdc/doe/FDC_S_TReview.ppt], [http://www.jlab.org/Hall-D/detector/fdc/doe/FDC3_2D_eff_doe.pdf], [http://www.jlab.org/Hall-D/detector/fdc/doe/FDC3_eff_doe.pdf], [http://www.jlab.org/Hall-D/detector/fdc/doe/wire_res_doe.pdf]
+
#* S&T review pictures [http://www.jlab.org/Hall-D/detector/fdc/doe/FDC_S_and_TReview.ppt], [http://www.jlab.org/Hall-D/detector/fdc/doe/FDC3_2D_eff_doe.pdf], [http://www.jlab.org/Hall-D/detector/fdc/doe/FDC3_eff_doe.pdf], [http://www.jlab.org/Hall-D/detector/fdc/doe/wire_res_doe.pdf]
 
# Electronics (Fernando, Chris, Nick)
 
# Electronics (Fernando, Chris, Nick)
 
# Engineering  
 
# Engineering  
 
# Other
 
# Other
 
<!--
 
  
 
= Minutes =
 
= Minutes =
  
Participants: Mike, Naomi(CMU), Sean(NU), Eugene, Luke, Chris, Nick, Simon, Beni, and Lubomir (JLab).
+
Participants: Curtis, Mike(CMU), Sean(NU), Eugene, Luke, Dave, Chris, Nick, Simon, Beni, Sergey, and Lubomir (JLab).
  
== CDC update ==
+
== FDC update ==
  
- Mike showed his latest CDC alignment results using Millepede linked above . The scale varies +/-200 microns, some "structures" in the offsets are visible. After applying the offsets, the chi2 of the track fits improves; Mike will show more results next meeting. Simon's results are different in scale (+/-600microns). Mike will compare the two results channel by channel to see if there's some systematics.  
+
- Lubomir presented two slides summarizing the FDC performance for the S&T Review. Comments about efficiency slide - the worst (package #3) is shown; each chamber reconstructs 3D hit which requires hits in both cathodes (with at least 3 strips fired) and the wire plane. Resolution - to show the average wire resolution; discussions about why two methods give different results for the cathode resolution and what is the effect of the multiple scattering. General remark - needs bigger axis labels.
  
- Beni looked at the pedestals in the CDC varying different parameters: number of samples used, integer vs real number averaging, position of the pedestal window - all this explained in the above link. The main discussion was about the fact that Beni sees some slope (increase), when averaging over all the events, in the region before the real detector signal comes in case the signals are late. Mike suggested this may come from cross-talk from the other channels, but Beni has looked at this. Late signals can happen in cosmic showers (Eugene). Beni will do more studies when the computers are back.
+
- Results with 55Fe source for the avalanche position w.r.t. wire reconstructed from the cathodes. Z (perp. to chamber plane) coordinate calculated from the up/down total charge ratio. X coordinate (in plane perp. to wire) from the crossing of the up/down centroids. After applying centroid corrections and calibrating the channels one can see a hole for the wire and amazing resolution of ~80 microns in X and ~30 microns in Z. Interestingly the avalanche develops at ~130 microns from the wire, while on the drift chamber books it should be below 60 microns.    
  
== FDC update ==
+
- Luke and Lubomir are setting up the spare package in the tagger to scan it with 55Fe source.
  
- Lubomir showed some plots demonstrating his studies of the effect of the centroid reconstruction. First plot shows the rhombus defined by two crossing strips and the wire position w.r.t. middle of the rhombus. If the wire is in the middle, the corrections for the reconstructed wire position cancel out at each point. If the wire is half-way distance from the middle the corrections are maximal, still on average along the wire, they should cancel if everything is symmetrical. Second plot shows the variations of the residuals, on top of them the red curve is the distance of the wire to the middle of the rhombus - we have minimum when the wire is in the middle and maximum when it's half-way from the middle. Obviously there's correlation. 
+
== CDC update ==
 
+
- The next plots show how different the variations of some quantities can be depending on whether the wire is in the middle or half-way from the middle. It turned out the wire picked up for the measurements in the Hall with 5Fe source was half-way from the middle, and the wire for the measurements with the spare package in EEL126 was in almost in the middle. One sees strong variations of the reconstructed wire position or up/down charge ratio for the measurements in the Hall, while there are no such variations with the spare package.
+
 
+
- We looked once more at the corrections (discussed at the last meeting) using different methods. Eugene is concerned for some of the methods the corrections change sign depending on the shape of the signal. He suggested adding also the simplest calculation - just using the middle of the strip.
+
  
== Electronics ==
+
- Mike on the CDC alignment with Millepede: using Monte Carlo he is varying the regularization parameter (that enters as an additional diagonal term) and looking how well the initial offsets are reconstructed. The attached slides above show the results, each slide for a different parameter value. When increasing the regularization term, there's a region where the offsets are best reproduced but then the diagonal term starts dominating forcing all the offsets to be zero. 
  
- All the electronics will be turned on on Monday. Before that Chris and Nick are going to clean the filters from the crates in the tagger; they are dirty. Eugene suggested consulting with the RadCon before duing this due to possible contamination.
+
== Engineering ==
  
-->
+
- No problems reported as a result of the power outage, however most of the equipment is not turned yet on.

Latest revision as of 15:57, 7 July 2015

July 2, 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 analysis update (Mike, Simon)File:MillepedeResults 2July2015.pdf
  2. FDC analysis update (Lubomir)
  3. Electronics (Fernando, Chris, Nick)
  4. Engineering
  5. Other

Minutes

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

FDC update

- Lubomir presented two slides summarizing the FDC performance for the S&T Review. Comments about efficiency slide - the worst (package #3) is shown; each chamber reconstructs 3D hit which requires hits in both cathodes (with at least 3 strips fired) and the wire plane. Resolution - to show the average wire resolution; discussions about why two methods give different results for the cathode resolution and what is the effect of the multiple scattering. General remark - needs bigger axis labels.

- Results with 55Fe source for the avalanche position w.r.t. wire reconstructed from the cathodes. Z (perp. to chamber plane) coordinate calculated from the up/down total charge ratio. X coordinate (in plane perp. to wire) from the crossing of the up/down centroids. After applying centroid corrections and calibrating the channels one can see a hole for the wire and amazing resolution of ~80 microns in X and ~30 microns in Z. Interestingly the avalanche develops at ~130 microns from the wire, while on the drift chamber books it should be below 60 microns.

- Luke and Lubomir are setting up the spare package in the tagger to scan it with 55Fe source.

CDC update

- Mike on the CDC alignment with Millepede: using Monte Carlo he is varying the regularization parameter (that enters as an additional diagonal term) and looking how well the initial offsets are reconstructed. The attached slides above show the results, each slide for a different parameter value. When increasing the regularization term, there's a region where the offsets are best reproduced but then the diagonal term starts dominating forcing all the offsets to be zero.

Engineering

- No problems reported as a result of the power outage, however most of the equipment is not turned yet on.