Range Stack Cable trimming Procedure

Procedure for Trimming the Bulk Range Stack Counters

History of the document

• This Web page
  Modified by Sash Kushnirenko
• File name: /e787d/e949/tr2/doc/E949-rc_trim.doc
  Modified by Takeshi K. Komatsubara on 17 Sep 2001
• Original file: /e949d/e787/tr2/doc/rc_trim.doc
  Created by RMcP on 9 Jan 1995

Procedure modifications

People and contacts

• The data sets were taken and analysed by Takeshi Komatsubara , Steve Kettell , and Sasha Kushnirneko .
• Step B was analyzed by Sasha Kushnirenko and Takeshi Komatsubara.
• Step C and Step X were analyzed by Takeshi Komatsubara.
• Step D and Step E were analyzed by Sasha Kushnirenko.

Procedure

Goals

Software

Step A: Establish the basic setup

1. Unplug the 6 cables from the BV TDC in slot 12, and plug in the 6 trigger test cables.
2. Use the SSP namelist with TDC (cs1) and RSTDC (cs2), without TDs (cs3,cs4) and without CCDs (in cs1). The current one is cs1_2_noCCD.cfg
3. We keep D3 on, B=10kG.

Step B: Calibrate the TDC

1. Use the pulser trigger with the beam off.
2. Connect a chaben cable to a free pre-scaler output of the pulser trigger to use as a test pulser.
3. Taking data to disk, run the pulser output sequentially in each test cable input (1-2 seconds is fine).

   Hint: Remember the rule white-to-the-right So when cable you testing is facing you, red cable is on top then the white wire of probe cable should be to the right.

4. Process the data
   • cd StepB
   • Create directory e.g. mkdir new_trim and get there.
   • Edit command script. You can use the following script as an example.
     Environment variable PREFIX defines the names of the output files.
   • Run the script. As a result you will get:
     $PREFIX.hbk Ntuple with all 96 channels. Look it through
     $PREFIX.klg KOFIA log.
     $PREFIX.dat 96 calibration constants for BV TDC
   • File $PREFIX.dat will be readout by programs in next steps.

Step C: Measure the MT2 arrival time at the Range Cards

1. Disconnect all the MT2 inputs from the RC's, and plug them into the first two test cables:

         MT2 Sectors  1-12  =>  Cable 1(1:12)
         MT2 Sectors 13-24  =>  Cable 2(1:12)
    

2. Set trigger to kbeam
3. Take 30K kbeam events to file
4. Process the data:
   • % cd StepC
   • Create directory and get there
     % mkdir new_trim
     % cd new_trim
   • Edit command script. You can use the following script as an example.
     Environment variable PREFIX defines the names of the output files.
   • Run the script. As a result you will get:
     $PREFIX.hbk histogram output
     $PREFIX.klg KOFIA log.
   • Fit the histograms in $PREFIX.hbk (replace PREFIX with actual name)
     % paw
     PAW> exec ../source/t2#init PREFIX.hbk
     PAW> exec ../source/t2#fit PREFIX.out
     PAW> exec ../source/t2#plotraw PREFIX.ps

     As a result you will get:
     $PREFIX.out Fit results
     $PREFIX.ps Pictures of the fit

   • Make the parameters
     % awk '$2>0' new_trim.out | awk -f ../source/shift_from_average.awk > new_trim.result
5. Mean value from new_trim.result plus 7 ns should be used as a Center of no-trim zone in StepD .

Step X: T and 2 time measurements on inputs of MT2A board

1. Disconnect the MT2A board signal cables to the test cables, following the record in Trigger Logbook #4, page 50:

   J1/ T:  1- 8 ---> test cable #1
   J7/L2:  1- 8 ---> test cable #2
   J2/ T:  9-16 ---> test cable #3
   J8/L2:  9-16 ---> test cable #4
   J3/ T: 17-24 ---> test cable #5
   J9/L2: 17-24 ---> test cable #6
   

2. Set trigger to kbeam
3. Take 30K kbeam events to file
4. Process the data:
   • % cd StepX
   • Create directory and get there
     % mkdir new_trim
     % cd new_trim
   • Edit command script. Run is processed with the same executable as in Step C, so you can use similar script as an example.
     Environment variable PREFIX defines the names of the output files. Usually PREFIX=new_trim
   • Run the script. As a result you will get:
     $PREFIX.hbk histogram output
     $PREFIX.klg KOFIA log.
   • Fit the histograms in $PREFIX.hbk (replace PREFIX with actual name)
     % paw
     PAW> exec ../source/mt2a#init PREFIX.hbk
     PAW> exec ../source/mt2a#fit PREFIX.out
     PAW> exec ../source/mt2a#plotraw PREFIX.ps

     As a result you will get:
     $PREFIX.out Fit results
     $PREFIX.ps Pictures of the fit

   • Make the parameters
     % awk '$2>0' PREFIX.out | awk -f ../source/mt2a_shift_from_average.awk > PREFIX.result
   • make a proposal of trimming by editing the file $PREFIX.proposal. To make a proposal, remind that 2Pmt2M (meantime of 2P and 2M) should be kept to 0.0. The meantime should be trimmed at the Tdot2 or for the Detector strobe. Example of proposal file prepared by Takeshi based on before trim data is here . It is based on PREFIX.result file. Do not delete extra line. Here Takeshi will share wisdom how to prepare proposal file
   • Check $PREFIX.proposal that you kept meantime of 2P and 2M intact (or changed it just a bit) by running
     % paste PREFIX.result mt2a.proposal | awk '$7 == "S"' | awk -f ../source/mt2a_proposed_trim.awk > mt2a.after_trim

Step D: Measure the layer T, 3-21 arrival times at the Range Cards.

1. Unplug discriminator cables from Range Cards and plug them into BV TDC cables.

   CABLE 1:   Odd Sector  3-10
   CABLE 2:   Odd Sector 11-18
   CABLE 3:   Odd Sector 19-T
   CABLE 4:  Even Sector  3-10
   CABLE 5:  Even Sector 11-18
   CABLE 6:  Even Sector 19-T

   Hints:

   • In first 5 cables colors of wires should match. On the sixth cable wire colors do not match. White wire goes to the red wire because it is the first wire. Purple wire is always last.
   • Adjust cables 1..6 cablelength-wise. So that Cable 1 is the shortest hanging from the ceiling and cable 6 is the tallest. Will save you tons of time locating the right cable.
   • Use little screwdrive to unplug cables 1..6
   • Be VERY gentle unplugginf last (3-rd) connsector from the RC modules. They only have half a cable in them. Do not pull the cable.
2. Take >=30K kbeam events to disk file kbeam_xx_yy.run, where
   xx = 01, 03, ..., 23
   yy = 02, 04, ..., 24
   run = run number

   If the name convension can not be followed then create symbolic links.

3. Repeat steps 1-2 for sectors 01-02, 03-04, ... 23-24.
4. Process the data:
   • % cd StepD
   • Create directory and get there
     % mkdir new_trim
     % cd new_trim
   • Edit command script new_trim.sh . You can use the following script as an example.
     Environment variable PREFIX defines the names of the output files. Usually PREFIX=new_trim
   • Run the script. As a result you will get:
     $PREFIX_sec01.hbk histogram output for sector 01
     ......
     $PREFIX_sec24.hbk histogram output for sector 24
     $PREFIX.klg KOFIA log.
     $PREFIX.dat Approximate fits of BV-IC histograms, this can be used as a reference. This file is not used in analysis, instead fitted values are used.
   • Fit the histograms in $PREFIX_secXX.hbk In PAW session replace PREFIX with actual name of directory.

     
     % cd .. - go to upper direcotry
     % paw
     PAW> exec source/trim#fit PREFIX
     PAW> exec source/trim#getfit PREFIX
     PAW> exec source/trim#plot PREFIX
     PAW> exec source/trim#view PREFIX
     alternatively
     PAW> exec ../source/trim#view PREFIX sec# layer# end#

     As a result you will get:
     $PREFIX.fit Fit results
     PREFIX_secXX.hbk files will contain now fitted histograms.
     $PREFIX.hbk Fit results in Ntuple for easy analysis
     $PREFIX.ps Pictures of the fit.

     Caution $PREFIX.ps is a 114 pages monstrosity. Don't blindly send it to printer, unless you love to look on endless gaussian fits. You may want to do the following to print few pictures of questionable channels:
     PAW> for/file 69 filename.ps; meta 66 -111
     PAW> exec source/trim#view PREFIX 24 1 2
     PAW> exec source/trim#view PREFIX 24 4 1
     PAW> exec source/trim#view PREFIX 13 6 1
     PAW> ....
     PAW> close 69

Step E: Generate trims

1. Goto StepE directory
   % cd StepE
2. Create directory and get there
   % mkdir new_trim
   % cd new_trim
3. Edit command script make_trims.sh . You can use the following script as an example.
4. Run the script. As a result you will get:
   rc_diffs.txt File with deviations from common mean
   rc_trims.txt File with suggested trims
5. Copy the file rc_trims.txt to something like suggested_trims_date.txt Edit this file to incorporate T-layer trims from Step X . Watch out for possible unreasonable trims.
6. Printout the file. Following command is useful (65 lines with 105 chars):
   a2ps -1 -L65 -l 105 -o - suggessted_trims_date.txt | lpr -Pe787lp2r
7. Make a graphic representation of trims.

   If this is the first time you do this procedure

   If this is the second time you do this procedure
   Now you have measuremnets before and after the trim, and you can compare how the things improve because of the trim.

   Edit command script trim_status.sh .
   You can use the following script as an example.

8. Start PAW
   % paw
   PAW> exec ../source/trim_status

   You will get a set of pictures and histogram file trim_status.hbk with Ntuple that describe arrival times before trim, after trim and suggested trims.

   On the plot check if Red solid triangles are inside black box

   
   Black solid circles connected with solid line represent arrival times measured before trim with BV TDC.
   Blue open circles connected with dotted line represent expected arrival times after trim . They are calculated as arrival time before trim measured with BV TDC + trim from file.
   Red solid triangles connected with dashed line represent measured arrival times after trim with BV TDC.
   Purple open triangles connected with dotted line represent measured arrival times after trim with RS TDC. This is calculated as BV time before trim + observed difference between after trim and before trim as measured by RS TDC.
   Black box represent area where arrival times supposed to end up. So the goal is to put Red triangles in black box.

Step F: Check trims using RS TDC information

This measurement compares RS TDC arrival times. So there are 2 measurements involved: first is the RS TDC arrival time when RS trimming was performed, and the second is some time along the run.

1. Goto StepF directory
   % cd StepF
2. Create directory indicating run number and get there
   % mkdir check_46380
   % cd check_46380
3. Edit command script check_46380.sh . You can use the following script as an example.
   Pay attention to which tables you are using
   use rd tof "/e949/etc/caldb/rd_tof.00001"
   use rd twk "/e949/etc/caldb/rd_twk.notwk"
   The rd_tof should be the same as used to get reference set of RS TDC arrival times in Step D
   The rd_twk should be rd_twk.notwk in all cases.
4. Run the script. As a result you will get:
   check_46380_secXX.hbk Files with RS TDC arrival time histograms
   check_46380.dat Average arrival times for reference. We are going to get better values after histogram fitting. But you could see dead channels and anomalities.
5. Fit the histograms (see StepD for more details it is the same kumac)

   
   % cd .. - go to upper direcotry
   % paw
   PAW> exec ../StepD/source/trim#fit check_46380
   PAW> exec ../StepD/source/trim#getfit check_46380

   As a result you will get:
   check_46380.fit Fit results
   check_46380_secXX.hbk files will contain now fitted histograms.
   check_46380.hbk Fit results in Ntuple for easy analysis

6. Go back to working directory ./check_46380 and edit command script trim_status.sh . You can use the following script as an example. Run the script. As a result you will get trim_status.dat file that contains comparisons of 2 sets.
7. Start PAW
   % paw
   PAW> for/file 66 check_46380_status.ps; meta 66 -111
   PAW> exec ../../StepE/source/trim_status 1 24 1001
   PAW> close 66

   You will get a set of pictures and histogram file trim_status.hbk with Ntuple that describe arrival times before trim, after trim and suggested trims.

   On the plot check if Purple open triangles are inside black box

   
   Black solid circles connected with solid line represent arrival times measured measured with BV TDC. for the first set of data. So those are result of lengthy procedure of StepC..StepE.
   Purple open triangles connected with dotted line represent measured arrival times measured with RS TDC. This is calculated as BV time of the first set + observed difference between first set and second set as measured by RS TDC.
   Black box represent area where arrival times supposed to end up. So the goal is to put Purple open triangles in black box.

8. There might be suspicious channels. To view arrival times for those channels go to upper directory and start PAW .
   % paw
   PAW> for/file 66 check_46380_suspicious.ps; meta 66 -111
   PAW> exec ../StepD/trim#view check_46380 [sec#] [lay#] [end#]
   PAW> close 66