Difference between revisions of "D2n/Expert Checklist"

From Hall A Wiki
Jump to: navigation, search
(Run information)
(Auxiliary DAQs)
 
(28 intermediate revisions by 3 users not shown)
Line 43: Line 43:
  
 
== Run information ==
 
== Run information ==
# Check scaler rate screen, pre-scale factors agree with earlier runs. [http://www.jlab.org/~adaq/halog/html/0811_archive/081126142648.html example of normal rate @ 12uA prodction]
+
# Check scaler rate screen, pre-scale factors agree with earlier runs. [http://www.jlab.org/~adaq/halog/html/0902_archive/090216091559.html example of normal rate @ 15uA prodction]
 
# Check online CODA rate and CODA dead time. (Run scaler script, verify deadtime)
 
# Check online CODA rate and CODA dead time. (Run scaler script, verify deadtime)
 
# Check start-of-run input. Target type, beam half wave plate, etc.
 
# Check start-of-run input. Target type, beam half wave plate, etc.
# Make sure Happex run is restarted every four hours or so
+
# Restart [[#Auxiliary_DAQs|Auxiliary DAQs]] (HAPPEX and both Comptons) every two hours or so.
 
# Look at shift worker's run-sheet records.
 
# Look at shift worker's run-sheet records.
# Confirm CODA not crashed, check DAQ-configuration: twoarm_RICH.
+
# Confirm CODA not crashed, check DAQ-configuration:  
 +
## LHRS: Lefthrs
 +
## BigBite: bigbite_buf
 
# Check Halog entries.
 
# Check Halog entries.
# Check target polarization history. [[Image:3Ext Polarization.png|thumb|(Old picture. Need a new one since there is no spin flip for D2n) Example plot of a stable starget]]
+
# Check target polarization history. [[Image:Hall A Target Polarization.png|thumb|Example plot of a stable target]]
 
## Go To [http://www.jlab.org/~jinhuang/TargetStatus/ Target Status History]
 
## Go To [http://www.jlab.org/~jinhuang/TargetStatus/ Target Status History]
 
# Check all HV on. BB chamber HV on before run starts.
 
# Check all HV on. BB chamber HV on before run starts.
Line 66: Line 68:
 
# Halog left arm VDC efficiency plots once a shift
 
# Halog left arm VDC efficiency plots once a shift
 
# Do [[Compton monitoring]]
 
# Do [[Compton monitoring]]
 +
# Look at Replays (See below)
 +
# Update [http://hallaweb.jlab.org/experiment/E06-014 d2n Run Database]
 +
#* Go to /adaq/e06014/expert_shift
 +
#* Run ./setAnalysisInfo.pl
 +
#** It will tell you what command line options are available
  
 
== [[Online_Replay#Expert_Info|Replay]] ==
 
== [[Online_Replay#Expert_Info|Replay]] ==
  
 
[[Online_Replay#HowTo_Quick_Check_Root_File_Replayed_by_Shift_Worker|Run show_all(<run number>) script]] after shift worker finish a replay. It will pop up a set of plots for expert, which would take ~10min to check through.  
 
[[Online_Replay#HowTo_Quick_Check_Root_File_Replayed_by_Shift_Worker|Run show_all(<run number>) script]] after shift worker finish a replay. It will pop up a set of plots for expert, which would take ~10min to check through.  
 +
 +
*If the replay number is less than 20000, the analysis is for BigBite.
 +
*If the replay number is greater than 20000, the analysis is for LHRS.
 +
**Be sure to run show_all(<run number>) for each.
 +
 +
* Once per left HRS kinematic, do a full physics replay / {{pass|replay_phys_L(run#,-1)}}
 +
** When prompted change the root file name from the default, so as not to overwrite the online physics replay root file:
 +
  e.g., ROOTfiles/e06014_phys_L_[run#]_full.root
 +
* Please post the "PID Detectors (Good Electrons)" page into the Halog indicating the number of good electrons from the replayed run.
  
 
A list of minimum check items : (Item / {{pass|related online plots}})
 
A list of minimum check items : (Item / {{pass|related online plots}})
Line 78: Line 94:
 
# Check scaler consistency  / {{pass|online("scalar",run#)}}
 
# Check scaler consistency  / {{pass|online("scalar",run#)}}
 
# Spot++ / {{pass|equavlent plots (raster current, BPMs) are included in online("detectorL",run#)}}
 
# Spot++ / {{pass|equavlent plots (raster current, BPMs) are included in online("detectorL",run#)}}
 +
# Check syncroc signals ("Sync check 1 (straight lines)" plot from online BB plots. '''If there are any gaps or discontinuities in the straight line, you or the shift worker should make a HALOG post.'''
 +
# Check L1A vs. clock. You can do this with the macro CheckL1A.C, which is located in the onlana directory:
 +
        .L CheckL1A.C
 +
        CheckL1A(run#)
 +
You're looking for a straight, flat line. It's okay if the first data point is a bit off (as in Fig. 2 [http://www.jlab.org/~adaq/halog/html/0902_archive/090223183153.html here]), or if there's a beam trip (as in the structures around 200 and 300 [http://www.jlab.org/~adaq/halog/html/0902_archive/090223143141.html here]; note that l1a drops all the way to zero). What's NOT ok is [http://www.jlab.org/~adaq/halog/html/0902_archive/090223141919.html weird] [http://www.jlab.org/~adaq/halog/html/0902_archive/090223144056.html behavior] that takes steady-beam running away from a straight line. HALOG plots like this.
  
 
''Hint : You can try to do_all() the golden run to find out what a normal plots should look like''
 
''Hint : You can try to do_all() the golden run to find out what a normal plots should look like''
  
== A list of known problems / possible bad runs ==
+
=== A list of known problems / possible bad runs ===
 
# BB chamber HV not on all the time.
 
# BB chamber HV not on all the time.
 
# Left-arm VDC gas problem, low efficiency.
 
# Left-arm VDC gas problem, low efficiency.
 
# Left-arm TDC shifts, symptom is current run and golden run shows significant difference in TDC times.
 
# Left-arm TDC shifts, symptom is current run and golden run shows significant difference in TDC times.
# Spin Flip is paused
+
 
 +
== Auxiliary DAQs ==
 +
# Stop and start the three auxiliary DAQs (Compton Counting, Compton FADC and HAPPEX) roughly every two hours. These are run through a VNC connection to the Compton machine. Currently you can access this through the HACSBC2 workspace on adaql6.
 +
 
 +
If you can't find a VNC viewer open, or if it has crashed (usually with an error message saying the RECT is too large), you can open a new VNC viewer on any workspace with the command
 +
 
 +
    vncviewer compton:1
 +
 
 +
The password is the same as the password for the adaq account. There is no username.
 +
 
 +
===HAPPEX DAQ===
 +
In the VNC window, this runcontrol window is in the workspace labeled "HAPPEX CODA" -- second column, bottom row. We are running in the "d2n" configuration.
 +
 
 +
# [[HAPPEX_Charge_Asymmetry_Feedback|General info]]
 +
# [[HAPPEX_Charge_Asymmetry_Feedback#How_To_Start|Getting started]]
 +
# [[HAPPEX_Charge_Asymmetry_Feedback#IF_SOMETHING_IS_WRONG.21.21|Troubleshooting]]
 +
 
 +
=== Compton Counting DAQ ===
 +
In the VNC window, this runcontrol window is in the workspace labeled CODA -- second column, top row. If Alexandre or Sirish is around, check with them before stopping and starting a run. The configuration is "Production."
 +
 
 +
# [[Compton_monitoring|General info]]
 +
# [[Compton_monitoring#Starting_from_Scratch|Getting started]]
 +
# [[Compton_monitoring#Troubleshooting|Troubleshooting]]
 +
 
 +
=== Compton FADC DAQ ===
 +
In the VNC window, this runcontrol window is in the workspace labeled CMU CODA -- next-to-last column, bottom row. If Diana is around, check with her before stopping and starting a run. The configuration is "CMUAccum."
 +
 
 +
# [[Compton_FADC_DAQ#Getting_Started|Getting started]]
 +
# [[Compton_FADC_DAQ#Troubleshooting|Troubleshooting]]
  
 
== Helpful Resources ==
 
== Helpful Resources ==

Latest revision as of 21:47, 24 February 2009

This is the check list for experts on D2n data checker shift.

Notice : Please add items of your specialty field, so every one will do the check

Beamline, beam charge, beam polarization

  1. BPM A and B (x,y) positions
    • Check the front room white board for default value
  2. Beam charge asymmetry is small (less than 1000ppm)
  3. Check that the beam charge at end-of-run makes sense ?
    • All BCM charge reading should be consistent
    • Example of Good Run:
APPROXIMATE BCM CHARGES (C)
BCM u1 [ 0]:0.03799        BCM u3 [ 0]:0.03798        BCM u10[ 0]:0.0336         
BCM u1 [++]:0.009217       BCM u3 [++]:0.009211       BCM u10[++]:0.008148       
BCM u1 [--]:0.009464       BCM u3 [--]:0.009462       BCM u10[--]:0.008371       
BCM u1 [+-]:0.009214       BCM u3 [+-]:0.009211       BCM u10[+-]:0.008148       
BCM u1 [-+]:0.009466       BCM u3 [-+]:0.009465       BCM u10[-+]:0.008373       

BCM d1 [ 0]:0.03786        BCM d3 [ 0]:0.03787        BCM d10[ 0]:0.03787        
BCM d1 [++]:0.009183       BCM d3 [++]:0.009198       BCM d10[++]:0.009188       
BCM d1 [--]:0.009428       BCM d3 [--]:0.009432       BCM d10[--]:0.009432       
BCM d1 [+-]:0.009182       BCM d3 [+-]:0.009185       BCM d10[+-]:0.009185       
BCM d1 [-+]:0.009431       BCM d3 [-+]:0.009434       BCM d10[-+]:0.009436    

Run information

  1. Check scaler rate screen, pre-scale factors agree with earlier runs. example of normal rate @ 15uA prodction
  2. Check online CODA rate and CODA dead time. (Run scaler script, verify deadtime)
  3. Check start-of-run input. Target type, beam half wave plate, etc.
  4. Restart Auxiliary DAQs (HAPPEX and both Comptons) every two hours or so.
  5. Look at shift worker's run-sheet records.
  6. Confirm CODA not crashed, check DAQ-configuration:
    1. LHRS: Lefthrs
    2. BigBite: bigbite_buf
  7. Check Halog entries.
  8. Check target polarization history.
    Example plot of a stable target
    1. Go To Target Status History
  9. Check all HV on. BB chamber HV on before run starts.
    • Example Start Run with All HV On
    • Check Section
      • Left S2 HV Left/Right PMTs
      • Left S1/Cerenkov/A1 Cerenkov HV
      • Left Pion Rejector Layer 1/2 HV
      • Map BB_SC_R Measured Voltage/Current
      • Map BB_DC Measured Voltage/Current
      • Map BB_PS_R Measured Voltage/Current
      • Map BB_SC_L Measured Voltage/Current
      • Map BB_PS_L Measured Voltage/Current
      • Map BB_SH Measured Voltage/Current
  10. Halog left arm VDC efficiency plots once a shift
  11. Do Compton monitoring
  12. Look at Replays (See below)
  13. Update d2n Run Database
    • Go to /adaq/e06014/expert_shift
    • Run ./setAnalysisInfo.pl
      • It will tell you what command line options are available

Replay

Run show_all(<run number>) script after shift worker finish a replay. It will pop up a set of plots for expert, which would take ~10min to check through.

  • If the replay number is less than 20000, the analysis is for BigBite.
  • If the replay number is greater than 20000, the analysis is for LHRS.
    • Be sure to run show_all(<run number>) for each.
  • Once per left HRS kinematic, do a full physics replay / replay_phys_L(run#,-1)
    • When prompted change the root file name from the default, so as not to overwrite the online physics replay root file:
  e.g., ROOTfiles/e06014_phys_L_[run#]_full.root
  • Please post the "PID Detectors (Good Electrons)" page into the Halog indicating the number of good electrons from the replayed run.

A list of minimum check items : (Item / related online plots)

  1. Check left-HRS VDC efficiency / online("detectorL",run#)
  2. Check left-HRS VDC time / online("detectorL",run#)
  3. Check BB chamber wire map / online("detectorBB",run#)
  4. Check BB shower vs preshower ADC plot, threshold not changed / online("physics",run#)
  5. Check scaler consistency / online("scalar",run#)
  6. Spot++ / equavlent plots (raster current, BPMs) are included in online("detectorL",run#)
  7. Check syncroc signals ("Sync check 1 (straight lines)" plot from online BB plots. If there are any gaps or discontinuities in the straight line, you or the shift worker should make a HALOG post.
  8. Check L1A vs. clock. You can do this with the macro CheckL1A.C, which is located in the onlana directory:
       .L CheckL1A.C 
       CheckL1A(run#)

You're looking for a straight, flat line. It's okay if the first data point is a bit off (as in Fig. 2 here), or if there's a beam trip (as in the structures around 200 and 300 here; note that l1a drops all the way to zero). What's NOT ok is weird behavior that takes steady-beam running away from a straight line. HALOG plots like this.

Hint : You can try to do_all() the golden run to find out what a normal plots should look like

A list of known problems / possible bad runs

  1. BB chamber HV not on all the time.
  2. Left-arm VDC gas problem, low efficiency.
  3. Left-arm TDC shifts, symptom is current run and golden run shows significant difference in TDC times.

Auxiliary DAQs

  1. Stop and start the three auxiliary DAQs (Compton Counting, Compton FADC and HAPPEX) roughly every two hours. These are run through a VNC connection to the Compton machine. Currently you can access this through the HACSBC2 workspace on adaql6.

If you can't find a VNC viewer open, or if it has crashed (usually with an error message saying the RECT is too large), you can open a new VNC viewer on any workspace with the command

   vncviewer compton:1

The password is the same as the password for the adaq account. There is no username.

HAPPEX DAQ

In the VNC window, this runcontrol window is in the workspace labeled "HAPPEX CODA" -- second column, bottom row. We are running in the "d2n" configuration.

  1. General info
  2. Getting started
  3. Troubleshooting

Compton Counting DAQ

In the VNC window, this runcontrol window is in the workspace labeled CODA -- second column, top row. If Alexandre or Sirish is around, check with them before stopping and starting a run. The configuration is "Production."

  1. General info
  2. Getting started
  3. Troubleshooting

Compton FADC DAQ

In the VNC window, this runcontrol window is in the workspace labeled CMU CODA -- next-to-last column, bottom row. If Diana is around, check with her before stopping and starting a run. The configuration is "CMUAccum."

  1. Getting started
  2. Troubleshooting

Helpful Resources