Difference between revisions of "CommissioningPlan"

Revision as of 12:30, 4 October 2022

<<SBS Main

Four days scheduled for commissioning program

Establishing Beam for the first time

Beam centering and raster calibration [8 hours]

This program is run by Simona M. Basic outline is below

1. BB, SBS, and corrector magnets are OFF. Target in empty position.
2. Accelerator will establish beam straight to the dump and use this straight trajectory to find BPM offsets. Accelerator will ramp up the SBS and corrector magnets.
3. Harp scans now done to measure beam intrinsic spot size >320um in x and y ([ https://sbs.jlab.org/wiki/index.php/Harp_Scans_for_Checking_Beam_Spot_Size_on_Target spreadsheet]).
4. BB is now ramped up (see the procedure) and Preshower, Shower (collectively BB CAL), and Hodoscope HV is ON. GEM group called to slowly ramp up BB GEMs so that we have basic track to determine target Z and resolve the foil position.
5. Carbon target foil with 1mm hole inserted and imaged. Position of beam verified and raster size calibrated.
6. Ion chambers now calibrated with foils
7. BPMs and BCMs calibrated.

=== BB and SBS Commissioning At this point we have good beam and characterized beam monitors are ready to commission our experiment. Procedure

1. Call GEM expert to make sure GEM HV is OFF.
2. Call expert on BigBite system to turn on HV (includes Preshower, Shower, and Hodoscopes)[1]
3. Ensure the target is empty/OUT of the beam path. BigBite and SBS magnets are at zero field.
4. First MCC centers the beam on the differential pumping (DP) apertures (0.75") and then delivers to the beam dump using the dump viewer.
5. Determine the intrinsic beam spot size by performing a HARP scan. Ask MCC to take harp scans on Harps A and B (IHA1H04A and IHA1H04B) and ask them to post the results to our logbook Halog. When the results arrive fill in the green blanks in "Beam spot size at target" section of this spreadsheet and report/halog the values of the spot size calculated which show up in the red blanks. If the average of the Xsigma and Ysigma spot radii (1 sigma) is significantly less than 100 microns. Call the RC and ask what to do. We take a run simultaneously (not necessary, but helpful). Compare BPM EPICS readings to HARP locations. Steps + script?
6. Ion chamber calibration on empty target position. This calibrates the dump ion chambers.
7. Ion chamber calibration on Optics 5-foil target. This calibrates the ion chamber at the target for our optics program.
8. Establish nominal 4x4 rastered beam.
9. Call MCC to stop the beam and request to move to the Carbon Hole (2-mm) target into the nominal beam path
10. Call MCC to request 1 uA of 4x4 rastered beam. Run the DAQ to take data. Analyze the data and produce raster X/Y plots by running the command spot_pp [run_number] [height] [width] from the a-onl@aonl2 account, where height and width are the raster size in MCC units.
11. Using the 2mm hole in the X,Y raster plot to calibrate the size of the raster and position of the hole. Iterate with MCC to center beam on the raster pattern by requesting beam position movements on the harps 1H04A and 1H04E. To calibrate the size of the raster, change the raster size until the X/Y raster plot is about twice the size of the hole. You should now have a beam centered on the target and a 4x4 raster. Record the positions as the new target lock location.
12. Ask MCC to set the raster to 2mmx2mm by dividing the 4mmx4mm units by 2. Take a new run to verify that the Carbon hole fills out the entirety of the plot produced by the spot_pp program. Record these values as the calibrated raster set units.

Raster Calibration and Beam Centering[1 hr]

NOTE: This happens for every new beam energy!

1. Establish stable 1 uA beam on centered on the Carbon 2mm hole target. Center of Carbon 2mm hole target is X=-0.8mm and Y=0.3mm.
2. Ask MCC for nominal 2x2mm raster (MCC units).
3. Run spot_pp to analyze the data. Estimate the size of the raster relative to the hole in the raster pattern plot. Tell MCC how much to increase the raster current to make the diameter of the hole approximately twice the size of the box in the raster plot. Check the new size to verify your estimation.

Initial BPM/Bulls-Eye checks

A bulls-eye scan is moving the beam to 4 corners of a square (2 mm in x and y) to calibrate the response of the BPMs in the DAQ using the positions given by the HARPs.

How to perform a bulls eye scan:

1. You need unrastered beam:

    - caution you should not do this with a target requiring rastered beam
    - use carbon, optics (single-foil), or in the worst case empty target.

2. Ask MCC to steer the beam to the nominal center of the target.

3. NOTE: We do this above!. Wait until beam is stable and have MCC perform a harp scan for the two superharps near the target (1H04A and 1H04B) and take a CODA run for both the left and right spectrometers during the same time. Start the coda run first before asking MCC. Request MCC to make an ELOG entry with the Harp results, you should see all three harp wires. Record ELOG entry numbers.

4. Ask MCC then to steer the beam to positions around the nominal center:

    - cover at least the area the raster will cover: (1,1), (1,-1), (-1,-1), (-1,1), and repeat (0,0)
    - repeat harp and CODA runs for each position

5. Record Harp scan run numbers and corresponding CODA run number for each beam position. As an example, see HALOG # 3324009.

6. Make a record of the harp scans and CODA runs in the HAlog.

Detector Commissioning Program

Maybe start with the Optics since Sieve is IN

Sieve plate is IN. Don't need it OUT to get GEMs properly working. .

UNDER DEVELOPMENT

AJRP: What is the thickness of the foils in the solid targets? Can't do rate estimate without those numbers.

Turn on All Major Detectors

Experts for each system turn on their systems for operation.

Trigger Rate Scaler Checkout

AJRP: If you want expected rates you need to tell me what the target foil thickness is. I can generate expected rates per unit integrated luminosity.

Personnel

1. HCAL: Expert driven: Scott Barus, Sebastian Seeds
2. BB CAL: Arun Tadepalli, Provakar Datta
3. Experts to draft plan

After centering the beam on the target and downstream DP apertures and beam dump, we examine whether the trigger rates seen on the BigBite calorimeter and HCAL make sense for various trigger setups.

Asking whether the trigger rates "make sense" at this stage is putting the cart before the horse. You cannot determine whether the trigger rates "make sense" until you have calibrated the signals in the calorimeters to energy deposits and compared them to "minimum bias" simulations with all possible physics backgrounds at the given thresholds. Realistic single-arm trigger rates were estimated for the original proposal kinematics here:

• Provakar's report on GMN trigger rates for original proposal kinematics PDF. Here the thresholds are given in units of energy deposit. Note that all of the trigger rates presented in this document assume 30 micro-amps beam current on a 15-cm LD2 target.

The only meaningful thing you can do at this stage is look at the trigger rate vs threshold in mV at some beam current on some target, from scalers. Once you know the rate vs threshold curve for a particular target, you can scale the threshold based on changes in target and/or beam current at the same kinematics to achieve a desired rate.

Some very rough and preliminary trigger rate estimates for SBS-1 kinematics from beam-on-target simulations with the 15-cm LD2 target were summarized in the following log entry:

Andrew's log entry with preliminary BBCAL and HCAL rates for the LD2 target in SBS-1 kinematics

Start with Carbon target, 1 uA beam. Roughly 1 kHz (QE, singe arm) expected for BB with high trigger threshold.

1. Do we do a threshold scan first? YES
2. Do we have a simulation for the present setup? BW: estimate for QE set up (threshold < 50% of QE peak, not sure of what the rate would be). R ~ e^{-9*thr/E_max}... AJRP: We do have a simulation but to use it to estimate single-arm trigger rates realistically is very involved (see Provakar's report above) and not simple
3. Look at xscaler GUI: quick feel for what the thresholds need to be; quickest way to do this kind of a measurement.
4. Need to pick suitable threshold to use for detector commissioning (i.e., gives a manageable rate for the DAQ). Then calibrate threshold to energy (need to know the energy). AJRP comment: Choosing a precisely optimized threshold for early stage commissioning is not essential. Needs to be high enough to give manageable DAQ rate and low enough to give reasonable efficiency for quasi-elastic scattering.
5. Choose a few thresholds and see if it makes sense. Vague, meaningless. AJRP: After we calibrate the calorimeters we can map trigger thresholds in mV to energy deposits in MeV or GeV, then they could be compared to simulations
6. Do this for both BB & HCAL. HCAL is not in xscaler yet (clusters); the trigger is, however..
7. Estimated rates for HCAL (as function of threshold).
8. Timing analysis between BB and HCAL. Is this software available? Cut on HCAL 10 ns wide => identify real QE events in BB. AJRP: What we CAN do quickly is use g4sbs to calculate electron and nucleon TOF for elastic and QE events, to be ready for time correlation analysis between BB and HCAL and to calibrate nucleon TOF/beta reconstruction. But one needs to calibrate all the offsets/walk corrections/propagation delays/etc in the precise timing detectors, particularly the hodoscope. Looking for the real coincidence peak in the HCAL vs. BBCAL trigger times can be done pretty early on.

For a given configuration, the steps are:

1. Call MCC to stop the beam
2. The TO moves the desired target into position (that is, along the central trajectory where the beam will be)
3. Set the DAQ to the correct configuration and set the prescales accordingly to select the appropriate trigger
4. Start a new run
5. Call MCC to deliver beam at the desired current and raster size (2x2 mm^2)
6. After X minutes, stop the run and check the beam setup using spot++. Work with MCC to optimize beam delivery on target.
7. Take data for X minutes and examine the trigger rates on the xscaler GUI (directions to bring up the GUI are here)
8. Perform offline/nearline analysis (if necessary)
9. Post a HALOG entry of the results

GEM High Voltage Efficiency Scans

People: Andrew, Holly, Ezekiel, Sean, Anu, John

This procedure should only be done by a GEM expert from the names listed above. Recommended but not required before optics runs. The full commissioning procedure can be found here

Electron Arm (BB) Optics Calibration

Andrew Puckett and Holly Szumila-Vance are the contact people for BigBite optics calibrations

The commissioning plan (not necessarily in the following order):

• Low current, zero-field runs: Purpose: align GEMs relative to BigBite Magnet and target center using straight-through tracks from with sieve slit and single-foil "point" target. Initial alignment comes from survey. Internal GEM relative alignment for tracking purposes is done separately using Andrew's script. The GEM HV scan/efficiency plateau should either already be done or can be done concurrently as part of this program.
  • BigBite and SBS Magnets are OFF.
  • Target = single-foil carbon.
  • Beam current = 1 uA or as low as we can go such that the BCMs still work. Note that GEMs can tolerate much higher charged particle flux than wire chambers and should tolerate magnet off conditions better than MWDCs as soft photon flux is unaffected by magnetic field
  • Beam is unrastered ("point" source of straight tracks for precise alignment)
  • Sieve slit is IN.
  • Trigger = BigBite calorimeter, threshold set for quasi-elastic from carbon, rate <~ 5 kHz
  • Time required probably ~1 h or less. Need ~few-hundred straight-through tracks per sieve hole.

• Optics data with sieve slit plus optics targets (and some LH2 data with sieve slit IN): Purpose: calibrate BigBite angle and vertex reconstruction (and also momentum).
  • BigBite magnet is ON, energized at full current of ~710 A with polarity set for up-bending electrons
  • SBS magnet is ON, at reduced current (~30% of max at commissioning kinematics)
  • Sieve slit is IN
  • Prefer unrastered beam on solid-foil optics targets. Always rastered beam on cryo-targets
  • Targets are:
    • Optics (9 foils at z = 0, +/-7.5 cm, +/-15cm, +/-22.5cm, +30 cm)
    • LH2 15 cm (without radiator)
  • Trigger is BigBite calorimeter
  • Need ~few hours data on each optics target plus LH2 target with sieve in. Possibly with lower threshold in BB or higher beam current, depending on rate. Need to collect ~500-1,000 events per sieve hole per target foil for each optics target position. Additional LH2 elastic data with sieve slit may also help with momentum calibration.
  • Starting optics model is from g4sbs simulation. Need to know actual magnet current to determine appropriate scale factor for simulation magnetic field to generate starting optics model. Approximate starting optics model helps to more easily identify which tracks come through which sieve holes from which foils. Unclear, but we might end up having to turn off SBS GEMs during BB multi-foil running depending on the good electron rate.
• LH2 elastic data without sieve slit: Purpose: Calibrate BigBite momentum reconstruction and BBCAL and HCAL energy reconstruction. Also calibrate BigBite and HCAL trigger threshold from mV to energy deposit
  • With BB magnet ON at full current
  • Sieve slit is OUT (require controlled access and trained/authorized personnel to insert/remove sieve slit)
  • Rastered beam at nominal
  • Target is LH2 15 cm without radiator
  • beam time/data requirements for this phase are driven by calorimeter calibration needs, not optics/momentum calibrations
  • Use angle-momentum correlation for elastic scattering to calibrate momentum reconstruction matrix elements, assuming angle reconstruction already calibrated. We need to know beam energy for this. Do we need a dedicated Arc and/or eP beam energy measurement?

HCAL Calibration

With the BB calorimeter calibrated (which gives us the electron momentum), we now can calibrate the response of the HCAL detector to determine the proton momentum, where we use the LH2 target.

• Target: LH2
• Beam: 5 μA, raster (2x2 mm^2) (?)
• DAQ Configuration: Name, BB CAL/HCAL coincidence trigger

1. Call MCC to stop the beam.
2. The TO inserts the LH2 target.
3. The sieve is pulled OUT (if it is in).
4. The BB and SBS magnets are ramped to their nominal currents.
5. Start a new run
6. Call MCC to deliver beam.
7. After X minutes, stop the run and check the beam setup using spot++. Work with MCC to optimize beam delivery on target.
8. Start a new run for X minutes. Shifters monitor the data using the online software
9. Experts perform offline/nearline analysis to produce plots to illustrate the calibration/success of the measurement.

BB GRINCH Commissioning

BB Hodoscope Commissioning

Next Steps

1. If we finish early, move to LD2/LH2 target. Look at proton/neutron separation. Slowly ramp to production/reasonably high current: Verify that the DAQ can handle the rates; verify the reconstruction works at higher occupancies.
2. BCM calibration: ABA measurements (beam on/beam off) Do we need the precision current source calibration now.
3. NOTE: Repeat commissioning steps at next beam energies.
4. Do we (can we) move up the energy/pass change if we finish early? Will affect other halls?
5. Note: Production data is elastic => refine calibration

Initial checklist

The following subsystems require updates prior to beam in the Hall on Sept 9.

Hall hardware required to be ready before taking beam

1. Hall A installation: Jessie Butler, Jack Segal
   • Things that require attention or are being finished up
   • Proposed lockup time
   • General instructions to collaboration
2. HCAL: Scott Barcus, Brian Quinn, Sebastian Seeds
   • Cabling and PMT performance verified. All connections complete.
3. Shower, Pre-shower: Arun Tadepalli, Provakar Datta
4. INFN GEMs: Holly Suzmila-Vance, Ezekiel Wertz, Evaristo Cisbani
5. Beam line: David Flay
6. Target: Dave Meekins
7. Gas system for GEM and GRINCH: Jack Segal
8. GRINCH: Todd Averett, Bradley Yale
9. Trigger: Mark Jones, Scott Barcus

Software and DAQ

1. Software: Andrew Puckett, Mark Jones
   • Analyzer
   • Event displays
   • Real time display
2. Computers and Counting House readiness: Ole Hansen, Alex Camsonne
3. DAQ: Alexandre Camsonne, Mark Jones, Bob Michaels, Ben Raydo, B. Moffit
4. Alarms
5. Control GUIs

Hardware required but not immediately

1. Hadron polarimeter: Brad Sawatzky, Kondo, David Hamilton
2. UVA GEMs: Nilanga Liyanage, Kondo Gnavo, Xinzhan Bai, Sean , Anu
3. Moller polarimeter: Simona
4. HRSL magnet: Jessie, Jack
5. HRSL detector/DAQ: Bob Michaels

Training and safety

1. Hall A walkthrough SAF110
2. Read and signed COO, RSAD, ESAD
3. Target training: Jian-Ping Chen, Silviu Covrig
4. Safety: B. Quinn, Jessie, Jack