Difference between revisions of "Beam tuning and recovery"

Revision as of 10:19, 15 April 2025

Beam Checkout Procedure for GEp (at the beginning of experiment or just after the pass change)

• Initially MCC will want to send tune beam into the hall to establish a good beam orbit by following this accelerator procedure: [1]. Before any beam is sent into the hall make sure the target ladder is in the "NO TARGET" position meaning all targets are out of beam. Also make sure the Moller target ladder is out of beam. All SBS and EArm detectors should be off. SBS magnet should be off as well. Also check, if the Moller Quads are OFF or not (field integral should be zero during GEp as in GEn-RP/K_LL [2]). If ON, ask MCC to degauss the Moller Quads and turn off power supplies.

• As part of the accelerator procedure mentioned above the intrinsic beam spot size (unrastered beam spot size) will be checked by performing harp scans with IHA1H04A and IHA1H04B. MCC will do harp scans with IHA1H04A and IHA1H04B with 5 muA TUNE beam with the target ladder in the "NO TARGET" position and raster OFF.

• Once MCC posts the harp scans to the elog check the beam width at the target in X and Y using instruction below To get the beam width from harp scans. If the beam width at the target is between 150 and 500 microns move on to the next step. If not, ask MCC to work on the beam profile. Make a HALOG with the final results (example here [3]).

• MCC/Hall A will then want to assess the level of beam scraping, if any, on the differential pump region small aperture in the downstream beam line right after the target [4], [5].

• At this stage MCC will need to do a functional test of the ion chambers following this procedure:[6]. I hope the downstream IC that's under the platform is easy enough to trip during this procedure so we don't waste time.

• Once the IC functional test is done we can start working on centering the beam on the Cross target with 1mm groove [7]. There is also one Carbon foil with a hole of 5mm at the center placed at z=0 if we need to use it.
  • Turn ON All detectors (at least ECAL) except GEM detector.
  • Select the GEp coda config without GEM (ECal-HCal-CDet) and set prescale PS12 to 0 while all other triggers are off in the PS gui.
  • Ask MCC to set the x,y beam positions at IPM1H04A and E to zero in the orbit locks gui. It is a good idea to check that the XSOF and YSOF have been returned to their original values [8].
  • Request 4x4 mm rastered CW beam at 5 muA and take a run on the Cross target. Based on previous experience, 2x2 mm raster in Hall-A is equivalent to 0.6x0.6 mm in raster GUI, so mention 1.2 mm (for 4 mm) and 0.6 mm (for 2 mm) while communicating with MCC. Keep in mind that we only use the first pair of raster coils to raster the beam for GEp.
  • Replay the run and see if you can identify the 1 mm Cross by checking the raster plots.
  • Adjust the beam positions until the 1 mm Cross-center is centered in the 4 mm raster pattern. (Hint: BPM +X is right and +Y is down while looking raster plot during GEn and GEn-RP/KLL beam checkout)
  • Beam positions for the production running are determined from the previous step by centering the Cross in the 4x4 mm raster pattern. These news positions should be communicated with MCC for orbit lock after the next step.
  • 4x4 mm raster size can now be verified by changing the beam position from the new value found above. Changing the beam position down by 1.4 mm, should show T pattern with 1 mm cuttings including the lower end (beam can be moved left/right/up which shows rotated T). The distance of upper side of horizontal cut to the lower end of vertical cut is 3.81 mm, so 4x4 mm raster should image T clearly. (3x3 mm raster can also be verified by changing the beam position with 1.9 mm where we should clearly see one side of the Cross in the raster. Each side of the Cross is 2.8 mm long while excluding the overlapping region at center)
  • Once the Cross is centered into the raster pattern and raster size is verified, ask MCC to set the beam position offsets at IPM1H04A and E in the orbit lock gui to the just found alignment numbers.

• Scale down the raster size by half from 4x4 mm and make sure the 1 mm Cross is concentric within the raster pattern. This will give the 2x2 mm raster needed for production. Note the size in raster GUI that corresponds to 2x2 mm raster measured on target and write on white-board.

• At this point, MCC can proceed to IC calibration of all target types for GEp

Beam Recovery Procedure for GEN-RP/K_LL

• Initially MCC will want to send tune beam into the hall to establish a good beam orbit by following this accelerator procedure: [9]. Before any beam is sent into the hall make sure the target ladder is in the "NO TARGET" position meaning all targets are out of beam. Also make sure the Moller target ladder is out of beam. All SBS and BB detectors should be off. SBS and BB magnets should be off as well. Also check, if the Moller Quads are OFF or not (field integral should be zero during GEn-RP/K_LL as in [10]). If ON, ask MCC to degauss the Moller Quads and turn off power supplies.

• As part of the accelerator procedure mentioned above the intrinsic beam spot size (unrastered beam spot size) will be checked by performing harp scans with IHA1H04A and IHA1H04B. MCC will do harp scans with IHA1H04A and IHA1H04B with 5 muA TUNE beam with the target ladder in the "NO TARGET" position and raster OFF.

• Once MCC posts the harp scans to the elog check the beam width at the target in X and Y using instruction below To get the beam width from harp scans. If the beam width at the target is between 200 and 300 microns move on to the next step. If not, ask MCC to work on the beam profile. Make a HALOG with the final results (example here [11]).

• MCC/Hall A will then want to assess the level of beam scraping, if any, on the differential pump region small aperture in the downstream beam line right after the target [12], [13].

• At this stage MCC will need to do a functional test of the ion chambers following this procedure:[14]. I hope the downstream IC that's under the platform is easy enough to trip during this procedure so we don't waste time.

• Once the IC functional test is done we can start working on centering the beam on the Carbon hole. There are 2 Carbon foils with holes at the center placed on different holders: one has a 2 mm hole while the other a 5 mm hole [15].
  • Turn ON the BigBite magnet and the BigBite detectors. If BB GEMs are needed, consult with the RC and GEM expert.
  • Select the GEnII_NoSBS coda config and set PS1 to 0 while all other triggers are off in the PS gui.
  • Ask MCC to set the x,y beam positions at IPM1H04A and E to zero in the orbit locks gui. It is a good idea to check that the XSOF and YSOF have been returned to their original values [16].
  • Request 4x4 mm rastered CW beam at 5 muA and take a run on the 2 mm C-hole target. Replay the run and see if you can identify the 2 mm hole by checking the raster plots. Keep in mind that we only use the first pair of raster coils to raster the beam for GEN_RP/K_LL.
  • Adjust the beam positions until the 2 mm C-hole is centered in the 4 mm raster pattern.
  • Once the Carbon hole is centered into the raster pattern ask MCC to set the beam position offsets at IPM1H04A and E in the orbit lock gui to the just found alignment numbers.

• Reduce the raster size until the 2 mm C-hole is concentric within the raster pattern. This will give you the 2x2 mm raster needed for production.

• This paragraph is written by Ciprian, feel free to contact him for questions: "The raster size can be x-checked with the harps: see https://logbooks.jlab.org/entry/3725647 for example. Have MCC turn on the raster, send up to 5uA tune beam and do a swipe of the 1H04A and 1H04B harps. The fits will look bad (see the example) but that is ok - make sure the MCC operator doesn't run multiple harp swipes because their fit is not converging. Take the width and height of the raster from the harp swipe (by eye) and put them in the same code that you used for the beam width to get the value at the target."

(!! Old !!) Beam Recovery Procedure (Followed during GEn for Glass target)

If beam has been down for over 30 minutes move target to "No Target"

This quick procedure applies to coming back from a shorter period of no beam. If coming back from machine tuning or any work that might affect the beam profile, the full procedure below should be executed.

1. Move to No Target position
2. Verify the beam setup

   Target BPM locks at (0,0)
   

   For He3 production target (and no target position) BPM 4A XSOF, YSOF = (-0.10, 1.20) 4E XSOF, YSOF = (-0.32, 1.07)
   

   Raster size: Vertical: 1.75V Horizontal: 1.50V.

   Check the beam position orbit locks are engaged with correct values

   Check Hall A BPM Target Protection is Enabled

   Check Hall A Current Ramp Protection is Enabled

3. Make a log entry like this: https://logbooks.jlab.org/entry/4110171
   
4. Ask MCC to send 5uA beam with the setup above
5. Monitor the target BPM positions. Once MCC can deliver stable beam with No Target, return to PolHe3 cell
6. Do not change the setup other than moving the target to He3. NEW Check the voltage encoder readback is approximately 7.08 +/- 0.01 in target motion GUI
7. Ask MCC to send 5uA CW beam and monitor all above, and additionally IC + trigger rate

If beam has been down due to machine tuning

1. Move to the "no target" position (detectors HVs OFF: especially GEMs should be off) before MCC starts to tune beam to the hall [consult with RC or sbs beamline experts to verify the procedure and numbers below]
2. If the raster is ON, call MCC and ask for the raster to be turned OFF
3. Request harp scans at IHA1H04A and IHA1H04B with 5 muA of TUNE beam with the target ladder in the "NO TARGET" position and raster OFF. Check if the Moller Quads are OFF or not (Field integral should be zero during GEnII as in [17]). If ON, ask MCC to degauss the Moller Quads and turn off Power supply
4. Check the beam width at the target in X and Y using the instruction below To get the beam width from harp scans. If the beam width at the target is between 200 and 500 microns move on to the next step. If not, ask MCC to work on the beam profile. Make a HALOG with final result (example here [18])
5. Once the beam width at the target in X and Y is between 200 and 500 microns, ask MCC to turn the raster ON and to set the Vertical amplitude 1.75 V and horizontal amplitude 1.50 V. We are using one pair of raster controlled by single generator 1, so Raster A should be unmasked and Raster B should be masked as in the example [19] . Please check that the amplitudes are as stated above in both the sending and readback fields for Generator 1. Also check that the Ix and Iy currents are close to what's shown in the picture posted at the link above. Also look towards the scope in the electronics room, you should see a nice circular pattern of the raster on the scope.
6. Now that you are sure that the raster is set up correctly you can call MCC and tell them that you want to move the target ladder to the Carbon hole target to check both that the raster size looks like what we had before and also the beam centering.
7. You are now on the Carbon hole target. Make sure MCC sets the Target BPM XSOF and YSOF

   a) The IPM1H0A,E are set to (0,0), and their XSOFs, YSOFs should be
   

   b) IPM1H04A (XPOS, YPOS) = (-0.10, 1.20)

   c) IPM1H04E (XSOF, YSOF) = (-0.32, 1.07)

   d) Make sure the target orbit lock is ON.
   

   e) The raster is already ON from the previous step but check anyway.

8. Take a CODA run Turn ON all detector HV (don't forget to turn ON specially B0igbite shower, preshower and BB GEMs). Use prescale ps1=0 for this run to accumulate statistics fast, and take a run with ~10 uA CW and the production BBCal thresholds. [note: start a run only after BPMs come closer to nominal values, as beam positions might be off at the beginning]. Please keep eyes on IC rates during step 8 above while ramping current, comparing with [ Please keep eyes on IC rates during step 8 above while ramping current, comparing with [20] [21] ][22] [23] Once you have 200k events, go to the next step while continuing to take data.
9. Analyze 50k and 200k events
10. Call RC or Hall A beamline expert to get approval. If the hole is no clearly seen, you can run this script[24]
11. Now you determined that: the intrinsic beam width is acceptable, the raster is rastering the beam as expected for data taking on glass cells and that the rastered beam is centered on the Carbon hole targets. You can move on to whatever the run plan requires. If moving to a glass cell first establish the beam position using the appropriate x/ysofs. Also check that the orbit locks, ramp protection, and target bpm protection are all on. Make a HALOG like this [25]
12. Before requesting beam, be sure to read below: What to monitor while running on glass cells [26]

    Please follow the beam positions carefully as well as the Ion Chamber readings. You can load the typical stripchart I am using by right-clicking on any stripchart and loading the config from SBS/simona/beam/instructions.

    Also make sure the ramp protection is on, the target orbit lock is on and that the raster is on. To verify that the ramp protection is on type "jmenu" in any aonl or adaq terminal. Once the jmenu gui pops up type "ramp" in the search field and click enter. Select from the new menu "Hall A ramp protection" and make sure everything is green like in this example [27]

What to monitor during production running on glass cells

1. Raster always ON when running on glass cell targets. Raster size is 5.2mm, V: 1.75 V H: 1.50 V. Cross check with the values on the white board.

2. Make sure the correctors that maintain the beam position close to the the ideals have changing Bdls meaning the target orbit lock is running.

3. There might be instances when although the target orbit lock is running the beam positions slowly drift from the ideal due to orbit instability in the machine: [28], [29]. In this case call MCC and ask them to investigate. Also call the Hall A beamline expert.

4. Ion chamber readings. Elevated IC readings compared to a baseline established by the beamline expert are indicative of scraping. The baseline for the target ion chambers for running on He3 at 3 pass is here: [30][31][32]

5. BBcal trigger rates. The BBcal trigger rate for a threshold of 57 mV as a function of current when running on He3 is posted here: [33][34] [35] [36] [37] and here after target parameter changes [38]. Rates from H2 running: [39]

6. Raster plots to ensure there's no scraping. A case of mild brushing of C foils holder by the beam envelop is shown here [40]. The way you want this to look it's like here [41]

7. "Hall A ramp protection" always ON, make sure everything is green like in this example [42]

8. NEW Check the voltage encoder readback is approximately 7.08 +/- 0.01 in target motion GUI

9. If beam has been gone for more than 30 minutes, always go to no target