Schematic of the spectrometer beamline


Energy Calibration

The function that defines the energy calibration is:

dnom and dy are constants for a particular screen. dnom is the distance between the zero-dispersion axis and the nominal position of the electron beam on the screen. For screens on the dump table this varies from about 60-70mm. dy is a measure of the offset of the screen from the zero dispersion axis. And finally, Ebend is the setting of the dipole in units of GeV. It is strange units, but it refers to the energy the electrons that will have a vertical dispersion equal to dnom at the screen location. yscreen is the position measured on the image.

While dnom and dy should be constant for a given screen, in fact they vary day to day, and for different final focus and spectrometer quad settings. This is a result of the beam not passing through the center of the final focus and spectrometer quads. This adds a dipole contribution to the beam trajectory, resulting in a different position (and angle) of the zero dispersion axis, on the scale of several mm. For this reason, it is recommended to perform an energy calibration for each S20 configuration change (including spectrometer change).

Energy Calibration on ProfMon_GUI

The "Energy scale" toggle should now work for FACET dump electron cameras! The calibration details need to manually (and carefully) entered, so caution should be applied.

To use:

  1. Select a dump camera. The energy scale will only work with DTOTR1, 2, LFOV, CHER for now. EDC_Screen is not calibrated, so will not give useful results
  2. Select "Use Calibration"
  3. Select "Energy scale"

Troubleshooting:

  • The DTOTR camera moves - so for the calibration to be correct, the 10 GeV should be roughly centered on the vertical FOV.
  • If the energy scale doesn't make sense, make sure the beam center is properly defined
    • i.e. with 'Use Calibration" on, the 10 GeV beam should be at y = 0mm
  • If it is not, then update the x and y image offsets:
    1. Make sure that the beam is at the nominal energy, and the dipole at the nominal value (10 GeV)
    2. Start profmon_gui and start acquiring images
    3. Select "Show Raw" on profmon_gui, and deselect "Use Calibration"
    4. Record the x and y locations of the beam centroid (in pixels)
      1. Applying a beam mark can help with this
    5. Update the following PVs (using the rotation/orientation of the raw image):
      1. CAMR:LI20:301:X_RTCL_CTR
      2. CAMR:LI20:301:Y_RTCL_CTR
  • Things can get weird if you zoom in or out, but it usually works.
  • If you start getting errors - try restarting the profmon_gui, or switching cameras, then switching back.

To add energy calibrations to new cameras, follow these steps:

  1. The vertical axis needs to be the energy axis
  2. Add a matlab support PV with the value of eta (~60mm at the dump table, ~30mm at the EDC), and reference energy (usually 10 GeV)
    1. The rest of the cameras are stored at SIOC:SYS1:ML00:A0330-339
    2. These is no more room here, so find some empty PV's close by.
  3. The camera ID should be added to profmon_grab.m:
    1. Line 284:    [isFDump,idFDump]=ismember(pv,{'CMOS:LI20:3505' 'CAMR:LI20:107' 'CAMR:LI20:301' 'CAMR:LI20:308' 'CAMR:LI20:309'});
  4. Add the call to the new PV's you added in step 2. i.e edit this code:
    1. Line 286:     energy=num2cell(reshape(lcaGetSmart(strcat('SIOC:SYS1:ML00:AO',{'330'; '331';'332';'333';'334';'335';'336';'337';'338';'339'})),2,[])',2);
  5. Push the changes back to the production CVS
  6. Update the x and y image offsets:
    1. Use the steps defined in the troubleshooting section to update the image offsets
  7. Select "Show Raw", and select "Energy scale" to check it all looks good.

Energy Calibration Routine

To perform the energy calibration:

  1. Images are acquired of the screen as the dipole is swept over a range that sweeps the beam to the top to bottom of the screen (or the relevant area).
  2. You MUST remove the CsI array before doing a dipole scan. If the beam strikes the CsI crystals they will be damaged.
    1. Motor controller MC01, LFOV channel position <75.
  3. Use the DAQ function scan_BNDS_LI20_3330
  4. Approximate ranges:
    • DTOTR1: 9.9-10.6 in 8 steps
    • DTOTR2: 8-13 in 7 steps
    • LFOV/CHER: 8-20 in 7 steps
  5. The centroid position of the beam is determined from the images and fit to the equation above to determine dnom and dy.


Typical values (but note that these change!):

DTOTR1: DAQ 4275 - dnom = 57.53 mm, dy = 59.31 mm

DTOTR2: DAQ 04388 - dnom = 59.31 mm, dy = 79.28 mm

LFOV: DAQ 04389 - dnom = 61.29 mm, dy = 148.45 mm


Example LFOV: (DAQ 04389)

Limitations

In no particular order:

  • The calibration depends on how accurately the beam energy is known.
  • The dipole component added by the spectrometer quads has an energy dependency that is not captured by this technique.
  • More to be added...


Energy Calibration Shortcut

A rough calibration can be determined from a single image if you know the position of the 10 GeV beam. This can be useful for making a rough estimate, but should not be relied on for accurate energy calibrations.

In this case use:

The more accurate your estimate of dnom, the more accurate this calibration will be. Using the previous method is much preferred over this.


  • No labels