SYAG

Energy Spectrometer Design

Details about the how the SYAG spectrometer works can be found in these documents:

Li_2011_BeamDiagnosticsFacet.pdf

Barnes_2005_LongitudinalPhaseSpace.pdf

Energy Calibration

Measured energy calibrations disagree wildly:

The design dispersion at the screen is 111.7mm which means the calibration should be 10.0 GeV / 111.7 mm = 89.5 keV/um. The difference between the measured and expected could be due to the screen calibration. No calibration images ever appear in the elog. The calibration starts at 17.94 um/px when the camera first appears on one of the FACET-II camera server. The calibration then changes to 9.91 um/px starting with this elog entry, but no calibration images appear in the e-log, this value as been used for all of FACET-II. The value matches the value used at FACET, most likely the old FACET value was used without every taking a new calibration.

The S20 energy BPM is not too far downstream of the energy BPM. Comparing centroid beam position on the two screens gives an idea of how similar the dispersion is:

Clearly they correlate very well and the slope between the two is 1.254, which implies that the dispersion is larger at SYAG than at the BPM (assuming the spatial calibration for both devices is correct). The expected dispersion at BPM 2445 is 120.6mm (83 keV/um), this implies that the spatial calibrations are not correct at either/both devices and/or the chicane is leaking dispersion into the IP. Measurements on a different day show a different slope between the two devices:

This is evidence that the dispersion has changed between the two days.

The two devices are separated by a quadrupole that focuses in x, resulting in the nearly identical dispersion at the two devices. SYAG is located right before the last positive dispersion peak and BPM 2445 is located right after the dispersion peak.

From particle tracking, the expected beam profiles at the two screens are shown below. Based off these images, the beam should have an energy spread of +-0.05 GeV spread over about 1 mm which would give an energy calibration of ~100 keV/um, which agrees with the calibration expected from the nominal dispersion.

In Doug's, the SYAG spectrum is compared to the spectrum on DTOTR1 and the results agree, qualitatively:

The energy calibration on SYAG implied by the image is 0.638 MeV/px. Using the image calibration of 9.91um/px this corresponds to 64.4 keV/um. Likely this means the spatial calibration on the screen is off by a factor of 1.38. Which is on the order of the difference measured between SYAG and BPM2445. This also implies that the energy spread of the beam is significantly larger than from the start to end simulations.