Parallel MD scheduled 17h after Elias finishes his tune scans. MD cancelled at 19h due to work with the SFTION2 beam steering.
Present physics beam SFTPRO2 only goes to 34 GeV (which is less than 200 GeV of MD1), so it does not excite any significant hysteresis…
MD 2024-11-26
A spontaneous MD was set up for 2024-11-26by inserting a 200 GeV MD cycle in place of the MD1.
We perturb the field by -5e-5 T, and watch the agent correct it back.
Logbook entry
We perturb the field by +5e-5 T, and watch the agent correct it back. Logbook entry
We perturb the field by -5e-5 T, and watch the agent correct it back in autoregressive mode. Logbook entry
We perturb the field by +2e-5 T, and watch the agent correct it back in autoregressive mode. Logbook entry
Observations
Field predictions are good, within 5e-5 T of the field measurements, as seen in the field predictions above in yellow.
Note that we forgot to export data during the MD, and the CycleData objects were constructed using measured and programmed current and field from NXCALS and LSA, respectively. This means that the programmed field is the true programmed field that was played for the cycle, including perturbation.
