RF summary

• Resonant excitation at flattop performed with a single sine-modulation at 97 % of the central synchrotron frequency ~20 Hz.
  • This induces on purpose dipole oscillations around 20 Hz and changes the bunch shape in a lossless manner
  • A 1 deg excitation amplitude is needed, as below this value, the resonant excitation does not take place
  • Similarly, the minimum length of the excitation is 30 s for it to be effective
  • In B1, this results already in 50 % of the dump threshold (which was probably the case during 2025)
• By applying voltage partitioning on B2, the full detuning algorithm converges to a different phase modulation in C3B2 than in the other cavities
  • Therefore, the RF bucket created by 3B2 is different from the other cavities, and the bunch length is modulated along the ring
  • The spread in bunch length seems to result in exceeding the dump threshold
• Actions: yesterday we reduced the partitioning from 1.87 MV / 1.0 MV -> 1.8 MV / 1.4 MV
  • See https://logbook.cern.ch/elogbook-server/GET/showEventInLogbook/4512747 and https://logbook.cern.ch/elogbook-server/GET/showEventInLogbook/4512688
  • The bunch length spread was reduced from 170 ps to 100 ps in B2, while we have 50 ps in B1 
• Further options: 
  • (1) reduce voltage on 3B2 e.g. to 0.5 MV and keep half detuning on this line
    • Should reduce the bunch length spread further
  • (2) remove voltage partitioning and keep 1.5 MV/cavity
    • Accept quenches on M1B2 from time to time