We had the 3rd IP-BPM meeting at 10:00-11:40, 6th June 2006, 425, 3-gokan, KEK.

Participants: P.Doublet, Y.Honda, T.Hino and T. Tauchi

Philippe reported his work for last two weeks. His presentation is here.

First he presented the MAFIA calculations with smaller mesh step of 50μm from 1mm . He calculated the mode voltages as a function of displacement in two directions. There is always an offset in the output voltage, while the voltage is linear to the displacement. The offset is artificial since it depends on the mesh step.@Also, he calculated the equi-potentials in X and Y dipole modes as Honda suggested. The result shows very small coupling between the two dipoles as expected. He found that MAFIA can not practically calculate properties in small distances of less than 100μm for increasing CPU time. He expressed that he would like to resolve this issue by re-scaling the cavity shape in the calculation.

There were comments on the MAFIA works. Major purpose is to design a cavity BPM with various error estimations of the resonance frequencies, gains, coupling between cavity and wave guide. It must be impractical to check the performance at nanometer level by the MAFIA calculation. The nanometer performance must be determined by the electronics noise. Also, he was suggested to talk to Alexei (UCL) for the calculation method.

Second, he reported results of the electronics test with answering the previous remarks. The converter gains of reference and sensor signals have been carefully measured to be same, i.e. 23dBM. The coupling between 0o and 90o detector outputs has been measured by two methods. Adjusting the two phases once at the beginning, the 0o signals were measured as a function of the 90o ones. The coupling was -6.656dB, which is too much. Next, inputting signals with 1MHz difference from the reference one, the 0o signals were measured at maximum of the 90o ones. The resultant coupling was -17.79dB, which is also larger than the desired value of -30dB. Also, he measured the band widths of the converter, limiter and detector modules. Results are consistent with the design values of 400MHz, 300MHz an 300MHz in the converter, limiter and detector, respectively, whose input frequencies are 5.712 (6.426)GHz, 714MHz and 714MHz, respectively.

Suggestions are listed below;

  1. Coupling of 0o and 90o signals in the detector; The phase should be adjusted with maximum input signal, while he adjusted it at minimum one. Also, phase difference between the maximum 90o signal and the minimum 0o one should be measured as a function of the input amplitude.
  2. Noise can be estimated from linear extrapolation to zero input.
  3. BPM signal can be simulated by the RF signals with external gate or fast RF switching device.

We will have a Tuesday night shift for first measurement of prototype IP-BPM at end of LINAC, for 21:00, 6/6 - 5:00,6/7 .

We will have the next meeting on 20th June 2006.