We had the 5th IP-BPM meeting at 10:00-11:30, 6th July 2006, 325, 3-gokan, KEK.

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

Philippe reported his work for last two weeks. His presentation is here and the beam test report ( here )

He completed the report of the beam test which is one of the homework.

Cavity design for an ILC-like beam

He is trying to modify the IP-BPM for the ILC-like beam, where the major difference is the bunch length of 300μm instead of 8mm and the bunch spacing of 307nsec. His assumptions are (1) the decay time must be short enough for the bunch spacing, i.e. 100nsec , (2) the dipole frequencies of 5.712GHz and 6.426GHz are the same for the same electronics. In order to get the decay length of 100nsec, the cavity length was shorten by half. Honda commented that the effective length of cavity field, therefore the decay time, does not change even with the shorter cavity length. He suggested to change the coupling to reduce Qext to about 400. He begun to study the coupling slots ( magnetic coupling to the dipole fields ) . He requested a help for S-parameter calculation.

Electronics study (5)

With help of Honda, he measured detector output signals as a function of converter input ones in the system of converter, limiter and detector for the noise estimation. The Detector outputs have been measured as differences between the minimum and maximum amplitudes by the oscilloscope, and the RMS has been measured in "the noise region" for a practical reason. Although this method does not effect the noise estimation, the outputs must be the amplitude which is the half. The measured noise was -98.2dBm, while the calculated one is -96dBm. Honda pointed out that additional attenuation of 3dB was expected in the detector module, especially at IQ hybrid. Therefore, the noise could be consistent within 0.9 dBm. It was suggested that the converter outputs can be measured for the noise study to remove ambiguity of 3dB attenuation in the detector.

The x and y, that is 0 and 90 degree's phases in the electronics, isolation has been investigated by following three methods; (1) RF 714MHz signal was divided into two which are input to the detector and the limiter with and without a variable attenuator, respectively, (2) location of the attenuator is exchanged, i.e. variable inputs to the limiter while constant one to the detector, and (3) two RF signals with 1MHz difference were input to the detector and the limiter. Results were large coupling between the 0 and 90 degree's outputs, modulus of the two outputs was constant at each sensor input and the two outputs differed by 1% in (1), (2) and (3), respectively. Confusing results came from phase variation by the attenuator. We may need to know how to evaluate accuracy of phase detection in the detector.

Suggestions and comments are summarized below;

  1. For the ILC-like design, modification should be coupling to reduce the decay time.
  2. Full optimization for the design must include the frequencies ( reference ). The optimization could be taken if time is available.
  3. Help is needed for the S-parameter calculation by MAFIA.
  4. The noise measurement should be done with the converter outputs.
  5. New method would be invented for accurate measurement of phase detection in the detector.

We will have the next meeting on 27th July 2006, 15:00 -. 425.