Minutes of 51st FFIR/BDSIM meeting on 4/23/2003

The meeting was held in a room of 325 at KEK, 13:30-15:00, 4/23/2003. We discussed on crab cavity, electron beam scattering at LINAC, BDSIM, FEATHER and others.

(1) NLC Crab Cavity Review (H. Hayano, KEK)

(transparencies, 15 pages, pdf ,829KB ) Hayano reviewed the NLC crab cavity studies based on "NLC-ZDR chapter 11.6", SLAC Report 17(1963), transparencies of J.Frisch and M.Ross 'Crab cavity phase stabilization'(1998) and J.Frisch'Crab cavity RF control'(nanobeam 2002). In the ZDR, following three types have been considered for the crab cavity ; (a) s-band LOLA-III, 3m long, 0.8~1.7MW power for beam energies of 250~750GeV, (b) x-band LOLA-X, 37.5cm long, 0.20~0.32MW and (c) x-band Lolita-X, 5.23cm long, 4.8~7.4MW. The Lolita-X is 6 cells of LOLA-X. So, the Lolita-X is the most preferable for shorter structure and smaller wake field although it must be driven with high power. The ZDR identified that the most difficult tolerance is "pulse-to-pulse phase difference jitter between cavities" to be less than 0.2 degree x-band phase within 1/5 seconds for less than 2% luminosity loss. For more than 1/5 sec, a steering corrector compensation will work effectively. Scaling to s-band, the phase difference would be 0.05 degree. Since the best SLAC Klystrons have about 0.1 degree jitter, this requirement must not ridiculous although it is tough.

J.Frisch and M.Ross proposed a system consisting of one klystron to drive both cavities, slow phase adjust with wave guide temperature control and fast phase adjust by coupling lower power RF source (i.e. an additional Klystron) to one of the cavities. The fast phases can be measured by mixers and the differences can be obtained by their comparison.

In the mixer phase measurement, typical signal is measured with 1mV/degree sensitivity for s-band. It corresponds to 50uV detection for 0.05 degree phase difference. With 1MHz band-width detection, S/N ratio between 50uV signal and thermal noise floor will be 35dB. Phase detector circuit noise can be smaller than 35dB easily. So that, the phase detection can be achievable and the fast feedback also can be achievable.

Hayono concluded that their estimation was very reasonable. So, the NLC crab cavity must be feasible.

(2) Electron Beam Scattering at Linear Collider (T.Yamamura, M2, University of Tokyo)

(transparencies, 13 pages, pdf ,610KB ) Yamamura is a graduate student(M2) at university of Tokyo. He studied electron beam scattering at the main linac. Major parameters are as follows; beam energy=500GeV, length of LINAC=13.51km, beam intensity = 0.75 x 1010/bunch, initial beam energy=8GeV, vacuum pressure=10n Torr and temperature = 300K in beam pipe. Residual gas is assumed to be Nitrogen and its density is 5.32 x 1014/cm2. There are three kinds of scattering processes to be estimated, which are (1) elastic scattering, (2) bremmstrahlung with the residual gas, and (3) Compton scattering with thermal photons. He calculated that 3.75 x 105 electrons have elastic scattering at the scattering angle of more than 20 n rad, while typical angular divergence of beam is O(10 u rad). A few tens electrons have bremmstrahlung process with the energy loss of more than 1%, where typical momentum acceptance is +/- 1 % at final focus system.@ Since the bremmstrahlung process has wider angular distribution than the elastic scattering, it must be carefully investigated for background in detectors. Finally, he estimated that only a few electrons would interact with thermal photons with Compton scattering. He will study these processes in terms of background in detectors by GEANT4 simulation@for the time being. Since the processes are major sources to make a beam halo, they are expected to be integrated into SAD too.

(3) FEATHER: Design of Movable Kicker (N.Delerue, KEK)

(transparencies, 10 pages, pdf ,4.2MB ) Nicolas begun a talk from a question of "Why do we want a movable electrode?" He shortly answered; kicker with a smaller gap can produce a higher field and thus give a bigger kick, however, if the gap is too small it may obstruct the beam. So, movable electrodes are necessary. In addition, since the FEATHER system will be installed in middle of ATF extraction line, the electrodes of kicker must be widely opened at no FEATHER experiment. He has investigated impedance performance of a general configuration consisting of a pair of shield and electrode with total six variables, which are length and thickness of shield, a gap between shield and electrode, length, thickness and position of electrode. He found a best set of variables; length and thickness of shield are 5 and 1mm, respectively, the gap is 8mm; electrode thickness should be as thin as possible (say 0.5mm which may be too thin to maintain the straightness ); electrode position should be as long as possible; electrode length is 6mm. However, acceptable working range ( 40 to 75 Ohm) is only 1 ~2mm for the electrode length.

To improve this small working range, he came up with two ideas which are (1) splitting the electrodes and (2) removing shields. For (1); if the electrode is split in 3 parts, the 2 outer parts can be applied with different potential when the electrode is moved for matching the impedance. So, we should be able to adjust the impedance within the acceptable range. For (2); removing the shields, we would have two working ranges in the electrode position, that is near to the beam and far from the beam.

(4) FEATHER: Status Report (N.Delerue, KEK)

(transparencies, 7 pages, pdf ,3.4MB ) Nicolas also reported recent beam test using a cavity BPM. He observed the cavity BPM signals with offset beams of 2~5 bunches. The offset values were 190, 200, 210, 220 and 230 um vertically. He also clearly observed an effect of kicker with 200 um offset beam. So, a feedback using cavity BPM seems to be possible, assuming position resolution of cavity BPM is nano meter in future. He implemented a GPIB communication over IP. Communication with the ATF_Control software is under study.

(5) BDSIM update (Tanabe, M2, University of Tokyo)

Tanabe showed that extraction line up to beam dump has been installed in the BDSIM based on GEANT4. So, he is ready to simulate background along the extraction line and the beam dump itself. Input data can be prepared by CAIN, which consist of disrupted beams and beamstrahlung photons.

CAIN informations are listed here;

The next meeting will be on 21May (Wed.), 2003,13:30 - 15:00 am at 3 gokan, 425 .