Gmail - RE: Meeting yesterday Yine Sun RE: Meeting yesterday Power, John Sun, Feb 18, 2007 at 9:17 PM To: "Kim, Kwang-Je" , "Gai, Wei" , "Harkay, Katherine C." , philippe.piot@gmail.com, Yine Sun Hi All, Here’s a short summary of the meeting. Please feel free to correct or add anything that needs fixing. (I’m sure I left out a few things!) *) Attending: Yin-e, Philippe, John, and Kwang-Je *) E-x Conceptual understanding. Yin-e is writing up her Matlab simulations that track individual particles through the first-order beamline. The goal of the simulations is to develop a simple explanation of the e-x process. The difficulty is in choosing the initial particle coordinates so that the results are clear. In particular, this means choosing initial coordinates that do not couple transverse and longitudinal coordinates. *) Deflector Cavity Sims: John showed MathCAD simulations of the TM110. The most significant result is that the cavity gives a transverse offset of the beam centroid on passing through the cavity. (Philippe used a Matlab-based code to verify this result so we now have high confidence that the effect is real.) The simulations also reveal that while the size of the transverse momentum kick does not depend on frequency for a given field, the size of the offset does. For example, a 4 mrad kick has a 5 mm offset at L-band, but a 40 um kick at X-band. This is likely the reason that this effect is either unknown or ignored by other researchers. Three options were discussed for dealing with the offset. 1) Design the back-half of the e-x beamline so that it can be translated. This is will not be trivial. Aside from the practical difficulties of using bellows and a translation system, we would also need to worry about re-aligning after each translation which may make the beam tuning difficult. 2) Use a pair of magnets to recenter the beam after the deflector cavity. This is probably the least attractive option since it will actually change dispersion which will affect the exchange. 3) Tilt the deflector cavity to cancel out the offset. This would have similar practical difficulties as ‘1’ since we would need to install bellows at the front and back of the cavity, develop a way to rotate the cavity in a controlled manner, and check the alignment after every rotation. *) Initial Phase Space. Kwang-Je suggested that we use an elliptical laser spot to generate a initial normalized phase space of (10, 3, 3) and e-x it to (3, 3, 10); thus avoiding the additional complexity of flat-beam generation for this PoP experiment. Ideally, we will do this at as low of a charge as possible (say, 100 pC) in the hopes of avoiding a space-charge dominated beam during the e-x. The only constraint of the lower limit of the charge is detection. *) Diagnostics. We agreed that the longitudinal and transverse emittance must be measured before and after the e-x beamline. Discussion of diagnostics is still in the very early stages and should be taken with a grain of salt at this point. While transverse emittance measurements should be manageable, the longitudinal emittance measurements are problematic. For transverse emittance we may even be able to use a quad scan since the charge is so low. Otherwise, a pepper pot or a slit is still straightforward. Options for longitudinal are: +The AWA has never done a measurement of the longitudinal phase space, but has planned (for several years) to mount a Be window on the spectrometer exit and use a streak camera to look at either aerogel or frosted-quartz plate for a true energy-time correlated meaurment. +Sweeping the phase of a 2nd accelerating linac (not currently installed) and measuring the energy spread to accomplish a “longitudinal quad-scan” ala Dave Dowell. Lastly, Philippe suggested that we use TOF monitors at the front and back of the e-x beamline in order to create a map of the longitudinal transfer matrix. Philippe suggested that we use an e-o based scheme for the TOF measurements. While we should consider this, we must consider that this technique is still in development at the AWA. John suggested that we get help from Bob Lill at APS to implement a stripline BPM-based system that uses the Analog Devices phase comparator chip which is capable of ~1 degree RF. Philippe will investigate what resolution the TOF system needs. *) The time of the e-x meetings have now been changed to Thursday’s since Philippe does have teaching obligations on this day. *) Next week’s agenda: Yin-e: look into the diagnostics John: look into the possibility of tilting the deflecting cavity in order to avoid the offset. Compare the first-order cavity kick (x,x’, z, delta) to the exact trajectory. Calculate the transverse shunt impedance of the MWS cavity to the TM110 pillbox. Philippe: Will make an estimate of the timing resolution needed from the TOF monitors in order to create a longitudinal transfer map. Cheers, John