FELSI Meeting 1 April 2008

Tuesday 1 Apr 2008, 14:00 → 15:30 Europe/Zurich

WBGB 020 (PSI)

1 Rene Bakker: Beam optics of the 250 MeV injector with alternative electron source configurations

FLP: what is in the 55 cm between the sections? VS: screen, BPM, 2 bellows with correctors FLP: the simpliefied diode provides less focussing, has this been quantified? RB: yes, it is up to a factor 2 VS: to ensure emittance compensation, is it necessary to measure emittance before the booster? RB: no, beam diameter is enough YK: why does the emittance decrease in the bunch compressor? RB: that is due to space charge effects (same effect as further upstream around 7 m) Discussion on alternative electron source configurations RI: why is there no diagnostic section in the case of the2.5-cell design (CERN gun)? RB: there is no space for it VS: in general, we should prefer solutions with fewer solenoids, they allow for more diagnostics YK: is the harmonic gradient of 30 MV/m for the 2-frequency cavity realistic? JYR: yes, we have verified that it's okay FLP: why do we keep the velocity bunching in the 2-frequency solution? RB: this is to keep the initial current low, as required for low emittance VS: what about a separate third-harmonic cavity after the first (one-frequency) cavity? RB: for the 2.5-cell case (CERN gun) this is not an option, the exit energy is too high; for the 1.5-cell case this could be a solution, but the matching will be very challenging JYR: what about a 12-GHz linearizer after the CERN gun? RB: this makes no sense MP starts a brief discussion on the use of FEAs in RF guns. We should revisit the effect of back-bombarding in the case of an RF gun. FLP points out that the time-structure of our current pulser is also oscillating. VS mentions that dark current will also be an issue when operating with FEAs. RB recalls a paper where the use of field emitters in an RF gun is described. Discussion on laser pulse shaping (VS), Gaussian versus rectangular. RB points out that in principle Gaussian profile can yield better slice emittance than rectangular profile, but the solution will be harder to find. JYR suggests to consider the option “pulser + 1.5-cell cavity”. VS: is the CERN 2.5-cell cavity side-coupled? At SLAC the performance improved considerably after symmetrizing the RF input. MP: there are two models, a side-coupled one and a symmetrized one. We will need the symmetrized one. FLP stresses that we should not only push for a 1-MV pulser but also reoptimize the pulser technology, which was developed for an FEA source but is now used for a photogun.

Slides FELSI-RB080401.odp FELSI-RB080401.pdf

2 Yujong Kim: Optimization of beam optics and layout for LOLA operation and emittance measurement at 250 MeV injector test facility

MP: for LOLA, did you use the exact SLAC design? YK: no, it is not the SLAC design VS: what does “no space charge effect” mean (slide 10)? YK (RB): it means it is not taken into account in the model RI: then why does the emittance change anyway? YK: chromatic effects in quadrupoles VS: the FODO cell on slide 11 seems not well matched. Why the beta beating at constant emittance? YK: there is a small beta beating effect from space-charge effects VS: then why does the emittance not increase? YK: there is a small emittance growth of 2 nm AA: at the number of particles you are tracking, you are probably not sensitive to such a small change in emittance! What space charge model are you using anyway? YK: cylindrical model, similar to BET Discussion on power supplies for quadrupoles: For the emittance measurement, it would be preferable to have one power supply feeding all quadrupoles, but for the slice emittance measurement we will need individual power supplies anyway. The power supplies should give an accuracy of 1% anyway, which suits all applications. VS: why is the LOLA cavity in the second FODO cell (as opposed to the first one)? YK: more quadrupoles in front of LOLA allow for a better matching of the beam. RB: so, is it okay to specify the quadrupoles to 20 T/m? YK: yes VS points out the need to look at slice length versus drift length (additional quadrupoles to make small spot in beam dump region) Magnet specifications: dispute between VS, who asks for two more weeks of studies before freezing any magnet specifications, and RG, who would like to have at least some preliminary (conservative) specifications to be passed on to the responsible groups.

Slides psi_xfel05_3FODO_05_final.ppt