WP11 meeting
Present: G Bellodi, A Grudiev, A Lombardi, R Scrivens, H Pommerenke, I Gamazo
No comments to the minutes of the previous meeting.
The meeting starts with a discussion about project scheduling.
RS clarifies that there are 3 distinct phases of the RFQ3 project :
1. Materials and beam dynamics study
2. Full design
3. Production
A conceptual design report should be drafted at the end of phase 1 detailing choices for material, construction technique, beam dynamics design and RF requirements as well as providing an estimate of costs and benefits (to PSB performance). This should be submitted to the management ideally by the end of the 2021 before moving to phase2. AG comments that RF studies of the baseline scenario (annealed Copper as material and same construction technique as for Linac4) will be ready by the end of the year, whereas a study for new materials and construction techniques would only be likely to be completed by middle 2022. Only by the end of 2021 it will be possible to confirm if promising new materials to build RFQ3 exist (CuCrZr)?
An update on the current status of the RFQ3 beam dynamics design was then given by AL.
Baseline assumptions for the study are an input beam current of 60mA in 0.4 mm mrad normalised RMS emittance, a maximum field on vanetip of 35MV/m, and a rho/r0 ratio of 0.75. The RFQ length has been split into a capture and an accelerating region. Only the first one has been studied to date. The accelerating part will be optimised in the following once the design of the first part has been fixed. Three different versions have been studied for different input beam energies: version 1 for 35keV, version 2 for 58keV and version 3 for 45keV beam energy. The vane voltage was also varied for each version, while staying below a maximum field of 35MV/m. Low values of the input beam energy ease the formation of longitudinal emittance and yield a shorter length design, but overall transmission is below 85%. Parametric resonance losses occur due to the fast phase ramp. High input beam energies yield very good capture and transmission (~95% on paper) at the cost of higher longitudinal emittance (for which there is some margin available in the DTL). Staying at 45keV input beam energy (as the current Linac4 RFQ) one faces a tradeoff between transverse acceptance and longitudinal emittance delivered to the DTL (namely a long version with good capture at lower RFQ voltage and a shorter version with lower transmission at higher RFQ voltage).
Next step will be to study the accelerating section of the RFQ for all cases having a transmission >90% and to track the beam forward to the DTL.
The two-term potential vane shape was used for these studies and the maximum voltage was intentionally kept below a maximum of 75kV in order to allow a margin to improve transmission and beam quality in operation. AG comments that it would be worth checking how much would be gained in performance using sinusoidal vane shape and higher voltages or trapezoidal vane shape, which is more efficient for acceleration. A varying transverse radius of curvature along the RFQ length should also be studied. A smaller rho/r0 would lead to an increased gap and different capacitance with effect on higher order multipoles. Modulation and phase would need to be re-adapted to have the same focusing.
RS asks about power requirements. For the present RFQ 500kW are needed for 78kV voltage. For the same shunt impedance power goes linearly with length and with the square of the cavity voltage. There are no particular limitations on the couplers, as the design will in any case need to be revised.
It would also be worth comparing to the design of RFQs in other labs.
The next WP11 meeting will be held on September 2nd.
Minutes by GB, 16/07/2021.