Choose timezone
Your profile timezone:
LIU Beam Parameter WG meeting #7
→
Europe/Zurich
874/R-018 - MEETING ROOM PRIORITY BE OP (CERN)
Follow up from last meeting
Large longitudinal emittances in PSB
- Scheme for longitudinal emittance blow up in PSB still quite complicated, relying on 1) C16 reproducibility, and 2) control of the phase between C02 and C16. Point 1) was relying on a firmware that should have been deployed by now. A few MDs are ongoing on this subject. Besides, also simulations needed further refinement (including intensity effects, ramp to 2 GeV). Bettina will check with Simon progress on this side and we aim at a follow up at next meeting
- There was an idea to benchmark the longitudinal emittance blow up scheme based on resonant excitation with the PS longitudinal emittance blow up. This was followed up immediately after the meeting with some simulations that seemed to reproduce what is done in the PS. To be followed up with Simon, Heiko, Elena.
Transverse feedback in PS (Guido Sterbini)
- The PS transverse feedback system is useful for many applications:
- Damping of injection oscillations induced by missteering (steering errors, ripple of injection kicker waveform, recombination kicker rise time). Applying a simple formula for emittance growth associated to a certain missteering, it turns out that to keep the emittance growth of LHC beams below the target 5% the missteering error should not exceed 1 mm. In daily operation the typical value of residual steering errors after correction of the transfer line steering are already about 1 mm. Furthermore, a radial jitter of about ~1 mm results from the B field reproducibility at injection in the PS, however this should correspond to injecting on a dispersive orbit, which is different from exciting betatron oscillations from an offset injection. For high intensity beam, e.g. TOF, the injection oscillation will not produce a relevant emittance blow up, but it usually results in beam losses due to the machine acceptance. It has been proven in operation that these losses can be reduced by using the transverse damper.
- Beam stabilisation against instabilities:
- Headtail instability at injection, this is helpful because it would allow keeping high working point and low chromaticity on the injection plateau for LHC beams.
- Instability at transition seems to be beyond the bandwidth of the damper.
- Electron cloud instabilities at flat top: it was demonstrated in the past that the H feedback can delay the onset of these instabilities.
- Kicking selectively one bunch at 2.5 GeV for the 80 bunch train production. Three windows for gating the feedback system are available. The technique could be either to excite a bunch out of the machine with the feedback system or to turn the system off on a selected bunch, so that it gets lost due to instability. Until now, the gating has been proven to work, but a significant fraction of the selected bunch still remains in the machine. Although presently the 80 bunch production scheme can also be realised by firing the extraction kicker at 2.5 GeV to lose one bunch (which has resulted into a remarkably clean procedure), the ultimate production mechanism is expected to rely on the feedback system in order to avoid firing the extraction kicker twice per cycle.
- Excitation for population of islands for MTE and a few more exotic applications envisageable for the future (e.g. slow extraction, quadrupolar excitation and correction of injection mismatch)
- Its amplifiers are used for the chirp exception to enhance the signal in the tune measurement.
- The different blocks of the transverse (H and V) feedback system determine the overall performance in terms of power and bandwidth. In particular, the system can be used to full power preserving the linearity of the response of the amplifier up to ~20 MHz, while for higher frequencies the available gain is significantly reduced. High power is needed for the damping of the injection oscillations, while against instabilities usually it is only the bandwidth that matters (the feedback prevents coherent oscillations from building up, therefore it only requires very little power). The H and V planes are very well decoupled (small resonance observed close to 80 MHz)
- Each amplifier has 3 kW peak power but is limited to 800 W RMS. Therefore, for a constant excitation the signal delivered by the amplifier decays with time in case the RMS power is too high.
- The system needs to sweep the harmonic excitation (open loop) and the delays (closed loop) automatically along the cycle to keep synchronism. This is done by the DSPU, which has been checked to correctly follow the revolution frequency of the beam along the cycle.
- New features of the DSPU are
- The option to have a quadrupolar configuration of the kicker
- Gated excitation/feedback. This can allow bunch-by-bunch feedback as well as the production of the 80 bunches.
- Multi-harmonic excitations (checked), sweeping excitation (via GFA) and external excitation (latter two to be checked).
- Synchronisation is crucial. This depends on the delay due to cables and electronics. This has been measured and the region for the optimal response has been identified. It also depends on the fractional part of the tune and needs to be robust against tune variations and coupling. Measurements show a rather flat response around the machine tune of 0.25
- For the future development of the transverse feedback system, the commissioning of the new DSPU needs to be completed (e.g. to have the GFA working), a new operational interface needs to be designed with the right knobs, new amplifiers (which are the present bottleneck) are expected before LS2 (5 kW instead of 3 kW peak power, 100 MHz instead of 20 MHz, the design is made at CERN within the RF group because the specifications are very demanding for commercial devices), a new matching network is expected by this year.
- Still lots of MDs to be done in 2016 (list in the slides)
- At the LIU beam parameter meeting on 22 July, we will resume this subject with Guido providing an update on MD progress and a list of functional specifications for the transverse feedback system, and Wolfgang B. summarising the studies on the emittance growth at PS injection for present and future LHC beams.
There are minutes attached to this event.
Show them.