Minutes of the ICE section
66th meeting on Wednesday 03/10/2012 (08:40-10:30, 6/2-004)
ICE members: Benoit Salvant (BS), Christian Hansen (CH), Carlo Zannini (CZ), Hugo Alistair Day (HD), Elena Benedetto (EB), Elias Metral (EM), Elena Wildner (EW), Frank Schmidt (FS), Giovanni Iadarola (GI), Giovanni Rumolo (GR), Jean-Luc Nougaret (JLN), Kevin Shing Bruce Li (KL), Nicolo Biancacci (NB), Nicolas Mounet (NM), Olav Ejner Berrig (OB), Serena Persichelli (SP), Tatiana Pieloni (TP), Werner Herr (WH), Xavier Buffat (XB).
Present/Excused: BS, CH, CZ, HD, EB, EM, EW, FS, GI, GR, JLN, KL, NB, NM, OB, SP, TP, WH, XB, Alexej Grudiev, Alexey Burov (from FNAL), Eirini Koukovini Platia, Georges Trad, Tatiana Libera Rijoff, Vittorio Vaccaro (from Naples), Simon White, Rama Calaga, Michi Hostettler, Stephane Fartoukh, Alexey Grudiev.
1) Newcomers / visitors
- 2 EPFL students will come to work with TP and beam-beam team (7 days in total until Christmas) .
2) Comments on the minutes of the previous 65th meeting + Actions
- No comment.
3) General infos
- SL meeting:
- Nothing special.
- News from LHC:
- New MKI8D => Has now the lowest temperature (see picture from Mike Barnes) => Excellent! HD will report about it next week.
- TCP heating => Info from BS: Inspection from Oliver Aberle and no pb could be found. It was nevertheless not possible to stay long due to the high radiation level (5 mSv/h) and they could not measure the flow in particular.
- TCTP ferrite outgassing => Info from BS: the ferrite outgassing results with the TT2-111R are not that good. Next step => Will be redone with higher temperature treatment (1000 deg C).
- 25 ns scrubbing run + MD 25 ns are postponed (due to new interlocks with very low thresholds which have been added on MKIs of point 8 but also point 2 to protect the equipments from flashovers and possible metallization) => New MD list with some MDs from MD4 which will be included => Do we have urgent MDs we would like to propose? For the heating, test with 1 ns? => To be discussed with the MD coordinators.
- BSRT status at the LMC yesterday by FedericoR: B1 mirror changed (with coating) => They are happy with it. B2 mirror changed (without coating) => Bad surprise looking at image quality. The amount of light (~ 2 times less due to the fact that there is no coating) seems OK but images are distorted. B1 heating ~ B2 heating => 1) Seems no effect of coating and 2) no effect of the new ferrite => Are we still above Curie temperature (as expected from simulations from M. Galasche?). To be followed up.
- Q20 optics is now used to fill the LHC.
- ADT damper => DanielV flattened the gain vs. frequency to have it almost flat until 20 MHz, i.e. really bunch by bunch. Ongoing discussions with NM, AlexeyB etc.
- New info from RalphS about new HEADTAIL monitor => They deployed their new HT monitor in the LHC (B1V) and got already some interesting data. It seems that we have very clean head-tail modes when arriving at beta*=60 cm with the mode 0 and 1 being clearly visible. Ongoing discussions to see what would be expected from theory /simulations when the same signals' treatment is applied.
- Info from GA yesterday: At the end of the fill Tobias has repeated a test to dump one beam after cleaning of the abort gap clean to avoid losses in the dump area that could lead to immediate dump of the second beam. Some pictures indicate a significant distortion of the orbit as already observed previously. The beam is still dumped because of the orbit distortion on the other beam => Being followed up by the beam-beam team.
- HL-LHC => Impedance model with crab-cavities (ongoing discussions with RamaC) and collimators (ongoing discussions with StefanoR) => Will be discussed at the future HiLumi workshop (https://indico.cern.ch/conferenceDisplay.py?confId=183635).
- Impedance meeting on MO => TDI recommendations sent by AlexejG for the impedance team.
- Discussion with Miguel Jimenez about some activity on the enamel flanges for SPS ionic pumps. It seems that the goal is to check the grounding of some pumps => To be followed up by impedance team.
- 2 AP fora organised by WH: 2-beam impedances on FR and circular modes and flat beams for the LHC on MO => See http://ap-forum.web.cern.ch/AP-forum/.
4) Recent observations of LHC beam instabilities and EOF studies (TP): pptx
- Instabilities of fills 3102 and 3108.
- EOF study on fill 3089.
- Fill 3102:
- Trimming octupoles down in collision some instabilities were seen => Only IP8 bunches affected.
- AlexeyB asked the BBLR tune shift => To be followed up with beam-beam team and add this information to the instability summary table of NM (see Action 1 below).
- Instability in the adjust when the octupoles' current is trimmed from 510 to 520 A. Why? Was something else moved at the same time?
- Fill 3108:
- 6 bunches with no beam-beam interaction at all.
- Would be good to compare with what we had before the collision process change (Action 2 below).
- EOF trimming down the octupoles without correction of the chromas:
- Lifetime decreased first and then increased.
- It was with 2 trains of 36 bunches and the special IP8 bunches were not unstable.
- Not totally clear on what we can we conclude.
5) Coherent effects in the presence of beam-beam and impedance (Simon White): pdf
- 2 models developed to study the combined effect of Beam-Beam (BB) and impedance:
- Hollow Beam Model (HBM): Based on the work by Perevedentsev et al. Allows to see all the modes and study their stability => Fast but there is no Landau damping.
- Described on slide 5, using a linearized BB kick.
- Macro-particle model: Based on BeamBeam3D code from J. Qiang. The impedance effect has been included in the code by SimonW => Slow but includes Landau damping.
=> These are 2 complementary tools to study how these effects couple.
- SimonW presented a status report of all his studies, step by step. This work was followed up in some beam-beam meetings.
- BeamBeam3D has been benchmarked against HEADTAIL with LHC impedance model looking at the TMCI with a single bunch and 0 chroma => Very good agreement.
- The next step was to compare the effect of the damper on BeamBeam3D against the new theory from AlexeyB (using an air-bag or Gaussian distribution):
- SimonW included the air-bag distribution in the code.
- Excellent agreement without damper => No instability for Q' = 0, and it is more critical for negative values than for positive ones.
- Very good agreement (less than ~ 10% in worst case) with damper => Instability now appears for Q' = 0 but disappears for Q' < - 4 (looking only until - 6). The shape of the curve of the instability growth rates vs. Q' is very different with damper than without damper and one can observe the beneficial effect of the damper as the instability rise-times are then much longer.- AB is Air Bag and not Alexey Burov => Simon generated the Air Bag in the code.
- With BB only and using the HBM, the synchro-betatron modes were then studied:
- For large ratios of beta* / sigmaz (case 100 shown) => No synchro-betatron coupling: side-bands deflected by the coherent BB tune shift + coherent modes at Q and Q - xi (linear BB kick). Yokoya factor is 1 due to linear model.
- For small ratios of beta* / sigmaz (case 1 shown) => The BB also deflects the side bands, resulting in a more complex picture (each line splits in 2). Note that in the case before there were also 2 lines but always on top of each other.
- The previous results (HBM with linear kick) have been then compared to tracking simulations with BeamBeam3D (6D Gaussian) in the case beta* / sigmaz = 1 (to enhance the coupling):
- Results qualitatively the same.
- With BB only, the beam is always stable.
- Looking at the case with BB (HO with single IP) and impedance, it was found that the beam can become unstable when mode 0 approaches mode - 1, i.e. there is a mode coupling. Note that sufficiently below the mode coupling and sufficiently above the mode coupling there is no instability anymore. The BB parameter x equal to 0.003 seems to be the most critical and this case was studied below:
- Instability seen on both beams, both Sigma and Pi modes rising.
- Chromaticity helps and the beam is stabilized for Q' ~ 2.
- For higher beta* a strong coupling seems to be obtained (the coupling range depends on beta* and wake). In this case there is really an overlap of the coupled modes (while for smaller beta* there seems to be a small gap, i.e. smaller coupling). Chromaticity still helps but need to go to higher values.
- The case of separated beams (in the H plane) was then studied:
- Same behaviour is observed when the Pi and Sigma modes cross the +/- 1 modes.
- The instability appears now in different places depending on the plane.
- The case of LR interactions was then studied:
- The separation was set to 10 sigmas and the intensity / b to 1.5E11 p/b.
- All LR are lumped in 1 interaction and the number of LR is increased by scaling the intensity => It is a first approximation.
- LR modes can also coupled to the head-tail modes.
- The alternate crossing cancels the tune shifts of the 2 planes => Always stable even for high number of LRs.
- In fact the 2 IPs are not fully symmetric (as mentioned by StephaneF) and the cancellation is only partial (as the phase advances are different) => To be looked at.
- Effect of transverse damper:
- For HO => Very effective, as even a very small gain (20 times smaller than nominal) is sufficient to damp the instability and preserve the transverse emittances.
- For LR => Damper is much less efficient. Coupling between - 1 and Sigma mode of mode 0. The instability rise time gets slower for higher gain and there is emittance blow-up.
- Effect of the octupoles' current:
- Octupoles can cure LR instabilities (here 10 LR and coupling between - 1 and Sigma mode).
- Stability obtained between - 350 A and - 450 A (with the old negative polarity).
- Stability obtained between 250 A and 350 A (with the new positive polarity).
=> The new positive polarity is more efficient by ~ 30%.
- This study will be redone assuming an impedance ~ 2 times larger than predicted as some past measurements pointed to this direction.
- For HO, octupoles have no impact on the beam stability.
- A more complex collision pattern was finally studied:
- At all IPs, 7 LRs with 10 sigmas separation in H plane.
- Adjustment such that the V Pi-mode couples with - 1.
- With BB only, all bunches are stable.
- When the impedance is included, the beam becomes unstable.
- Past simulations done with COMBI by TP revealed that with many bunches many peaks appear, the LR interactions can add up and build large tune shift and LR modes are always observed even with strong HO => Understanding the complete picture may be difficult with a single bunch approximation and there is really a need to include the impedance in COMBI (See Action 3 below) and make some experiments.
- StephaneF mentioned that in the case of more LR and more separation, the same result as for 10 LR with less separation could be recovered.
- StephaneF proposed to check the octupoles strength vs. tune split between the 2 beams during MDs. SimonW said that it was done at RHIC with HO and they had some emittance blow-up. But here it is for LR. To be followed up => A dedicated MD should be proposed to check some of the results.
- WH reminded his preferred scenario to try and collide HO before the squeeze to provide all the necessary Landau damping during the critical phases => An MD will take place during MD block #3.
- Summary of this study:
- Coherent modes in the presence of BB and impedance can couple with much lower threshold than TMCI (from impedance only) and drive the beams unstable.
- It appears to be driven by mode - 1 in all cases.
- HO (driven by the core particles):
- High chroma helps.
- Damper very efficient.
- LR (driven by the tails particles):
- Chroma still needs to be investigated.
- Damper much less efficient than for HO.
- Octupoles can stabilize the beams (acting mainly on the tails) => Case with 2 times the nominal impedance to be simulated.
- Next steps:
- Crossing angle dependency.
- Check with AlexeyB's theoretical approach.
- Try and verify some of the results experimentally.
6) Actions
- Old actions.
- New actions:
- Action 1 (NM, TP and EM): Maintain the LHC instability summary table adding all the interesting and relevant info (beam separation, LR tune shifts, etc.) and put it available for everybody on the ICE web page.
- Action 2 (TP and XB): Compare the recent instability observations with what we had before the change of the collision process and also with we had before the change of the octupoles's sign.
- Action 3 (TP): Introduce the impedance in COMBI (to be able to simulate the interplay between BB and impedance with all the bunches).
7) Miscellaneous
- Deadlines and important dates for ICE.
- The next (67th) meeting will take place on 17/10/2012 => Agenda:
1) Impedance simulations of the updated design for the PS dummy septum (SP).
2) Review of MKI8_D heating (HD).
3) Highlights from the last MD block #3 (all MD participants).
- See preliminary agendas for the next meetings.
Minutes by E. Metral, 12/10/2012.
