![]() |
LPC meeting summary 21-10-2024 - final |
![]() |
Main purpose of the meeting: Schedule planning, final HI run plans and BDSim for FASER/SND
LPC minutes 21st October
Introduction (Federico Alessio)
(Catrin Bernius) Is the intensity stated at the start of the run or at the start of stable beams. It is at the start of stable beams.
(Jorg Wenninger) We stop at around 13:15 for a visit in ATLAS (not the tunnel) 13:30-14:30 then half an hour patrol and then back at 15h.
(Roderik Bruce) So in the timetable with the VIP visit being less than half a day we should only just fit the 2.5 days of PbPb setup and the last bit of the pp ref setup as we haven't extended the setup period for the VIP visit being included, as it is less than 1/2 a day.
(Reyes Alemany Fernandez) Can you confirm what is meant by +/+. This is solonoid and dipole magnet. Neither are the external crossing angle. This is the preferred one for the setup and first running.
(Catrin Bernius) For ATLAS the crossing angle sign should be positive, like in ion running.
(Jorg Wenninger, Witold Kozanecki) What is the desired configuration for the vdM? 8e10 ppb and 3um emittance, but the main important thing is to have the beam tailoring as in the main pp vdM. The lower the intensity the better to get a mu lower than 1.
(Chris Young) These plots don't contain the 2026 luminosity so even after just 2025 IP5 ends up at the radiation limit for the triplet, with no margin for any 2026 configuration.
(Roderik Bruce) It should be stressed that this is increasing the background with respect to this year which was already significantly higher, by a large factor, than 2023.
(Catrin Bernius, Jorg Wenninger) What would flat optics look like, what b* would it become flat? Probably round until 60cm then flat until something like 60 cm / 18 cm.
(Jorg Wenninger, Catrin Bernius, David Stickland) We have investigated a little further on the D1 replacement. The issue is that replacing the magnet would disrupt the survey alignment system for the triplet. This adds significantly complications/problems and the triplet would need to be re-aligned and the bypasses to the experiment would also need to be opened to redo the alignment. Would this extend the time? There is probably time within the 6-7 weeks as some things like the bakeout could be done in parallel to the alignment. Would this be in this shutdown. Ideally, yes.
(Filip Moortgat, Jorg Wenninger) The motivation for having different configurations for the two years would be the triplets which go beyond 30 MGy? The limits are almost respected with nom-H both years; 30.7 MGy, but it is obviously not as good as RP-H.
(Michi Hostettler, Jorg Wenninger) Would this D1 exchange be done preventive-ly? Ideally yes. If this happens in the middle of the summer it would mean a large downtime in the middle of the year.
(Chris Young) If we go for the nom-H scenario then we go over the moderate damage limit in 2025 which motivates the replacement this YETS.
(Catrin Bernius, Jorg Wenninger) So if the replacement is not done then we will definitely do the RP-H scenario? No, there is also the box suggesting having the same as this year at IP1 for one of the years. The RP-H scenario is highly disfavored by the forward physics experiments so avoiding it if possible is the idea. It for example it would be a compromise to the forward experiments to only do RP-H for one of the years. But this does mean extra commissioning in 2026 as the configuration would be different.
(David Stickland, Jorg Wenninger) For IP5 the decision is made. The discussions are more around IP1 but it isn't final what will be done at IP5 either. For the IP5 forward physics there is a necessary
(Chris Young) Next week we will go through this in a bit more detail with the details from the LBOC.
(Jorg Wenninger, Chris Young) On the 6th November there will be a presentation at the LMC by Matteo and then there will be a decision by the end of November. From the experiments side we can give our opinions. Last year there was thought to be little effect on the experiments which model was chosen so no opinions/preferences were stated.
(Eric Torrence, Chris Young, Catrin Berinus, Roderik Bruce) Have the radiation models been updated to account for the different/increased backgrounds observed in the experiments with respect to the FLUKA models? No, these are still from FLUKA. The dose on the triplet is radiation coming from the IP outwards rather than radiation from the collimation system seen in ATLAS. But AFP also see higher radiation. The simulation should account for the changed polarity but the particles are not tracked out to AFP or FASER/SND so we don't have predictions for the radiation so far downstream. When the simulations are extended then down to FASER. Are the simulations thought to be reliable? It is the best we can do but won't be perfect.
(Federico Alessio, Jorg Wenninger) Would you want an experimental opinion by the 6th November or can we wait until after the LHCC where we can also bring what the machine says at the LMC? For example, for replacing the D1 magnets we need to decide soon, but this is a ATS decision. In any case the decision should be made in November.
BDSim simulations of FASER/SND Background (Alex Keyken)
(Chris Young) Just to clarify is your label of "2025 proposal" is nominal polarity with horizontal crossing? Yes. So this is the one we can only test in simulation, as opposed to the RP configurations which we tested in MD and the 2023 nom-V scenario.
(Chris Young) I had a look at the FLUKA results in comparison to your slides and they also see a large reduction of the problematic high E positive muons compared to 2024. They see slightly lower overall flux, and lower high E flux compared to 2023. You see slightly higher overall flux and lower high E flux compared to 2023. So the two simulations show similar things but don't agree on everything.
Schedule Slides (Federico Alessio)
(Roderik Bruce, Robert Helmut Munzer) There is a small optimization of doing PbPb and pPb in the same run, but not the full 7 days, as the Pb beam would be already setup. The most physics days is to do only PbPb and all in one period, but if you are doing pPb then it is still better to do it within the same period as PbPb. We should roughly quantify what would be gained in such a scenario.
(Andrea Ferrero) It is not only scenario B where pPb can be done, it can also be done in scenario A by splitting the period between PbPb and pPb.
(Roderik Bruce) It is certainly the case that we don't want to do risky MDs before.
(Andrea Ferrero) Why is pPb preferred for 2025 cool-down rather than 2026? We estimated that pPb is still much lower than low-mu pp (which is excluded for cool-down), but it is much higher than PbPb. CMS need multiple weeks of cool-down after high-mu pp running. We need to evaluate if pPb counts as cool-down.
(Georges Trad, Catrin Bernius) There is also a limited number of MDs that can be used as cool-down as many MDs require significant number of collisions. Usually RP consider more than ~100 colliding as not counting as cool-down. We cannot sacrifice half of the MD program to just be cool-down at the end of the year. Additionally we need MD before the TS in advance of the ion run for the ZDC installation.
(Reyes Alemany Fernandez, Chris Young) In scenario B is it considered when the pp intensity would be raised above 1.6e11 ppb? For next year we plan to go to 1.8e11 ppb. Going beyond this is not planned, and will suffer from other limitations of the machine such as electron-cloud. For pushing right to the limit there is a bit of a competition as it is risky so would be done at the end, but doing it just before the ion run is not desirable, and it cannot be after due to radiation.
(Chris Young) At the last LS3C it was looking very much like the YETS will be 11 weeks rather than 9 weeks.
(Reyes Alemany Fernandez) When you say 2024-like for ions, you mean the projected performance? Yes, exactly it is the predicted performance.
(Reyes Alemany Fernandez) Why is one scenario in green? This is because it is the only one we don't think that there will be at least one experiment which says that they wish to veto that scenario. It is not a recommendation per say. The final decision depends on the recommendations of the LHCC and the decision by the research board or research director.
CMS (Filip Moortgat)
(Chris Young) How long would it usually take to get the magnet back after such an incident, without the regeneration. Probably about 12 hours, but this is the 13th one in the lifetime of the magnet, it is a major incident and there is significant follow-up. It should not be thought of something that happens regularly.
(Federico Alessio) What do you mean by not optimal on slide 3? By this we are referring to the total ion physics days rather than the amount of cool-down. For cool-down it is fine.
(Federico Alessio) By large PbPb dataset what do you consider is large? They are looking at dataset sizes above the target for example 6 nb-1 but we can get back to you on when we would be fine to switch to pPb.
(Chris Young) On page 3 I'm confused. At the LHCC CMS we fine with stopping several months later than the end of June, but now it is absolutely essential to be in on the first day of LS3? Has something changed? CMS were always unhappy with the length of LS3 and wanted a larger extension, but were willing to live with it. They do not wish to waste the extended time by not being able to start work on the first day.
(Federico Alessio) On the risk assessment our perspective is that it limits the testing that the LHC can be done at the limits of parameters and the choice of the optics needs to be very conservative. At least this needs to be folded into the decision.
(Chris Young) The other consideration on the risk assessment is a problem which occurs in 2025 which takes several months, like the vacuum leak in the triplet, but then we still have the run in 2026. In this case we can re-allocate days. If such a failure happens in 2026 and all the ions are in 2026 then we will lose it all. If there is an accident then we will lose something, either ions or protons, and extending the run could be considered. Whoever is still to go loses for a late problem.
(Catrin Bernius, Jorg Wenninger) Can we consider what the impact of different failures are? For magnet problems there is a table in Matteo's presentation, but in general there are too many components that could fail and also unknown unknowns such that it isn't possible to outline all possible scenarios.
ATLAS (Catrin Bernius)
(Federico Alessio) Similarly can you define significantly larger? We were looking at the 5.3 nb-1. This makes sense are we have recovered almost all the days expected at the start of the run. However, this also included a pPb run. ATLAS is not interested in pPb. The point we are more concerned with is the ratios of luminosities to those from Run 2.
(Andrea Ferrero, Roderik Bruce, Federico Alessio) Different ratios for PbPb and pp luminosities is by design as Run 3 is HL-LHC for ions while Run 3 is not HL-LHC for protons. This is why the luminosity projections are significantly better fractionally for PbPb compared to pp. This you can also see in the targets where the targets at the start of Run 3 are larger fractions of the Run 2 luminosity for PbPb. Also, it should also be noted that we are removing the pPb run from the original Run 3 plans. From ATLAS's point of view we are not interested in pPb so didn't include this as a loss. If we reach the PbPb targets with a period still to go then the experiments need to decide what is done in the additional period, it could even be pp. ALICE pointed out that these days were allocated to ion physics so it shouldn't be pp.
(Federico Alessio, Chris Young) Do you veto scenario A? No, we don't as the limitation is the work by the vacuum group on the beam-pipe shielding and we are waiting on an update. Is this considered then an LS3 committee issue as it is the vacuum group rather than ATLAS who need to do this work? Yes, exactly.
LHCb (Elena Dall'Occo)
(Chris Young, Federico Alessio) Is the desired split in lumi half/half for the SMOG2 gasses? No unfortunately not. The statement is under the assumption of two very long fills. LHCb will work out the desired target lumi for each configuration. Can this be done at any time? In principle yes, but it needs the vacuum group and RP. It should be more efficient if done during the day. Optimistically with long fills dumped by OP we should know when it will be dumped. For 1 hour it is almost in the shadow of the ramp-down.
(Chris Young, Federico Alessio) By 0.3 pb-1 do you assume this would be 0.15 pb-1 for each of p-Pb and Pb-p? Yes, we always assume we would do both direction. Is the species reversal a condition of the run being of interest to LHCb? This can be confirmed with the community, and which direction we would be interested in.
(Chris Young) Is the data output the same for pp, pPb, PbPb so there is no difference between them for a disk/tape point of view. They are all the same as it is driven by the trigger output.
(Catrin Bernius, Federico Alessio, Chris Young, Reyes Alemany Fernandez) You mentioned the pO/OO run in the Intro slides. Is its position in time being discussed? No, it is basically fixed as 2026 is too short to do it then switch to ions, and ions are also required in the north area. It is more that we were asked to mention it in the LMC presentation.
ALICE (Andrea Ferrero)
(Chris Young) If it is found the pPb doesn't work for cool down then if we wait until we make the PbPb target before considering switching then we will end up doing all PbPb. It is key to work out with RP if pPb can be used for cool-down. With 2024 we will know a bit more about the performance but the priority is still to take the full PbPb dataset.
(Federico Alessio, Filip Moortgat, Catrin Bernius) Do you still want Scenario A shown at the LMC? It is fine to show it and state that CMS vetoes it. We will show all three scenarios. ATLAS also favour C over B so desire all three are shown.
(Reyes Alemany Fernandez, Jorg Wenninger) Scenario C might be vetoed by the injector complex as it takes 6 weeks to stabilize the ion source and propagate the beam through the complex. This makes it tight to get the ions to the LHC in time and ensure the performance is good. Between 2025 and 2026 we are not meant to re-commission the machine as we won't be starting everything from scratch, particularly in the PS and SPS, but not LINAC and LEIR. There should be enough weeks for this to work in the schedule after the proton chain is complete. This is why we show these at the LMC from the experiments point of view and see what the reaction from the injectors and LS3C is.
(Federico Alessio) If ALICE and LHCb are both keen on pPb it would be good if they could agree on what is needed in numbers of days and p-Pb / Pb-p or both. We will work out nb-1 per day for some different beam scenarios.
(Federico Alessio) What is the pp ref target based on? The 5.5 pb-1 is based on the 6.5 nb-1 of PbPb.
(Federico Alessio) On Wednesday morning, before the LMC presentation, we will discuss the schedule at the spokesperson meeting also including the research director.