Minutes and Summary

Main purpose of the meeting: pp ref and PbPb preparations and 25/25 optics discussion

LPC minutes 28th October

Introduction (Federico Alessio)

(Catrin Bernius, Dragoslav Lazic, Mirko Pojer) Can you clarify what the flexibility is for the ALICE/LHCb scan? There is a request for an access before the 400b either after 75b or after the vdM. We therefore we want to have the flexibility to place the access during daytime by doing it in either slot. Will AFP access be possible during the CMS access? Yes. The access is scheduled to be 1h(RP)+2h assuming it is done before the 400b step. There is a parallel intervention in the SPS to setup the cavities for the ions such that we need to ensure that we are not wanting injection during this period.

(Witold Kozanecki) In the 400b ATLAS/CMS do the LSC. Maybe we need to turn down the injected intensity and ensure that one of IP1/5 is head-on at all times.

(David Stickland) CMS were thinking of doing the LSC in the 400b but the CMS scans don't hit any of the limits.

(Roderik Bruce) The net angle is 71 urad due to the 210 urad external angle. For ATLAS/CMS the crossing angle will be 150 urad.

(Andrea Ferrero) Does this mean we are not interleaving the ALICE/LHCb scans in PbPb. No, each gets their own fill.

(Catrin Bernius) For the first step of the ramp-up we agreed that we wanted stable beams and do we know how many bunches there will be? 4 bunches doesn't make sense so the same one

(David Stickland, Roderik Bruce) What are the bunch intensities at the end of fills expected as last weekend they were reaching levels that were below the thresholds for registering in CMS. It isn't expected that fill will be kept as long as they were in this test and injection should be at higher intensity. We should only get to these levels if there are problems in the injectors and have to keep the beam longer. When they get this low and BPTX cannot register them CMS loses some data-taking efficiency.

(Roderik Bruce, Chris Young) If pPb cannot be used as cool-down how does it affect the planning? It would lead to more discussion. It could be that in scenario B pPb could be taken in 2025 if this was seen as a priority. Also if it is decided that PbPb is done in both periods then the efficiency arguments are stronger. Currently we are looking at recovering the full Run 3 program. You could also consider putting everything in 2026 and starting with pPb and then having PbPb afterwards.

(Andrea Ferrero) Why is it such a strong constraint to start LS3 on a Monday? This is because there is a plan for the "first week" in place so it cannot be a fraction of a week.

2025/2026 Optics configurations (Chris Young)

(Brian Cole) For ZDC we want vertical upward crossing in IP1. It is surprising the LHCf request something different. This will be checked.

(Matteo Solfaroli Camillocci, Michi Hostettler) On a different topic the vdM filling scheme doesn't work due to the AGK length and this is about to start. There is a version with double the number of injections that will work but this means more time at flat bottom. If the scheme is changed the the trigger menus need to be changed. This also applies to ALICE/LHCb. The source of this is because the AGK is set to 3x36b which is a very short train. Changing the AGK and then back again takes quite a long time. In parallel to the meeting people will discuss a solution.

(Catrin Bernius, Matteo Solfaroli Camillocci) When will we know about the possibility of replacing D1 this YETS? There is also the possibility to do it in the next (25/26) YETS, with some limited risk. Is it acceptable to be in the middle of the moderate damage band? These thresholds for the D1 are much more believable than for the triplet but we are still only hitting a single point, so it is probably an acceptable risk to be in the moderate damage zone. It is probably better to do it this YETS, as it is longer. However, the issue is really the prioritization of the teams rather than any financial implication.

(Andrea Ferrero, Matteo Solfaroli Camillocci, Chris Young) What is the impact on the other IPs of a failure of D1? There would be significant re-commissioning and we would need to use higher power to the other dipoles. This could result in more trips ie. beam dumps, but the performance at the other IPs would not be affected after returning to operation.

(Filip Moortgat, Matteo Solfaroli Camillocci) Why does it take so long to replace the dipole? It is extremely hot, and additionally the triplet alignment system needs to be dismantled and then re-installed afterwards as it sits on top of the magnet. It is a more complicated situation than the picture suggests as these are the other D1 dipoles.

(Roderik Bruce, Chris Young, Matteo Solfaroli Camillocci) Are you coupling the discussion of the schedule and the optics as with riskier optics you put the ion program at the end of 2026 at a higher risk? We were not particularly considering this seeing as it is only one of the possible sources of a problem with the machine. However, in the case that all the ions are at the end of 2026, 2025 pp running becomes significantly longer such that the 2025/2026 YETS is after significantly more radiation meaning replacement this YETS becomes more important in this scenario. We would take much more than 100 fb-1 in this case. For the ion program we can still operate the other IPs but if it happens eg. 1 week before the ion run then it will probably still be lost due to the re-commissioning time. It also depends a bit on the failure mode. It could be that the magnet starts to trip more often rather than abruptly failing.

(Jamie Boyd, Chris Young) From the FASER side the obvious preference is to replace D1 and run with the nom-H configuration. However, the requirement of having the same configuration in 2025 and 2026 to save 1 week of commissioning seems too strict when you are considering canceling significant fractions of the FASER/SND programs as well as the forward physics program of ATLAS. Why is it not considered to do 1 year in each scenario to ensure some data to these programs in the case that D1 cannot be replaced? Would it be preferred to have 2025 as the good one for FASER? Yes. For 1 week of data-taking it seems an extreme choice to close so much of the physics program.

(Mirko Pojer, Matteo Solfaroli Camillocci, Chris Young) It is also possible that we will see signs in advance of more frequent trips and then we can replace the magnet rather than trusting the number that might have a large safety factor. It should be possible to do the replacement in the 25/26 YETS as it will likely be 11 weeks. We could also argue that we built a spare so we may as well use it as it will be thrown away before the HL-LHC as it is not used in the HL-LHC.

(Catrin Bernius) Should the experiments provide feedback next week on preferred scenarios? Yes, and it will be discussed at the LMC on the 6th November where we can voice any opinions that we have. There will also be a more technical discussion on the 5th November at the LBOC.

(Filip Moortgat, Chris Young) It has to be a message from the machine as the experiments are not experts in this field so the discussions are just circular. We can make a statement from the experiments side that we are happy/not happy to trade 1 week of commissioning to have one of the years in a good configuration for the forward experiments. This doesn't require magnet expertise. It is clear that no one from the magnet group will make a strong statement on the reliability of these numbers from previous conversations.

(Roderik Bruce, Matteo Solfaroli Camillocci) For nom-H in IR1 the triplet gets right to the limit is this thought to be a problem. The 0.7 over the limit is presumably within the error as all the numbers are very round but it obviously does pose more risk. This does slightly change the risk profile for Option C on the schedule.

(Roderik Bruce, Chris Young) At the same LMC meeting I will give a talk summarizing the background studies and there will likely be an LBS beforehand to wrap everything up. We should coordinate such that information is not repeated. The LMC talk on the optics will be by Matteo, mainly summarizing the discussions from the LBOC rather than given by LPC.

(Catrin Bernius) In the ID and Muon systems of ATLAS there are also increased backgrounds that could also be reported in that talk.

2025/2026 Optics configurations -- feedback from PPS (Mario Deile)

(Roderik Bruce) In your acceptance plots what n_sigma did you assume? The geometry shown is not to be taken seriously as they are very close.

(Matteo Solfaroli Camillocci, Chris Young, Roderik Bruce) If the 27 degrees changes to another angle can this be adapted? No the parts are being ordered and they are for 27 degrees, they are not adaptable/adjustable. They are custom made brackets that are cut with water jets. They are too expensive to order a variety of angles. If we have the non-optimal angle the acceptance might still be acceptable even if it is not optimal. It is also possible to slightly adjust the optics to alter the optimum. The current optics is still under discussion but it is the best of our knowledge what it will be. What is the loss in acceptance for different angles ie. the tolerance? This is not known, but for 10% it is probably ok, but depending on the crossing angle there might be regions where we are quite non-optimal. It is possible that certain periods of the fill due to the different crossing angles will be good/bad. 10% is likely able to be digested.

(Roderik Bruce) Will PPS likely operate in the pO run? This might have no crossing angle at all. How would this affect your acceptance? In principle this could be quite nice, but better if we weren't rotated. However, likely not too bad. It isn't decided if PPS would join this run, and timing is presumably not required. Therefore it looks like we would be able to operate but it wouldn't be completely optimal. Valentina confirms that timing would not be used as only one arm would be used.

CMS (Giulia Negro)

(No questions)

ATLAS (Catrin Bernius)

(Roderik Bruce, Andrea Ferrero) So 15urad tilt is your optimum angle. So for no tilt you are 0.5mm off as measured by the detector. But this is the raw detector measurement. How bad would it be to stay the same? It is difficult to say but to turn it around how easy is it to put in the bump? If there is additional commissioning then we will need to do a cost-benefit analysis, otherwise it can be put in. For the ALICE background no variations were seen across the scan to +/-50urad such that we don't think this will make a big difference to ALICE.

LHCb (Rosen Matev)

(Federico Alessio) The SMOG2 intervention can be done when ALICE flips and we can keep you informed when this will happen.

(Witold Kozanecki) Yes, it is standard to separate all the other IPs during the PbPb vdM scans.

ALICE (Andrea Ferrero)

(Witold Kozanecki) Inbetween scan pairs you can change the leveling. As it will be at night it is easier to do between the two experiments rather than within a scanning program.

(Filip Moortgat, Roderik Bruce) What were the on_disp settings? These were varied but as the data is still being analyzed a fuller report will be given next week. As the TCTs did not make much difference is this a different background from last year? The large background from last year doesn't appear to be the same as this one but it is all very preliminary.

(Witold Kozanecki) For the vdM issue Michi has changed it so that it will be 4b per injection rather than 8b. This means there will be emmitance growth at flat bottom so we need to be careful that they don't get so large that the scans can't cover all of them. This will be done by monitoring what the SPS provides.