Minutes and Summary

Main purpose of the meeting: Data taking progress (including plan for increased intensity); First look towards PbPb; 2026 questions

LPC minutes 06 October 2025

Present (P = in person): Chris Young, Chiara Zampolli (P), Eric Torrence (P), Roderik Bruce (P), Andrej Gorisek (P), Flavio Pisani (P), Giulia Negro (P), Archie Sharma (P), Filip Moortgat, Witold Kozanecki, Dragoslav Lazic, Andres Dellanoy, Tomasz Bold, Stephane Willocq, Lorenzo Rossini, Riccardo Longo, Mirko Pojer (P), Silvia Pisano (P), Paula Collins (P), Juan Esteban (P), Jorg Wenninger (P), Michi Hostletter (P), Cedric Hernalsteens (P), Robert Muenzer, Georges Trad, Peter Steinberg, David Stickland (P), Delphine Jacquet, Ivan Calì

Introduction (Chris Young)

Witold Kozanecki: is it planned to start with 50 ns bunch spacing?

Chris Young: yes, we will go straight to 50 ns. 

Witold Kozanecki: this might be an issue for the VdM calibrations. 

Andrej Gorisek: the VdM filling scheme is still 100 ns.

Roderik Bruce: the plan is to prepare both the 50 ns physics beam and the 100 ns beam for the VdM and we can then switch like last year.

Chris Young: yes, but the physics will go straight to 50 ns. For the VdM we’ll use 100 ns.

Roderik Bruce: some more comments. On one hand we will have a joint meeting between the collimation working group and MPP on 17.10) to discuss in detail collimation settings, intensity ramp up, loss maps, qualification… From a pre-discussion with MPP, it seems that we can possibly remove one step of the intensity ramp up. There will be a talk at the LBOC tomorrow, for those interested in the details.

Roderik Bruce: in the yellow book, the target for Run 3 and Run 4 was 13 nb-1 delivered. Halfing that, we’d have 6.5 nb-1 delivered in Run 3, and 6.5 in Run 4. 7 nb-1 recorded is much more challenging. I am curious to know why the target has changed and if the Run 3 + Run 4 target is still the same, or if it has increased too.

Giulia Negro: I can ask CMS if the total target should also increase. For sure we asked for this target in Run 3 to be sure to have all PbPb before moving to pPb. 

Roderik Bruce: so you made this impossible target to make sure that we do PbPb next year :-)

Filip Moortgat: to be fair, there are discussion within the ion group in CMS to see if this will change again. 

Roderik Bruce: it is not impossible that we can get more than 2.25 nb-1 this year if you have no downtime, and can reproduce the good injectors performance of last year, but to get to this value here is quite challenging. 

Chris Young: I think that as Giulia said the ATLAS and CMS targets might have been set such that they basically denote a preference for doing PbPb both years, no matter the performance. Note that we have one dedicated heavy ion experiment, two large collaborations, and smaller ones. If there is a disagreement, if there is no consensus leading to pPb, the decision might be taken by the Director General and the Director of Research. It is not necessary that you need all four to say we want to do the same thing. If there is a disagreement between the four experiments, then they can draw what conclusion they want.  

Tomasz Bold: about ion next year: is it the only option to have the entire year next year with PbPb or pPb, or is it possible to consider light ions?

Chris Young: it does not make sense to make both pPb and PbPb. It is not crazy but then you have to pay for the setup of the two configurations. So it is probably not a good idea to make 8 days of pPb and 8 days of PbPb, which makes the choice between the two kind of one or the other. The idea of doing light ions has never really been discussed, it would be quite tricky since the injectors want lead later in the year, and therefore setting the Pb in the full injector cycle as it has been agreed. I think it would not be feasible to have light ions and then have Pb in the injectors later in the year, due the amount of time that it takes to switch between light ions and PbPb. This would also need to be in total agreement from everyone, because the agreed plan was to do heavy ions and not light ions.

Eric Torrence: when do you need an answer to the pPb question? By November? I mean in terms of preference, lumi target…

Chris Young: it would be good before the lead ion run to know if the numbers on slide 9 are correct. 

Andrej Gorisek: the point was to see how this run goes.

Chris Young: yes, but you could still say what is your total, before seeing how it goes. We would need to know the Run 3 target, and till what number you would need to continue taking Pb data. It would be good to do that before the start of the ion run. It would be preferable by the end of November, when we have to present and get the approval of the DG of the 2026 timetable, if we had a first version of whether we’d do PbPb or pPb. If we can’t decide by then, it might be that they decide for you.

Robert Muenzer: in ALICE we have a discussion ongoing, also considering that we might run at lower luminosity this year.

Roderik Bruce: about the table on s9. In the past, there were discussions about what is a target and what is a preference. And it was said at some point that 7 was a preference. It would be good to clarify. 

Chris Young: for the machine where what matters is the target, the last column is the one that matters. If we do do Pb in all 2026, the Pb target becomes very easy if it is 6.5 delivered.

Paula Collins: for LHCb the 2 nb-1 is absolutely the target on which the IFT (Ion and Fixed Target) group is counting on. We’ll talk to them to understand the details of the questions. about it internally. The 0.75 nb-1 is only PbPb or also ref run?

Chiara Zampolli: it is only PbPb, in the whole Run 3, 2023, and 2024. 

Roderik Bruce: I think that also LHCb target has doubled, it used to be 2 nb-1 for both Run 3 + 4, now only for Run 3. 

Paula Collins: yes. We’ll also check the symmetry for pPb: the requirement should be for symmetric but we will confirm it.

Roderik Brice: if we have pPb in 2026, we might have less in Run 4. Maybe only 1. Then there is no chance to get to the pPb target in one run. 

 

ATLAS (Andrej Gorisek)

Roderik Bruce: for the question about the MD in December, it depends on the MD itself. If it is a quench test, there should be no problem. But if it is something different, e.g. the one with the new beam (25 ns ion colliding) from injectors, it might be better not to switch off.

Andrej Gorisek: if you could let us know as soon as possible, it would be good. We would like stable conditions for the inner detector calibrations starting as soon as possible when the beams are dumped. 

Chris Young: we should consider that these are floating MDs, which means that we typically do not dedicate the full 24 h to MD as in the table. 

Andrej Gorisek: sure, just let us know when you know the MD program.

 

CMS (Archie Sharma)

Jorg Wenninger: for the Pots in the high pileup MD, if it is Stable Beams, the pots can go in. 

Michi Hostletter: yes, it will be in SB. If the issue is solved that was reported today, we can insert them. And for the emittance scan, we’ll do it if there is time.

Andrej Gorisek: AFP will not be inserted because it will not be in the data taking.

Filip Moortagt: wasn’t Sofia in touch with ATLAS for the emittance scan?

Andrej Gorisek: yes, it was not about an emittance scan, but we asked 5 minutes steps in leveling at the end of the MD. Sofia expects that the mu will naturally reach 90, and to bridge this 90 to whatever we are running now, we would like to have a 5 minutes at 80, 70, 60, 50. This is a good test for our luminosity detector for the upgrade. 

Filip Moortgat: if you’re asking that, also CMS would support it. 

Giulia Negro: for the emittance scan, we did not ask for it, Sofia proposed it, and we would be in favour of that.

Filip Moortgat: we are not asking for anything, but we’d like to know what is going to happen since we need to prepare a trigger menu. E.g. for the emittance scan, we would need to know when it happens, at which pileup etc.

Andrej Gorisek: will you for sure remove the ZDC before the HL-LHC lumi tests?

Giulia Negro: not for sure, and now it seems that there will be no time, we’d need to take some from physics, so probably not. We will update the slides.

ALICE (Silvia Pisano)

Roderik Bruce: the 2024 level was also the maximum in the fill?

Silvia Pisano: yes, it is always at the beginning of the fill, averaged over the whole 2024.

 

LHCb (Paula Collins)

Jorg Wenninger: for flipping the magnet, we’d need 2 days notice, once we know the week. 

Paula Collins: what is needed in terms of bkg checks in PbPb? What we can do in ADJUST is limited, since we’d not have the VELO and would rely only on the main tracker and the calorimeters, which are not very good to check bkg.

Rodeirk Bruce: the point is that since we have smaller b*, we’ll have to tighten the collimators in IR8, and this gives a risk for background. The earlier we catch it, the better.

Flavio Pisani: the main issue is that in ADJUST we cannot either turn on or close the VELO.

Paula Collins: we’ll do the best we can. 

 

AOB

 

2FA (All)

General complaints for the 2FA implementation.

- Such a change in the middle of the run is very risky. LHCb has 2 examples when the experts could not connect in time due to this change. 

- Discussion on a bypass for the new server. 

- CMS and ATLAS also complained. 

- The test procedure that was advertised was not sufficient. It even stopped working from 29.09.

- Technical coordinations could be involved. They are involved in CMS. 

- IT was warned that this might have created issues for operation and it did happen.

- For lxplus it works, but for rDesktop is a hack, which does not even work.

- So there are two points: technical → it does not work; philosophical → this should not happen in the middle of data taking.

- The workaround that was proposed by IT is not known by everyone. 

- Idea to propose a piquet to help with issues. 

- Delay when issues happen due to the fact that people waste time to try solutions, also because the errors might be different.

- It seems a load problem, the server sometimes works and sometimes not.

- IT gave some solutions that worked only for a limited time.

- Data taking should not debug the solution.

- Workaround to connect to the old gateway will be shared.

 

For light ions, there is an offset about the inst lumi in NXCLAS and what is reported offline (Rodeirk Bruce)

Roderik Bruce: Is it something related to the calibration factors? Or can we just take the ratio of what we’d get from the integrated lumi from NXCALS and the offline one? And then we apply this to all the NXCLAS data?

David Stickland: it might be that we changed the calibration during the run. I need to check. Since we did not know what calibration we would have, we might have had a wrong one for some fills. For the second fill we tried to do better. 

Roderik Bruce: could we get the calibration factors for a given fill?

David Stickland: we can open the Massi files with the correct calibration. We have some numbers now, but there are still some few percent of systematic. If you have a need, we need to see how to help. 

Roderik Bruce: for now we can use the overall ratio.

Michi Hostletter: the massi files also foresee some versioning to allow having multiple versions, so if you have one now, you can add a new better one later. Maybe you changed the online. 

Chiara Zampolli: do some experiments show more discrepancy?

Roderik Bruce: seems ATLAS and CMS have values that differ more from what we expected from simulation with these beam parameters.