Minutes and Summary

Main purpose of the meeting: LHCC feedback, Intensity Ramp-Up, Crossing-angle anti-leveling

LPC minutes 17th March 2025

In person: Andrej, Eric, Robert, Silvia, Paula, Rosen, Michi, David, Giulia, Filip, Andrea, Jorg, Cedric, Juan, Chiara, Chris.
On zoom: Matteo, Joanna, Andres, Anna, Benoit, Gerardo, Ivan, Jaime, Lorenzo, Qipeng, Reyes, Riccardo, Roderik, Witold.

Introduction (Chris Young)

S4:

Chris Young: will we keep the ALICE x-ing angle the same as in 2024?

Jorg Wenninger: still being discussed, the latest version was to keep it as in 2024, maybe in one week it will be frozen. 

Robert Muenzer: for us, both are fine.

 

S7:

Jorg Wenninger: concerning going beyond the safe beam limit: this is very likely to happen.

Roderik Bruce: offline discussions with MPP already took place, there is no fundamental showstopper to do this. One thing to overcome is that if we do this, then we might not do the loss maps on the fly because we cannot mask interlocks. We could dump the beam then in the stable beam fill. We are trying to look for ways around this. 

Chris Young: concerning what Roderik Bruce means by “loss maps on the fly”: in the original plan we were going to have one of the bunches in the fill to be used for the loss maps, it was going to be blown up and not used for physics. 

 

S8:

Jorg Wenninger: concerning the possibility to run at low mu, pp ref…): the only interest during the heating test would be to collide at 6.8 TeV. Then it depends on the radiation level, on what is accepted by the [CMS] collaboration. Everybody in the machine would be happy to collide at 200, but maybe we should wait :-) 

Chris Young: Whether collisions are possible at all or a useful number of them is a different question. It depends on whether we are talking about hours of collisions, days… If you’re sitting at injection for all the period, then the amount of data is very very small.

 

S11:

Michi Hostettler: (emittance scans during the ramp up): emittance scan is not possible at P2 due to separation. At IP8 we can only if we can go head one. Maybe in the intensity ramp up but not at the very beginning of the fill, and if it is useful for you. It would mean that we first optimize for head on and then do the emittance scan. Then if you want we can separate again or we dump, if it is at the end of the fill it is easy to dump. If it is in the middle of the fill, we need a few minutes to restore the leveling.

Chris Young: this does not affect the other experiments, since it is done on demand by LHCb when they find it useful. So they can decide on the spot, with LHCb calling the CCC. We just need to warn the operator that this can happen.

 

S12:

Jorg Wenninger: it must be clear that initially it will look like a decay that just goes a bit slower. We’ll start improving a bit the slope and then we see, because if the lifetime decrease too much, you kind of burn it to the collimators, or to the detectors.

Michi Hostettler: this is the reason why we don't do this based on your lumi value, so we have more flexibility to tune it to make it more or less aggressive. 

Jorg Wenninger: it will start like the middle plot. We cannot use the pileup of the experiments because it can give some strange effects. 

Filip Moortgat : so at the end of the beta* leveling you don’t look at the pileup of the experiments and you do your own wiggling to keep it flat. 

Jorg Wenninger: the idea is indeed to keep it flat, but it is not clear if it is not too aggressive. 

Chris Young: also not clear what happens if the two experiments level at different pileup levels: will one get to the end before the other one? Then the lumi would drop off till both of them are stopped leveling and then you do the manipulation. So there are some subtleties to be worked out. But we hope for a 3.5% lumi gain for IP1 and IP5, and it makes the fill time ~1 hour longer, which is good for IP2 and IP8.

Rosen Matev: is there a chance that our leveling is exhausted?

Jorg Wenninger: no. 

Michi Hostettler: you see it also from the plots that the decrease starts very late and very slowly.

Chris Young: the extra hour does help you because the fraction in beam time vs in turnaround is improved. 

Jaime: from FASER it is fine to go to 100 urad. Are the steps in x-ing angle defined? 

Jorg Wenninger: between 1 and 3 urad. 1 could be a bit too small. We’ll try to make it as smooth as possible.

Michi Hostettler: at the beginning we’ll give a range (e.g. minimum will be 2 and maximum 5) and then the tool will do the trick with a pre-programmed curve of value vs intensity.

Jaime: should the value be the same value in IP1 and IP5 at any time?

Jorg Wenninger: it would be better.

Michi Hostettler: maybe there could be a 5 urad difference, but not the full range. 

Jaime: one thing that could be interesting: to have a plot like this with also the x-ing angle, how it is changing. In addition, to have an idea of what fraction of lumi is taken with what x-ing angle. 

Michi Hostettler: the plots shown come from the LPC tool for the luminosity simulator. The pre-set is for the left plot. You can then tune it, and try yourself. 

Chris Young: what can be useful for Jaime and can be extracted from the tool, and we should work with Michi Hostettler on this is how much time in the exponential decay and how much in the leveling. This is then the amount at 100 and at 160 urad.

Michi Hostettler: you can export a json and feed it to any tool like a Jupyter notebook for further inspection.

 

S13:

Jorg Wenninger: we should not expect too many details on the high intensity test, it would be much easier to say something when we are at 1.8. Also some of the tests might be done without filling all the machine, so they can go in regular MDs. We’ll have to find a balance about what one wants to do early and what can be done late.

Chris Young: it could be useful to have a rough idea of whether most of the tests can be done with long trains at injection or they’d be before the ramp?

Jorg Wenninger: at ramp we can do only 8b4e.

Chris Young: yes, but if the majority of the time is spent not in ramp, maybe it is interesting.

Jorg Wenninger: with 8e4b we cannot go to more than 1900 bunches but we want to go to 2400 or more, so this would be done at injection. 

Chris Young: you mean that most of the time would be not in ramp.

Jorg Wenninger: maybe most of the time not in ramp.

Chris Young: if you want to do W mass measurement, you might not get the statistics you need if you’re not all the time in collision. 

Jorg Wenninger: we can assume that at least 1/3 is run at ramp. 50% high intensity, a bit of fraction for tests, and 1/3 with low mu… The low pileup mu could be done in this time, if it makes sense.

 

Paula Collins: a question about the spreadsheet with the planning (https://docs.google.com/spreadsheets/d/1GKSzG9yVtknviEWBun4ZuqdUQdKcvR7pMqhB1jFyRRY/edit?gid=1991621601#gid=1991621601): on 25th April it says ion cycle optics. Is this the moment when the b* 1 m is commissioned for LHCb for the HI running? 

Jorg Wenninger: the spreadsheet is not really precise.   

Roderik Bruce: it is not sure that we will do the full cycle the first time. Whatever cycle we will decide, it will be commissioned, probably we will start with a ramp, and then add a squeeze at a later point. Maybe Jorg Wenninger wants to comment. 

Jorg Wenninger: I would go to the end from the beginning. Knowing that anyway with optics corrections we have to redo it. 

Chris Young: in the spreadsheet there are three blocks with ion optics, so maybe the squeeze will be in the second one. 

Jorg Wenninger: I have one early, and then I think that we do 2/3 of the commissioning during the ramp in May.

Roderik Bruce: ideally we should decide on the ion optics well before the commissioning starts, and there are some questions which we still need to clarify. 

 

CMS (Andrea Massironi)

 

S7: 

Chris Young: remember to follow up on the 50 ns, whether you can do this with having more INDIVs in a single fills. E.g. in one of the 100 bunch step of the ramp we have some space, and we could put in 6 more INDIVS instead of taking the request. 

Giulia Negro: we let you know, it should be fine. 

ALICE (Robert Muenzer)

    Nothing to be reported.

ATLAS (Andrej Gorisek)

S3:

Chiara Zampolli: what makes the request of delivered lumi to 0.8 if you then expect that this brings the recorded to 0.5? It means a 62.5% efficiency, seems low.

Andrej Gorisek: there is some contingency.

Filip Moortgat  (since the same is for CMS): it is also a special run. We can expect 80% efficiency [even if this would be 0.625 nb-1 and not 0.8 to get to the 0.5].

Eric Torrence: the lumi calibration is also included in this time.

LHCb (Paula Collins)

S4:

Michi Hostettler (about the request of early flip of the LHCb magnet). The fill needs to be short because you cannot take physics data. Anything beyond 30 minutes woul dbe lost for you. What can make it efficient is considering that MP typically asks is to have a fill which goes to the end of the cycle, otherwise it does not count for the intensity ramp up. With the extensive leveling and the fact that we cannot overshoot too much the pileup in P1/5, and we cannot go at any speed, so any fill of this ramp up should be ~5-6 hours, otherwise it is an extra fill.

Chris Young: we can make it as short as possible, and say that 5 hours is the minimum. Otherwise if we lose two fills and we are shorter than 15 hours and we have just a few hours left.

Michi Hostettler/Jorg Wenninger: we cannot anticipate.

Chris Young: yes, but we can consider this.

Rosen Matev: does this count for MPP?

Jorg Wenninger: usually we made it count. It is likely that a short fill of 5 hours will make it need 2 more to get to 15. 

Chris Young: we’ll try to make it as short as possible.

Jorg Wenninger: you say you want to change polarity after one week of full machine? The first flip will be quite demanding because we don’t know where we land.

Paula Collins: we want to first complete our calibrations which we cannot do before we get full machine. 

Jorg Wenninger: so we have full machine, then 1 week, then you flip?

Paula Collins: when we have full machine we’ll do our calibrations which can take between 1 fill and 2 and we need the experts around, so we conservatively said 1 week.

Rosen Matev: but the sooner the better.

Jorg Wenninger: which polarity do you want to start? 

Rosen Matev/Paula Collins: not known yet.

Jorg Wenninger: Please, let us know, so that when we switch we do the right one. 

Chris Young: when you say full machine, do you mean the full machine, or the 1200 bunches that we start counting for physics?

Paula Collins: the full-full machine, with 2400 bunches.

 

MoEDAL: 

Chris Young: MoEDAL is re-installing in your cavern. Do you and your technical coordinator (LHCb) follow their activities?

Paula Collins: yes.