1 / 10

How to further improve muon system performance for the high-luminosity phase

How to further improve muon system performance for the high-luminosity phase. Alessandro Cardini and many others. Some ongoing work.

kareem
Télécharger la présentation

How to further improve muon system performance for the high-luminosity phase

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. How to further improve muon system performance for the high-luminosity phase Alessandro Cardini and many others

  2. Some ongoing work • Giuseppe has implemented a ToyMC which uses the TDR rates in every region/station, and estimates inefficiencies and ghosts by reconstructing the readout detector scheme and adding CARIOCA and DIALOG dead-time • Violetta (+ Rolf, Andrea and Ale) uses the .sim data and by knowing the readout scheme in every region/station estimates “true MC” occupancies, ghosts, etc… the low energy background, parameterized by Alessia in Boole, has been added in Violetta’s analysis by Andrea. There is still no cross-talk description in this work, and the spillover effect is not included yet • Alessia and Giacomo are performing MC/Data comparison • Matteo is estimating muon performance at high-luminosity from data summary by A. Cardini

  3. The current critical points • The muon detector will not operate optimally at high luminosity: • The OR between the different gas gaps increases the total rate due to the uncorrelated low energy background (LEB) • The hardware OR between 2 adjacent gaps increases the CARIOCA dead-time • The logic strip creation increases the DIALOG dead-time • The crossing operation, needed to recover the logical pad information, creates ghost hits • All this reduces the muon system efficiency and increases the combinatorial background, with a worsening of the muon identification performance • Additional thing to keep in mind: how things could be at 3E33? summary by A. Cardini

  4. Some possible solutions • New pad detectors in most irradiated regions, M23R12 • Reduce the size of logical strips by reducing the number of channels in OR: this could be achieved by removing some IB • Reduce the number of active gaps by switching OFF the HV on the 2 external gaps (in the stack of 4), or by performing and AND between the gaps (could create geometric inefficiencies; station efficiency loss should not be a problem: in the fully SW LLT we could even do 3/4) • Perform a fine tuning of the chamber working point: if we are operating at a too high gain (or at a too low threshold )we could have a high dead-time on the CARIOCA • All this solutions are described in a draft document attached to this meeting INDICO page summary by A. Cardini

  5. Pad detectors: some details (1) • M2R1 • Rates as high as 600 kHz/cm2 • Currently mixed readout  384 logical pads • Difficult to have for 2018 a new highly integrated FEE  we could envisage to readout 192 pads (24 FEBs, like in M1R1, will fit) with current FEBs, not many cables to be added because it’s only 12 detectors • This should bring approximately a factor 2 reduction in occupancies, with a possible another factor 2 that could be obtained by switching OFF 2/4 gaps • Feasible in triple-GEM technology (but needs GEM gas and HV) • M3R1 • Rates as high as 200 kHz/cm2 • Currently mixed readout  384 logical pads • As in M2R1 we could envisage to readout 192 pads (24 FEBs, like in M1R1, will fit), not many cables to be added because it’s only 12 detectors • Will 96 pads be enough? Do we want to have a different R/O in M3R1 with respect to M2R1? To be studied • Feasible in triple-GEM technology (needs GEM gas and HV) summary by A. Cardini

  6. Pad detectors: some details (2) • M2R2 • Rates as high as 100 kHz/cm2 • Currently mixed readout  192 logical pads • 192 pad detector possible but many cables to be added  we should check if 96 pads are enough that would imply NO cables to be added  check muon performance • MWPC technology • M3R2 • Not critical from the rate point of view • Currently mixed readout  192 logical pads • 96 pads OK for dead-time, check muon performance • Easier to debug if we make it similar to M2R2, but not absolutely necessary • MWPC technology summary by A. Cardini

  7. Removing additional IBs • We currently have 104 ODE in our system. According to PID Upgrade TDR, the removal of IBs in M5R4 will bring to 108 the number of needed nODEs • M2R3 • it’s already a pad detector  no important CARIOCA dead-time • It seems however that there is a large contribution for dead-time coming from DIALOG due to the large number of channels OR-ed together to make the logical strips (to be carefully verified) • If we remove all M3R2 IBs  + 16 nODEs • M3R3 • Not so critical, if we want to maintain it similar to M2R2 we need +16 nODEs • If we could afford it, we could also think of removing IBs in M2R4 (+16 nODEs) and in M3R4 (+16 nODEs) • We should careful consider the following points: • Where to put additional nODEs? I would not recommend to use already cabled crates! • Power consideration • Components: we have already ordered optical transmitter/receivers • Optical fibres: they should be available, but we should check summary by A. Cardini

  8. Some ongoing MC studies • Alessia and Giacomo are looking into the muon MC and are comparing MC and data • The current situation is that M1 is approx. OK, while the rest is not • In particular, for M2-M5, too few MC hits in R1 (data/MC ~ 1.6) but also too many MC hits in R4 (data/MC ~ 0.3) • Alessia has checked the 2007 LEB parameterization with new G4 by lowering various thresholds but situation does not change substantially • Found that beam plug in M2 is not well described, too much material in MC  could be the reason of too few MC hits, Alessia is checking this • We need also to check the M3 beam plug description in MC and correctly describe the new shielding behind M5 summary by A. Cardini

  9. Studies to be performed (mostly with beam) • Take data in AND/OR at various collision energies • Estimate the effect of switching OFF 2/4 gaps in a chamber • Check the CARIOCA/CARIOCAGEM dead-time vs. input charge (partially started already) • Check if an HV reduction could improve overall muon efficiency in high luminosity conditions • Understand the LEB and how the muon detector is efficient in detecting it (could be done in lab) • … summary by A. Cardini

  10. Conclusions • A lot of work ongoing, many ideas, starting to focus on a feasible solution • Muon MC tuning urgently needed • It’s important that we define soon what is the detector configuration that we need to simulate in official MC • It’s a huge effort: new detectors/system description, new coding/decoding routines, new reconstruction, new trigger, new muonID algorithms, … • Time is running fast: I believe we should present the performance of the upgraded muon system to the LHCb week in December 2014 if we want to propose (and install by 2020) something in addition of what’s was described in “PID Upgrade TDR” summary by A. Cardini

More Related