 |
|
|
| |||
Muon answeresFrom: CommentsDear LHCb colleagues, Here are the outline of answers presented to the LHCC
referees during the meetings on Wednesday next week. Please e-mail any comments
to Brad Cox
Original from Brad; Question 1: The requirement of 5 out of 5 stations in the muon trigger implies very high efficiency and reliability of the system. Assuming a risk to operate in condition 4 out of 5, could you quantify a deterioration of the performance of the muon detector. Answer: First, muon station 1 has four planes of CPC and we plan to operate asking for three of four planes to fire for a hit. Thus we are not asking excessive efficiency from any one plane in the station. For muon stations 2-5, we have either four planes of CPC in the small angle region operated in the three of four majority logic condition or two planes of MRPC operated in OR in the wide angle region. Once again, we are not making excessive demands on efficiency of any one plane of CPC or MRPC We have run minimum bias events operating the muon system in the 4 of 5 (which to all intents and purposes means defining a triple coincidence as 2 of 3 of Muon 3 firing and either M4 or M5 firing). We find that the resuling trigger rate is to all intents and purposes the same as when operating with a three of three condition. Question: Could you quantify a misidentification probability you expect in the Muon System? The number of hadrons firing only 0.2% of the total trigger rate (page 108) says that you have very clean muons already at the trigger level with quite modest pt requirement.
Answer: The statement on Page 108 is misleading. The 0.2% refers to the ratio of triggers caused by hadrons to the total minimum bias interation rate. In actuality, the trigger rate due to hadrons is approximately 10% of the total level 0 muon trigger (1.6% -> 1.8% increase with large errors). There is already some information about muon misidentification in the technical proposal. The main problem is not with the muon detector identifying a hadron or a random combination of shower hits as a muon. Rather the problem is linking the muon as seen in the Muon Detector with the proper track upstream in the tracking system. We have estimated from our Monte Carlo studies (as reported in Section 11.5.2) that muons from B->mu decay with pt above 1.0 GeV/c will be associated with the wrong upstream track 2% of the time. For B->mu muons below 1 GeV/c this percentage rises to 9%. If we consider any muon (not necessarily from B->mu), the overall misassociation is 18%. If we restrict ourself to muons above 1 GeV/c the misassociation fraction is 9%. Question: What is the pt resolution of the Muon System? Answer: The pt resolution of the muon system is given in Fig. 12.11 of the Technical Proposal. The distribution of true pt minus trigger pt of the detected muon from B semimuonic decays is fit by two Gaussians with sigmas of 234 and 891 MeV/c. The Gaussian of larger sigma is mainly needed to fit the tails of the distribution. Question: If I read Fig. 11.2 for M1, the occupancy is expected to be below 2-3% at any radius. Is it correct?
Answer: As indicated by Fig. 11.2, the actual number of tracks traversing a given 1 cm**2 area is less than a few % everywhere for the stated inner aperture of 25 mrad x 25 mrad. However, the pulse widths of 100 ns for the CPC's increase this occupancy. In spite of this we expect worst case occupancies (which will be in M1) to less than 10%. ========================= And some comments from Andrei ========================= Q1 Efficiency requirement can be slightly less for M4,M5 (requiring either of them to be fired). But the requirement for M2 and M3 are very high. Therefore, we may consider having 4 layers of RPCs and 6 layers of CPCs in M3 which is the most crucial. This can be also very useful as the majority logic (3 out of 4) can be applied everywhere in this station thus reducing spurious background and lessening load of the trigger logic. Note also that possible inefficiency for CPC can be due to timing properties (not catching the signal within the 25 ns window). Since the M3 signal will tag the bunch, we can consider relaxing timing window for other CPC chambers to 50 ns. In order words, it is rather a question of extra occupancy rather than efficiency. Q2 The results on muon (mis)identification obtained by Brad and Gloria were not really presented and discussed before putting to the TP. In TP it is said that a track was associated to a muon track if it is the nearest at M1. This is certainly not enough requirement and slope difference should be also checked. This check can result in less misidentification. In any case, it seems to be not correct to have different misidentification probability for muons from B->muX decays and other muons with the same Pt. In Brad's answer the misassociation probability for muons with Pt>1GeV/c is 9% for any muon and 2% for B->muX decay products. What is this difference due to? Is a muon from B decay so nicely isolated? I have not seen that to be demonstrated clearly. I would leave then just numbers for B->muX decays for example and say that the work is going on. We don't have track finding yet, so this study can not be completely justified in any case. Q3 In a bit different presentation of the Pt distribution (deltaPt/Pt_true) The resolution is <30% in the pt region 1-2gev/c where it is most interesting.
| ||||||||||