From BUPHYC::KEARNS Sun Oct 24 07:10:06 1993 Return-path: Date: Sun Oct 24 07:10:06 1993 To: VAXGS::KEARNS From: BUPHYC::KEARNS Subject: monopole/wfd stopping Status: RO From: BUPHYC::KEARNS 28-MAR-1993 21:45:46.09 To: CITHEX::GL,CITHEX::MICHAEL,CITHEX::NOLTY,CITHEX::PECK,CITHEX::RZL,CITHEX::WALTER CC: KEARNS Subj: stop meeting 3/26/93 notes This is the "minutes" of the informal stop/readout meeting that we had at Caltech on Friday, March 26th. These notes are mostly a transcript of the blackboard we used while discussing the problem- I have not (yet) checked any numbers; in fact, I think the plan of action is for us all to firm up some of these estimates. 1) We concluded that we had allow for the signature of a slow monopole catalyzing proton decay. This leads to a prompt muon signal preceding the slow monopole. This is shown schematically below for a monopole entering from above MACRO- the muons from proton decay in the tunnel roof can penetrate to the center layer. Keep in mind that the muons and monopole trace need not be in the same tank: T ---- --------- ----------------------------------------------- | |___| C ---- -------------------------- ------------------------------ | |___| B ----------------------------------------------- ------- ------ | |___| The exact details of this picture also depend on the cross section for proton decay catalysis: this effects the number of muons and their time spread and whether or not they can overlap the monopole trace itself. This signature puts us in a difficult situation. Basically, for all triggers we must delay the stop by 1000 us. The rest of the discussion pertained to the impact on deadtime and data size. 2) Some benchmarks: 500 = number of channels per MACRO 64 K = total memory per channel 160 us = minimum time stored in 64K buffer of wfd 60 ms = typical time stored in 64K buffer, assuming 100kHz bg. 3 Mb/s = transfer rate for straight reads .3 Mb/s = crude guess at transfer rate including extra code 11 Mb/s = Block transfer rate The transfer rate for my MVME167 is 50% faster than for the VME-uVaX + VIC; this difference is ignored because of the uncertainty in .3 Mb/s. 3) This is a table of triggers sorted by the amount of interesting information. The second column is the amount of information we wish to record; for reasons of code efficiency, I hypothesize that it will actually be quantized by 160 us (it is faster to count rollovers than to calculate exact time windows). The third column is the rate of each trigger. The next column is the deadtime fraction. The last column is the amount of the detector to read-out (I somehow neglected to record this, so I am going by memory to some extent). SLOW MONOPOLE (1 face) 1000us high 1.6e-3 a few channels SLOW MONOPOLE (2 face) 1000us 10/hr 2 SM S.T. MONOPOLE 1000us need 10/hr 2 SM FAST MONOPOLE 10us 1/hr trigger channels (?) LIP with S.T. 2us 600/hr 1e-2 trigger channels ERP MUON 2us 60/hr 1e-2 2 SM CSPAM 200ns 600/hr 2e-2 trigger channels (?) LIP w/o S.T. 200ns 1000/hr 1.5e-2 trigger channels