I forward an email that was sent out sometime ago as (back then it was
already a reminder). I've been asked endless times to "see" this email
ever since. It was in the archive of course
The more pressing point is that many of you will need to use over-determined
elastic scattering for your analysis. Let me iterate this again:
You are expected to spend no time looking for that ideal proton cut. We
have done a lot of work in phase2, we have a working ntuple.C, cuts.C and
init.C (for which there are emails, and info on the code). There is already
a good proton cut and many people use it. We'll hear also at the meeting.
If you do things completely on your own you will spend a lot of time
hardcoding and recoding things in your macros. But at this point is your
own free time I guess !
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^---------- Forwarded message ---------- Date: Thu, 7 Nov 2002 13:37:04 -0500 (EST) From: Tancredi Botto <tancredi@mitlns.mit.edu> To: "[Blast_Anaware]" <blast_anaware@rocko.mit.edu> Subject: Analysis with the eep ntuple. How to get a clean proton ?
Here is an example root session for analyzing data with the eep ntuple, with my comments. This is the fruit of the labour of many of you. So I recollected this knowloedge hoping to making more accessible to people not so familiar with the analysis (yet, they are frustrated because they are expected to do analysis).
This sample session produces the attached result were, even for a "bad" run I get a clean proton peak. The four plots (from top left to bot right) are as follows. Everything can be typed at the root prompt. You can type root from anydirectory. ** are my comments
TFile * run = new TFile("/net/data/1/Analysis/commis/eep-2637.root") ** thats the name and location of the nutple
TCanvas *c5 = new TCanvas("c5","stuff", 30, 30 , 800, 600); c5->Divide(2,2);
c5->cd(1); eep->Draw("ttr-ttl","ntl==13&&ntr==3");
** plots the ( electron time(tof=R3) - proton time(tof=L13) ). This is the ** proton time of flight. With this definition the proton peak is on the ** left, corresponding to ca 300 later arrival from a "fast-fast" ** coincidence peak, where both L13 and R3 were hit by 2 particles close ** in time.
c5->cd(2); eep->Draw("ptr+ptl:ttr-ttl","ntl==13&&ntr==3");
** In the vertical axis I plot "coplanarity", i.e. the sum of TDC ** differences in the two tofs. This should be zero (in real time). You ** can see that the proton peak corrsponds to those vents narrowly ** correlated in coplanarity
c5->cd(3); eep->Draw("atl:ttr-ttl","ntl==13&&ntr==3");
** The vertical axis is the rear tof (L13) adc sum. You can see where ** the protons are...
c5->cd(4); eep->Draw("ttr-ttl","ntl==13&&ntr==3&&atl>7500");
** Back to the first plot. There is only a cut on the rear tof adc. The ** proton peak is clean
ABOUT THE EEP NTUPLE =====================
The eep ntuple can be generated for every run with
root -l ntuple.C ### where ### is the run number
It can be loaded in a root session as in the example above. The eep ntuple performs a strict data reduction requiring that 1 and only 1 particle hit each sector. By "hit", *in each sector*, we mean that only one tof scintillator had top/bot signals, there were 0 or 1 cerenkovs firing, there were 0 or 1 neutron bars firing. There is no Wch information.
The ntuple provides several fields
* for the TOFS
ntr = the number of the R-sector tof hit (0-15), atr = for that tof, the sum of top/bot adc's ttr = for that tof, the sum of top/bot tdc's ptr = for that tof, the difference of top - bot tdc's. This is positive for events above the machine plane
and similarly for the left sector (ntl, atl, ttl, ptl)
* for the Cerenkovs
ncr = is the number of the C hit on the Right sector. 0-3, is =-1 is no Cerenkov was hit
acr, tcr are the (right sector) cerenkov adc and tdc (there is only 1 adc, 1 tdc)
likewise, ncl, acl, tcl in the left sector
* for the neutron bars
nnl is the bar hit in the left sector tnl is the tdc sum (lef/tright) of that bar anl is the adc sum (lef/tright) of that bar pnl is the tdc difference (left - right) of that bar (if I am right). Positive means downstream
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:28 EST