Dear Djamel, Your results are very constitent with what a got for analysis run 5940 (which is in DB), namely around 113-> 85 ps average for slewing applied. It is though also clear that individual tubes exhibit much larger spread with a range of 70->95/100 psec. The results for H2 are similar. This can been seen easily by looking at the 'diagnostic histogram' timediffsCorr that applies the slewing correction on the 'data being used to generate it plotting vs slat # . Try e.g. to make a FitSlicesY() and plot the results. Notice also that the timediff histogram likewise checks the offsets as in the DB. I found that for at least the set of runs I looked at the existing database offsets are reasonable but not good being off up to +-50 psec and sometime much more. One would also think that the slewing parameters should really not change from run to run - assuming the initial time-offsets are good. This is particular cruical observation for MRS where the slewing only gets stable fits if one adds many runs (havings ~80K MRS tracks at 60 deg less at 90, and I doubt it can be done at the more forward angles. As I am sure you have noticed the offset code and the slewcode have the same cuts, and selection to determine constants. At FS it is a bit easier to get the statistics to achieve good slewing constants except for the first slats in both H1 and H2 where trackdensities are low. My suggestion would be at first to get slewing from a good high statistics run (one early and one late), but first getting time-offsets (high stat) and then the slewing and test that the slweings are run independent. Bu this method it does not matter either what the offsets are (eg in the pp run while the slewing should stay). When you run the slewing check the values for athe A(up) and A(down) if they are close to 4 it is a sign that the statistics is low or the cut selection bad (it does reqiuire |t-tcalc|<fTimeCut say 300 psec, so offsets have to be good. Flemming ------------------------------------------------------ Flemming Videbaek Physics Department Brookhaven National Laboratory tlf: 631-344-4106 fax 631-344-1334 e-mail: videbaek@bnl.gov ----- Original Message ----- From: "Djamel Ouerdane" <ouerdane@nbi.dk> To: "Flemming Videbaek" <videbaek@bnl.gov> Cc: <brahms-dev-l@bnl.gov> Sent: Monday, June 17, 2002 2:06 PM Subject: slewing correction (new procedure) > Hi Flemming, > > I've made a slewing correction test on run 5592 (pol A, 8 deg. I think). > The results are improved, I get a time res. of about 87 ps, > (without, it gives ~ 105 ps). > > Plost can be checked out here: > > http://www.nbi.dk/~ouerdane/h1PidSlewCorr.gif > http://www.nbi.dk/~ouerdane/h1TofRes.gif (with all momenta, should be less > with small momentum bin) > > Now, since the time offsets are needed for a slewing correction, I would > think we should make a slewing corr. for each time offset cal. revision > but I'm not sure. If so, one has to use the calibtool and check time > offset revisions from run 5361 or so (unless we want to use earlier runs). > > Djam > > > > -- > Djamel Ouerdane ------------------------------------------o > | Niels Bohr Institute | Home: | > | Blegdamsvej 17, DK-2100 Ø | Jagtvej 141 2D, | > | Fax: +45 35 32 50 16 | DK-2200 Copenhagen N | > | Tel: +45 35 32 52 69 | +45 35 86 19 74 | > | http://www.nbi.dk/~ouerdane | > | ouerdane@nbi.dk | > o---------------------------------------------------------o >
This archive was generated by hypermail 2b30 : Mon Jun 17 2002 - 14:56:29 EDT
