[BLAST_ANAWARE] ed elastic asymmetry? etc.

From: zhangchi (zhangchi@general.lns.mit.edu)
Date: Thu Jul 17 2003 - 23:29:01 EDT


Hi folks,

I have three things to say here.

1st. atatched, please see two figures of ed elastic tensor asymmetry from
our data.

ed_asym_1121_1323.eps is made with old cycle3 runs
ed_asym_1353_1476.eps is made with new cycle2 runs

the numbers in file name indicates the runs used

I am not sure how conclusive they are to say we taht have polarization and
asymmetry. it is all open to your interpretation. Just keep in mind that
the analysis assumes 25% target polarization.

2nd. As of cross section and ABS target flow, which I am more interested
in at the moment, here is the deal:

unpol system dialed at 0.05 sccm in fact provides a flow of 0.026sccm. a
factor of 2 lower.

I have two pieces of evidence to support this claim:

ligit read our for 0.1 H2 is 5.1e-7, for 0.05 D2 is 1.7e-7, empty target
at one moment was 0.5e-7. if we trust the readout at 0.1, then the flow
for 0.05 D2 on the dial scales to 0.026sccm.
Assume 0.05sccm flow in Monte Carlo, I need an extra factor of 2.8(36%) to
bring cross section data to match with MC. I think it is reasonable to
assume the detector and reconstruction give 70% (Tancredi would say even
lower) then gas flow deduced this way is 0.026 sccm.

For ABS.

earlier cycle3 runs:
flow is 0.02sccm (9e15 atoms/sec). ligit read out is on average 1.34e-7
compare with 1.7e-7 for unpol, it scals ABS flow to 0.021sccm. extra scale
factor against MC is 7.5 (13%), which puts ABS flow to 0.019sccm.

new cycle2 runs: it is a confusing.
trigger rate per coulumb is not changed, 11,500,000 triggers/5038.42C vs.
12,500,000 triggers/5974.39C. yet ligit reading is 35% different.
1.34e-7 vs. 0.98e-7.

If I scale the elastic rates with MC, need a factor of 4.1. Assuming
detection efficiency 70%, we have a flow of 0.035 sccm(1.6e16 atoms/sec)

but if I scale with ligit readout, the gas flow would be 0.012sccm. seems
the vacum background is different. This seems especially possible when
noticing that the unpol data were taken interweaved with the old cycle3
data.

3rd. It would be nice if we could: First take a few more runs unpol data
and book a ligit number with empty target.

right now, we have unpol D2 data with cover off:
1132 1133 1135-1139 1142 1143, 1228 1230, 1262 1265 1266-1273 1275-1277
however, 1132 1133 1135-1139 1142 1143 1228 1230 does not have cell
tempreture records in lr, or say epics->cellt is 0 all the time.

I looked at the few runs with tempreture, cell temp average to 106.5. I
looked at again all those runs assuming cell temp is 100 all the time.
the extra scaling to match MC did not change.

It would also be nice if we could add cell tempreture in to lr ntuple.
Then we could scale each event with tempreture to cancel out target
density dependance on T.

Finally, I see large tail above 1e-7 on ligit read out histogram in runs
1451 and 1452. do lr->Draw("ligit", "ligit>0") on a few runs, you ll see
what I mean.

1475 does not have tempreture. Epics down?

Chi







This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST