zhangchi wrote:
>Hi all,
>
>I would like to bring to your attention to the yields from recent unpol
>runs.
>
>since 22nd, we v been running with unpol gas with Mass Flow Controller
>dialed at different values, and I analyzed the yield from these run.
>
>Again I present the yield by equivalent gas flow assuming 100% detection
>efficiency:
>
> gas flow dialed flow from edelastic yield runs
>-------------------------------------------------------------
>4.48 atoms/sec (0.05sccm) 1.67+-0.17 atoms/sec 1262-1277(June)
>9.96 atoms/sec (0.1sccm) 7.82+-0.21 atoms/sec 1642-1725(7/22-
>7.97 atoms/sec (0.08sccm) 3.69+-0.36 atoms/sec 1726-1730
>4.48 atoms/sec (0.05sccm) 2.83+-0.09 atoms/sec 1731-1799
>
>the errors on the yield are estimated from the number of counts in the
>fastest counting bin Q^2 = 0.125-0.185 GeV^2
>
>yield/(dialed gas flow) varries from 37%(0.05sccm runs: 1262-1277) to
>79%(0.1sccm runs: 1642-1725) and the two series of 0.05sccm runs one month
>apart produce rather different yields.
>
>Detection efficiency could well be varying during these days, due to the
>raise and lower of left box 6, however, there is definitely no evidence
>that the efficiency changed by a factor of 2.
>
>Again, this indicates the limit of the current flow control system. these
>1sccm controlers provides errors of 1% full scale, calibrated with N2 gas.
>for D2 gas which has smaller mass, the uncertainty is 0.025sccm. when
>working at 0.05sccm, the error could be as large as 50%.
>
>It may be possible to calibrate the linear relation between real gas
>output and dialed value, but seems the validity of such a calibration over
>a long period of time is still questionable.
>
>It demonstrates again the need for a reliable and accuate gas flow system.
>Otherwise, calibration and monitoring of detector efficiency and target
>density would always be entangled and hard to resolve independantly and
>definitively.
>
>
Chi,
As far as unpolarized target is concerned, you have a very reliable
(at least relative) measurements
of gas flow using LIGIT. You may compare the LIGIT readings with MFC
settings for the runs in
question in conclude if the flow was really different in these runs. I
am very confident that LIGIT
is proportional to the real gas flow.
Genya
>
>
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST