On Thu, 5 Feb 2004, Chi Zhang wrote:
>
> Hi guys,
>
> crunches are under way using calibration WITH k factor, took a look at the
> events already processed in run 4213, 4217, 4222, 4226. about 450k
> triggers available to analysis at this time, resolutions ARE improved:
>
> p: <=40MeV
> theta: <=0.8deg
> phi: <=1deg
> z: <=2.5cm
> total energy: <= 60MeV
>
> close to 700 series level, momentum recon on left and right still
> asymmetric, with left momentum 50MeV below the calculated line, but right
> side right on top of the calculation.
>
> I have checked the Wire.Cal file with the k factors and updated codes in
> BlastLib2 into CVS.
>
> I agree with Chris that ultimate k factor calibration needs to be done hit
> per hit. But that requires a lot of statistics.
Ah ! I feel better.
We still have to apply many consistency checks of the calibration data, like
I tried to say a few days ago and indeed chris has been also proposing
this a long time ago. We just have to agree on what we want to see when
imaging back the wire chamber..
_ cell sizes
_ 3 points on a line in a cell (for certain momenta) ?
can we also look for "correlations"
For instance: P_e - P_e(calc) vs phi_e, vs z_e ? and other similar
combinations. And we should of course check if our reconstruction accuracy
is more or less uniform across a cell length , by selecting different
Xdrift regions.
> Chi
>
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:30 EST