[Frost] g9b target polarization

[Michael Dugger]

Hi,

I think that, for some runs, the signs of the target polarizations from 
the table might be incorrect.

We have been having difficulty getting good T observables for pi+. For the 
heck of it, I changed the sign on the polarization values for runs > 
62490. I figured that the sign should change each time the target was 
repolarized. Now, the target asymmetry plots look very good.

The latest plots can be found at:
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t675.gif
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t775.gif
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t875.gif
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t975.gif
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t1075.gif
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t1175.gif
http://www.jlab.org/Hall-B/secure/g9/ASU/tPrelim/t1275.gif

Note:
* Only 66 million events!

* Hard cuts. No fitting routine for background subtraction.

* We no longer use an arbitrary dilution factor, the new plots have 
dilution factors calculated for each bin.

* Allowed target azimuthal offset to vary. I'll talk about those results 
on Thursday.

* Ross and I think that the problem with the cosThetaCM = -0.5 bins at 
eGamma = 675 and 775 are due to efficiency issues and I'll talk about 
that on Thursday.

Take care,
Michael