[G8b_run] Tagger Sag and Proton Momentum Corrections

[Michael Dugger]

Hi,

I think that we are now in agreement. Please see the plot:
http://www.jlab.org/~dugger/tmp/EcorrComp.gif

This plot shows the energy correction (up to 2.2 GeV) determined by ASU 
with the new FSU energy correction (red line), full ASU fit (green line) 
and first order polynomial fit to ASU (black line). The first degree 
polynomial fit parameters are stated on the plot and are consistent with 
the new energy correction suggested by FSU.

It would be good to see if the full fit (green line) would produce 
a reasonable kinematic fit over the full energy range.
(Full energy range plot: 
http://www.jlab.org/~dugger/tmp/EcorrCompFull.gif )
The full fit function is given in equation 15 (page 6) of the pdf:
http://www.jlab.org/Hall-B/secure/g8b/ASU/egCorrASU.pdf
The a0, a1, and a2 parameters are given on page 5.
The \nu and b parameters are given near the bottom of page 6.

It would be best to use the energy dependent function. We can show this 
over the full range of the tagger and obtain a better systematic error 
than if we used the straight-line solution.

I think that we are almost done with the energy correction :)

Chuck and Volker: Thanks for all your hard work on this issue.

Take care,
Michael

On Mon, 19 Oct 2009, Charles Hanretty wrote:

> Dear colleagues,
>
> We apologize for the delay with regard to the requests from the last g8b 
> meeting. I've invested more efforts to estimate the required photon energy 
> correction due to the tagger sag using the kinematic fitter. We have
> also double-checked again our pull distributions for the double-pion reaction 
> gp->ppi+pi- (all particles detected) including error adjustments, and 
> (hopefully) finalized the proton momentum corrections.
>
> The first page of the pdf file in the attachment shows again the tagger sag; 
> this time, the photon energy is plotted versus the (fitted) mean of the 
> photon energy pull. The typical structure that we have been discussing for 
> weeks
> is clearly visible. The lower left plot shows the photon pull integrated over
> all energies (essentially the projection of the upper plot onto the y-axis).
> Ignoring a possible energy dependence of the tagger sag, we tried to center 
> the
> fairly symmetric pull distribution in order to get a feeling for the required
> energy correction. The original (black) distribution (no tagger photon 
> correction) requires approximately a 0.2% energy correction to become the 
> (much better centered) blue distribution. The mean and sigma are given on the 
> slide following the same color convention.
>
> Now we went back to the proton momentum corrections that were previously 
> determined without a photon energy correction. This is shown on the second 
> page.
> The mean of the (global) proton momentum pull has become worse and shifted 
> from
> almost zero to about 0.18. To correct for this shift, an additional 0.4% 
> proton momentum correction would be required. This is much less than the 
> original proton momentum correction, which is of the order of a few percent.
>
> So far, we have not applied any pion momentum correction. The status of the 
> pion pulls is shown on the third page. While the error adjustments (pull 
> widths) are all very acceptable (the comparison with Mike Williams values for 
> the g11 data is shown on the fourth page), the pion means are somewhat off 
> center. The displacement is of the same order as for the proton and only a 
> very small correction would be required to center them. Unfortunately, this 
> is a very tricky business. Further proton corrections shift the pions farther 
> away from zero; we observe the same effect with the proton when we try to 
> correct the pion momentum.
>
> For this reason, we think that the current status of all the corrections
> seems to indicate a very good compromise: a very small 0.2% photon energy 
> correction (a few MeV without any energy dependence) and proton momentum 
> corrections as determined before. Given the small energy correction, we feel 
> that ignoring the energy dependence is justified for energies below 2.2 GeV, 
> which covers the important photon polarization of the g8b data.
>
> Thoughts?
>
> -Chuck and Volker
>