[G8b_run] Please check your code

[Michael Dugger]

Hi,

Ken found a problem that everyone should be aware of.

On the 6th of August, 2009, I sent an email (forwarded below) that gave a 
link to the tagger-correction function. This function at
http://www.jlab.org/Hall-B/secure/g8b/ASU/egCorr/cmuErgCor.f
has not changed, is correct, and should be the one used in your analysis.

On November second of 2009, I uploaded the CLAS-note detailing how the 
tagger correction was determined. Included in this document was the 
tagger correction equation. The parameters in the CLAS-note were 
incorrect. I have uploaded a new version of the CLAS-note that has the 
proper parameter values.

Please make sure that, either, you use the parameter values given in the 
FORTRAN function at
http://www.jlab.org/Hall-B/secure/g8b/ASU/egCorr/cmuErgCor.f
OR from the most recent CLAS-note at
http://www1.jlab.org/ul/Physics/Hall-B/clas/public/2009-030.pdf .

In short:

Parameter values of tagger correction from old CLAS-note -> BAD

Parameter values of tagger correction from new CLAS-note -> GOOD

Parameter values of tagger correction from FORTRAN function -> GOOD

I apologize for any inconvienence this may cause.

Thanks for your time.

Sincerely,
Michael




---------- Forwarded message ----------
Date: Thu, 6 Aug 2009 15:26:38 -0400 (EDT)
From: Michael Dugger <dugger at jlab.org>
To: g8b <g8b_run at jlab.org>
Subject: [G8b_run] Tagger sag correction


Dear g8b collaborators,

Stuart showed an energy dependence in the photon energy correction (for
g8b data) that was not a simple multiplicative factor. I decided to verify
Stuart's results. I found that there is an energy dependence in the photon
energy correction that is not a simple multiplicative factor. It appears
that the tagger sag correction applied to g8b data was not sufficient.

The method I had been using previously (massX^2 vs. incident photon
energy) can not be used to determine an energy correction that is
different from a common multiplicative factor. Therefore, I am now using
an iterative routine that capitalizes on the overdetermined reaction
gamma p -> p pi+ pi- (as is standard in this type of study). This
iterative routine has been successfully tested against the known energy
correction obtained by the CMU group for the g1c data set.

The g1c CMU energy correction plot (figure 9 on page 12):
http://www1.jlab.org/ul/Physics/Hall-B/clas/public/2004-017.pdf

compares well to the ASU iterative routine
http://www.jlab.org/Hall-B/secure/g8b/ASU/egCorr/g1cEgCorr.gif

The CMU energy correction has one free parameter. This parameter is an
offset to the fractional photon energy (E_gamma/E_e). For g8b data, in
addition to the standard CMU energy correction, I have included a term
that is a simple multiplicative factor:
energy_correction -> CMU_correction + E_gamma*multiplicative_factor

The result of this 2 parameter fit can be found at
http://www.jlab.org/Hall-B/secure/g8b/ASU/egCorr/g8bEgCorr.gif

The three bump structure of the tagger sag is clearly seen.

The fit results:
multiplicative factor = -0.00351(6)
offset = 0.00134(3)

The function is very simple and a FORTRAN version of the g8b photon
energy correction can be found at
http://www.jlab.org/Hall-B/secure/g8b/ASU/egCorr/cmuErgCor.f

I have decided not to produce a stack of slides fully describing the steps
taken to obtain the results shown in this email. Instead, I will write a
CLAS note that can be easily referenced from analysis notes. I hope to
have the note completed in the next week or two.

Sincerely,
Michael

_______________________________________________
G8b_run mailing list
G8b_run at jlab.org
https://mailman.jlab.org/mailman/listinfo/g8b_run