[d2n-analysis-talk] Discontinuity in momentum plots

[Diana Parno]

Hi all,

Yesterday, we discussed discontinuities in the momentum plots, and  
whether they were due to the low-momentum corrections that kicked in  
for p<0.9 GeV/c (negative optics) and for p<0.75 GeV/c (positive  
optics). I've re-replayed Run 1449 (five-pass production) with the low- 
momentum corrections commented out. The results are in the first  
attachment, lowp_onoff.png. The red shows the final momentum spectrum  
(the BB.tr.p variable) with the low-momentum corrections turned on.  
The black shows the momentum spectrum with the low-momentum  
corrections turned off (commented out). You can see that commenting  
that section out removes discontinuities in both negative- and  
positive-optics data.

The remaining attachments show the result of subtracting the red  
histogram from the black histogram for each set of optics. You can see  
the step-function discontinuity very clearly in each case.

By the way, the figure I showed yesterday of the intermediate momentum  
stages for the positive optics was misleading: it turns out there was  
a subtle bug in where the variable for the purple histogram (the stage  
right after the low-momentum correction) was picking up its value.  
This resulted in inflating the amplitude of each bin on the left side  
of the discontinuity line, and pulling in negative-optics data for the  
bins on the right side of the discontinuity line. This is why the  
visual scale of the discontinuity did not match the scale of the  
momentum changes in the code. I've fixed the bug but don't yet have a  
replayed run to produce a corrected version of the plot. However, I  
think the below results show pretty clearly that the discontinuity  
problem is due to the low-momentum corrections, and not to other,  
hidden problems elsewhere in the code.

Best,
Diana
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lowp_onoff.png
Type: image/png
Size: 14091 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20101015/b5001545/attachment-0003.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lowp_diff_neg.png
Type: image/png
Size: 13450 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20101015/b5001545/attachment-0004.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lowp_diff_pos.png
Type: image/png
Size: 12288 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20101015/b5001545/attachment-0005.png