[d2n-analysis-talk] Partial solution to in-plane acceptance asymmetry

[Diana Parno]

Hi all,

I played around with some cuts this afternoon to try to understand  
Vahe's in-plane scattering angle cuts better. I thought you'd be  
interested in the results. All plots below show only events from the  
same vertex region Vahe was plotting, within 5 mm of z=0. I have not  
placed any cuts on the out-of-plane scattering angle. Also, this is  
from Run 2150 (the same run Vahe was using in his analysis), a single  
run from late in the 4.7-GeV dataset.

In today's meeting, we looked at a plot from my thesis where the range  
of horizontal track positions, projected out to the Cerenkov plane,  
looked symmetric around zero. This projected position is based on  
BB.tr.ph (after bending through the BigBite magnet), not on  
BB.tr.tg_ph, so I thought it would be interesting to look at the  
relationship between those two variables. If BigBite were an ideal  
magnet, phi would be orthogonal to the dispersive direction, so the  
angle would be exactly the same between those two tracks.

The first attachment (phi_vs_tg_phi_basic.png) shows this relationship  
for just two basic cuts: a charge of -1 and a T2 trigger. There's  
certainly a strong correlation between the angles, but the bowtie  
scatter shape is disturbing.

In the second attachment (phi_vs_tg_phi_opticsvalidity.png) I've added  
two cuts to ensure we can trust the optics. The first, labeled "optics  
validity", is BB.optics.vzflag[]==1 && BB.tr.tg_th[]<0.2. This removes  
parts of the acceptance that were flagged as unreliable during our  
optics calibration. (The vzflag variable is set based on where the  
track intersects the magnet bend plane.) The second, labeled  
"horizontal track projection", requires (BB.optics.bendx +  
0.23*BB.tr.ph)>-0.097 && (BB.optics.bendx + 0.23*BB.tr.ph)<0.13. This  
is the rescattering cut that we identified fairly late in the game,  
which removes a large percentage of the anomalous stiff tracks  
(presumably due to rescattering somewhere inside the magnet). These  
two optics cuts nearly eliminate the bowtie scattering and thus  
tighten the correlation between the phi angles of the front and back  
tracks.

The third attachment (tgphi_acceptance_cutsequence.png) shows the  
distribution of BB.tr.tg_ph (that is, the in-plane scattering angle)  
after successive cuts have been applied. The legend shows which cuts  
have been *added* for each histogram: each histogram also incorporates  
the cuts of the ones that have gone before. You can see that adding in  
the horizontal track projection cut -- the one that removes events we  
think may have been rescattered -- dramatically reduces the acceptance  
on the small-angle (positive phi) side. No other cuts have an effect  
on the acceptance range. (The dark purple histogram at the very bottom  
shows events that passed *all* our cuts and would be used in  
asymmetries.)

Applying the rescattering cut doesn't eliminate the asymmetry in the  
in-plane-angular acceptance range, but it does reduce it by a factor  
of two. This is a very reasonable consequence of a cut that is  
designed to eliminate a certain region of the magnet from the  
horizontal acceptance. (It's interesting that its effect is so  
asymmetric, though.)

For completeness, the fourth attachment  
(phi_acceptance_cutsequence.png) shows the same thing for BB.tr.ph,  
the in-plane angle for the back track. This shows the same effect,  
although the boundaries of the final range are different. There is one  
difference that I think is interesting: that "dip" in the middle of  
the acceptance is pretty pronounced in BB.tr.tg_ph even when the only  
cuts are charge and T2 (black histogram), but the dip is almost  
totally absent in the black histogram for BB.tr.ph. I would have  
thought it would go the other way, since BB.tr.ph should be more  
directly related to the position of the particle in the actual  
Cerenkov detector.

Best,
Diana
-------------- next part --------------
A non-text attachment was scrubbed...
Name: phi_vs_tg_phi_basic.png
Type: image/png
Size: 18884 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20111201/85a6b5e5/attachment-0004.png 
-------------- next part --------------

-------------- next part --------------
A non-text attachment was scrubbed...
Name: phi_vs_tg_phi_opticsvalidity.png
Type: image/png
Size: 16125 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20111201/85a6b5e5/attachment-0005.png 
-------------- next part --------------

-------------- next part --------------
A non-text attachment was scrubbed...
Name: tgphi_acceptance_cutsequence.png
Type: image/png
Size: 19526 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20111201/85a6b5e5/attachment-0006.png 
-------------- next part --------------


-------------- next part --------------
A non-text attachment was scrubbed...
Name: phi_acceptance_cutsequence.png
Type: image/png
Size: 18873 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20111201/85a6b5e5/attachment-0007.png 
-------------- next part --------------