[Clas_lmd] lepton pion PID

Fri Jul 10 13:51:18 EDT 2015

Dear all,

Although it is not clear to me what are we going to learn from the requested plot I created it from g11 data.
See it here:

http://www.lions.odu.edu/~mamaryan/all_mxp.pdf
The upper panel is invariant mass of e+e-\gamma squared, the middle one is for M_{x}(p)^2 with electron assignment,
the lower one is the same with pion mass assignment.

It would have been more interesting to compare the same plots from g12 to understand which dataset is
more powerful statistically in particular channels.

Best regards,
Moskov.

________________________________________
From: Clas_lmd [clas_lmd-bounces at jlab.org<mailto:clas_lmd-bounces at jlab.org>] on behalf of Michael C. Kunkel [mkunkel at jlab.org<mailto:mkunkel at jlab.org>]
Sent: Thursday, July 09, 2015 7:02 AM
To: clas_lmd at jlab.org<mailto:clas_lmd at jlab.org>
Subject: Re: [Clas_lmd] lepton pion PID

Greetings,

Can you overlay the MxP^2 with pion mass setting on this plot? Using the same kinematic cuts except with pion pid?

BR
MK
----------------------------------------
Michael C. Kunkel, PhD
Forschungszentrum Jülich
Nuclear Physics Institute and Juelich Center for Hadron Physics
Experimental Hadron Structure (IKP-1)
www.fz-juelich.de/ikp<http://www.fz-juelich.de/ikp><http://www.fz-juelich.de/ikp>

On 7/9/15 1:11 AM, Amaryan, Moskov wrote:
Dear all,

I attach a plot with Mx(p)^2 and M(e+e-\gamma)^2 distributions overlaid. Statistics of both histograms are the same,
however, as one can see there is no peak of eta in the invariant mass squared, data are from g11.
Invariant mass has no eta in it similar  to what Michael showed for g12.
One can think of two possibilities, either these are true eta’s decaying to e+e-\gamma, or these are  remnants of eta->pi+pi-gamma events.
The latter hypothesis was ruled out by plotting invariant mass of pi+pi-\gamma or missing mass of proton for pion mass assignment
for the same events that make Mx(p)^2 under lepton assignment.

Best regards,
Moskov.

On Jul 8, 2015, at 11:02 PM, Amaryan, Moskov <MAmaryan at odu.edu<mailto:MAmaryan at odu.edu><mailto:MAmaryan at odu.edu>> wrote:

Hi Michael,

Thanks for your message, looks interesting.

However we have shown in the CAA and also Georgie has shown it in our meetings that invariant mass of pi+pi-\gamma
resolution is much worse than missing mass of  proton.  The reason is energy resolution of the final state photon and
vertex resolution as momentum of the photon is defined through both of these, energy and vertex.
Therefore we use missing mass of  proton with other cuts, which use pion mass assignment to \pi^+ and \pi^-.

Now what I have shown on Tuesday, was missing mass of proton with electron (positron) mass assignments.
There we clearly see a peak of eta. The mass assignment is arbitrary, of course, but requirement of missing mass squared
of pe+e- to be around zero (<0.01 GeV) and missing energy of pe+e- to be E_x>0.2 GeV makes this selection definite, not arbitrary.
And the fact that we see peak of eta is a proof that some of pi+pi- tracks are actually e+e- pairs

The plot you sent around doesn’t show a peak of eta very clearly in the best topology with EC/CC, but one can imagine that with more statistics
the  eta peak will be visible. However without EC/CC there is no signature of eta whatsoever. Would it be visible in the missing mass of proton in g12?

It wouldn’t make sense for me to argue that PID on leptons improves invariant mass e+e-\gamma, of course it does.

The question was which fraction of events that make eta in the missing mass of proton with e+e-\gamma assignment is due to events that were
from pi+pi-\gamma final state.
I think we have shown that this fraction is negligible.

It will be interesting to compare missing mass of the proton squared from g11 and g12 (for pe+e-gamma final state)  to see if because of different
acceptances these two data sets have similar or very different numbers of eta’s.

It appeared that ntuples we are using from g11 have been eloss applied. In this case eloss for pions and leptons will be the same, I am not sure how different they should be, but the fact that we see peaks of pi0 and eta in the missing mass of proton is indicative that in the first approximation we can neglect eloss uncertainties.

Best regards,
Moskov.

On Jul 8, 2015, at 1:33 PM, Michael C. Kunkel <mkunkel at jlab.org<mailto:mkunkel at jlab.org><mailto:mkunkel at jlab.org>> wrote:

Greetings,

So I was curious to reproduce my study on pion contamination without the use of CC or EC quantities.
I am attaching a plot that shows G12 data that shows 3 lines.
Red: G12 data with eloss on pions, pion mass setting, invariant mass PipPimGam; conditions Ngamma > 0 && abs(mm2_PPipPimGam)<0.01 && mE_PPipPim > 0.2

Blue: G12 data with eloss on leptons,lepton mass setting , invariant mass EpEmGam; conditions Ngamma > 0 && abs(mm2_PEpEmGam)<0.01 && mE_PEpEm > 0.2

Cyan: G12 data with eloss on leptons,lepton mass setting , invariant mass EpEmGam; conditions Ngamma > 0 && abs(mm2_PEpEmGam)<0.01 && mE_PEpEm > 0.2 && lepton PID through CC and EC. Histogram is scale by x10

As I concluded years ago, and is seen in the histogram, without CC or EC, the eta peak cannot be reproduced in the proper place.

Total data used:1/463 of g12 data set.

Hope this helps.

BR
MK
----------------------------------------
Michael C. Kunkel, PhD
Forschungszentrum Jülich
Nuclear Physics Institute and Juelich Center for Hadron Physics
Experimental Hadron Structure (IKP-1)
www.fz-juelich.de/ikp<http://www.fz-juelich.de/ikp><http://www.fz-juelich.de/ikp>

--
BEGIN-ANTISPAM-VOTING-LINKS
------------------------------------------------------

NOTE: This message was trained as non-spam.  If this is wrong,
please correct the training as soon as possible.

Teach CanIt if this mail (ID 01ONRzDqD) is spam:
Spam:        about:blank
Not spam:    about:blank
Forget vote: about:blank
------------------------------------------------------
END-ANTISPAM-VOTING-LINKS

<Lepton_Pion_mass_comparision.pdf>_______________________________________________
Clas_lmd mailing list
Clas_lmd at jlab.org<mailto:Clas_lmd at jlab.org><mailto:Clas_lmd at jlab.org>
https://mailman.jlab.org/mailman/listinfo/clas_lmd

Prof. Moskov Amaryan
Department of Physics
Old Dominion University
4600 Elkhorn Avenue
Norfolk, VA 23529

phone:757-683-4614

________________________________

Spam<about:blank>
Not spam<about:blank>
Forget previous vote<about:blank>
_______________________________________________
Clas_lmd mailing list
Clas_lmd at jlab.org<mailto:Clas_lmd at jlab.org>
https://mailman.jlab.org/mailman/listinfo/clas_lmd

Prof. Moskov Amaryan
Department of Physics
Old Dominion University
4600 Elkhorn Avenue
Norfolk, VA 23529

phone:757-683-4614

_______________________________________________
Clas_lmd mailing list
Clas_lmd at jlab.org<mailto:Clas_lmd at jlab.org>
https://mailman.jlab.org/mailman/listinfo/clas_lmd

________________________________

Spam<https://www.spamtrap.odu.edu/canit/b.php?i=03OOb59Fn&m=0aef9bbda7d3&t=20150709&c=s>
Not spam<https://www.spamtrap.odu.edu/canit/b.php?i=03OOb59Fn&m=0aef9bbda7d3&t=20150709&c=n>
Forget previous vote<https://www.spamtrap.odu.edu/canit/b.php?i=03OOb59Fn&m=0aef9bbda7d3&t=20150709&c=f>

Prof. Moskov Amaryan
Department of Physics
Old Dominion University
4600 Elkhorn Avenue
Norfolk, VA 23529

phone:757-683-4614

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/clas_lmd/attachments/20150710/04958e4a/attachment.html>