[G12] Target Position

[Volker Crede]

Lei,

Bad TOF paddles were knocked out. The z-dependence cannot be properly taken care of by MK's efficiency map since it depends on the kinematics of the final-state particles. It can provide an average correction which we have been using but it comes with a large systematic uncertainty.

- Volker


On Jul 12, 2016, at 1:34 PM, Lei Guo <lguo at jlab.org> wrote:

> We should expect that the acceptance is z-dependent since it’s such a long target. MK’s efficiency map also is z-dependent. Has the map been applied, and the photon multiplicity been dealt with correctly? Was the bad TOF paddles knocked out?
> 
> 
> Lei Guo
> Assistant Professor
> Physics Department
> Florida International University
> Miami, FL
> 
> email: leguo at fiu.edu or lguo at jlab.org
> Office:305-348-0234
> 
>> On Jul 12, 2016, at 1:04 PM, Volker Crede <crede at fsu.edu> wrote:
>> 
>> Hi Carlos and Michael,
>> 
>> We perform a proper acceptance correction of the g p --> p ω reaction, but we do what almost everybody else does in the earlier meson reconstruction. We look at the invariant 3π mass distributions for each (energy & angle) ω bin, subtract the background either event-based or with side bins, and then we do the same thing with ω --> 3π  MC events to determine the acceptance for each kinematic bin. However, if the c.m. angular distribution of one of the decay-π has holes, there will be a mistake from integrating over unknown territory. Each pion is associated with three experimental parameters, so in principle (the neutral pion missing), it will require a six-dimensional acceptance correction. This is of course not very practical and sometimes, with actual holes in the acceptance, even impossible. You cannot correct for something that is unknown. This could be a problem for many analyses that study meson decays and try to determine branching fractions.
>> 
>> We looked at the double-pion channel for the lower energies < 2 GeV to study baryon resonances. We do see Δ's (by simply taking the lab four-vectors) but not as pronounced as in g11a. You see the effect in the mass distributions when you distinguish between the forward and backward half of the target.
>> 
>> - Volker
>> 
>> 
>> On Jul 12, 2016, at 12:24 PM, Carlos Salgado <salgado at jlab.org> wrote:
>> 
>>> Hello Volker
>>> 
>>> This an interest phenomena. Angular acceptance will be very different in g12/g11 as the target was moved.
>>> But , this different should be taken care of by acceptance corrections? Is that discrepancy remains in cross section, is this  showing that the acceptance correction is not right?
>>> 
>>> Also for PWA, which frame are you using?  how the Gottfried-Jackson or Adair distributions look like respect to z_vert?
>>> At higher energies (>3.6GeV) the Δ-peaks are very noticeable in the data I have seen?
>>> Will be great to see Zulkaida's slides next week. See you.
>>> -Carlos
>>> 
>>> 
>>> On Jul 12, 2016, at 10:44 AM, Volker Crede wrote:
>>> 
>>>> Good Morning Everyone,
>>>> 
>>>> I would like to raise another challenging aspect of the g12 experiment that we are currently dealing with at FSU and that may possibly affect other people. For a long time, we have considered the double-pion channel a by-product of Zulkaida's ω analysis. It is the same measured final state and given the good agreement of the g12-ω cross section with the published g11a-ω cross section, we did not expect any serious problems. Note that Zulkaida's g12-ω cross section is however about 7-8% larger on average. The ratio distribution of corresponding g12 / g11a (cross section) data points shows an almost perfect Gaussian; it is shifted by 7-8%, though.
>>>> 
>>>> We have accepted that we may not be able to understand this remaining discrepancy. While working with the Bonn-Gatchina group on the PWA of the double-pion channel, we noticed that we are loosing the (usually) dominant Δ-peaks in the invariant pπ mass distributions. This was puzzling since the Δ's show up as pronounced features in the g11a data. Upon investigating this, Zulkaida discovered that the target position plays a crucial role. The center-of-mass angular distributions of our final-state particles (in the ω and ππ channels) depend strongly on the z-vertex; we see huge holes when we plot c.m. angles versus the z-vertex. Cutting out these holes when analyzing final states with more than two particles is not an option (since it does not remove the effect of integrating in higher dimensions and will require a higher-dimensional acceptance correction). I think this was briefly discussed yesterday (but I had to leave the meeting early).
>>>> 
>>>> We looked back at the ω channel and compared the cross sections from the forward half of the target with the cross section of the backward half. The angular distributions are in good agreement (the ratio distribution looks Gaussian) but unfortunately, we observe a 10-15% normalization effect. It appears that this could be the missing piece in understanding the discrepancy in the g11a / g12 ω comparison.
>>>> 
>>>> Zulkaida will prepare some slides for next week's g12 meeting. I am afraid we will have to substantially increase the systematic uncertainty of the ω cross section (unless we apply an event-based acceptance correction using some kind of PWA). It also remains to be seen if and how the two-pion channel can be partial-wave analyzed. This is an issue that could affect all multi-particle final states, e.g. η, η', ω, …
>>>> 
>>>> Has anybody else noticed the problem?
>>>> 
>>>> Best wishes,
>>>> 
>>>> 	Volker
>>>> 
>>>> 
>>>> 
>>>> 
>>>> _______________________________________________
>>>> G12 mailing list
>>>> G12 at jlab.org
>>>> https://mailman.jlab.org/mailman/listinfo/g12
>>> 
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> G12 mailing list
>> G12 at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/g12
> 






-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/g12/attachments/20160712/81d849dd/attachment-0001.html>