[Erd108] [External] Re: Additional simulation request

[Sourav Tarafdar]

Hi Alexander,

Point well taken but I am not sure who is involved in DIRC simulation. If possible we probably should get together during collaboration meeting with available DIRC expert (if anyone is attending) to get their insight in this matter.

Cheers,
Sourav
________________________________
From: Kiselev, Alexander <ayk at bnl.gov>
Sent: Wednesday, July 17, 2024 12:22 PM
To: Sourav Tarafdar <stara at jlab.org>; Kondo Gnanvo <kagnanvo at jlab.org>; Shyam Kumar <shyam055119 at gmail.com>
Cc: erd108 at jlab.org <erd108 at jlab.org>
Subject: Re: [Erd108] [External] Re: Additional simulation request

  Hi All,

  Given how serious this problem potentially is for the hpDIRC, I'm still a bit puzzled why no one from the experts looked into a feasibility of PID event reconstruction with it without using tracking information for any other purposes but a track-to-ring matching.

  I'm not talking about writing a full reconstruction suite. Just a conceptual consideration with a simple outcome: the approach is feasible (then one can think about estimates and / or implementation) or outright excluded (then minimizing the material budget is of a paramount importance). From a technical perspective, it is a matter of systematically shifting reconstructed track parameters by few mrad, and checking whether a chi^2-like measure of how well a PID hypothesis matches the measured hits changes substantially or it does not. Big deal.

  Everybody understands that if you have N points on a 2D plane, which are supposed to be distributed along a circumference of a circle of unknown radius, one can solve the problem either with or without an a priori knowledge of where the center of the circle is. In one case you have a problem with 2N-1 degrees of freedom. In the other case - with 2N-3. Of course, this cannot work well if all the hits are sitting on one side of the circle. Is it really the case for hpDIRC images (which are not circles, but then what exactly is the difference?). Otherwise, an ultimately correct hpDIRC reconstruction code may actually want to add track parameters as Lagrange multipliers with a cov. matrix given by tracking to the problem, and re-tune them along with obtaining a PID hypothesis. My 2 cents.

  Cheers,
    Alexander.

________________________________
From: Erd108 <erd108-bounces at jlab.org> on behalf of Sourav Tarafdar <stara at jlab.org>
Sent: Wednesday, July 17, 2024 11:57 AM
To: Kondo Gnanvo <kagnanvo at jlab.org>; Shyam Kumar <shyam055119 at gmail.com>
Cc: erd108 at jlab.org <erd108 at jlab.org>
Subject: Re: [Erd108] [External] Re: Additional simulation request

Hi Kondo,

You are right for the momentum range for both DIRC and ToF. At momentum of >= 1 GeV is where the tracking angular resolution will help with DIRC PID , if I am not mistaken. 1-3 GeV is where the problem with multiple scattering is coming up.

best regards,
Sourav
________________________________
From: Kondo Gnanvo <kagnanvo at jlab.org>
Sent: Wednesday, July 17, 2024 11:53 AM
To: Sourav Tarafdar <stara at jlab.org>; Shyam Kumar <shyam055119 at gmail.com>
Cc: Matthew Posik <posik at temple.edu>; Gnanvo, Kondo (kg6cq) <kg6cq at virginia.edu>; erd108 at jlab.org <erd108 at jlab.org>
Subject: Re: [Erd108] [External] Re: Additional simulation request

One quick question,
Is the tracking needed for PID with DIRC at low momenta (<1GeV) or rather a higher momenta (6GeV)
I would expect the ToF to basically cover the low momenta region
Best regards
Kondo

Get Outlook for iOS<https://urldefense.com/v3/__https://aka.ms/o0ukef__;!!P4SdNyxKAPE!HwUui4Fq2fQ8QMOXD3Vjwyg41KFxfbKN_xS0SpnIfo5w3oubzDs-0NSY0TSr1caZMJW8YYMThQ$>
________________________________
From: Sourav Tarafdar <stara at jlab.org>
Sent: Wednesday, July 17, 2024 11:48:46 AM
To: Shyam Kumar <shyam055119 at gmail.com>
Cc: Matthew Posik <posik at temple.edu>; Kondo Gnanvo <kagnanvo at jlab.org>; Gnanvo, Kondo (kg6cq) <kg6cq at virginia.edu>; erd108 at jlab.org <erd108 at jlab.org>
Subject: Re: [Erd108] [External] Re: Additional simulation request

Thanks Shyam. The final numbers with 5% ToF material budget will be helpful in justifying not to increase the ToF material budget further. It will also affect the electron line shape at EMCAL (but anyway, EMCAL is not my or this mailing list main concern). At some point one need to see if one can do better than current 3.5 mrad based on Matt's study in reality to improve DIRC performance (may be not at the level of "high performance" )but reasonable pi/K separation (again that has to come from PID group).

Best regards,
Sourav

________________________________
From: Shyam Kumar <shyam055119 at gmail.com>
Sent: Wednesday, July 17, 2024 11:25 AM
To: Sourav Tarafdar <stara at jlab.org>
Cc: Matthew Posik <posik at temple.edu>; Kondo Gnanvo <kagnanvo at jlab.org>; Gnanvo, Kondo (kg6cq) <kg6cq at virginia.edu>; erd108 at jlab.org <erd108 at jlab.org>
Subject: Re: [Erd108] [External] Re: Additional simulation request

Dear Matt and Sourav,

The idea of Matt of the formation of fast tracklets can help there. As I understand, it may help but will need to do some studies.


Regards,
Shyam


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/erd108/attachments/20240717/72109d26/attachment.html>