sPHENIX EMCal discussion

["Osborn, Joe" <osbornjd AT ornl.gov>]

Hi all,

 

Thanks for the discussion – it sounds like the easiest starting point is just the face of the calorimeters. I can add an optional function to let users define what radii they want to project to, but for now I think the easiest thing is just to default to the front face of the calorimeters (then if advanced users want to do something else, they can choose whatever they like).

 

Thanks all,

 

Joe Osborn

 

---------------------------

 

Joe Osborn, Ph.D.

Postdoctoral Research Associate

Oak Ridge National Laboratory

osbornjd AT ornl.gov

(859)-433-8738

 

 

From: Kistenev, Edouard <kistenev AT bnl.gov>
Date: Monday, February 22, 2021 at 2:54 PM
To: Lajoie, John G [PHYSA] <lajoie AT iastate.edu>, Xiaochun He <xhe AT gsu.edu>, Osborn, Joe <osbornjd AT ornl.gov>, sphenix-emcal-l AT lists.bnl.gov <sphenix-emcal-l AT lists.bnl.gov>, sphenix-hcal-l AT lists.bnl.gov <sphenix-hcal-l AT lists.bnl.gov>
Subject: [EXTERNAL] Re: Track-Calorimeter Projection Radii

John, thanks for the comment.

 

I may add only this - the only points which are coupled to physics (indirectly) are collision, decay and conversion points (sorry if I forgot something). Everything else are vectors. If you start your analysis with vectors - they are not different in tracking detectors, PID or calorimeters, the latter are just attributes - any scientific library will do the further technical work for you. Propagating those vectors through tracking detectors is very involved, calorimeters are a bit easier. In case of calorimeters the zero approximations to actual physics values are vectors pointing to mechanical gravity centers. Depending on the total assembled energy of the physical object - gravity center for a different objects can be corrected to reflect expected object behavior in the matter - its shape, fluctuations etc. It is then the process which is probably better made using properly trained AI - outside my scope.

 

Anyway - good luck trying to everyone in analysis.

 

Edward

 

---

From: Lajoie, John G [PHYSA] <lajoie AT iastate.edu>
Sent: Monday, February 22, 2021 2:18 PM
To: Kistenev, Edouard <kistenev AT bnl.gov>; Xiaochun He <xhe AT gsu.edu>; Osborn, Joe <osbornjd AT ornl.gov>; sphenix-emcal-l AT lists.bnl.gov <sphenix-emcal-l AT lists.bnl.gov>; sphenix-hcal-l AT lists.bnl.gov <sphenix-hcal-l AT lists.bnl.gov>
Subject: RE: Track-Calorimeter Projection Radii

 

Hi Joe,

 

As a starting point we would certainly want to have projections to the approximate radii of the entrance face of the EMCal, iHCAL and oHCAL.  I’m not pretending that there is enormous physics value to these particular projections on their own, but they are a logical starting place when trying to develop algorithms and “figure things out”.

 

I like Edward’s suggestion very much; I can certainly see where it would have great value when you have projective geometry, but my feeble mind has a hard time seeing how it works for the tilted tile HCAL.  In any case, I think that is a much deeper layer of discussion.

 

John

John Lajoie

he, him, his

Professor of Physics

Iowa State University

 

(515) 294-6952

lajoie AT iastate.edu

 

From: sPHENIX-HCal-l <sphenix-hcal-l-bounces AT lists.bnl.gov> On Behalf Of Kistenev, Edouard
Sent: Monday, February 22, 2021 12:22 PM
To: Xiaochun He <xhe AT gsu.edu>; Osborn, Joe <osbornjd AT ornl.gov>; sphenix-emcal-l AT lists.bnl.gov; sphenix-hcal-l AT lists.bnl.gov
Subject: Re: [Sphenix-hcal-l] Track-Calorimeter Projection Radii

 

To whom it may concern.

 

Since the day when hadronic calorimeters were first proposed to look like they look today I was trying to make it clear that the only sensible way to assign the energy in particular tower (same in EmCAL - this is the "feature" of nearly projective geometry) is its center of gravity (mass center). This basically means that towers in analysis are 3-vectors pointing to gravity centers with deposited (measured) energies being the vector values. For simplicity (or for the physics analysis) I would suggest to deal with 4-vectors of zero mass. Then the whole analysis can be  based upon vector algebra identical to all (EM and Had) components of calorimeter system from the moment when measured energies are known to your program. 

 

This approach to treating the calorimeter sensitive elements (EM and Had towers) on a common basis allows to collect all information related to calorimeter shower candidates in a very straightforward way. It also has one more beautiful feature which is still ignored by existing analysis chain (at least I am not aware of attempts to use this idea) - it allows (physics dependent) to computationally redefine the shower tracked in the calorimeter  into more then 3-longitudinal sections (remember that in the Outer HCal shower longitudinally always seen by more then two towers (towers overlap). For the shower tracked in EMCal and IHCal the OHCal tracking may result in two correlated crossing points at a different depths in two neighbor towers. It may require a bit of iterating but improved resulting depth dependent energy leakage corrections deserve the effort. 

 

This is my two cents to this discussion.

 

Edward

 

---

From: sPHENIX-EMCal-l <sphenix-emcal-l-bounces AT lists.bnl.gov> on behalf of Xiaochun He <xhe AT gsu.edu>
Sent: Monday, February 22, 2021 12:28 PM
To: Osborn, Joe <osbornjd AT ornl.gov>; sphenix-emcal-l AT lists.bnl.gov <sphenix-emcal-l AT lists.bnl.gov>; sphenix-hcal-l AT lists.bnl.gov <sphenix-hcal-l AT lists.bnl.gov>
Subject: Re: [Sphenix-emcal-l] Track-Calorimeter Projection Radii

 

Dear Joe,

 

Having the track projection is indeed very helpful. Thanks for the effort.

 

Given the intrinsic shower fluctuations in each of the calorimeter subsystems, I would think that the only meaningful layer locations are the entry points to each of the calorimeter system. It should be up to the calorimeter person who will take this info for the subsequent analysis needs. This is even more true when there is no PID info for the tracks.

 

With this said, I would think that it may be useful if the tracking package can provide a function call for the calorimeter analyzers for a projection at a given depth.

 

My two cents,

Xiaochun

--

 

From: sPHENIX-HCal-l <sphenix-hcal-l-bounces AT lists.bnl.gov> on behalf of Osborn, Joe <osbornjd AT ornl.gov>
Date: Monday, February 22, 2021 at 8:20 AM
To: sphenix-emcal-l AT lists.bnl.gov <sphenix-emcal-l AT lists.bnl.gov>, sphenix-hcal-l AT lists.bnl.gov <sphenix-hcal-l AT lists.bnl.gov>
Subject: [Sphenix-hcal-l] Track-Calorimeter Projection Radii

Dear EMCal and HCal experts,

 

Following some discussion in the software meeting last week regarding track projection to the calorimeter layers, I am prioritizing getting a new track projection module functional for broader analysis use.

 

One point of discussion was to include the track projection global position with the track, so that analyzers could potentially match this track position to calorimeter clusters or towers in their own specific way depending on what kind of analysis they are performing. The track projection software takes a set of track parameters and projects them to an arbitrary cylindrical surface at a fixed radius from the beam axis.

 

So, my question is, what is the best radius to use for each of the calorimeter layers? I would assume this should correspond to the average maximum shower depth, but perhaps the experts think something else would be best.

 

Thank you,

 

Joe Osborn

 

---------------------------

 

Joe Osborn, Ph.D.

Postdoctoral Research Associate

Oak Ridge National Laboratory

osbornjd AT ornl.gov

(859)-433-8738