sPHENIX HCal discussion

[Edouard Kistenev <kistenev AT bnl.gov>]

Well, this whole discussion seem based on  misconception. One can’t measure something what is not defined or not measurable. e/pi can be defined only for infinite calorimeter with constant density (can be W-Sc fibers or Fe-Sc sampling calorimeter, it is important that structure is the same everywhere). If this condition is not satisfied then comparing e and pi is more or less the same as comparing apples and oranges.  If calorimeter is MIP calibrated then measured e/pi is at best equal to that of one structured like EMC (or whatever the first segment is) with minimal corrections due to the differences in e/h (real purpose for all such measurements) in device longitudinal sections. Showing pictures for e/pi in the whole hybrid calorimeter serves no purposes unless it is compared and found consistent with simulation. This will be good G4 test (minimal corrections above) but I don’t think it can be used to serve physics.

I also disagree with the idea of multiple different calibrations for different parts of calorimeter in different applications (jets/hadrons/em). There is only one way to calibrate our calorimeters in situ - cosmic muons (produces one set of calibration constants). The electron calibration can’t be different from muons for anything but e/mu. Any deviation from c{e)/c(mu)=e/mu is an indication of a problem in data or analysis. We may correct the values assigned to different objects in calorimeters depending on physics which we assume dominating any particular subset of data (be that direct photons or jets). Such corrections needs e/h, measured e/mu  which may be extracted from electron data in standalone hadron calorimeter will help but only to the point - final scaling must be either data driven or from simulation (if proved right). An example of such data driven estimate are  scaling coefficients used to balance section contributions in test beam data analysis to insure linear response. Most likely this balancing reflects exactly this -  e/mu variations in individual sections - will be interesting to check.

Edward

On Sep 22, 2016, at 3:49 PM, woody <woody AT bnl.gov> wrote:

Hi Abhisek,
   I was not referring to a purely EMCAL e/pi but rather the e/pi of the combined three calorimeter system (EMCAL+Inner HCAL+Outer HCAL). Of course, each detector has it's own e/pi ratio, which may be interesting parameters for us to look at, but we would never actually use any of these detectors alone to measure hadronic energy. We really only care about the combined detector when we measure hadronic energy and we would somehow like to characterize this. What this really boils down to is how to add the energies from the three longitudinal compartments. As Jin pointed out, there is no absolutely right or wrong way, or even any unique way, to do this. It will depend on what we want to do with the number or curve we come up with. In terms of a stand alone detector (and I'm still referring to the combined detector), I think coming up with a set of weighting factors for each compartment that minimizes the pion resolution is fine, and I guess this is what you've done. However, I would show this as a curve of e/pi response versus energy, applying the optimal weighting factors at each energy. This would then be the sort of "intrinsic" e/pi ratio of the combined detector, and this is what I would put in the paper. However, as Jin also pointed out, we may want to use a different set of weighting factors for doing jet analysis later with the full detector. I don't think we want to try and show anything about this in the paper, but we can say in words that this is something we would consider doing in the actual experiment.
  I hope this is clearer now what I was referring to, but if not, maybe we can sit down with Jin and talk about it.

Cheers,
Craig

On 9/22/2016 2:15 PM, Abhisek Sen wrote:

Dear Craig and Jin,

  I think what you are saying is purely EMCAL e/pi. I dont know if it is right to call it EMCAL/HCAL e/pi because they are different for two sections.

   I looked into e/pi for both section separately. They are quiet difference. Please see the slides:

http://www.phenix.bnl.gov/WWW/publish/abhisek/sphenix/eoverpi_sPHENIX.pdf

HCAL e/pi is done by standalone data where electrons and pions can be separated in testbeam data. If you want HCAL to assist EMCAL by measuring electron energy leaking from EMCAL to HCAL, then the e/pi we need is shown in slide 1. 

EMCAL e/pi is done by balancing EMCAL with respect to HCAL. It is actually ~1.55. I see no big energy dependence between 8-28 GeV where hadrons dominate. At lower energies, I do see a hint of rise but low energy hadron data quality is poor. 

Jin: I do use c2_sum < 20 (inner+outer) for all energies as hadron cut. It was a typo in my slides yesterday.

Cheers,

Abhisek

On Thu, Sep 22, 2016 at 10:29 AM Huang, Jin <jhuang AT bnl.gov> wrote:

Hi, Craig and Abhisek

 

Here is my understanding from the meeting:

 

The e/h ratio of the whole calorimeter system will be depending on the choice of energy scale between EMcal / HCal. At the choice of optimizing pion energy resolution (as in abhisek’s talk), e/h shower response would be higher than 1, e.g. 1.4 ~ 10 GeV. One can also choose the scale on EMcal / HCal so that it optimizes the energy resolution for a typical jet which has a fluctuating fraction of EM (pi0->gamma) and hadrons. Then we could bring the e/h ratio of the whole calorimeter system to approximately 1 by scale down EMCal energy scale relative to HCal. It would not produce as good resolution for pions, but should give a better resolution for jets and better jet energy scale.

 

Therefore, to produce an e/h ratio plot, we need to choose an energy scale between EMcal / HCal first and state the choice clearly in the paper. I would prefer this choice is to optimize for pion energy resolution, as used in Abhisek’s analysis. Then we can produce the e/h plot simply by taking ratio of the electron and hadron linearity curve ratio. The e/h event sample separation at or below ~12 GeV/c should be very clear in the joint data set just using c2 inner + outer sum. It is also important that our simulation tunes reproduce this ratio, so we could trust its prediction for the jet performance in sPHENIX.

 

Cheers

 

Jin

 

 

______________________________

 

Jin HUANG

 

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: sphenix-hcal-l-bounces AT lists.bnl.gov [mailto:sphenix-hcal-l-bounces AT lists.bnl.gov] On Behalf Of Craig Woody
Sent: Thursday, September 22, 2016 7:45 AM
To: Abhisek Sen <sen.abhisek AT gmail.com>
Cc: sphenix-hcal-l AT lists.bnl.gov
Subject: Re: [Sphenix-hcal-l] HCAL Meeting 9/21 NOTE NEW TIME

 

Hi Abhisek,
  I don't remember how we left things at our meeting yesterday about including a plot of the e/h ration for the combined  3 part calorimeter, but I think it would be good to include a plot like this in the final paper. We talked about just quoting the number of 1/0.7 ~ 1.4, but that's not what people usually quote about calorimeter performance. The e/h ratio surely has an energy dependence and I think we should show that. It may be as simple as making your balancing plot for the hadronic energy in the EMCAL vs the HCAL at various energies and plotting the ratio of that to electrons (or, rather one over that). Hopefully that wouldn't be that hard to do, and I think it would provide a valuable piece of information about our combined calorimeter system.

Cheers,
Craig

On 9/18/2016 3:41 PM, John Lajoie wrote:

Dear HCAL'ers:

    We've been on hiatus for a while, but it's time again for our HCAL bi-weekly meeting. We will meet this coming Wednesday, 9/21, at 2:30PM BNL time, immediately following the sPHENIX electronics meeting:

https://indico.bnl.gov/conferenceDisplay.py?confId=2395

PLEASE NOTE THE NEW DATE, TIME, AND BLUE JEANS RESERVATION.

If you would like to make a presentation at the meeting, please let me know.

Regards,
John

 

--

John Lajoie

Professor of Physics

Iowa State University

 

(515) 294-6952

lajoie AT iastate.edu

Contact me: john.lajoie

_______________________________________________

Sphenix-hcal-l mailing list

Sphenix-hcal-l AT lists.bnl.gov

https://lists.bnl.gov/mailman/listinfo/sphenix-hcal-l

 

_______________________________________________
Sphenix-hcal-l mailing list
Sphenix-hcal-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-hcal-l