sPHENIX discussion of physics

[Anthony Frawley <afrawley AT fsu.edu>]

Hi All,

Following up on the question raised in the meeting about the need to remake the DST seeds after final drift velocity and TPC time offset calibrations are available.

If the initial seeding is done with an incorrect z scale, the seed projection to the beam line (Z0) is wrong. The preliminary distortion correction uses this Z0 value to move the clusters so they point to z = 0 at the beam line, after which the clusters are distortion corrected and moved back by the same amount. The distortion corrected clusters are refitted by the ALICEKalmanFilter, and the seed parameters are replaced on the node tree.

Bottom line: The distortion corrected seeds are fully refitted, including a new estimate of Z0 - so the track matching is done with a fully refitted (preliminary corrected) seed. However the preliminary distortion correction is made to the clusters at the original, incorrect z. That means that when track matching is done using updated drift velocity and time offset calibrations, it is only affected by the fact that the (approximate) distortion correction was made at the original cluster z position, not the properly calibrated one.

So plots made with the existing seed DST's are not as badly affected as I thought. But the new seed DST's will be better, and will include some other improvements too.

Best regards

Tony

---

From: sphenix-hf-jets-l-request AT lists.bnl.gov <sphenix-hf-jets-l-request AT lists.bnl.gov> on behalf of Gregory Ottino <gottino AT lbl.gov>
Sent: Wednesday, March 5, 2025 3:32 PM
To: sphenix-physics-l AT lists.bnl.gov <sphenix-physics-l AT lists.bnl.gov>; sphenix-hf-jets-l AT lists.bnl.gov <sphenix-hf-jets-l AT lists.bnl.gov>; sphenix-upsilons-l AT lists.bnl.gov <sphenix-upsilons-l AT lists.bnl.gov>
Subject: Re: [[Sphenix-hf-jets-l] ] Next Joint Heavy Flavor/Quarkonium Topical Group Meeting 3/5 1:30pm ET

 

Dear collaborators,

Please find below the minutes for today's meeting. Additionally, the convenors have discussed creating a central, single mattermost channel in anticipation of QM 25 that all active analyzers can use for up to date feedback. As there are currently many channels for many analyses (and separate mattermost channels) consolidating in the short term will be helpful to solve common issues. The purpose of this channel will be to post any updates, in between our weekly meeting, for discussion. The channel is linked here.

Best,

Greg Ottino, on behalf of the convenors

03/05/25

Photon conversion - Xudong

• AF - DST seeds need to be rerun when reproduced to fix t0 matching issues

• XY - redo matching at macro level and load ACTS geom 

• AF - seeds record z proj of track as a number that is not corrected

• JO - Does prelim track distortion module update track params?

• AF - unclear but don’t think it does

JH - From 2 tpc halves, ExB changes sign from N->S, wouldnt you see two peaks,one from each tpc half (slide 4). 

• AF - effect is same on both sides of tpc, not expecting two peaks

AF - small number of conversions

• Seemingly smaller than expected

JH - sl 11, has this energy dependent correction been applied?

• Blair - not applied so residual is expected

• JH - in this case we have 2 towers in each block, in test beam we saw 2 bumps, but here we only see 1 bump, so we probably need more data to map out these structure

• JO - how are the uncertainties on the black dots determined

• XY - determined by slice by slice fit, uncertainty is the uncertainty of the mean value

• JO - look at hodoscope testbeam analysis note 

• Blair - we see two hump structure, caveat x axis is center of gravity of detector, so its not an unbiased position distribution, not real incident position on the block

Blair- sl 3. Little combinatorics, any loose matching requirements on plots or is it all conversion candidates?

• XY - not all clusters, loose requirement on matching, require conversion candidates ie one positive and one negative charge for each candidate

Blair - looking at sl 6, what is the dominant source of bkg?

• XY - combinatorial pion pair background

• Blair - so for 2 tracks randomly intersecting, would those be candidates or would they be excluded for extending back to the beam spot?

• XY - no requirement on associated vertex

• Blair - how do you get the conversion candidates?

• XY - see slide 8 for cuts, and track pair must have same charge sign

D0 production and analysis

• CD - 200M collisions is a small amount of the data set? 

• JO - yes but we should be able to see something

AF - What large scale analyses will we complete in the next couple of weeks

• CD - D0, Lambda_C, Ds/D+. Can run multiple KFP instances in same processing. Can attach 3 instance to KFP to end of tracking production stream

• GO - Can we add D+ Kpipi as well

• AF - should also reconstruct Ks since its a physics signal we can see as a sanity check

• JO - agree about Ks, 100% failsafe option, very powerful plot to show as function of crossing number

• MR - do we need some kind of way to count events? Ks/#events as a QA quantity as a calibration

• AF - yes will add some QA to it 

• JO - for internal consumption, use BCO counting as a sanity check

JO - KFP over track DSTs is 10ms/event so this will be really fast, can combine many segments into a job to prevent I/O throttling. Not as CPU arduous as slide suggests

AF - produce microDSTs with fairly loose windows that we can then use for detailed analyses

• JO - minimize running on condor

On Tue, Mar 4, 2025 at 1:40 PM Gregory Ottino <gottino AT lbl.gov> wrote:

Dear collaborators,

We will have the next HF/Quarkonia Topical groups meeting tomorrow, Wednesday March 3rd at 1:30pm ET. The agenda so far is as follows:

1. Intro/D0 Discussion - Convenors

2. Lambda dE/dX - Alex

3. Photon conversions - Xudong

4. AOB

The indico can be found at this link, including zoom information for attendance. Please let me know if you would like to add a contribution.

Warm regards,

Greg, on behalf of the convenors