sPHENIX tracking discussion

[Anthony Frawley <afrawley AT fsu.edu>]

Hi All,

This is an attempt at summarizing what (I think) we concluded in today's meeting about tracking with distortions in Au+Au collisions in MDC2. This does not apply to p+p collisions, which are more complicated.

Joe showed today that running seeding on distorted TPC clusters works, but the seed track parameters are not good for matching to the silicon tracklets. Good silicon matching requires correction for both static and average distortions, when using the current matching windows. I do not believe that increasing the matching windows in AuAu collisions is a realistic option.

Therefore it seems that we will need to make a preliminary average distortion correction before we can do silicon-TPC track matching.

It is true that we can make distortion corrections using the static map and an average map for clusters without regard to the track trajectory they are on. However we cannot move the corrected cluster back to the readout layer it came from unless we have the trajectory. Therefore we have to find the seed tracks before we can fully correct the TPC cluster positions back to the readout layers.

So, a possible workflow is:

1) Assume that all tracks and clusters of interest are from the triggered bunch crossing, so they have the correct (albeit distorted) z values.

2) Run CA seeding on the distorted clusters.

3) Make static and (preliminary) average distortion corrections on all track-associated clusters.

4) Project the corrected clusters to their TPC readout surface, parallel to the (distorted) track trajectory.

5) Rerun KFProp on the corrected clusters to recalculate the TPC seed track parameters.

We are then ready for silicon-TPC track matching.

Proceed with pass 1 from there.

Is this a fair summary?

Let's talk more about it in the distortions meeting tomorrow.

Tony