sPHENIX tracking discussion

[Thomas K Hemmick <hemmick AT skipper.physics.sunysb.edu>]

Hi everyone

Let me add a comment before this conversation vectors too far into speculation.  Please look at this zoom-in of an event display since it is VERY instructive for the conversation.  Furthermore, based upon experience, this event display is completely plausible and within expectations of what one should see.

The wide cyan-blueish bands are the layers in the TPC gas.  The thin lines are the G4-hits.  From this display, one can learn very rapidly what goes on and what are the reasonable conclusions.  First, let's concentrate on a few "normal tracks":

These pics are from Jin Huang and I'm told that the yellow dots represent G4 hits.  Basically, this looks completely reasonable.

Now let's look at the worst spot in the diagram:

Here you see what a delta does.  The delta ray spirals along the field in a VERY tight spiral, bound to the field lines.  What you will also notice is that this delta has made an ENORMOUS number of G4Hits, while actually being confined to a pretty small region of the gas (single layer or perhaps spiraling across the boundary between layers).

Let's look carefully at Tony's plot keeping in mind the info from the event display.

Here you see that the primaries (summed over these events) make a bit over 12M clusters (mostly pions).  You see that the secondaries contain about 5M delta-clusters (e- minus e+).  I would assert based upon the event display (and common sense) that these counts of clusters are MUCH higher than the actual number of delta rays, because of the deltas typically forming the tight spirals shown at several places in the event display.  

Therefore, Tony and I are in agreement, the number of deltas is small, and the number of clusters from them is tricky to interpret as a straight "occupancy" since these are overwhelmingly tightly grouped as a zed-spiral.  Cluster is an algorithm output of our tracking code, not a simple count.

FYI...we can report Friday about a meeting in which PM is looking somewhat favorably at the suggestion from the TPC group that we commission a SAMPA version that re-institutes the time constant that was the basis of our initial design  (original SAMPA switch-selectable between 80/160/320 nsec, current SAMPA 160/320, sPHENIX version 80/160).  This move is also supported by the work of the simulations group

Tom

On Thu, Sep 21, 2017 at 12:23 AM, Marzia Rosati <marziarosati AT gmail.com> wrote:

Tony,
if you could keep a count of the number of clusters per each primary and secondary particle in your plot you could divide your plots by the average number of clusters in each particle flavor to extract the actual flavor multiplicity.
I suspect that the average number of clusters for delta electrons is very high since they are low momentum and might spiral in the detector or have weird sideways trajectories crossing the whole detector....
Best regards
Marzia

On 9/20/17 11:03 PM, Anthony Frawley wrote:

Hello All,


Following a discussion in the TPC electronics meeting this morning, I have been trying to understand the origin of the clusters in the TPC that do not come from primary tracks, by looking at clusters and G4 hits from evaluator ntuples made from 100 central Hijing simulations.


The answer is that they come from a variety of secondary particles. Please see the attached three plots.


The first plot shows the particle flavor (gflavor) of clusters associated with (left) primary tracks and (right) secondary tracks. The plot on the left shows pions and kaons. The plot on the right shows entries from:

a) secondary pions (+/- 211)

b) secondary muons (+/- 13)

c) secondary electrons (+/- 11)

d) delta electrons (-11 only, the excess of -11 over +11)


The second plot shows (left) the truth energy deposit for each reconstructed cluster and (right) the truth pT distribution for the reconstructed clusters. These are shown separately for the primary pi+and p-, and for secondary mu+ and mu-, secondary pi+ and pi-, and secondary e+ and e-.  The delta electrons appear in the left plot as the excess of e- over e+, and in the right plot they can be seen to sit at essentially zero pT.


The third plot is the same as the second plot except that it is made from the G4 hits ntuple. It just shows that there are a large number of delta electrons recorded as G4 hits, but most of them are not reconstructed as clusters.


I should point out that these plots should not be read as showing the relative yields of the secondary particles. They show the number of clusters due to different types of particles. The secondary particles are at low pT and If a low pT muon (for example) loops in the magnetic field it will make many more clusters than a higher momentum track.


The conclusion is, I think, that Tom is correct in his belief that delta electrons could not double the occupancy in the TPC. They do not. The occupancy in the simulation is doubled by a combination of decay and conversion particles, combined with delta electrons.


Unfortunately, the evaluator does not contain truth track information for secondary particles. Therefore I can't extract from my ntuples the relative yields of the secondary particles. I need to figure out how to get that info into the ntuples in future.


All comments welcome.


Tony

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

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