HDGeant4 Meeting, March 9, 2021

From GlueXWiki
Jump to: navigation, search

HDGeant4 Meeting
Tuesday, March 9, 2021
3:00 pm EST
BlueJeans: 968 592 007


  1. Announcements
  2. Review of minutes from the last meeting (all)
  3. New wiki page: Particle Gun Collection (Richard)
  4. Issues on GitHub
  5. Pull Requests on GitHub
  6. Action Item Review


Present: Tegan Beattie, Mark Ito (chair), Richard Jones, Zisis Papandreou, Churamani Paudel, Lubomir Pentchev, Justin Stevens, Simon Taylor, Nilanga Wickramaarachchi

There is a recording of this meeting on the BlueJeans site. Log into the BlueJeans site first to gain access (use your JLab credentials).

Review of minutes from the last meeting

We went over the minutes from February 23rd. Richard has already made the proposed changes to the tagger-hit-to-energy conversion. Pull requests await merging in the halld_recon and halld_sim repositories.

New wiki page: Particle Gun Collection

Richard described an upcoming historical catalog of single particle gun simulations. See his new wiki page for the complete description.

Issues on GitHub

We went over the issues.

Hits in both CALs from the same track

Lubomir led us through a discussion of Issues on GitHub. Two ideas that came up:

  • There might be an effect where a charged particle's trajectory is extrapolated beyond what is reasonable, allowing it to hit the FCAL and create a spurious match there.
  • Particles that clip the corner of the BCAL and send parts of the shower into the FCAL could explain some of the double matching. This behavior could be different between G3 and G4.

For all the details, see the Issue itself, linked above.

Proton timing in the BCAL

Again, Issue #179 has mainly to do with late hits in the BCAL from charged hadrons. Tegan presented results from a study to determine the physical mechanism for the late hits, as usual,comparing the effects among G4, G3/HADR1, and G3/HADR4. It turns out identifying parent particles for the particles creating late hits was not possible in an unambiguous manner. He was able to plot the timing distributions for the various physical interaction mechanisms separately, again comparing the different Geant engines. In fact, some of these interactions happen quite late, tens of nanoseconds after the event.

For all of the details see Tegan's plots on the DocDB. They are the ones with "MAR04" in the name.

G3/G4 Difference in FDC wire efficiency at the cell boundary

Richard reported on the root cause of Issue #181. Interestingly, the G3 simulation agrees better with the data because of the requirement that only one cell in a layer is hit. This artificially eliminates hits from tracks that spend a small amount of time in a cell, but happily since those trajectory have lower detection efficiency, their elimination mimics what is seen in data. The G4 simulation has no such mechanism for eliminating corner-clipping tracks and so hits with long drift times are produced at a rate higher than seen in the data.

See the Issue, linked above, for a more detailed explanation, with diagrams.

We seemed to be forming a consensus that Richard should allow multiple hits per layer, but implement an efficiency function that reduced the number of large DOCA hits.

[Added in press: in the issue itself you will see Richard's description of adding actual wires to the FDC simulation geometry.]