FWD Software

["Brandenburg, Daniel" <brandenburg.89 AT osu.edu>]

Hi All,

 

Please find several updates below. I would like to avoid meetings entirely devoted to me reporting what I’ve done, so unless I hear of an update, please consider this email in place of a meeting today. However, if there is anything to discuss I am more than happy to have a meeting.

 

PicoDst updates

• PicoDst now has “FwdVertices” that contain a position, cov matrix, and nTracks associated with the vertex.
• FwdTracks mVtxIndex is now split 6 bits vs. 2 bits to contain the track type (2 bits) = [0=Global, 1=Beam line Constrained, 2=Primary Vertex constrained, 3=FWD vertex constrained] and the 6 bits provide the index of the vertex depending on the type.
• mGlobalTrackIndex should now be set correctly in the PicoDst. For Beam-line, primary, or FWD vertex constrained tracks it is the corresponding global track index (in FwdTracks array), for Global tracks it is set to UShort_Max

 

Tracking Performance:

• There has been some confusion about the p_z=10 etc. However, I have investigate both mathematically and via pythia/single particles. If we have ONLY FST measurements + PV, we have essentially no measurement/constraint of the track sagitta. The fits are conducted with q/p_T and angular variables, but if the sagitta is not constrained it means that we only measure (to first order) a ratio between pT and pZ within a significant range.
• It would be nice for someone else to verify this with single particle or pythia -> compare track refit ON/OFF to see what happens when FST+sTGC vs. FST only.
• To this end I started working (again, since we started this last year) to investigate the viability of utilizing EPD hits into tracking.
• Unfortunately EPD was somehow not included in our recent productions (Not available in MuDst) – I am trying to see if this can be recovered or if a new production would be needed – I am trying to determine BEFORE initiating the production file reconstruction request

 

sTGC:

• Sam (OSU) produced a version of the sTGC reconstruction code that provides cluster level information to the FWD Tracking – instead of combining into spacepoints.
• Since each sTGC has potentially 4 measurement layers, this should greatly improve the chance of having at least one valid sTGC measurement along a tracks trajectory  - which greatly improves the sagitta measurement and therefore the pT/pZ constraints as discussed above.
• This is partially integrated and I will give further updates on its progress.

Misalignment:

• I found the (or at least a) major error in the misalignment that was applying the rotations twice causing the hits final position to be badly distorted.
• With this fixed the misalignment-enabled tracking is much more consistent with the previous “spacepoint” mode.
• With this I am ready to try with/without survey tables applied.

 

Productions:

• Due to the uncertainty related to the EPD information, I have not requested a production run from Gene, but I will try to close the loop on this today.
• I have groups of ~30M events running. In each folder I am putting a README file with description of any important details – e.g. FST ONLY, or misalignment mode ON, etc.

 

Tasks that could use help:

• Update StFwdTrackMatchMaker to use (R, phi) instead of (X,Y) for distance matching
• Plots showing correlations between mid-rapidity detectors (TPC tracks, TOF, BEMC, etc.) and FWD
• Plots showing correlations between EPD or EEMC and FWD
• Plots comparing FWD vertex positions with Event Vertex positions
• Summary of event stats vs. triggers (# of FST hits, # of tracks, FCS info) – I currently process all events regardless of triggers.
• Summary showing availability of vertex information(vs. trigger) – TPC, VPD, EPD, ZDC, BTOF…
• Just to name a few 😊

 

 

Best,

Daniel Brandenburg

Assistant Professor

College of Arts and Sciences

Department of Physics

Ohio State University