Feature or Bug? No hits in the strawtubes for HNLs in low momentum range. (EventCalc-SHiP)

[anupama-reghunath]

Carrying over this discussion from the software channel:
Hi!
I've noticed a significant difference in the reconstruction efficiency between the generated and normalised samples produced with EventCalc-FairShip

*correction in the table headers, it should be /nSim in %

on inspecting the HNL momentum of the samples(blue) and the ones that were successfully reconstructed (red), and the ones which have hits in all four tracking stations (green), there seems to be a correlation between the momentum and the reconstruction efficiency.

many of the events where all four stations are not hit, it seems there are no hits?

I tried to plot simplified trajectories of these daughters ( orange being the case where the reconstruction was successful) and blue otherwise.

at the intersection of these "tracks" ( I call them so because these are essentially a line connecting the truth of decay vertex till the farthest StartZ() position available of tracks from any subsequent interactions of the daughters, sort of an ugly hack to paint a simplified picture)

and their intersection densities at the fourth Tracking station z position(z=3538cm).

Maybe its better I also share the path for further diagnosis:
/eos/experiment/ship/user/anupamar/Signal_EventCalc/mupi/12435731/HNL_1.000e+00_5.000e+01_3.333e-01_3.333e-01_3.333e-01/
the files used here is from job_{0...4} (used for these plots).

can someone help me understand what could be happening here? and if the difference in reconstruction efficiency % is a bug or a 'feature'?

[mferril]

can someone help me understand what could be happening here? and if the difference in reconstruction efficiency % is a bug or a 'feature'?

Hi @anupama-reghunath , have you compared the HNL momentum distribution* for the same final state you are interested in btw EventCalc/OldGenerator? I suspect that some difference might arise from a softer/harder spectrum, but that is to be seen.

*Excluding events below 100MeV

[anupama-reghunath]

Unfortunately since I no longer work on software tasks in my current role, I leave this for the other software experts to resolve for now. :)