[Halld-offline] Truth info from BCAL hadronic showers

[Justin Stevens]

Hi Will,

I'm wary of trying to draw too much from a single event (so take this with a grain of salt), but if I assume the event you give below is "typical" in the BCAL TruthCell it looks like there might be 3 classes of hits:

1) very late in time (2-4 microseconds) 
2) late in time (30-70 ns), 
3) close in time (5-10 ns)

1) doesn't seem very relevant since they're so far away (and happen to be on the timescale of the muon lifetime as Richard suggested).  Depending on the source of class 2), they might be able to be removed from the reconstructed shower time calculation by only using hits within XX ns of the first hit.  Do you have a requirement that the hits used to calculate the shower time have to be in some XX ns window from the first hit?  For class 3) two of the outer layer hits have t = 8 and 10 ns, is that typically expected from propagation or are these hits "late"?

Maybe you've already tried it, but if you were to form clusters and time averages from these TruthCell hits, similar to the algorithm you use in the offline, it might tell you more about what kinds of hit selection could be useful in offline reconstruction to prune late hits.

-Justin

On May 29, 2014, at 6:32 PM, Will Levine wrote:

> Hi offline,
> Following up on the issue of "late" energy deposition in hadronic
> showers in the BCAL.
> 
> There are two different ways that BCAL truth information from hdgeant
> is stored: these can be accessed in JANA from the DBCALTruthShower and
> the DBCALTruthCell objects. The code that stores this information is
> in hitBarrelEMcal() in
> sim-recon/src/programs/Simulation/HDGeant/hitBCal.cc
> 
> As I understand the code, DBCALTruthShower records every particle that
> geant tracks thru the BCAL (with energy greater than 1 MeV). One
> DBCALTruthShower for each unique particle in the BCAL. This includes
> all secondary shower particles. The particle type, initial energy, and
> initial position in the BCAL is recorded for each particle.
> 
> DBCALTruthCell just records the energy deposited in a given BCAL cell
> without recording which particle was responsible for the energy.
> (Again this is "truth" information, before any detector resolution or
> readout effects are incorporated).
> 
> These two sources of information are not always tremendously helpful.
> For example in one single-pion event that has some characteristic
> "late hits", we see the following DBCALTruthShower's:
> 
> DBCALTruthShower:
> ptype: track: primary:    phi:      r:     z:     t:     p:     E:
> -------------------------------------------------------------------
>     8      1        1  -1.863  65.042  185.3  4.780  0.571  0.588
> 
> Only the primary pion is recorded. There is no record of any secondary
> tracks. But when we look at the TruthCell's in the same event:
> 
> DBCALTruthCell [truncated]:
> module: layer: sector:     E:       t: zLocal:
> -----------------------------------------------
>     34      1       1  0.006    71.31    -6.0
>     34      1       3  0.027     4.87   -24.4
>     34      1       4  0.002     4.79   -26.5
>     34      2       1  0.006    33.59   -16.9
>     34      2       2  0.018     5.07   -19.6
>     34      2       3  0.013     4.98   -21.7
>     34      3       1  0.020    28.91   -15.4
>     34      3       1  0.005  4301.28   -19.1
>     34      3       2  0.023     5.19   -16.6
>     34      4       1  0.029     6.45   -13.1
>     34      4       1  0.019    73.07   -13.0
>     34      4       1  0.004  2219.77   -13.3
>     34      4       1  0.039  4301.12   -15.0
>     34      5       1  0.017     8.24    -9.6
>     34      5       1  0.001  4301.19   -17.1
>     34      7       1  0.011    10.44   -21.4
> 
> The time is in the second-to-last column. The "on-time" hits have t ~
> 5 ns. Where are all the late hits coming from? It is not clear. There
> must be some tracks that are not recorded in the TruthShower info.
> 
> Even in events with a more extensive record of secondary tracks it's
> not clear what is responsible for a late hit:
> 
> DBCALTruthShower:
> ptype: track: primary:    phi:      r:     z:     t:     p:     E:
> -------------------------------------------------------------------
>     8      1        1  -2.072  65.042  186.4  4.812  0.572  0.589
>    13     22        0  -2.083  65.969  188.5  5.185  0.284  0.982
>    14     21        0  -2.083  65.969  188.5  5.185  0.040  0.939
>    13     20        0  -2.083  65.969  188.5  5.185  0.173  0.955
>     3     19        0  -2.081  65.814  188.1  4.879  0.005  0.005
>     3     19        0  -2.083  65.851  188.2  4.884  0.002  0.002
>     1     19        0  -2.083  65.852  188.2  4.884  0.001  0.001
> 
> Any ideas?
> 
> Will
> _______________________________________________
> Halld-offline mailing list
> Halld-offline at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-offline