On Jul 3, 2018, at 12:39 AM, Kyle McCarty <[log in to unmask]> wrote:

Hello hps-software,

While performing hodoscope studies, Rafo and and I have stumbled on some odd output when attempting to obtain track momenta. I'm not sure if this is an issue with what I'm doing to obtain it, or if it's something odd happening with the momentum calculations. Note that this is done on the iss166 branch with the hodoscope detector, so it is also possible that something here is causing an issue. To see the issue, consider the momentum distribution for the initial MC particles for all events in the data set (pure tridents). We see the following:

<MCParticleMomentum.png>

This is reasonable for 2.3 GeV data. Now, consider the momentum of the observed tracks (using track state "0"):

<ActualMomentum_X.png>   <ActualMomentum_Y.png>   <ActualMomentum_Z.png>

We see that virtually all of the momentum is in the x-direction. This is reasonable - we expect most of the momentum to be in the direction of the beam line. However, we can see that the track momenta are dramatically lower than what we would expect, with almost nothing above 500 MeV.
 
I checked to make sure this wasn't a field map issue, and verified that all steps in the process (SLIC, readout, and recon) used the correct field map (209acm2_5kg_corrected_unfolded_scaled_1.04545.dat), so this is not the issue. Rafo suggested manually calculating the momentum from the track parameters using the following method:

            final double aa = 3.e-4;
            final double B_PSpec = 0.5242301;
            final double a_Bz = aa*B_PSpec;
            
            double phi = track.getTrackStates().get(0).getPhi();
            double omega = track.getTrackStates().get(0).getOmega();
            double tanLambda = track.getTrackStates().get(0).getTanLambda();
            
            double magP = a_Bz / Math.abs(omega);
            
            double px = magP * Math.cos(phi);
            double py = magP * Math.sin(phi);
            double pz = magP * tanLambda;

Using this method, we obtain a more reasonable total momentum distribution.

<ActualMomentum_X_Fixed.png>

This suggests to me that there might be something weird going on with the momentum calculation in HPS-Java (or at least this branch of it). Another sign that this may be the case can be observed by looking at the track state at the calorimeter. I obtain this with the following method suggested by Miriam:

             TrackUtils.getTrackStateAtECal(Track)  

When calling "Track.getMomentum()", I see two classes of outcome. One where the momentum is absurdly high, and one where it is absurdly low. Some examples from actual data are included below, with the format "EcalState(px, py, pz) << >> State0(px, py, pz)":

             (  0.004,   0.001,  -0.000) << >> (0.315, 0.007, -0.005)

             (  0.003,  -0.001,  -0.000) << >> (0.208, 0.011, -0.006)
             (  0.004,   0.001,   0.000) << >> (0.285, 0.006,  0.006)
 

             (458.765, -133.711,   9.243) << >> (0.160, 0.005,  0.003)

             (453.831,  149.604, - 9.234) << >> (0.174, 0.005, -0.003)

             (447.438,  167.760,  11.122) << >> (0.149, 0.003,  0.004)
             (459.146, -132.398, -18.623) << >> (0.165, 0.004, -0.006)

Clearly, something quite odd is going on, since it is completely unreasonable for a track to have the greater part of 500 GeV momentum in a 2.3 GeV beam. Likewise, I would not really expect us to see tracks with nearly no momentum at all. This suggests to me that the track state calculations for states other than "track state 0" (the target?) may be incorrect.

I'm hoping someone more knowledgeable on the SVT side of things can give some insight into what is going on here, and correct me if I'm doing something wrong.  Also, if there are any tests I can do or files that would be useful to have access to, let me know.

Thanks,

Kyle

---

Use REPLY-ALL to reply to list

To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1