Efficiency Study Continues with New High Voltage Scan

HV scan results on 4/20/2012 

(Now there are two identical files attached due to me sending out a wrong link initially)

Since the above linked result on HV scan shows clear signs that in many cases the unmatched cluster is still "half right", it is interesting to see what efficiency we get if we match only P or R cluster to the track without requiring both be in the right place.

I used +250V run (13109039), which had the highest efficiency. 

For this to make any sense at all, first we need to make sure that the cluster multiplicity is low. 

Fig. 1. Cluster multiplicity for "good track" events

So already it seems that we have an issue with the Phi plane. The distribution looks strikingly like an exponential starting from 0. But the R plane looks reasonable. For our purposes, this will have to do. 

Now we look at the efficiency for P and R plane separately. The cutoff is still 1cm. For P plane, I need to translate the radian into cm, which is normally done by using the R cluster coordinate. But since now I don't always have the R cluster, I use average value of 20cm when missing then R cluster.

Fig. 2. R-cluster only cluster matching rate and efficiency

So for the "good" region, we get almost 60% efficiency, and the cluster matching rate is ~80%. These are all quite similar to the simulation (70% and 95%, respectively), which is not fully realistic to begin with.  

Fig. 3. Phi-cluster only cluster matching rate and efficiency

As expected, phi plane is not quite as good. 

At this point I do not know if the difference between the Phi and R plane is something that can be improved upon off line.