Kunsu

Slide 1 and 2 are Kunsu’s plan for NPE analysis. 

Slide 4 and 5 are DCAz distribution and blue points are obtained by subtraction mirrored distribution itself.

Slide 6, left plot is mean and sigma distribution for main (blue) and 2nd (black and red) peaks. 2nd peak has pT dependence. Right plot is raw yields. hadrons and pions mean and sigma distributions are same but raw yields are different.

comments:
1. Check conversion electrons. Since the 2nd peak seems to depend on pT and after pion selection cut the ratio of 2nd peak is redueced, it maybe is from converted electron by beam-pipe or detector materials.
2. Pile-up events which can be checked with Vertex (Vx, Vy, Vz) distributions.
3. How about its behavior with separation of charges (positive and negative tracks, respectively).
4. Change pT bins to wider for hadrons.
Xiaozhi’s conclusion about 2nd peak: I never said  the second little peak was from conversion electrons, as you can see, the peak was more clear from in hadrons than electrons samples, however, i would like to suggest it is from some of  events or runs with the bad vertex quality, you can apply the vertex Vx and Vy to confirm, if the  little peak is still there, you can  check the run dependent.