Page History

I ran the standard electron calibration code over the W production, resulting in the saving of 217,866 tracks that were suitable for calibration study. In the current analysis, some cuts will be relaxed since the goal is not calibration here. Runs before 10085131 were not used because Jan said there was some sort of trigger problem.

In these plots:

E = Energy E_T in 3x3 using ideal gains

p = p from outer helix of track (global if available, primary otherwise)

...

I then remove tracks with another track pointing to one of the surrounding towers, leaving 13696,806 946 tracks.

At this point, there's no clear E/p peak so things are looking a little troubling.

*Fig 1:*
!all_histo_0.pdf!

However, in the BSMDP total ADC spectrum there is a clear peak at the saturation boundary, which should correspond to electrons.

*Fig 2:*
!bsmdp_0.pdf!

I then apply a cut on BSMDP total ADC > 800.

This leaves only 3,987 160 tracks, put gives a very significant E/p peak around 1.1.

*Fig 3:*
!all_histo_1.pdf!

When I look at BSMDE total ADC vs E/p, I see that the remaining background is below around 500 ADC, so I add another cut there and remove the rather strong cut on BSMDP.

*Fig 4:*
!bsmdevep_1.pdf!

I'm left with 20,100 and a very strong E/p peak at just above 1.

*Fig 5:*
!all_histo_2.pdf!
Here are some more plots of these electrons

*Fig 6:* p vs E/p:
!pvsep_2.pdf!
This band in the upper right looked a little peculiar to us, but we found that it corresponds to the L2 trigger threshold, which was at \~13GeV E_T (the L0 threshold is also visible):

*Fig 7*: E_T spectrum: !te_2.pdf!
The shape of the band can be explained by the track p smearing:

*Fig 8*: !tevsp_2.pdf!