- Based on the VT group "final" primary collimator design, I studied the possible choice of the shielding wall position and the size of the opening on the wall. Firstly, I ran a few GEANT simulation to check the elastic electron envelope at different Z positions along beam-line. The electron envelope is obtained by plotting x projection and y projection histograms of the electrons that hit the Cenerkov bar at Z=570 cm (331> x >313 cm, 100 >y >-100 cm). In the simulation, 1 million elastic electrons are triggered (ep elastic). Below is a table of results (unit: cm).
Z X Y Figure 330 213.5< x <250.5, x = 232 +/- 18.5 -68 < y < 68 X and Y proj 380 234.5< x <267, x = 250.75 +/- 16.25 -74 < y < 74 X and Y proj 430 255.5< x <284, x = 269.75 +/- 14.25 -81 < y < 81 X and Y proj 480 276< x <300.5, x = 288.25 +/- 12.25 -88 < y < 88 X and Y proj 530 297< x <317.5, x = 307.25 +/- 10.25 -95 < y < 95 X and Y proj
According to the table, I fit the X_mean as a linear function of Z position to define the central ray angle of the scattered electrons. The angle is 20.57 degree, as shown in figure below. My old result is 20.85 degree which was obtained based on the old Boston collimator design.
- Based on the obtained electron envelopes and the possible dimension of the region 3 drift chamber (at least 1 m long along beamline), I put a 0.5 m thick shielding wall at Z = 380-430 cm, and selected an opening of x = +/-17.25 cm , y = +/- 75cm at the front, and x = +/-15.25 cm, y = +/- 82 cm at the back, with center at x = 260.25 cm, z = 405 cm.
1. View of the shielding opening along the beamline
2. Sideview of the shielding opening
I ran the simulation with/without the 0.5m wall for 1 million events (ep elastic reaction, tracking primary and secondary e- only, with all secondary reactions turning on). As seen in Fig.1,
without the wall, number of e- (energy > 5MeV) = 56681;
with a 0.5 m wall, number of e- (energy > 5 MeV) = 56767.
The results are close within the statistic uncertainty. With the selected opening, some electrons (leas than 0.2%) hit the edge of opening and bounced to the Cerenkov bar. Also, based on the Fig.1, there is no shower electrons (The electrons generated by the electron hitting the edge of the shielding opening) hitting the Cerenkov bar.
- I also checked the inelastic background for with/without 0.5m shielding wall case. Below are the results I got for different inelastic reactions and moller scattering. Fig.2 and Fig.3 showed the results of the first two inelastic reactions.
Without wall
(a) e + p -> e + n + (pi+), # of e = 64 (0.07%, when weighed by the cross section)
(b) e + p -> e + p + (pi+) + (pi-), # of e = 0
(c) e + p -> e + p + (pi0), # of e = 0
(d) moller scattering, # of e = 0
(e) e + p -> e + X (inclusive), # of e = 0
With 0.5m wall
(a) e + p -> e + n + (pi+), # of e = 122 (0.2%, when weighed by the cross section)
(b) e + p -> e + p + (pi+) + (pi-), # of e = 0
(c) e + p -> e + p + (pi0), # of e = 0
(d) moller scattering, # of e = 0
(e) e + p -> e + X (inclusive), # of e = 0
