Klaus's Status Report
Yongguang's Status Report
Qweak Simulation Forum
Dilution factor notes (Dave Bowman)
Previous Reports
Virginia Tech Update for Tracking Group,
Summary of previous working group:
Please click here for the working group summary
Aluminum Backgrounds:
Dave Mack asked us to look at elastic e-Al events. The distributions from the upstream and downstream
endcaps are shown in Figure 1. The elastic e-Al rate from the two windows totals about 12 MHz.
Figure 1 - This plot shows the distribution of electrons elastically scattered from the downstream (top) and
upstream (bottom) Al target endcaps. The ep elastic peak is shown in the middle for reference.
GEANT geomtery changes:
Version
Elastic Rate (MHz)
Inelastic Rate (MHz)
Percent Inelastic (%)
< Q2>
FOM
Reference Design
801
0.212
0.026
0.02994
0.7182
Updated Reference Design
727
.191
.03
.03070
.6858
Klaus discovered that the location of the Region II chambers interfered with the support structure of the main
magnet, so Mark proposed moving the second half of the primary collimator downstream and placing the chambers
between the two halves of the primary collimator. There are several issues that have come up as a result of
this change.
Impact on Region II Chamber Design
- increased chamber size, shape change?
Impact on Primary Collimator Design
- second half moved further downstream
- openings in second half appropriately updated
- certain beamline elements appropriately updated
"Housekeeping" changes needed in the geometry
- two halves of collimator entirely separated
- collimator material goes all the way down to beampipe sheilding
- target endcaps now included in the geometry
- second half of primary collimator has larger phi range than first half
- beamline mother volume large enough to include all beam elements
Table 1 - Comparison of rates with various changes to the reference design.
The target material in the reference design was beryllium. Changing it back to aluminum, along with all of the
changes listed above, and the new elastic ep rate is 727 MHz (See Table 1, second row). By far the largest loss
in rate was due to changing the target cell back to aluminum. The tube around the target cell is 20 mils thick
in the radial direction. Changing the thickness of this tube for either material affects the rate, as shown in
Figure 2. We believe this difference in rate is mostly due to bremsstrahlung in the target tube which alters
the energy enough that the electrons no longer make it into the detector acceptance at the focal plane.
|
|
Figure 2a - A plot of rate vs. target tube thicknesses for beryllium and aluminum target cells. |
Figure 2b - A plot of FOM vs. target tube thicknesses for beryllium and aluminum target cells. |
Greg then asked about the distribution of the ep's at the end of the target cell. It might be possible
to "design a target cell that keeps the cell wall perpendicular to all our accepted trajectories" to
minimize the effect of the bremsstrahlung in the target cell. The plots in Figure 3 were made by projecting
the trajectories to a plane at the end of the target.
|
Figure 3 - Plots of the ep distribution at the end of the target. |
Because of the various housekeeping changes that I had to make, I started to make the GEANT geometry
summary page geant_geometry_summary.html. This page summarizes every component that exists in the
simulation and shows a drawing of each. I also included the kumacs that I used to make each drawing
and the euclid file that I used. The euclid file was modified slightly for the purpose of making some
of the pictures. The page is NOT meant as the defining geometry of the experiment.
BFIL=1 Detector Update:
Figure 4 - The ep peak for BFIL=1.0 with the chosen detector shape superimposed.
BFIL
z (cm)
Elastic Rate (MHz)
Inelastic Rate (MHz)
Percent Inelastic (%)
< Q2>
FOM
Cuts
1.040
570
727
.191
.03
.03070
.6858
a= .13
1.000
590
733
.289
.039
.0303
.6749
a= .09
1.000
590
726
.138
.019
.0303
.6648
a= .10
1.000
590
718
.056
.008
.0302
.6540
a= .11
1.000
595
734
.357
.048
.0304
.6794
a= .09
1.000
595
727
.191
.026
.0304
.6704
a= .10
1.000
595
720
.081
.011
.0303
.6598
a= .11
BFIL is a factor to multiply the main magnet field strength for the simulation. It was set to 1.025
and was increased to accomodate the minitorus and to move the ep peak out to a larger radius in
order to make room for PMTs, etc.
The possibility of changing this factor back to BFIL=1.0 was discussed at the last meeting. The
concern was finding a workable detector location and size/shape with the minitorus on and the
lower magnetic field. I set BFIL=1.0 with detectors spaced 5cm apart near the reference design
z location. I looked at the ep peak with the inelastics superimposed for all of these locations (see
detector_study.html). I chose z=590 cm because the ep peak has a "nice" shape there.
Admittedly this is rather qualitative; more specifically the ep peak was getting skinny at this z but
had not started to develop the "horns" which are becoming evident in the plot for z = 595 cm. I then
tried several slopes after centering a 16cm wide bar on the ep peak. The rate information is compared
to the updated reference design in Table 2 below. Figure 4 shows the chosen z location with the
detector shape superimposed.
322.5 < x < 338.5
-105 < y < 105
324 < x < 340
-105 < y < 105
324 < x < 340
-105 < y < 105
324 < x < 340
-105 < y < 105
326 < x < 342
-105 < y < 105
326 < x < 342
-105 < y < 105
326 < x < 342
-105 < y < 105
Table 2 - Comparison of different detectorsfor BFIL=1.0 to the updated reference design.
Primary Collimator Update:
Figure 5a - Plot of mean x vs. mean y in bins of theta and phi at the focal plane. Figure 5b - Plot of mean x vs. mean y in bins of theta and phi after the primary collimator.
This is being presented as a work in progress, but the conclusion that I am coming to is that sculpting
of the primary collimator to lower the inelastic percentage is not desirable. As the inelastics are
sculpted away, so are elastics, and the loss in FOM could outway the benefit of scuplting.
Below are plots of mean x vs. mean y for the focal plane detector (left) and the first detector after
the primary collimator. The squares are the elastics and the triangles are the inelastics. The
different colors represent different theta_o values in two degree bins andthe increasing size of the
markers represents increasing phi values (also in bins of two degrees). Only half of the octant is shown,
but the plots should be symmetric about the vertical axis. The plot on the left is essentially a map of
where the various theta_o and phi_o events end up on the focal plane. The low theta events had low
statistics and should be ignored on this plot.
Basically, these plots show that in order to cut the inelastics at high theta and phi values, we would
have to cut out some elastics as well.
Future plans:
It looks like it is possible to find a workable solution at BFIL=1, so we would set BFIL=1 and continue
trying to optimize the primary collimator, including sculpting.
