Primary Collimator Live Page
Katherine's Status Report
Tony's Status Report
Klaus's Status Report
Michael's Web Page
Burnham's Web Page
Burnham's collimator pics
Keep Out Zones a la Allena
Previous Reports
The following pages require a username and password:
Qweak elog at JLab
Qweak Simulation Forum
Virginia Tech Update for Tracking Group,
Collimator simulation/design:
At a previous tracking group meeting Burnham showed us great pictures of the primary collimator that he had rendered using SolidWorks (see figure 1). Unfortunately the pictures revealed some strange gaps in the downstream face of the collimator. At first I thought that there may have been a problem with the translation from GEANT 3 to SolidWorks, but upon re-examining the collimator in GEANT I found I could see the gaps too (see figure 3 below figure 1).
Then I realized what the problem was - the very last step of the collimator design! After the acceptance had been finalized and trimmed to fit the bars and to pass through QTOR and its support structure with some clearance, I tried to make a more "machineable" shape by using octagonal volumes to define the upper and lower edges of the collimator opening. In the process of doing this I had to choose a radius for the center of each edge. I found that the best radius was somewhere between the octagon that would circumscribe the circle and vice-versa (see figure 2).
This is the origin of the gaps at large and small radius of the opening. Figure 4 shows the GEANT picture of the wireframe of the collimator when the proper definitions of the box and wedge are put into the euclid file.
|
|
Figure 1 - SolidWorks picture of downstream face of primary collimator. |
Figure 2 - Diagram comparing an octagon and a circle. |
|
|
Figure 3 - Wire frame drawing of primary collimator with gaps at inner and outer R at downstream face. |
Figure 4 - Wire frame drawing of primary collimator with proper dimensions. |
Since there weren't actually holes, but more like depressions, I suspected that the profile and the rates would be unaffected by their presence. So I went ahead and re-ran the simulation without the fix and then checked the rates at the Region IIa detector and at the cerenkov bars. There is a slight difference because once one random seed is different, the events after that will all be different. Other than that, it appears that the gaps had no effect on the profiles after the primary collimator or on the rates.
Collimator |
Region IIa |
Cerenkov bar |
Original |
1224 MHz/octant |
875.2 MHz/octant |
"Fixed" |
1226 MHz/octant |
876.3 MHz/octant |
So the "fixed" euclid definition of the row 20 collimator has:
RINNU = 31.4
ROUTD = 62
VOLU 'LPHI' 'BOX' T_LEA 3 (ROUTD-RINNU)/2 6. 10.83
POSI 'LPHI' S_TRG+CID_TRG 'PCOL' (ROUTD-RINNU)/2+RINNU 0 0 0 'ONLY'
SATT LPHI COLO 46
CID_TRG = CID_TRG+1
!
RADIUS = 26
!42~62/cos(22.5)-RADIUS
!VOLU 'TORD' 'TUBS' T_LEA 5 0 40 10.83 -22 22
VOLU 'TORD' 'TUBS' T_LEA 5 0 42 10.83 -22 22
POSI 'TORD' S_TRG+CID_TRG 'PCOL' RADIUS 0 0 0 'ONLY'
SATT TORD COLO 46
CID_TRG = CID_TRG+1
Beamline and shielding:
Pavel produced some results for power deposited in the plugs and neutron production rates in a simple geometry for the two plug case. Mark is still going over the results; he'll send out a report via email before our next tracking group meeting so people can digest it before the meeting.
Drift chamber:
Currently we have a half-chamber + an x plane. We plan to use the half chamber to find the position on the extra x plan and get out a resolution.
Now we are working on getting full readout for the 4 planes.
We replaced our switching +-5V supply with a linear supply. It is much quieter than the switching supply but provides only enough power for one chamber.
We need +5V and 44 Amps and -5V and 62Amps to power all of the cards for four full chambers.
