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:
|
Katherine and I have successfully managed to come up with a definition that uses TRAP volumes (Figure 1). With the new definition, hopefully it will be relatively easy to sculpt the sides in phi. In general it is a much neater way to define things than previously, and should be easier to translate into SolidWorks or to provide dimensions to the machinists. I found a company in Radford that claims to be able to water-jet cut most materials to better than .02cm, or about 5 times the precision of the water-jet prototype from the jeopardy proposal (see p. 53). They are limited to slabs less than 8cm thick, but Roger mentioned that the collimator may have to be made of thinner slabs anyway. Their table is about 70cm x 130cm, so they may be able to do the collimator in 2 pieces. I haven't been able to reach anyone yet, but this sounds promising. Metal Processing, Inc. MPI Waterjet |
|
Beamline and shielding:
Based on discussions with Roger, we have decided to stop pursuing the 2-plug solution, at least with the second plug in the downstream cleanup collimator. So we have begun concentrating on the 1-plug scenario, with the plug in the upstream cleanup collimator as the only one. In either case the important angle is the largest angle that can make it through the beamline inside the QTOR support structure. That angle is approximately 0.9° . We originally were hoping to see up to this angle from the whole target for the downstream lumis, but now we will see up to this angle only from the downstream end of the target. The maximum angle these lumis would be able to see from the upstream end of the target is 0.52° .
So far we have confirmed the preliminary geometry for that case as well as updating it with any changes implemented using the 2-plug geometry. We are still using the primary collimator as defined by row 20 from the famous table. We have not yet begun to optimize the shielding, because we would like to calculate new numbers for the power deposited in the plug, as well as getting a first estimate of the backgrounds with this case.
The basic beamline design is telescoping - it increases so that "nothing" will hit the beamline inside the beam after the plug, and the number of sections is kept to a minimum. There is a 50cm thick concrete detector shield wall, and as much lead in region I and II as will fit without interfering with the envelope of electrons that make it through the first collimator.
Optimization will consist of removing lead where it seems unnecessary to shield the main detectors, and adding it where needed. Of course, the electron profiles both inside and outside the beamline will have to be considered.
|
When Tony sent out his drawings for the GEMs I was knee deep in beamline geometry calculations, so I noticed right away that the frame he describes wouldn't work with the beamline I have currently. Figure 2 shows the outline of the GEM "active area" as well as the outer frame. At low radius, the outer frame interferes with the beamline. This was not a problem based on this drawing of the GEM frame which I guess is only the inner frame. I may be able to decrease the radius of that beampipe slightly in order to accomodate the outer frame, if it is placed as in this figure, but there would be no room for any shielding of the beamline there. If the active area must be placed so that is more centered on the electron profile, then that severely limits the size of the beamline at that location. This makes the defining element for the beamline the GEMs rather than the QTOR support structure, with a maximum angle of about 0.3° from the downstream end of the target, and only about 0.2° from the upstream end. This would impact the lumis as well as increasing the backgrounds created in the plug and the power deposited in it. |
|
Roger arranged for Pavel to calculate some neutron rates from the tungsten plugs and a reasonable approximation to the beampipe. Here are his results:
The rates for neutrons passing through the quartz bar are: (for a 18 cm x 200 cm quartz bar at 1.165 GeV and 180 uA):
E_k > 1 MeV: 648 MHz (most of these are from the tunsten plug and downstream beampipes)
E_k > 50 MeV: 29 MHz (Most of these are directly from the target)
Of course, the actual detected rate is much smaller and even the ones that have an interaction in the quartz bar likely won't produce any particles with enough velocity to generate Cerenkov light. So this is probably not a problem, but we should have him take one more look at it when we have all shielding etc. finalized.
Drift chamber:
There was a problem with one of the TDCs, so we don't have enough channels to instrument all four planes of the prototype chamber. Right now parts of each plane are instrumented so that some tracking code development can begin. It has been stable with the cathode plane and field shaping wires set to 1800 V. It is drawing about 50nA/plane. Norm is doing a series of runs with various threshholds.
