Assuming my beam optics and collimator details are correct (someone should see if they reproduce the results), the first diagram shows the unweighted distribution of angle of scatter, theta_o, in the target for the downstream collimator as of our last meeting on March 17, with detector at z = 590 cm and BFIL = 1.0, MFIL = - 0.078175, ISOT = 1. The black curve is with no cut on events on the Cerenkov bar, red is with the handlebar cut and green with a straight bar cut (319 < x < 334 and -100 < y < 100).
The second figure shows the distribution of angle phi_o at the target with the same colour-coding as for figure 1.
The collimator is not entirely defining the acceptance. The loss of events at large theta and small phi indicates that the top centre of the moustache is being cut.
The variation of rate on a Cerenkov bar as a function of the position of beam on target is well-represented by:
R(x, y) = a(1 + bx + cy2 + dxy2)
Here are the coefficients, using centimetres as the unit of length, from plots 100 channels by 100 channels covering an area at the target of 2 cm by 2 cm (BEAX = BEAY = -1 cm). The error bars were estimated by switching off cross-section weighting so that the PAW fitting routine was using pure numbers of electrons surviving through the collimator to the Cerenkov bar. The reduced chi-square in each case was about 1.02. Results for the Boston collimator are added for comparison.
| No cut | Handlebar cut | Straight bar cut | Boston | |
|---|---|---|---|---|
| a | 0.0976 +/- 0.0001 | 0.0736 +/- 0.0001 | 0.0702 +/- 0.0001 | 0.0606 +/- 0.0002 |
| b (x) | 0.1058 +/- 0.0017 | 0.1262 +/- 0.0019 | 0.1465 +/- 0.0019 | 0.2659 +/- 0.0050 |
| c (y2) | -0.0950 +/- 0.0019 | -0.0985 +/- 0.0022 | -0.1004 +/- 0.0022 | -0.1247 +/- 0.0060 |
| d (xy2) | 0.0245 +/- 0.0033 | 0.0230 +/- 0.0038 | 0.0247 +/- 0.0037 | -0.053 +/- 0.011 |
There is an increase in the slope in x going from no cut to handlebar to straight bar. The sensitivity to position motion averaged over all eight bars is proportional to the y2 term, which increases by about 6%. The total rate for unrastered beam is 976, 736 and 702 MHz per bar for the three cases for 10 mil Al walls and windows on the target. Average Q2 = 0.03.
The following tomographic plots show the origin of secondary electrons that are generated by primary electrons hitting collimators and thus show where the electrons go. The secondary photons have to reach a large detector plane at z = 590 cm in order to be counted. Rates are calculated as for elastic electrons as elastic electrons are being used to generate the events. Mark points out that Juliette has removed I believe all but the primary collimator until it is finalized.
The first pair of plots shows z, horizontal, as a function of y (non bend direction) and x (bend direction) in the upper and lower plots. The target is seen centred at z = -650 cm, followed by a very narrow source of secondary photons and three more substantial sources at the three main stages of the collimator. The following plots show x-y distributions at those three collimator locations.
The next pair of plots shows the region of the first part of the collimator at -580 < z < -576 cm. The lower plot shows rate in MHz/bin at the right side. The most concentrated source of secondary photons is at the inner (small angle) side of the collimator (total 77 MHz), although the broad band at the top sums to 60 MHz.
The next pair shows (x, y) at -564 < z < -558 cm. Electrons hit mostly the inner surface and the lower part of the sides of the collimator. The band at the bottom contributes 74 MHz, and 144 MHz from the sides. The total for -564 < z < -530 cm is 243 MHz.
The next is for -350 < z < -320 cm. The strongest source of photons is the inner edge of the collimator (315 MHz).
