Latest revision as of 16:53, 21 June 2023

Back to Main Page >> Optics Meetings

>> following meeting

>> previous meeting

Agenda

1. Table of present Sieve Hole angular acceptances (Kate) [1]
2. Sieve Hole Tilting: energy loss in sieve. 5-pass simulations with Sieve (Vassu)[2]
3. Sieve Hole Tilting: r,r', phi, theta distributions, rates (Zuhal), here
4. Initial results using hole centers in fits, rather than Monte Carlo truth values (Tyler)

Attendance

Kate, David, Vassu, Kent, Zuhal, Ryan, Tyler, Ciprian, Mike D.

Minutes

1. From Kate's table of present sieve holes: we cover a theta range from 0.313 degrees (US foil with hole 11) out to 1.10 degrees (DS foil with holes 13 & 73), which covers almost all of the lH2 target acceptance.

1. (Vassu): Slit scattering measured by looking at the fraction of events that lose > 10 MeV in energy in the sieve. With the 9 mm non-tilt hole we typically get less slit scattering than for smaller (5mm) diameter holes; statistics are marginal, and the effect is not dramatic: average (over all holes, all sectors, all 3 beam energies) we go from 1.6% slit scattering with 9 mm holes to 2.6% slot scattering with 5 mm holes. If we tilt the 5 mm holes, then the slit scattering goes down from 2.6% to 2.1%. The rate of clean events (delta E<10 MeV) decreases roughly proportional to relative areas between 5 mm and 9 mm, but is better for tilted than non-tilted. Conclusions: slit-scattering does not appear to be a big effect, non-tilted 5 mm holes seem a feasible option. Would be interesting to look at 1D distributions for r (and r') with and without selection on slit scattering to see if the "radiative tail" on r has large by slit scattering component.
2. (Vassu): 5-pass plots from 12C. Moller events from most of the different holes appear separable at 5-pass. Should make plots that look to see if the "close" holes can be isolated using (r,r') or zoomed-in (r,phi) plots. Note that the 12C elastic events at 5 pass are mostly not in the GEM acceptance (highest energy electrons, appear at too small GEM radius, smaller than for the hydrogen elastics (should double-check this statement with the lH2 elastic generator). Conclusion: adding additional holes are likely to make it hard to separate the Moller events from different holes at 5-pass.
3. (Zuhal): Looked at (r,phi) and r and r' 1D plots for tilted and non-tilt versions of largest radius 5mm holes (holes 13 and 73, with DS 12C foil). Tilting does not seem to improve the "tails" in the r distributions - no motivation seen here for tilting. Looked at the theta_lab distributions - there is a systematic but small (1.5%) shift in the mean theta between non-tilted (2.032 degrees) and tilted holes (2.063 degrees), which should be included in fits using sieve hole centers as "truth" data. However there is not an obvious change in the sigmas of the theta distributions (although I would have expected the non-tilted holes to appear "smaller" in diameter, i.e. smaller sigmas. However, the distributions in R_GEM are clearly narrower for non-tilted than for tilted holes.
4. (David, all) Tentative conclusions: modulo additional plots requested above, we should probably go ahead with present hole pattern with 5 mm diameter non-tilted holes all-around.
5. (Tyler): Initial results on using sieve hole centers as truth values for theta and phi are promising. Need to select only one foil at a time. Should ignore holes 21 and 41 in one-pass, because they are hard to separate 12C from Moller events. Will look at the mean and width of the residuals of the theta and phi distributions, compare to those from earlier fits to the Monte Carlo truth values, and also look at whether we get similar values for our polynomial fit parameters.
6. (Ciprian): Suggests that in our fitting algorithms we use rate-weighting on our simulated events, so that unlikely events are not biasing our fits in ways that won't happen in real data.

Meeting link information

See email invitation, or contact David Armstrong, Kate Evans or Jennifer Finch for Zoom link