Hi,
This is a followup for the accidental study. Although, it is
not certain whether the event-mixing method will work for Blast
or not, the method in general works pretty good for accidental
studies.
I had used this method to extract the accidentals for my thesis
experiment (np -> NN \pi , np -> np \gamma), where the yield
of accidental events were comparable in size to the true events
thanks to the noisy environment (lots of low energy neutrons, both
from our target and neighboring experiment) of LANSCE-WNR. I like
to share some of the results to demonstrate this, and a simple
simulation that I had used to understand the process behind it
which gives some insight how the neutron TOF is distributed for
random coincidences.
I attached 3 figures, and a small code:
1) accidental.m: This is a Matlab code that simulates the
accidental TOFs. I changed this to adapt it to the Blast.
TOF is calculated differently here. There are plenty of
remarks in the code.
2) simulation.pdf: This is a sample result for the above code
that shows the triangular shape of the distribution.
It is generated by accidental(2e6,2e6,200e-9,200e-9) call
(See the code for the explanations of the parameters)
The rates are inflated to see the effect quickly. The slope
is probably smaller for blast as one of the channels
is not counting as high as the other.
3) true.pdf: This is the data from my thesis, that shows the
neutron TOF from the LH2 target (basically the time
difference where times are calculated a little
indirectly). The sharp peak is the gamma flash from the
target. The stuff at the right of the peak is (expected to be)
the neutrons. And, there is a huge amount of accidentals
below this which spans both the physical region (t_n>t_gamma),
and the unphysical region (t_n<t_gamma).
4) accidental.pdf: This is the same plot generated AFTER
event mixing. The accidental background is perfectly
visible here when we compare it to the true events,
and "as" expected when we compare it to the simulation.
Then we determine the scaling factor using the unphysical
region in t_n plot. Then we subtract the accidentals
using the "final quantity of interest" scaled by that amount.
Although, the effect is expected to be small at Blast both
because the accidentals are not huge (~5-7% ?), and it caries the
asymmetry, it may be worth to try something similar to this just
to show that it is NOT large.
Eugene, if you'll try this, and have questions, we can sit
together sometime, and look at the details (which may not be
trivial.)
Regards,
Taylan
-- ---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--- Taylan Akdogan Massachusetts Institute of Technology akdogan@mit.edu Department of Physics Phn:+1-617-258-0801 Laboratory for Nuclear Science Fax:+1-617-258-5440 Room 26-559 ---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=---
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:32 EST