I do substitute low threshold TDCs when the high threshold ones are
either overflow or underflow. I'm currently defining underflow as <400 and
overflow as >4000. I had to fix a small bug with this recently, so I
know that the low threshold TDCs do get used frequently.
Also, to keep track of TDC problems, I use the energy reconstructed
using the ADCZ if both sets of TDCs are bad or if the TDCZ is more than 5
meters from the box center. Since most of these hits are cut anyways, I
don't think it changes anything for the hits that are kept, but it allows
me to count the number of hits that would have reconstructed at a certain
energy but were tossed because of bad TDCs.
Below is a summary of my cuts for run 10015. Note that cuts are applied
sequentially in the order indicated. Approximately 1/3 of all events are
reconstructing with |TDCZ|>700, regardless of energy, and also about 1/3 of
the remaining events have |ADCZ-TDCZ|>200.
All E>6MeV E>8MeV E>10MeV E>12MeV
Total GC events: 167786 60675 25147 17201 14665
Multi box hits: 14162 13552 13159 12853 12571
------ ----- ----- ----- -----
Single box hits: 153624 47123 11988 4348 2094
TDC Over/underflow 18016 2552 906 417 118
------ ----- ----- ---- ----
135608 44571 11082 3931 1976
TDCZ Cut: 45086 14472 3799 1366 658
(ABS(TDCZ).LT.700) ------ ----- ----- ---- ----
90522 30099 7283 2565 1318
ADCZ Cut: 25551 8227 2411 1006 438
(ABS(ADCZ-TDCZ).LT.200) ----- ----- ---- ---- ----
64971 21872 4872 1559 880
Trigger Coinc. Cut: 145 144 143 140 139
----- ----- ---- ---- ---
64826 21728 4729 1419 741
Hits >=6 cut: 83 81 81 79 79
(>5 hits in fast view) ----- ----- ---- ---- ---
64743 21647 4648 1340 662
Near hits cut: 83 80 80 79 79
(>3 hits in fast view, ----- ----- ---- ---- ---
one or more hit within 64660 21567 4568 1261 583
50 cm of box axis)
If people think it would be interesting, I can play around with the
order of the cuts or any of the parameters. Let me know what you think.
As for the ADCZ cuts for Thursday's burst, the discrepancies were much
more than 2 meters.
Hit TDCZ ADCZ Diff
2 -308.7 276.8 585.5
7 -439.4 7.1 446.5
8 -507.7 402.3 910.0
12 421.0 -288.8 709.8
In all these cases, the two positions are on opposite sides of the tank
even. Also, the energies reconstructed from the ADCZ are significantly
lower than the TDCZ reconstructed energies. For these 4 hits, I'm sure
the ADCZ cut is good physics.