PMT hits with large negative timestamps

[metelko]

In LessSimpleDAQ2Proc.cc
calls to
mcpmt->GetMCPhoton(i)->GetFrontEndTime()
mcpmt->GetMCPhoton(i)->GetHitTime()
occasionally give a large negative timestamp eg -1.68079e+08.
All hits in an even may have simular large -ve timestamps.

As the timestamps are all between -10^6 & -10^9, and there's nothing else outside -10 to 100, I think these timestamps can't be physical.

Are the entire hits unphysical or just the timestamps?

[MarcFBergevin]

For what type of events do you see these time stamps? There was a problem at one point with certain events with long livetime (e.g.: Li-9) and the time was greater than the Double cast.

[metelko]

Hi Marc,
just done some tests. I don't see any spurious events when running
/generator/add decaychain 40K:regexfill
but do when I run
/generator/add decaychain 232Th:regexfill

Just doing couts from rat I get

CountProc: Event 3 (3 triggered events)
New Event====================================
(time,charge,id): (39.7368,1.12994,137)
(time,charge,id): (67.6649,0.883934,2304)
(time,charge,id): (-6.29576e+07,0.692617,172)
SubEvents 2

and sometimes
CountProc: Event 5 (6 triggered events)
New Event====================================
(time,charge,id): (-1.15667e+07,1.1369,2321)
(time,charge,id): (-1.15667e+07,0.6273,2405)
(time,charge,id): (-1.15667e+07,0.892157,2405)
(time,charge,id): (-1.15667e+07,0.84658,2405)
(time,charge,id): (-1.15667e+07,1.12662,2199)
(time,charge,id): (-1.15667e+07,0.49498,1381)
SubEvents 1

Reading about the decay chain generator, the last decay is centred about t=0,and previous decays have negative timestamps,

Doing a large run for 232Th I can see that the timestamps follow an exponential, and I think I'm seeing part of the decay chain. So this is probably not a bug then, but then I don't understand what generating an event actually does. Assuming secular equilibrium does it simulate a decay from each isotope in the chain? If so, why don't we see hits years in the past?

[MarcFBergevin]

I may let someone more in tune with the SNO+ version of rat answer the philosophy behind the negative time. My understanding was that this was relevant for the shorter decay ones, a look back on a possible delayed-prompt scheme, but my explanation would simply be pure guess.

So, for Th-232, Rn-222, and U-238 rat-pac will attempt to simulate the entire chain. That will include time stamps and the likes. However, if you request a specific decay in the chain, say Bi-214, then rat-pac will only simulate that single event.

We have been simulating the single element and assuming secular equilibrium, since, like you say, it would not be very relevant to have events days or years in the pass.