Different channel trace start times

[christophwelling]

Currently, some modules (for example the trigger modules) assume that all channels have the same trace start time. This will not necessarily be the case, for example if we do the double trigger to catch both DnR pulses.

We should modify the modules in question to be able to handle different trace_start_times

[cg-laser]

This is a tricky question. So far our definition of trace start time is that it is the same for all channels when the signal is being recorded. The efieldToVoltageConverter sets all channels to the same trace start time. And also all the remaining module executed before the triggerSimulator don't change the trace start time. This makes sense because the data that we get from a detector does not have any trace start time information. This is part of the reconstruction process to correct for delays that are different between channels.

Also current trigger realizations (ARA, ARIANNA) don't allow for time offsets between channels when calculating the channel coincidences (although this would probably be possible as this happens within the FPGA).

I don't think allow multiple traces per channels is the right way to go (see #162 ).

I rather think if we want a trigger with different time offsets for the coincidence logic between channels, this should be an argument passed directly to the trigger module. Then it is the same as it would be implemented in hardware.

BTW: If you have really long cables, you need to make sure that the prepulse time is long enough to contain the delay from the cables. In the forward simulation of the hardwareresponse, the traces are shifted in time according to the cable delays!