re: trigger timing issues

NEXT
Make New Entry, Make Followup Entry

User name R. Michaels

Log entry time 08:34:33 on November 11, 2008

Entry number 248104

This entry is a followup to: 248061

Followups:

keyword=re: trigger timing issues

I talked to Xin by phone and I *think* I understand the problems better now. Let's see ....

Originally they had a problem that the coinc. time peak was not well centered, so they
tried to fix this by delaying T3, but those involved perhaps did not realize the bad implications
for L-HRS. Please keep Vince Sulkosky involved, he knows about L-HRS issues.

I've said this before but it's worth repeating: Take the circuit from T3 down to BB, through
the coinc. circuit and trigger supervisor and back to L-HRS as L1A. Call the total time of
this circuit "T". This T is very delicate; unfortunately due to the fixed analog delays we don't
have a lot of headroom. If you delay T3 by more than ~15 nsec then "wierd" things can
happen (ask Vince).

I think Xin understands this and has a good solution: reduce T1 delay at T5 formation,
but leave T3 alone as much as possible and don't delay L1A either.

Also I agree that BB retiming is a good idea.

All triggers and all RT signals should go to TDCs (individual channels), in both hi-res and
multihit TDCs. With this multiplicity of info you can sort out what all the peaks mean.
Use the trigger latch pattern (triggers appearing on TDC channels) to decide the event type.

A copy of this log entry has been emailed to: xqian@jlab.org,moffit@jlab.org,vasulk@jlab.org,camsonne@jlab.org