Ah, that’s here:
It’s a bit of a hack– untested, as I said.
The changes that you’re asking for now are a little more invasive, and should be tested.
If I understand correctly, the changes (versus the standard program) that you are asking for are as follows:
(1) Change PB3 from an input to an output.
(2) Add a 3 second delay after receiving a trigger.
(3) After the 3 second delay, turn PB3 high.
(4) After PB3 goes high, add a 0.5 second delay.
(5) After the 0.5 second delay, turn PB3 low.
(6) Start the “real” delay, which is to the programmed input time delay, minus one second.
(7) When the delay time expires, turn PB3 high.
(8) After PB3 goes high, add a 0.5 second delay.
(9) After the 0.5 second delay, turn PB3 low. Then, wait for the next trigger.
Is that all correct?