That's right; I had to check. Once triggered, it will ignore other inputs until C1 discharges.

If the input is still high when C1 discharges, the output will remain high until the input drops. It will take a sharp rising leading edge to latch the output once again.

 

Excellent, thanks Sarge, in normal operation there will be 1second between events so a trigger length of 44ms is ideal and will deal with the event. the real pulse of 10ns was far too short to detect so taking the pulse of 10ns and outputting 44ms is perfect as we would ideally like to sample the input at 120Hz so need 16ms sample pulse minimum (to apply Nyquist theorem) I think ...

(where the my brain is melting smilie?)

 

Have you discarded the non-working prototype?
If not, try connecting a 1MEG resistor across C1 and see if that causes the output to return low after a period of time.

 

I left it with the guy who mounted the components to have a look at, I'm not sure if he still has it or not. I'll see if he still has it and see if we can try adding a resistor to it.

Writing the QA procedure is going to be interesting.