I am thinking that the CMOS IC would be working by now. And the discreet parts circuit needs one extra battery while the CMOS woulld need two extra batteries.

why not design for a single 9v battery?
The TS showed us an arduino attached to the machine in an earlier post. I dont think it matters to the TS if it requires 3v 5v or 9v(?)

 

why not design for a single 9v battery?
The TS showed us an arduino attached to the machine in an earlier post. I dont think it matters to the TS if it requires 3v 5v or 9v(?)

Quiye a lot of posts back, when I first suggested using a CMOS IC to generate start and stop pulses. it was knocked down because the timer ran on a single AA cell, and so it was proposed to use a rather creative discreet component circuit, which is quite a brilliant design.
But then the timer changed to one with a single button for start and stop, at least I think it did. That change makes ot a bit more complicated to do with CMOS. In fact I am still thinking about that one.

 

Right, was a bit of a PITA this one but its alive!

Slight mods to circuit.... dumped all the diodes except D4 - they were part of the problem rather than the solution...
Works ok down to 1.4v, triggering a little wayward below that, will probably need lower values of R13/R14, some experimentation may be needed. I think we've gone as far as we can on a 1.5v solution,



1 second switch closure simulation...



1 second switch closure, for real... red = switch in, blue = collector of Q2 (Vout inverted, I didn't breadboard Q3)




10 sec switch closure, for real...



The breadboard...

 

Fantastic, it works now consistently 100% of the time.

Can't thank you enough for all your work, looking forward to keeping an accurate time of my espresso shots!

 

You're welcome. Interesting challenge, and yes, there are easier solutions. As the old saying goes, "if I wanted to go there, I wouldn't start from here!" but we did


Enjoy!