Two years ago I got to do an all analog design professionally for a fast-acting high energy voltage clamp to protect delicate battery packs from misbehaving chargers - it had to be able to absorb full fast-charge current for the few milliseconds it took to shut down the charger, in the event that the charger continued to try to charge the pack after the pack protection circuit activated. You might say "that sounds inherently analog" and you would be mostly right, except that it needed to follow a specific sequence of measuring a transient, timing it to see if it exceeded a time-voltage threshold, tripping a clamping circuit for a specified length of time, pulling down on a notification line for a specified longer length of time, and then inhibiting itself from re-triggering for another specific length of time. I used an "analog state machine" or "analog sequencer" composed of two capacitors with charge and discharge resistors, two voltage dividers, one with two taps, and three comparators. In the end, as much as I loved doing it, I realized that it would have been much easier to use a programmable part, since it took a lot of tweaking to make sure that the whole thing worked within tolerances over the entire range of component values, temperature, ageing, supply voltage, etc. I didn't use a micro because a) it was considered a safety component and would be subject to a more lengthy review if it used software, b) I would have been obliged to get someone else to write the code because of the division of departments in my company, and c) I would have either had to make it remotely updateable or make a very good case why it shouldn't be - which is a silly policy given that an analog solution is certainly not updateable! Sean On Wed, Aug 24, 2016 at 9:07 PM, Denny Esterline wrote: > I say "bah humbug" to all of you. Real programmers program with resistors= .. > :-) > > I did an actual _all analog_ design last week. Current loop driven, 3 > switches to frequency multiplexer. Used a 555 and had the switches play > with the timing resistors. > > Gotta say, it felt weird to not start a project with a software flowchart= .. > > -Denny > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .