I need a circuit that will accept a pulse of width x (x being 1-10ms) and output a pulse of width nx, where n is a fixed whole number in the range 2-24. Ideas?
Use the technique suggested in Wal's post but use an EPLD. You will be able to get much better resolution because you can build counters of almost any length. If it's a one off project you can use an eval. kit.
In my opinion Altera's Quartus (free download) is easier to use than Xilinx design suite if you're used to designing stuff with more traditional logic chips.
__________________
If you spend all your time looking for people and things to complain about: trust me, you will find plenty to complain about.
Yep. I got concerned about the size of the pulse when I found the meter movement, which I assumed to have a very low resistance, turned out to be 50 ohms. This meant I needed a 60 volt supply and a pulse of 1.2 amps. Even tho the pulse is narrow, I wondered what 1.2 amps would do to the movement, and I really wasn't keen about the 60 volt supply either. So my thought was to stretch the pulse so it wouldn't have to be so powerful. But now I have another idea: instead of just using one cylinder, I'll use all 8 of them. By taking the firing signal from the coil and putting 4 marks on the torque converter (since 4 holes fire each revolution), I should be able to reduce the voltage to about 8 volts and the current pulses to about 150ma. I'm rewriting my circuit paper and will post it soon. And I've finished reworking the meter and will post a pic.
Good Answers: