Originally Posted by Joe Perez

Obviously there are a hell of a lot of Miatas running around on 1k pullups, as that was what DIY called out in their original build document. And they run fine. We're gilding the lilies here, I admit, but I like these things to be perfect.

What happens when you use a high value resistor is that it takes longer for the coil primary to charge. The igniter circuit seems to load its trigger input pretty heavily at turn-on, such that the voltage on the trigger line gets pulled down. This seems to correspond to a a decrease in the rate at which current through the primary winding builds.

Here is an example of what I'm talking about:



The yellow trace is voltage on the trigger line, and the blue trace is current through the coil primary. See how when the trigger first turns on, the voltage starts out very low and then builds up, almost as though the igniter itself is reaching a saturation point? As you use smaller value resistors (and thus increase the available drive current) the rise of this waveform is much less retarded.

Where this becomes important is that I observed a correlation between the rise time of the trigger voltage and the rise time of the primary current. It's not a linear relationship, but it does happen. As you make more current available to the trigger (and thus increase its voltage) the current in the primary builds more rapidly. In this picture, you can see that the primary current never did reach full saturation (which would be indicated by the blue trace leveling out and becoming horizontal). When trigger drive was increased, the primary was able to become saturated within the time allowed.

(Keen-eyed viewers will note that the dwell time on this capture is less than what's specified for an NA. I believe that this particular trace was actually taken back when I had my EMU, I just can't find any of the more modern stuff right now.)

Ok, now the practical upshot of all this is that what I have illustrated here (limiting trigger current too much) is roughly equivalent to decreasing your dwell time. Both have the effect of limiting the amount of charge which the coil can build prior to firing. If you have the ability to measure DC current with a scope, you can tune the dwell time of an MS built with the 1k pullups until you reach coil saturation.

Or, you can just install smaller resistors.

Originally Posted by jbelanger

I just wanted to add that turning on the ignition driver is similar to charging a capacitor

Originally Posted by jbelanger

It also means that my previous concern about the power rating of the resistor is somewhat unfounded since the actual current will reach the theoretical maximum value only at turn on and go lower from there (and would be almost null at the fully saturated point).

Originally Posted by richyvrlimited

I have a scope, but it's 1 channel and can't pause stuff easily, nor can it dump to a PC, I'll give scoping the coils a go, (as they're coils I doubt anyone else has used so I'm not gonna get the Dwell for them easily!).