Well that would be an easy way to do it
Just for the sake of investigating an alternative am I right in thinking that the pulse width of the transistor should be less than the coil saturation time? I think I'd need a one shot to keep the smiths rc circuit happy.
At 60 degrees dwell / 720 at 7000 Rpms I'm figuring about 3ms is the smallest pulse width if the coil isn't fully saturated at that speed. I forget how to calculate coil saturation time.
You are way over thinking it.
The current pulse occurs at the instant the ignition transistors open. The dwell time doesnt factor much into anything. The coil current does.
If you do modify the smiths tach circuit (which makes no sense to me at all), you would want to trigger Q1 with a very short PW signal (like the transformer does).
The circuit works by having the transformer provide bias to Q1 with the current impulse. This triggers Q2 through the RC network forming a one shot flip flop (monostable?). Q2 then holds off Q1 till the RC has charged.
This works becuase the transformer becomes essentially a very low value resistor when there is no change in coil current.
Put a transistor in the mix and you will need to make the transistor on all the time except for a very brief interval (200micro seconds or so) as the coil dumps to the secondary.
You will need a pretty sophisticated circuit to keep the smiths tach otherwise happy and in the end, using it as it was designed works and is dirt simple.
The only reason I would entertain modifying the tach would be if you were using an MSD ignition and then I would gut the smiths circuitry and make my own digital one shot circuit to drive the movement.