So if you were to look at the coil drive wire with an oscilloscope you would see +12v up until the ECU wants to create a spark, then 0v while the ECU is charging up the magnetic field in the coil, then a rapidly rising voltage up to say 400v followed by a return to +12v. The spark accompanies the rapid rise up to 400v (which would be <10 microseconds in width).
With my system attached, the pattern on the scope would look the same, but the pulse would be extended. I see the ECU start the cycle. If I want to retard this particular spark, I turn my FET on (so both ECU and my box are allowing current to flow through the coil to 0v). At some point, the ECU will release it's drive, thinking that it will create a spark. Because my FET is still turned on, it won't - instead current keeps flowing through my FET until we reach the mapped spark time, then I release my FET and we see the rapid voltage rise and a spark.
So you can see that the time for which the current flows through the coil is extended by this scheme (which might also make for bigger sparks, but wil certainly make for more heat in the coil windings).
This wouldn't work for cars with the coil drive FET built into the plug-top coil pack as these are self timing I think (the 5v pulse from the ECU says "spark now" and it doesn't matter how long the pulse lasts for).
For such cars you would need to cut the original wire, feed the ecu pulses into the box and have the box drive out new pulses to the coil drivers.
A power commander would indeed work, but will cost a lot more than it will cost me to build the above, and where's the fun in buying something when you can build it instead
Cheers,
Robin
