So if it is a straight up .5ms delay, wouldn't it have the most advance at idle, and retard as the rpms rise?
actually, i think you have that backwards. timing is most retarded for starting and advances as the rpms rise so that combustion can be completed at tdc.
from the wikipedia article on ignition timing:
"Spark timing, relative to piston position, is based on static (initial or base) timing without mechanical advance. The distributor's centrifugal timing advance mechanism makes the spark occur sooner as engine speed increases.
Yes it would, but it probably has a delay that is fixed in degrees, not in seconds. And more modern ignitions will have a delay that varies with rpm.
doesn't really matter whether you think in degrees or time. the figure of merit is the amount of time it takes to burn a particular grade of fuel vapor. at a higher rpm you have to start earlier because because the crank rotates in a shorter period of time. any part of combustion that occurs after tdc is wasted.
Yes, but it likely would be using some of that delay just for the supervisory circuitry. My assumption is that circuitry has been added to the standard CD design that varies the delay based on rpm, and this circuit may have a additional delay of its own. So if you want to vary the advance from say, 0-20deg, you might instead vary it from 10-30deg, moving the magnet ahead another 10deg and giving you a little more time to work with. I'm only guessing there though.
i was "only guessing" when i hypothesized that some of the delay was to disable detecting of rpm until switching noise settled down, but i still think that it's a good possibility. i've seen it before in analogue to digital circuits. one thing has come of this discussion, though. it's time to order a new copy of the bosch automotive handbook from abe.com.
and i'll stick to my guns. the increased failure rate of ignition modules is likely due to their more complex designs, not component quality. it's the electronic version of the stihl flippy cap.