Well I'll put my 2 cents worth in and it's long winded. First Don I'd like to commend you on your research, it's nice to see. The only real problem I have is that your results are more to do with the ignition coils than the actual CDI units.
I have been researching ignition/s, combustion, and deflagration (
Wikipedia), for over 20 years. This thread appears to deal with all 3 subjects and I'd like to touch on them.
First combustion:
Basically it is the burning of a fuel. It's not an explosion but, rather, hopefully a controlled process of combining heat, fuel, and oxygen. Dave in that motor.com article I have to disagree when they say it is the separation of a fuel molecule because first you have to combine the oxygen with the fuel and that only happens with heat (ionization). It's only after they combined with a sufficient amount of heat (current now) that the fuel burns and breaks down. I live in Canada on the cold prairies and trust me on a very cold day neither a diesel or a gasoline engine will burn fuel if they get to cold. Oh in electronics and electrical we say voltage hurts but current burns you.
Second deflagration:
Because of the racing I do I've spent a lot of time researching this subject. It's more than just cylinder and combustion chamber design. It's also more than just the spark at the end of the sparkplug. I don't want a slow burn, I want one that's as fast as possible to build as much cylinder pressure as possible before the piston goes past 45 degrees from TDC. As a fuel/oxygen mixture burns it expands creating a pressure wave in front of it. It is this wave that compresses the air/fuel mixture even more. Because of the pressure the air/fuel molecules are closer to each other, they combine faster and burn faster, split up faster. Think of gunpowder poured on the ground as compared to being in a gun barrel. When the pressure wave goes supersonic this is when we get what we feel is an explosion. All things from simple burning of black gunpowder to a thermal nuclear device go through the deflagration process. I have found that a very strong inductive spark with multi strikes is the best to ignite the fuel mixture allowing me to take ignition timing of up to 10 degrees out of an engine. Which leads me to ignition.
Ignition and Ignition systems:
So we have basically 2 types of ignition systems here not counting thermal ignitions, inductive and CDI. CDI still uses induction just differently. Since we are looking at CDI I'll try and stay on that topic. As most people know with CDI we have to charge up a capacitor with high voltage, well somewhat high voltage of around 100 volts or higher first. We then send that through a step-up transformer to get our HV spark. Sounds simple right. Well it's not if we want it done right. This type of ignition was first invented by a Canadian Air Force pilot back in the 50's. We have different types of CDI ignition systems also such as AC-CDI and DC-CDI. AC-CDI is more like the CDI ignitions we find on our lawn mowers which, is a pulsed charged CDI. The DC-CDI has a DC to AC to DC system where it takes the DC current and converts it to stepped up AC voltage and then rectifies that back to HV DC which goes to the capacitor. Now both capacitors and inductors have what is know as 'time constants' of charge and discharge. There are 6 time constants but, because the 6th is so small of a change we ignore it and go with just 5. This brings us to 'how fast can I charge a capacitor to the energy level that I want'. There basically 3 things we are concerned with here. 1. The bigger the capacitor the longer it takes to charge it. 2. Most CDIs quit sparking at voltages below 100 volts. 3. The ESR or the internal resistance of the capacitor, the lower the better. 1 & 2 have more to do with the RPM of the engine, more RPM less time to charge the cap. 3 has more to do with spark quality because the faster the discharge the better the spark. But because of the charge/discharge time constants we may not get all the energy in or out of that cap in time. So to compensate we have to use a smaller cap but, now we are decreasing the amount of stored energy.
The primary side inductor or the primary winding of our ignition coil has it's own time constants and we need to or at least try to match the time constants of the capacitor. After that we need to take into account the XC & XL (reactive impedance) of the capacitor and coil to match our RPM. For example the old 4 wire GM HEI coil-in-a-cap circa 1975 was known as a system that couldn't go above 5,500 RPM. Well that was wrong but, it did have a output droop between 5,300 and 5,700 caused by that XL of the coil. I've seen it on my test bench that can take a GM-HEI to 15,000 RPM. So we need not just any coil but, one that has low resistance but, also high enough inductance to create a high current transfer to the secondary side and match our RPM. Not like your going to find one just anywhere. Oh I should mention, you can use a regular automotive induction coil with a CDI but, not a CDI coil with an inductive driver, you'll blow the coil, been there, done that. Because the current transfer of a CDI is so weak, compared to an inductive ignition, the secondary winding is also of low resistance and kind of high inductance.
When a CDI ignition fires it creates a very high voltage spark at the sparkplug but it is low on current. Because air and fuel have a high resistance (dielectric strength) the more we compress that air/fuel mixture the more resistance and the more voltage we need to jump the gap and start the ionization. With all ignition coils the energy stored in it once the spark has jumped the gap gets converted into current, ionization creates low resistance and the field collapses faster. Question is "how much energy is left to convert into current." Remember I said voltage hurts and current burns, I have a few old acquaintances that can attest to that. So is our capacitor big enough? Can we charge it up enough? Can it discharge fast enough? Can our coil work with the capacitor? There are a ton of calculations that can be made to get you in the ballpark but, nothing beats a good old testbench. A proper safe one with a pressure test setup.
Well some have but, remember GM's ignition lock screw up when they only wanted to spend $1 instead of $1.50. Well 1,000,000 cars times $0.50 was some ones Xmas bonus of $500,000. It's the same thing with ignitions, CDIs cost more than Tx inductive ones. It all comes down to cost and "just good enough wins out".
End of part 1 - The basics.
Ray