Fully agree that wire enamel coating, winding tightness, expansion from heat etc is the prime cause of coil failue where the coils set on the motor to 'enjoy' all this heat.
The coils usually take between 5 and 6 amps peak current 'times' the number of times per second they must do this. The average current is low but increases as RPM goes up as well as the heating effect besides that from the moptor heat..
This is one reason why the waisted spark system was abandioned but the effects of heat vs design was not well researched what with the decision that if the coils will last an average high mileage overall, that it would not be a big issue for the conpany.

 

Here is an example of coil failure and how it was detected.
Conditions: engine skip and surging in OD light throttle crusing, no CEL lamp, no DTC code set.
Tool used is a scanner able to do live data recording while driving.
Set to record 20 frames of data when triggered by PCM detection of a fault (not a DTC) that does not set a code or CEL in the 'normal' way.
Live data history showed a large difference between Long Term Fuel Trims for both banks. ((((suspect))))
Small differences between Short Term Fuel Trims that could be missed by many.
What this indicates is a cylinder is passing un-combusted air on one bank but which cylinder?
A long drive cycle triggered on fault, recorded data and wonder of wonders set a 301 code without the CEL being turned on. Why; because this coil fault is not a 'hard' fault or failure.
Changed the coil, cleard the fault.
A look afterward at the same LTFT and STFT again, now shows they are the same for each bank; motor runs as it should.
Moral of this story, every ACCEL coil I have run in the set has failed one by one over short time intervals for even reasonable life expectations.
I carried on with this just to learn the extent and how to deal with it.
So much for "PRC" Peoples Republic of China parts.
BTW a note. Way back in about 1997 we had an ACCEL coil failure tried on a 5L.Mustang.
Anything better for coils? How would we know except over the long term of use of them.? A this point, the OEM coils are apt to be about as good as you will get and they to will fail at some point if you keep the vehichle long enough. Just a maintaince item same as alternators, belts and tires or other outside parts..

 

[quote=Bluegrass 7;8767475]Here is an example of coil failure and how it was detected.
Conditions: engine skip and surging in OD light throttle crusing, no CEL lamp, no DTC code set.


Bluegrass,
This describes my engine exactly, before AND after installing a new set of COPs. I thought it had to do with something completely unrelated but now I know what to look for but I need to get a code so I know what coil is the problem.

Also, describe to me how the HOT starts work as you suggested I try earlier in this post. I keep forgetting to do that when I get going in the mornings but I will remember and have time this weekend for sure. I'd just like to know the theory behind it.

 

Hot engine start: When the ignition switch is put into Run before cranking, the fuel pump runs for about 1 second to prime the fuel rails.
Also the PCM takes a 'look' at the value of the coolant temperature sensor as the only way it has, to know what parameters the motor needs for operation under the conditions.
This sets the fuel pulsewidth, the ignition timing and IAC opening conditions for Hot start.
At cranking, the PCM detects the crank movement from the crank sensor,and restarts the fuel pump.
The Cam movement is detected by the Cam sensor. This finds the cylinder fire order reference for ignition fire order.
And your on your way.
Included in all this is full time diagnostics of many kinds.
The coil low voltage issue we have was not well designed for in the original PCM program as you see, as far as easy code detection goes, but there are ways of looking for the issue as long as someone will take the time to do it and has some background knowledge of system operation.
The cold start works the same.
This method has been in use since at least the first 5L fuel injection from 1985.
All the operations have their parrellel to a carbed motor since the motor fuel/air needs are still the same. Just that FI does it better.