The problem was indeed an ignition misfire. The header tube had melted the 90 degree plug wire boot. The boots on these MSD wires are real thick. I had great difficulty removing the wire off of the plug because the tube was so close. The only way I would have been able to replace the entire plug wire would have been to remove the supercharger. Instead I replace the boot only with a smaller one and wrapped heat cloth around it. Poor header design.

Like Edwin mentioned, I didn't think this was a single cylinder igniton misfire based on the trims. Another thing is, it didn't feel like a typical ignition misfire. My initial thought was a fuel delivery issue which is why I started testing injection. When those tests passed, I scoped secondary. When I looked at that capture, I figured this had to be a leak to ground. I'm thinking of one of two possibilites. Either the plug ceramic is carbon tracking, or there is a leak to ground from the wire itself. A clue that really helped was the first series of strikes. Two strikes outside the combustion chamber, one strike inside, and the last one outside.

You guys are sharp. There were some really good suggestions.

I thought it was impressive to still see the rotor notch with this much time on the screen. There is some serious sampling going on there.

I have no idea why the rotor KV was higher than the plug KV. Maybe a bug got caught between the rotor tab and the cap tower?   

Thanks for playing along guys.

Lean trim is what I first thought of on your #5 Carl.  Similar except yours didn't have an upward (lean)
burn line curve before the high resistance as in this example.

http://www.members.aol.com/bgoode007/leantrim

The next thing I thought of was high resistance in the plug or the wire (snap at the top) Still not a match
to yours though since your firing line kv was fairly normal.
(These are known conditions of secondary idle and snap I did some time ago for my own satisfaction.)


The bottom idle and snap on these was from a spark outside the cylinder and is really confusing.
You would assume that a spark outside the cylinder would be low resistance and the idle capture on
this one seems to confirm this. The confusing part
of this is the snap outside the cylinder.

Edwin has seen this previously and pointed out that the snap outside the cylinder should not
have shortened the burn line and caused higher resistance than the idle wave.  I totally agree with
him. Theory tells you it shouldn’t but for some reason in this case it did.

Carl I know you and believe that what you found (spark outside the combustion chamber) was causing
your waveform high resistance on #5. Theory tells you if it’s a leak to ground it should cause low not high
resistance but for some reason in your case it did. Although maybe what I am calling high
resistance doesn't fit the definition. Maybe I should say varying resitance?


Quote:

Yes. IMO too many variables means secondary analysis doesn’t always fit the theory  

Quote:


  Unless the pcm detects a misfire and shuts down an injector
(and on a 96 GM I seriously doubt that to be the case)
I don't see an ignition misfire causing a trim that high either Edwin.
I suspect this engine had another issue totally unrelated to the
misfire driving the trim that high.

Carl,
                     No doubt about it my notes were correct. I specifically created and noted these conditions for the purpose of comparing the results and all the captures were made within about 15 minutes on the same engine.
           
         I wasn’t really interested in the firing kv at the time, more the burn lines. The reason the voltage range is wrong is because I was using Pico’s presets which the version I was using at the time scaled in volts rather than kv’s I guess maybe they prefer that in Europe? I can probably dig up the original psd’s if you really want to examine the settings I don’t know what they were other than whatever Pico had them set to.

         The small notes at the bottom of the high resistance pics are more specific as to the high resistance wire. It was actually two wires in series with a thexton spark tester in between to increase resistance. Snapping with the high resistance wire in place obviously did have the effect of using up the available energy and shortening the burn time. Why? I dunno. Doesn’t fit with theory huh? I guess this was actually two fixed environments however, one in the combustion chamber and one inside the thexton unit. Don’t see why that should have an effect though. At least no more so than a wire with high resistance.

         The middle “no problem” captures are labeled correctly . As to why would there be a longer burn within a fixed environment with snap than at idle, again, I dunno. You tell me. That’s part of my point. Maybe it’s an anomaly that’s engine specific (don’t know why it would be) but that doesn’t seem to fit with theory of this is how it “always” is either.

         I wouldn’t expect turbulence at all with a spark firing outside the cylinder. What’s there to cause it? True there’s no HC but to jump a gap without cylinder pressure you don’t need them. There certainly wasn’t any on my lower left idle capture with a thexton tester with a spark outside the cylinder. But then again there was turbulence when snapped. Of course maybe it depends on the gap it has to jump to find a ground. (But I didn’t change the gap on the thexton). Again part of my point. Too many variables I feel to name “absolute” circumstances.

         I’m not trying to argue any of this with you. I am sure that you do much more secondary analysis than I do. I’m not from Missouri but I do like to see things for myself. Maybe I just got lucky on this vehicle on this day with the controlled testing I did. Although this engine did run fine and showed perfect lambda on a 5 gas could it have had had another problem unknown to me that caused some of this? Could the spark tester gap rather than an actual wire caused some of these non-fitting results? I dunno, all I know is that I am still very skeptical that when it comes to secondary testing every engine will “fit” a set theory every time.

       Part of my problem is that I don’t believe in too many “absolutes”. My point is only that they are many variables that can affect secondary patterns. What if you have a vehicle with a high resistance wire and an over fueling problem at the same time? What is that secondary pattern gonna look like high resistance or low resistance? Again I dunno, do you? I guess it could depend on the severity of the resistance and the over fueling. I’m not trying to discourage anyone who is comfortable with secondary analysis from using it. I think most of the time the theories can and do fit and give excellent direction, just not “all” of the time  Sorry for the long winded response, just my opinions. 

Quote:


I'm sorry that you are still struggling with it Edwin. I don't quite know quite what to say other than I guess it's like the old Ripley show. I didn't just say the spark was outside the combustion chamber, I know the spark was outside of the combustion chamber. I was there.

Yes.

Carl,
LOL!!!! Actually after I asked the question, I already knew the answer, because you posted the injectors wave form and it looked perfect.  You are correct this vehicle does not disable injectors.

Edwin,

How would the PCM be able to identify which injector to disable? Is it possible to have a misfire monitor without a CKP sensor?

There are two CMP sensors which are of the optical variety. One disc has two sets of slots in it. There are 360 slots for high resolution which are equally spaced apart. This is used for timing control. The other eight slots which are of different widths tells the PCM which plug is being fired.

It seems to me that it knows what cylinder is being fired by your description. Right?

Yes, you could monitor for misfire without using the crank sensor, by ionization.

Edwin