Failure analysis

mosk

New Member
Well, after almost a month of happy motoring, it looks like I'll be pulling my motor to re-ring it. :( I think I have a grasp of what went wrong and why, but I want to run it by the folks here to make sure I have the right understanding of the cause and it's effects.

First of all, I should mention that I was running in closed loop mode, with a large potential fuel enrichment number. This was done at my tuner's suggestion, and "seemed like a good idea at the time", even though we had it tuned to the point where, in practice, very little O2 correction was needed.

Secondly, I need to mention that I had my WBO2 sensor mounted in the mid-pipe, in front of the cat. My exhaust has four components: downpipe, mid-pipe, cat, and cat-back. The cat has V-band clamps attaching it to the pipes on either side, but the downpipe-to-mid-pipe connection is with a more traditional three bolt header flange.

Here's what happened: several of the downpipe-to-mid-pipe bolts loosened on me, causing the connection to separate by a half inch or so. Since the O2 sensor was located in the mid-pipe, on the wrong side of this big ol' exhaust leak, it was suddenly reading leaner than it should have (or so I'm theorizing). The ECU compensated for this by kicking in a lot more fuel. Since this happened at freeway speeds and positive boost, things got ugly quickly: the motor began to ping quite a bit. I had to nurse it along a little ways before I could pull over, and by that time I believe the damage was done.

After having the truck towed home, I pulled the plugs. Detonation had almost closed the gap of the plug in the #1 cylinder and cracked/split its insulator. I changed the plugs (the other plugs looked fine), bolted the exhaust back up, and tried starting the motor; it ran very rough, as if at least one cylinder wasn't firing (actual A/F was around 10.3:1 while it was running in this state, while the target is more like 13:1). I did a compression test, and my compression in the #1 & #2 cylinders was 90 psi, and compression in the #3 & #4 cylinders was 62(!) psi. Yuck! I'll be doing a leakdown test before I pull the motor, but I am pretty sure I am looking at broken rings in each cylinder.

Does this sound right? I have an o-ringed deck, and really don't think I simply took out the headgasket (somethign I have experienced). I haven't had this happen before, so I'm trying to learn as much as I can from this -- everything from where NOT to mount the O2 sensor to how much fuel correction NOT to dial in ;) But I am most curious about the connection between excess fuel, detonation, and broken rings.

Thanks for whatever additional insight anyone can share.

Jeff
 
Jeff,

If you overfuel the engine to an extreme, the fuel will actually run past the top ring and burn between the first and second ring. Been there done that... but it was with a carb. :D

that's about all I can offer on that... sorry to hear the bad news. I hate to hear about broken stuff. I guess that in the future for anyone that reads this, maybe keeping the positive correction limits a little lower than 25% would be a good idea, given that your VE table is well calibrated.
 
Thanks, Craig. Maybe I did get fuel between the rings, where it just blew things apart. It all happened very quickly, in traffic with the stereo up, so I don't know exactly what happened. I scan my EGTs and coolant temps regularly, and neither rose significantly. weird...

What I don't exactly understand is the pinging *after* doing whatever damage I did. Pinging is the result of fuel pre-igniting, right? Unless the broken rings are (or maybe broken other bits??) were creating some serious hot-spots, I would have simply expected power to be way down.

I don't know...guess I'm going to have to pull it apart to see what happened.

Jeff
 
Well there's detonation and then there's pre-ignition. Pre-ignition is from a heat source, and occurs before normal spark (hence the "pre" prefix :-) Detonation as we usually ref it occurs after the mix is lit off, where a pocket(s) of unburned air/fuel charge suddenly auto-ignites from heat and pressure, not from the usual, progressive flame front burn.

You may just have a blown gasket(?) Detonation generally results in excess cylinder pressure, which could blow out the gasket. Or worse :-( Maybe you can do a leakdown test first, where listening for where the pressure is escaping may tell you the source of the low compression.

TurboTR
 
BLOWED UP??

Jeff, could be that it is the gaskets. The 4 problem cylinders are adjacent to each other. 1-3, 2-4. Before I got too far into a major teardown, I'd consider pulling a head and taking a look.

HTH,
 
Thanks, guys. I had noticed that both pairs of cylinders were low, and with the same readings. I had ruled out a HG failure because each of the previous times I've had a HG go on this engine it has always involved coolant leaks, but yeah, it's possible that the HG let go between 1 & 2 and 3 & 4. I still need to do a LD test and see what sort of info I can glean from that. Thanks for the suggestion; I have had some suspicions about the rings in the past, but maybe this is just a HG.

Jeff
 
And don't forget that pinging is an audible noise that could be caused by a mechanical knocking (piston on valve), it doesn't have to be pre-ignition.

I had this knock once that wouldn't go away, found out (after having all of the machine work done) that the block had a crack and the main caps were knocking against the block (sorta, I'd have to have a photo to explain better) where I had just thought it was a rod-knock sound.

Anyhow, I've heard of rich-running motors before, but never so rich that you cause substantial damage that way. Usually I associate detonation with running lean, running too rich tends to supress detonation. I suppose it's possible, but I've had my A/F pegged at 9:1 (and I know it was more like 7:1) on quite a few occasions at WOT, and no damage resulted. Sounds like a weird problem- I have a hard time believing the Fi was dumping in enough fuel to do that.

Dont forget that head gaskets have several failure modes- they can leak cylinder-to-coolant passage, cylinder to cylinder, or cylinder to atmosphere. Also, cylinder to crankcase.

And I'll second what others have already said -once you get your VE numbers close, you shouldn't need that much correction, perhaps like 12% or so would be a good correction limit. I have my limits set to 10%, except for WOT in which case I have it set for a 25% because I don't want it going lean...

-Bob Cunningham
 
After reading Bob's post I think it makes the most sense. Yes, I have broken piston rings before from putting in too much fuel, but I had nitrous on the car when it happened. I think that kinda changes the rules. I guess I'll wait to hear from you when you get the motor pulled apart. :(
 
Back
Top