What is destroying sparkplugs in my engine?

[Tsavo]

Problem is happening over time, thus ruling out anything smacking into the plug, which would cause instant (and sometimes spectacular) results.

Problem is happening over time, something making the plug too hot. So, either the wrong heat range of plug and/or cylinder running lean. I'm not familiar with Mazda engines, so I don't know what their sensor set-up is like, but something is causing that cylinder to run lean (if the problem isn't the wrong plug). Be it fuel injector, or a sensor causing leading the computer to run that cylinder lean. It's not the coil.

Either way, take it to a good Mazda dealership or GOOD shop and have em sort it. Also, ask on a Mazda specific forum.

 

[leper84]

First off I would put the colder plugs in. Ignore whatever you pulled out first, everywhere says that motor uses the tr6, before anything else put the right plugs in. Even one step too hot a plug and really bad, weird stuff can happen. I would do that before you try anything else. I would also really, really try to get a look inside that cylinder to see what three or four ground straps might have done. Cant vouch for it but I think harbor freight sells a digital scope for sorta cheap.

 

[boomerang]

I would say that the seat for that plug is damaged and it cannot transfer heat to the cylinder head. Look carefully all the way around the mating surface of the plug you removed to make sure it is making contact all the way around. The carbon buildup in that spark plug tube is not normal and leads me to believe the plug is not seating properly. Combustion gases are leaking past the plug and depositing themselves on the spark plug tube. The side electrode is probably being vaporized away over time because it is getting so hot. I highly doubt the weld that holds it to the shell of the plug is consistently failing across several plug changes (and in the same cylinder every time, no less) and it is actually "falling off".

The plug indicates a lean condition but it is actually running way too hot because it can't transfer the heat away.

There is an issue here with nomenclature. You're calling the spark plug tube a cylinder. What you're calling the tip is actually the side electrode. I'm not trying to be a meanie, but it would be helpful if you could use the correct terms from this point forward now that you know them. It's confusing everybody. We understand that you are learning the ropes. We were all there once.

Here is a picture of the seat area of a tapered seat (which is what you have) spark plug. The pencil is pointing to the seat which is the angled surface.

You can cause a spark plug to overheat by not tightening it properly but you are only seeing it in one out of four plugs and it is consistently the same cylinder. So I don't feel that's the issue here. There are tools that can be used to recut that seat. Based on your level of experience, I would suggest you not do it yourself. It might necessitate some further disassembly to remove the spark plug tube to get the job done.

 

[Tsavo]

^ Would be kind of funny if he wasn't using the spark plug crush gasket. That wouldn't make the plug run hot, but it would explain carbon in the plug tube.

 

[LTC8K6]

That insulator discoloration looks worse than just a corona stain. I think it's getting way too hot and I would be wondering why.

Then again, it's run a long time that way...

 

[boomerang]

^ Would be kind of funny if he wasn't using the spark plug crush gasket. That wouldn't make the plug run hot, but it would explain carbon in the plug tube.

There is no gasket, it's a tapered seat plug. He's shown us pictures of it in both this thread and an older thread. And if it were a crush gasket plug, why would he consistently remove just one over the course of several plug changes and take pictures of a different plug with the side electrode missing?

 

[LTC8K6]

So this time you know the ground electrode was gone fairly quickly. Maybe put new plugs in and run the engine for a bit, then pull the plugs and examine them? Look to see if that one electrode is either bent or already gone?

If it looks okay, then it's probably heat and not contact that is damaging the electrode.

 

[BUTCH1]

An engine with that many miles and producing that kind of problem needs thorough inspection and testing. The sensors, fuel injectors, cylinder compression, emission levels, etc. should be checked/gauged.

There is a possibility of bad fuel injector or air leak resulting in hot spot/cylinder over temp (note the almost too clean porcelain insulator). (If there is one bad injector, can the others be far behind?)

Addendum

Carbon build up within a cylinder has been know to result in misfire and/or detonation.

Misfires should set a CEL code, my GM had a bad wire that made the car stumble on WOT, set a misfire code, changed wires, all id fine now..

 

[shabby]

how do i clean the carbon build up? just stick a wet paper towel down the cylinder then stick a dry paper towel?

Go to a mechanic and and cough up the money to diagnose and fix the problem, because this is obviously out of your league.... paper towel? Come on...

 

[JEDI]

Misfires should set a CEL code, my GM had a bad wire that made the car stumble on WOT, set a misfire code, changed wires, all id fine now..

yeah, i'm going to change the coil pack for cyl #4.

coil pack = ignition coil?

 

[LTC8K6]

Misfires should set a CEL code, my GM had a bad wire that made the car stumble on WOT, set a misfire code, changed wires, all id fine now..

A misfire was recorded.

odb II scanner says PO304: Cylinder #4 Misfire.

Not surprising with the sparkplug looking like that.

I would swap coil packs and see if the misfire moves with the pack.

I wouldn't blame the coil pack first in this case.

 

[boomerang]

yeah, i'm going to change the coil pack for cyl #4.

coil pack = ignition coil?

Change them all - after you figure out what they're called. And change the generator, the battery and everything else electrical under the hood. Change the headlights too.

I've found this to be a very interesting thread. It's come down to you having a misfire being your problem. This is laughingly absurd. The misfire is the result of the spark having to jump a gap 100 times bigger that normal. It's not the cause of the problem the misfire is a consequence of the problem.

You had your mind made up what the problem was in your first post. You've gotten your confirmation and now you're off to throw money at it. Best of luck.

 

[onza]

Seek professional help.

 

[LTC8K6]

Yeah, the coil pack is not removing the spark plug electrode...

It's either burning off or breaking off...

Since it happened in 500 miles, breaking off would seem to be the leading candidate...

Severe misfiring or knocking could cause the overheating, but one would tend to notice that while driving.

 

[Ferzerp]

Like I said, sounds like that cylinder isn't fueling properly leading to a very lean situation and detonation.

 

[LTC8K6]

Like I said, sounds like that cylinder isn't fueling properly leading to a very lean situation and detonation.

A very lean cylinder should trigger a check engine light as well, I'd think.

I'm going to assume the OP was not ignoring a light and the misfire was the first thing that triggered the light.

Assuming things is problematic though...

 

[Ferzerp]

A very lean cylinder should trigger a check engine light as well, I'd think.

true


vvvvvvvvvvvvvvvv also true and maybe truer.

 

[SparkyJJO]

A very lean cylinder should trigger a check engine light as well, I'd think.

I'm going to assume the OP was not ignoring a light and the misfire was the first thing that triggered the light.

Assuming things is problematic though...

Not always.

There is only one O2 sensor monitoring those 4 cylinders. If it sees the exhaust go a bit lean, it calls for more fuel for ALL cylinders to counter it. As long as the extra fuel added doesn't go above a certain threshold to bring the exhaust out of the lean state, no check engine light is set. However, that one cylinder could still be burning lean while the others are now burning a little rich.

If we had a sensor per cylinder you could get much more accurate cylinder fueling and problem detection, but that would make it far more expensive and complex for very little gain over all.

Definitely problems with that cylinder. Temperatures (that plug ceramic is too clean), the plug isn't even sealing (carbon in the plug well), and of course the broken/missing electrode.

OP, sometimes car manufacturers update the plug specs after the car was released for various reasons. If everything is now saying to use the colder plug, I'd use them. Especially since you're having issues with that cylinder which is at least running too hot.

 

[boomerang]

There is only one O2 sensor monitoring those 4 cylinders. If it sees the exhaust go a bit lean, it calls for more fuel for ALL cylinders to counter it. As long as the extra fuel added doesn't go above a certain threshold to bring the exhaust out of the lean state, no check engine light is set.

Definitely problems with that cylinder. Temperatures (that plug ceramic is too clean), the plug isn't even sealing (carbon in the plug well), and of course the broken/missing electrode.

Exactly right in your first paragraph.

It's important for everyone to keep in mind that this problem has occurred over the course of many miles and several spark plug changes. The most important aspect of this problem and you picked up on it too is the carbon in the plug well. That points to the plug not seating properly and combustion gases coming past the threads.

At first I thought it was a lean condition affecting that cylinder, probably an intake manifold leak but once the carbon in the plug well was brought up, I changed directions.

I also discount the side electrode being repeatedly broken off. What would be doing that in an engine that runs well and gives good mileage? Carbon buildup on the piston crown? I'd say it's safe to say that carbon is softer than steel. The steel would win. Did the side electrode come off on the factory plug, which then embedded in the top of the piston and is breaking off the electrode on subsequent plug changes? Possibly, but I think the chances of no damage being done to the center electrode over the course of several plug changes is infinitely small.

Inserting a new plug and running the engine a short time (one revolution would do it) and then pulling the plug for inspection is not a bad idea. Easy and very quick. I might do it to rule out some form of mechanical interference, but I keep coming back to the carbon in the plug well. It can't be ignored.

Throwing parts at a problem is an expensive way to troubleshoot something. If your only recourse is to do that, it would be more cost-effective to take it to somebody with the knowledge and know how to fix it.

 

[LTC8K6]

IIRC, this one occurred after only 500 miles on new plugs.

My understanding is that new plugs were installed. At 500 miles the code came up. Plugs were pulled. Electrode is already missing.

We don't know when the previous one occurred in the life of the plug, iirc. We just know the electrode was missing when he pulled that plug...

I think the OP is thinking that a bad coil is causing misfiring and damaging the plug. To that end, I suggested swapping coil packs first before buying a new coil pack.

 

[LTC8K6]

I'm looking at the earlier spark plug pics and at the new ones and I'm going to jump to the conclusion that the side electrode appears to be broken off and not eroded away.

http://imgur.com/PXEGM.jpg

 

[LTC8K6]

http://forum.miata.net/vb/showthread.php?t=426158

This guy mentions at the end that the plug was loose...

 

[Ferzerp]

I'm looking at the earlier spark plug pics and at the new ones and I'm going to jump to the conclusion that the side electrode appears to be broken off and not eroded away.

http://imgur.com/PXEGM.jpg

Sharp edges like that would indicate this to me as well.

 

[Meghan54]

I might do it to rule out some form of mechanical interference, but I keep coming back to the carbon in the plug well. It can't be ignored.

Originally I thought incorrect heat range might be doing it, but with all the carbon, gotta be a few things conspiring to burn off the electrode.

Bet it's something like: wrong heat range plug (hotter) combined with an aluminum head that's had that cylinder's plug tightened too much several plugs ago, overtightening distorted the plug's seat in the head, now feels tight but isn't and leaks. Or a piece of crap was stuck down there on the seat and a plug put in without cleaning, scoring the seat and making a seal impossible now.

Dunno, you're correct, the root cause has to be heat, not mechanical.

 

[LTC8K6]

Would the heat completely remove the side electrode at the weld/braze in 500 miles though?