What on Earth could cause a spark plug to do this?!
Discussion
Hi all, hoping for some advice as this one's gone beyond my capabilities!
Basically, we've got a Nova kit car in the workshop, to which we have fitted a Subaru 2.5 flat-four from a Legacy Outback.
The engine was fitted a couple of years ago, and though the car hasn't been on the road, it's been running for at least a year, if not longer. Has been running perfectly happily up until now.
First sign of problem: Had the car booked in for MoT on a Monday, and on the Friday beforehand, it developed a misfire last thing in the day. Saturday we planned to remove plugs and clean them (as it hadn't been driven or allowed to get up to temp, so presumed plugs might be fouled). Saturday came and the car wouldn't start at all.
For the following week, we checked everything. The timing gun AND spark plug testers were registering a spark. It had fuel pressure, and flow. It had all signals needed from various sensors, and injectors were firing.
Eventually we removed a plug to visually check the spark, as we were well into clutching-at-straws territory, and had a long-shot idea that the spark may be there, but too weak (though if it's managing to spark under compression, it can't be that weak, right?
We removed the plug, and held it to earth while cranking. We were astonished to find it was sparking, but between the threaded outer section of the plug, and the earth. When holding it tight to earth, there was no spark at the tip! All four plugs, exactly the same.
Fit an old set of used plugs from another engine (not even correct grade) and it starts up fine.
Second sign of problem:
Order in a new set of plugs (thinking it's slightly too coincidental that ALL four died at the same time previously) and fit them. Car runs fine, book another MoT.
MoT morning comes, misfire appears after approx. 30mins of running, and then nothing. Dead. Pull new (now f
ked) plugs out, and bosh used ones back in (the ones that went in at the end of the first stage) and it runs.....but not long enough to get to the MoT centre. Dead.
All plugs dead, all with same problem.
It now seems that rather than the original set of plugs slowly being killed over the course of a year or two's worth of intermittent running and testing in a workshop, they probably started failing 30mins before they failed completely!
Here's a video I just took (apologies for wobbly camera - I was leaning on the engine!):
https://youtu.be/Xy-3bKSt80M
This is one of the dead plugs fitted into the HT lead of another car. As you can see, the spark is making its way straight to earth. I'm assuming the ceramic's fractured inside and it's shorting out, or something along those lines. On the Subaru engine, the spark is also weaker than this - probably about half as strong.
Now I can keep buying new spark plugs and tell the customer that he only needs to stop every 30mins to renew them, but I don't think he's going to be too happy about that, so we're now into the territory of trying to figure out what can kill spark plugs like this, but in such a short space of time. Obviously issues like excessive compression can cause spark bridging, but this engine's fine. And in 30mins?
There are no obvious deposits on the plugs.
A lot of Subaru chaps have suggested that the earths need to be A1 on a Subaru EJ engine. We've got a fat earth cable from the engine direct to the chassis, and again from the chassis to the battery. We're getting milliohms of resistance across the earth points! That said, we also had a spark on the spark tester, so I do plan on fitting another earth on the heads, but I don't want to do this until I've got some ideas of what tests I could carry out once the engine is running again, otherwise I'm just running it until it (possibly) kills a set of brand new plugs.
The car has a new coil pack fitted (I thought this was the issue, initially) but that made no difference. Otherwise we've changed nothing, as it's very easy to loose track of where you started and muddy the waters!
I had thought about an ignition module, but they're all £150+, and not in the UK. And to be honest, I can't logically figure out in my head how that could cause it.
How can it run fine for so long, and all of a sudden develop such an odd fault? Oh yeah, right, I forgot....it's a car. They're bds.
Any advice from the board?
Thanks
Basically, we've got a Nova kit car in the workshop, to which we have fitted a Subaru 2.5 flat-four from a Legacy Outback.
The engine was fitted a couple of years ago, and though the car hasn't been on the road, it's been running for at least a year, if not longer. Has been running perfectly happily up until now.
First sign of problem: Had the car booked in for MoT on a Monday, and on the Friday beforehand, it developed a misfire last thing in the day. Saturday we planned to remove plugs and clean them (as it hadn't been driven or allowed to get up to temp, so presumed plugs might be fouled). Saturday came and the car wouldn't start at all.
For the following week, we checked everything. The timing gun AND spark plug testers were registering a spark. It had fuel pressure, and flow. It had all signals needed from various sensors, and injectors were firing.
Eventually we removed a plug to visually check the spark, as we were well into clutching-at-straws territory, and had a long-shot idea that the spark may be there, but too weak (though if it's managing to spark under compression, it can't be that weak, right?
We removed the plug, and held it to earth while cranking. We were astonished to find it was sparking, but between the threaded outer section of the plug, and the earth. When holding it tight to earth, there was no spark at the tip! All four plugs, exactly the same.
Fit an old set of used plugs from another engine (not even correct grade) and it starts up fine.
Second sign of problem:
Order in a new set of plugs (thinking it's slightly too coincidental that ALL four died at the same time previously) and fit them. Car runs fine, book another MoT.
MoT morning comes, misfire appears after approx. 30mins of running, and then nothing. Dead. Pull new (now f
ked) plugs out, and bosh used ones back in (the ones that went in at the end of the first stage) and it runs.....but not long enough to get to the MoT centre. Dead.All plugs dead, all with same problem.
It now seems that rather than the original set of plugs slowly being killed over the course of a year or two's worth of intermittent running and testing in a workshop, they probably started failing 30mins before they failed completely!
Here's a video I just took (apologies for wobbly camera - I was leaning on the engine!):
https://youtu.be/Xy-3bKSt80M
This is one of the dead plugs fitted into the HT lead of another car. As you can see, the spark is making its way straight to earth. I'm assuming the ceramic's fractured inside and it's shorting out, or something along those lines. On the Subaru engine, the spark is also weaker than this - probably about half as strong.
Now I can keep buying new spark plugs and tell the customer that he only needs to stop every 30mins to renew them, but I don't think he's going to be too happy about that, so we're now into the territory of trying to figure out what can kill spark plugs like this, but in such a short space of time. Obviously issues like excessive compression can cause spark bridging, but this engine's fine. And in 30mins?
There are no obvious deposits on the plugs.
A lot of Subaru chaps have suggested that the earths need to be A1 on a Subaru EJ engine. We've got a fat earth cable from the engine direct to the chassis, and again from the chassis to the battery. We're getting milliohms of resistance across the earth points! That said, we also had a spark on the spark tester, so I do plan on fitting another earth on the heads, but I don't want to do this until I've got some ideas of what tests I could carry out once the engine is running again, otherwise I'm just running it until it (possibly) kills a set of brand new plugs.
The car has a new coil pack fitted (I thought this was the issue, initially) but that made no difference. Otherwise we've changed nothing, as it's very easy to loose track of where you started and muddy the waters!
I had thought about an ignition module, but they're all £150+, and not in the UK. And to be honest, I can't logically figure out in my head how that could cause it.
How can it run fine for so long, and all of a sudden develop such an odd fault? Oh yeah, right, I forgot....it's a car. They're b
ds.Any advice from the board?
Thanks
I'm glad you don't get to work on my car. Holding a plug in mid air like that with the engine running means the HT voltage has nowhere to earth to and it will then start to break down the HT leads or any other part of the HT ignition system. Chances are you've already damaged something.
Is it possible that the plugs are being damaged mechanically? If the ceramic is being cracked by something you or the engine are doing to it, that could cause a short to ground.
Also, have you confirmed that there is still a current path from the terminal to the center electrode in the failed plugs? If that's failing for some reason, the spark might be forcing its way through another route.
Also, have you confirmed that there is still a current path from the terminal to the center electrode in the failed plugs? If that's failing for some reason, the spark might be forcing its way through another route.
kev b said:
The plugs are fouled, current is finding it easier to run from the centre electrode, over the fouled insulator and onto the outside of the plug rather than jumping the gap to the earth electrode.
They're really not that fouled though. They might appear it in the video, but up close they're not. A very thin film perhaps.Mignon said:
I'm glad you don't get to work on my car. Holding a plug in mid air like that with the engine running means the HT voltage has nowhere to earth to and it will then start to break down the HT leads or any other part of the HT ignition system. Chances are you've already damaged something.
Calm down, I'm testing that plug on my own car. It's fine though, I expected at least one response of that nature from somebody in this particular forum, just wasn't sure who it'd be.GreenV8S said:
Is it possible that the plugs are being damaged mechanically? If the ceramic is being cracked by something you or the engine are doing to it, that could cause a short to ground.
Also, have you confirmed that there is still a current path from the terminal to the center electrode in the failed plugs? If that's failing for some reason, the spark might be forcing its way through another route.
They're being torqued up correctly, and the access to them is good in this car. There's no continuity through the plug after they've 'died', though.Also, have you confirmed that there is still a current path from the terminal to the center electrode in the failed plugs? If that's failing for some reason, the spark might be forcing its way through another route.
Mr2Mike said:
They are absolutely black with carbon deposits. Suggest you check for overfueling and definitely keep the spark plug grounded during testing unless you enjoy forking out for new coil packs and potentially new ECUs.
They're not that black though, it may just be poor lighting on the video. As mentioned before (and in the video) that plug is being tested on another car (mine) and I'm fine with the risks 
I will look more into the fuelling aspect though, and inspect the plugs a bit closer. One thing I forgot to mention was that there was a trace of white powder at the base of the ceramic, which I believe (and I could be wrong) indicates a blown seal inside? Could a blown resistor cause that I wonder? And what could blow all the resistors on all the plugs?!
Kitchski said:
I forgot to mention was that there was a trace of white powder at the base of the ceramic, which I believe (and I could be wrong) indicates a blown seal inside? Could a blown resistor cause that I wonder? And what could blow all the resistors on all the plugs?!
I was just wondering whether they were resistive plugs. Have you compared the resistance of the faulty plugs against that of new plugs? I remember reading about people finding there was a low voltage resistance with some Champion plugs although they passed on a high voltage tester - the theory was this was caused by a dry joint / air gap within the plug and some erratic running was attributed to this. Anything unusual about this ignition system which could cause the internal resistance to behave differently? I understand some dragster engines run very high ignition currents and I imagine that might not go down to well with resistive plugs.The plugs are fked ! just replace them, with NEW plugs of the CORRECT type for the engine.
And waving a plug about like that is not how you test, the plug body needs to be properly earthed, as indeed then does the engine block/heads, same as any engine.
It's apparent from your waving the plug about the place there is sufficient spark energy to fire sparks all over the place at very large gaps, so it's quite clear there is no issue with the coil pack or triggering setup.
ked ! just replace them, with NEW plugs of the CORRECT type for the engine.And waving a plug about like that is not how you test, the plug body needs to be properly earthed, as indeed then does the engine block/heads, same as any engine.
It's apparent from your waving the plug about the place there is sufficient spark energy to fire sparks all over the place at very large gaps, so it's quite clear there is no issue with the coil pack or triggering setup.
GreenV8S said:
I was just wondering whether they were resistive plugs. Have you compared the resistance of the faulty plugs against that of new plugs? I remember reading about people finding there was a low voltage resistance with some Champion plugs although they passed on a high voltage tester - the theory was this was caused by a dry joint / air gap within the plug and some erratic running was attributed to this. Anything unusual about this ignition system which could cause the internal resistance to behave differently? I understand some dragster engines run very high ignition currents and I imagine that might not go down to well with resistive plugs.
I haven't compared the resistances with new plugs, no. I do have a set on the way though, so will try that. A multimeter won't buzz out between the cap and the electrode, so they're sufficiently borked that it's broken the connection. But worth measuring the resistance on a new one, yes. Gives me a line in the sand, if nothing else!The plugs used are the correct ones for the engine, though. They're an NGK platinum type, though by all accouts the only reason people use platinum is because of the increase service intervals, as access to the plugs when the engine is residing in the normal Subaru engine bay is really poor.
stevieturbo said:
The plugs are fked ! just replace them, with NEW plugs of the CORRECT type for the engine.
And waving a plug about like that is not how you test, the plug body needs to be properly earthed, as indeed then does the engine block/heads, same as any engine.
It's apparent from your waving the plug about the place there is sufficient spark energy to fire sparks all over the place at very large gaps, so it's quite clear there is no issue with the coil pack or triggering setup.
I'm guessing you've scan-read everything (I'm guilty of doing that too!)ked ! just replace them, with NEW plugs of the CORRECT type for the engine.And waving a plug about like that is not how you test, the plug body needs to be properly earthed, as indeed then does the engine block/heads, same as any engine.
It's apparent from your waving the plug about the place there is sufficient spark energy to fire sparks all over the place at very large gaps, so it's quite clear there is no issue with the coil pack or triggering setup.
The original plugs fitted WERE the correct type. Likewise the new set that went in before the second MoT attempt. The only set that weren't were a set taken from an old AX GT engine on the floor (that I have run) and tested in the Subaru engine purely to prove a theory. Once the theory was proven, they were removed and the proper set put in. Those died, and then the old AX ones were refitted in a vain attempt to not muck the MoT station around for a second time, but alas we could not limp it through as it died again.
As for the waving around, once again that is not on the car that has the issue. That's on one of my cars (which I know inside out) and was purely being used as an apparatus to test the spark plug (ironic that it's got such a good spark, when some of the comments on here are slightly derogatory in tone - it's almost like I know how to look after it!) I appreciate the risks to the ignition system on that car, the ECU, my health, the ozone layer and all the baby chickens, bless 'em.
I concede it's possibly bad practice to do what I did, however. Like I say though, it's my car - I'll kill it if I want to
Right, let's cut through a bit of the hot air in all this. The video you posted shows the plug not earthed for most of the time and waving about in fresh air which proves nothing. When you do finally touch it to the dirty bracket it still sparks from the plug body to the bracket meaning it didn't have a good solid earth even then. You need a clean (emery or a file) bit of well earthed metal to touch the plug body to if you want to even start proving what's wrong.
However as this test was on a different car you say it still doesn't go very far towards what is wrong with the Subaru engine. Get a known good plug and test it like you did before but earthed on freshly cleaned metal. If it sparks properly between the two electrodes then retest the subaru plugs. If they then don't spark properly then ok there's a fault with them. Until you do that we aren't even sure those plugs have anything wrong with them.
However as this test was on a different car you say it still doesn't go very far towards what is wrong with the Subaru engine. Get a known good plug and test it like you did before but earthed on freshly cleaned metal. If it sparks properly between the two electrodes then retest the subaru plugs. If they then don't spark properly then ok there's a fault with them. Until you do that we aren't even sure those plugs have anything wrong with them.
Mignon said:
Right, let's cut through a bit of the hot air in all this. The video you posted shows the plug not earthed for most of the time and waving about in fresh air which proves nothing. When you do finally touch it to the dirty bracket it still sparks from the plug body to the bracket meaning it didn't have a good solid earth even then. You need a clean (emery or a file) bit of well earthed metal to touch the plug body to if you want to even start proving what's wrong.
However as this test was on a different car you say it still doesn't go very far towards what is wrong with the Subaru engine. Get a known good plug and test it like you did before but earthed on freshly cleaned metal. If it sparks properly between the two electrodes then retest the subaru plugs. If they then don't spark properly then ok there's a fault with them. Until you do that we aren't even sure those plugs have anything wrong with them.
And as he mentions, the test video isnt even on the Subaru engine !However as this test was on a different car you say it still doesn't go very far towards what is wrong with the Subaru engine. Get a known good plug and test it like you did before but earthed on freshly cleaned metal. If it sparks properly between the two electrodes then retest the subaru plugs. If they then don't spark properly then ok there's a fault with them. Until you do that we aren't even sure those plugs have anything wrong with them.
At a risk of sounding like a broken record, that spark plug was only fitted into the engine in my car for a brief moment, purely to see if they worked on a car that I know was working perfectly. It was an experiment, nothing else. I just happened to think "I know, I'll film it to show people what I mean, as I was worried I wouldn't be able to explain the problem in a manner that 'stuck'. Seems I was right!
That issue is the same when fitted to the Subaru engine. I hear the comments regards earthing the plug fully, but at the end of the day we managed to prove the theory back on the first day even with a used plug from different engine: When held tightly against an earth, there was a strong spark at the electrode, exactly as it should be. This was the same when the second batch of new plugs arrived. The fact I initially stuck it in another car to confirm the failure just shows we're on the right path. The plug HAS failed, and they continually FAIL within the space of 30mins running. I'm stating that as a fact, and with respect if that's not good enough then feel free not to add any more input to the thread. I'm trying to figure out why a spark plug might be failing, and within minutes of posting the thread, I get comments from people like "I'm glad you don't work on my car!" That's great, thanks. It's not like I flog my arse off trying to run a business, and rely on it as my family's living. As I said, I know what this forum can be like - I should have been prepared for something. But if it makes you feel better, I felt st about that for the rest of the day.
My question through all this has been centred around how a spark plug could fail like this. Over rich mixture causing carbon deposits is a good theory, and as a result of this thread I'll take that more seriously, but honestly the plugs aren't that fouled. It's probably just poor light on the video. In any case, I'll fit a new set later and test it for emissions/AFR and see if anything's amiss.
Seriously though, if anyone doesn't like how I'm coming across, you don't have to reply. I'll figure it out eventually, I was just seeing if anyone already had experience of issues like this, as after all, that's all knowledge is - experience.
That issue is the same when fitted to the Subaru engine. I hear the comments regards earthing the plug fully, but at the end of the day we managed to prove the theory back on the first day even with a used plug from different engine: When held tightly against an earth, there was a strong spark at the electrode, exactly as it should be. This was the same when the second batch of new plugs arrived. The fact I initially stuck it in another car to confirm the failure just shows we're on the right path. The plug HAS failed, and they continually FAIL within the space of 30mins running. I'm stating that as a fact, and with respect if that's not good enough then feel free not to add any more input to the thread. I'm trying to figure out why a spark plug might be failing, and within minutes of posting the thread, I get comments from people like "I'm glad you don't work on my car!" That's great, thanks. It's not like I flog my arse off trying to run a business, and rely on it as my family's living. As I said, I know what this forum can be like - I should have been prepared for something. But if it makes you feel better, I felt s
t about that for the rest of the day.My question through all this has been centred around how a spark plug could fail like this. Over rich mixture causing carbon deposits is a good theory, and as a result of this thread I'll take that more seriously, but honestly the plugs aren't that fouled. It's probably just poor light on the video. In any case, I'll fit a new set later and test it for emissions/AFR and see if anything's amiss.
Seriously though, if anyone doesn't like how I'm coming across, you don't have to reply. I'll figure it out eventually, I was just seeing if anyone already had experience of issues like this, as after all, that's all knowledge is - experience.
Its a known problem in the 'classic' world that the insulation on modern spark plugs isn't fully glazed. Not a problem if you have an efficient modern injection system which doesn't allow 'flooding, but if you have an old carb model, the excess fuel can quickly soak into the ceramic insulator and (especially with modern fuel additives), contaminate and knacker the insulator.
I'd suggest you may have a fuelling problem which is allowing the engine to flood.
The solutions the classic guys use are: 1, use an older brand of plug. 2,preheat the engine with old plugs before fitting new ones for the first time.3,fix the fuelling problems.
Hope this helps.
I'd suggest you may have a fuelling problem which is allowing the engine to flood.
The solutions the classic guys use are: 1, use an older brand of plug. 2,preheat the engine with old plugs before fitting new ones for the first time.3,fix the fuelling problems.
Hope this helps.
brrapp said:
Its a known problem in the 'classic' world that the insulation on modern spark plugs isn't fully glazed. Not a problem if you have an efficient modern injection system which doesn't allow 'flooding, but if you have an old carb model, the excess fuel can quickly soak into the ceramic insulator and (especially with modern fuel additives), contaminate and knacker the insulator.
I'd suggest you may have a fuelling problem which is allowing the engine to flood.
The solutions the classic guys use are: 1, use an older brand of plug. 2,preheat the engine with old plugs before fitting new ones for the first time.3,fix the fuelling problems.
Hope this helps.
Thanks for that. This is an modern(ish....mid 90's) electronic MPFI system though, so unless something's gone properly Pete Tong with the air flow meter or temp sensor et al., I don't think it can be that. In any case, we checked all the resistances and connections when it wouldn't start the first time round.I'd suggest you may have a fuelling problem which is allowing the engine to flood.
The solutions the classic guys use are: 1, use an older brand of plug. 2,preheat the engine with old plugs before fitting new ones for the first time.3,fix the fuelling problems.
Hope this helps.
It's another vote for overfuelling though, so I'll look at it later and see if I can spot anything. I don't think it's a question of the plugs not being suitable, as the original ones were fitted in 2015, and it's run absolutely fine ever since then.
So exactly what measures have you taken to find out if it is overfueling or similar ?
You give no real detail about the engine, control system, whether all aspects are there ( given it is in a swap vehicle ) or whether the engine has ever even ran correctly ?
So how is anyone even meant to attempt to diagnose without some good history on the engine ?
So what are the mixtures ? wideband or 4 gas would be fine.
What does the OEM lambda report ? fuel trims ? Do all sensor readings appear normal ? etc etc
The first basic steps you'd take when you can see the plugs are potentially rich fouled. And again...has the engine ever ran correctly ?
You give no real detail about the engine, control system, whether all aspects are there ( given it is in a swap vehicle ) or whether the engine has ever even ran correctly ?
So how is anyone even meant to attempt to diagnose without some good history on the engine ?
So what are the mixtures ? wideband or 4 gas would be fine.
What does the OEM lambda report ? fuel trims ? Do all sensor readings appear normal ? etc etc
The first basic steps you'd take when you can see the plugs are potentially rich fouled. And again...has the engine ever ran correctly ?
stevieturbo said:
So exactly what measures have you taken to find out if it is overfueling or similar ?
Kitchski post 5 said:
In any case, I'll fit a new set later and test it for emissions/AFR and see if anything's amiss.
stevieturbo said:
You give no real detail about the engine, control system, whether all aspects are there ( given it is in a swap vehicle ) or whether the engine has ever even ran correctly ?
Kitchski post 1 said:
we've got a Nova kit car in the workshop, to which we have fitted a Subaru 2.5 flat-four from a Legacy Outback.
The engine was fitted a couple of years ago, and though the car hasn't been on the road, it's been running for at least a year, if not longer. Has been running perfectly happily up until now.
The engine was fitted a couple of years ago, and though the car hasn't been on the road, it's been running for at least a year, if not longer. Has been running perfectly happily up until now.
stevieturbo said:
So how is anyone even meant to attempt to diagnose without some good history on the engine ?
I stated what car the engine was from, and that it was running fine. What else do you want to know?stevieturbo said:
So what are the mixtures ? wideband or 4 gas would be fine.
As answer further up regards my plans to look more into the mixture.stevieturbo said:
What does the OEM lambda report ? fuel trims ? Do all sensor readings appear normal ? etc etc
The only OBD it has is an OBD I system. But there are no flashing lights, if that helps. Haven't got any software to tap into it and get any readings. We have an emissions tester, and an AFR meter. We specialise in restoring chassis' and building cars up to IVA standard. We don't carry much in the way of diagnostic equipment. But in post #6 while replying to 'brraapp's suggestions, I stated we'd gone round and checked resistances and circuit continuity across all the sensors and ECU pinouts while trying to highlight the problem initially.stevieturbo said:
The first basic steps you'd take when you can see the plugs are potentially rich fouled. And again...has the engine ever ran correctly ?
Kitchski post 1 again said:
The engine was fitted a couple of years ago, and though the car hasn't been on the road, it's been running for at least a year, if not longer. Has been running perfectly happily up until now.
Here's a better shot of the plug. My thoughts are that it looks a bit too dirty for something that ran for 30mins, but at the same time no where near dirty enough to prevent it working completely.
By comparison, the plug that I pulled from another engine and put in this one (just to prove the theory) looked like it had been sprayed with Schutz! And it worked perfectly.
By comparison, the plug that I pulled from another engine and put in this one (just to prove the theory) looked like it had been sprayed with Schutz! And it worked perfectly.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff