Misfire

Author
Discussion

PetrolHeadPete

Original Poster:

743 posts

188 months

Sunday 17th June 2018
quotequote all
Hi All, need some help/advice
I have a misfire that has developed quite suddenly over the past week. Only significant thing was a short period of high temp running and idling.

The car starts and idles well at all temps.
From cold there is a subtle flat spot at part throttle...nothing audible you just kind of know from your right foot that there should be more.
Once warmed up, there seems to be a trigger point (temp? could be, could also be faster running, not sure) where things get ugly.
If you accelerate really gently then not that noticeable. If you try and accelerate normally you hit a wall, nasty loud rumbling shaky misfire. If you floor it you can sort of drive through pretty well...almost like a magic point where throttle opening and fuel delivery is enough to just swamp the problem and off it goes. On the Mway once cruising in 5th, it pull fine but you can still hear a misfire rumble. I have checked by going through gears etc that it has no obvious link to TPS position or engine revs as-such...it can be horrible at 1500rpm in 4th, 3500rpm in 4th and 3500rpm in 2nd etc etc.

Here's what I know/have done:
1. Changed HT leads...nope
2. Changed coil pack...nope
3. Checked plugs...all look rather lean...not severe, just not much light brown to see. All 6 very similar even after idle
4. Compression test shows 220ish psi on all cylinders
5. If I use a hand over exhaust tip, the left one (rear bank) has a definite random strong puff-pop i.e. misfire. Right hand has nothing and is definitely quieter than the left side overall
6. Smelling the fumes, left is almost nothing, right has that very mild but slightly sweet burnt petrol smell that I'd expect
7. If I pop off each injector plug one at a time, each one has an effect on the running, so all are doing something
8. If I pop off any from 4,5,6 the left pipe makes a strong regular puff-pop misfire in time with engine cycle and interestingly then doesn't seem to have a random one superimposed on top...may just be masking it I suppose, but odd
9. I have checked the injector wiring and voltages with DMM, all look fine (thinking of the corroded wiring loom found on this forum some years back)
10. Diagnostics seem ok-ish but adaptives do look a bit low, heading to -10% ish and did notice rear bank heading down to -20% at one point. Seems at odds with plug colours...if it was correcting a rich mixture then surely it would tend to look rich on the plugs...dunno, perhaps not
11. Diagnostics show both lambdas working and indeed, if I unplug one injector briefly on a bank, that bank's adaptive's change
12. Idle is "fine" all through this (obviously there is a misfire as described) and there is no popping in the throttle bodies etc, so dont think there's an air leak

So now I'm wondering if I could have an injector problem, or even overall fuel delivery problem ? For sure it feels like fueling to me. The fact that I can press the gas peddle hard down and blast through the problem might be significant.

That's war-and-peace complete. Any ideas folks? I'm into unknown territory.

Chim450

1,452 posts

260 months

Sunday 17th June 2018
quotequote all
I’m no expert, but from what you’ve described it does sound like a fuelling problem on the rear bank. Lamda or throttle pots?

PetrolHeadPete

Original Poster:

743 posts

188 months

Sunday 17th June 2018
quotequote all
Lambda is possible.
Tps is shared between both banks on mine smile so think we can rule that out.

Basil Brush

5,061 posts

262 months

Sunday 17th June 2018
quotequote all
I had a similar misfire that I was convinced was ignition related but turned out to be a closed up exhaust valve clearance.

PetrolHeadPete

Original Poster:

743 posts

188 months

Monday 18th June 2018
quotequote all
Do you mean exhaust valve staying open?

Basil Brush

5,061 posts

262 months

Monday 18th June 2018
quotequote all
The clearance between the valve and follower had closed up so that, as the engine warmed up, it was not sealing properly. Cold it was ok as there was just enough gap until everything expanded.

PetrolHeadPete

Original Poster:

743 posts

188 months

Monday 18th June 2018
quotequote all
OK, that's what I thought you meant. I did the compression test at quite hot engine temp. I think that would have shown something. But for sure valve clearance test is looking closer and closer.
Just been playing and using old diags to watch the lambdas...both seem normal but there's deffo a kind of cyclic effect going on when it riches up, both lambdas go high, adaptives go more -ve, then both lambdas ping pong for a few secs then they both settle low for a few secs, then they ping pong again for a while, it riches up again etc etc. Engine rpm follows the same cyclic effect changing by perhaps 50rpm (audiable)...which could be the cause or the effect of course.
Also now got new plugs. Same.

Do injectors ever fail ??? Really ? I run on premium unleaded btw.

PetrolHeadPete

Original Poster:

743 posts

188 months

Wednesday 20th June 2018
quotequote all
Quick update
Hot valve clearances all fine. Everything looks fine at the top of engine.
Lambda switch next
Some replacement injectors coming soon
Has anyone ever had a corrupted base map in the mbe?


Oneball

842 posts

86 months

Wednesday 20th June 2018
quotequote all
Have you considered a blocked cat?

I know you said you don’t think it’s a vacuum leak but may be worth spraying some easy start around the manifold.

Reading your description it seems it has to be something that only effects the one bank.

PetrolHeadPete

Original Poster:

743 posts

188 months

Wednesday 20th June 2018
quotequote all
Good shout, hadn't thought of that.
Spanners to the ready !

twinreal

300 posts

154 months

Friday 22nd June 2018
quotequote all
Hi Pete,

i'm having these misfires for a longer time now. sometimes more, sometimes less.
i did almost everything of your above list except injectors.
So i'm very interested in your experiences.

Two years ago or so i put in andys throttle bodies with bearing mod.Therefore i could take a direct look at the inlet valves. They were partially very coked. i think this is affecting the misfiring problem and shunting at driving under partial load. i'm consideríng ice blasting or something else to clean the valves.

regards,
thomas

PetrolHeadPete

Original Poster:

743 posts

188 months

Friday 22nd June 2018
quotequote all
My misfire is pretty nasty, would not want to drive on it too much. Its like a small earthquake under your right foot (the whole car vibrates) and the amount of "go" is very poor...until you press your foot down further and then off it goes with gusto. Not saying the misfire has gone when you press on, just that you suddenly get most of the acceleration back (quite specifically when you have opened the throttles pretty wide, there it is!).

Mine fault started after some hours of tinkering when the engine got really hot. Soon after that it started. Up til then it had been behaving really well!
You've got to luv 'em!

I'm pretty convinced its fueling related. The fact the rear bank adaptives are v low and kind of cycling is suspicious.

twinreal

300 posts

154 months

Friday 22nd June 2018
quotequote all
Ok, your misfire seems a bit different to mine then.
i can feel the vibrations and even hear the misfire with window open, but is not that bad.
When i put foot down a little to get it more under load, the misfire is not noticable anymore.
My adaptives and lambda are ok, the mixture seems ok, BUT: when i unplug my lambdas to test, the mixture is going to rich up to +30 and the misfires are almost gone.

Oh, and i tested the decat eprom some time ago --> much less misfire. Think it has to do something with the more static mapping and less lambda when using decat eprom.

s6boy

1,613 posts

224 months

Friday 22nd June 2018
quotequote all
Have you had a look at the fuel pressure regulator?

PetrolHeadPete

Original Poster:

743 posts

188 months

Friday 22nd June 2018
quotequote all
Not yet. But its on the list

Sagi Badger

589 posts

192 months

Friday 22nd June 2018
quotequote all
Pete,

The memory chip, forget the name of the little buggers right now, can cause all sorts of wild problems but normally these show at idle as well.
Doubt fuel pressure regulator from what you described. Cat melted and blocking or melted and blown down the exhaust is likely. Powers had a 350 in with exactly that and it was diagnosed by the uneven pressure from the tail pipes and spitting back through the throttle bodies.
You could use a borescope, you can hire one easily enough, and prod up the exhaust from the front to the cherry bomb effort.

Good luck and keep us posted.

J

PetrolHeadPete

Original Poster:

743 posts

188 months

Saturday 23rd June 2018
quotequote all
Cats look pristine...no sign of meltage or damage. Also changed the crank sensor just in case...no change.
Tried running with lambdas disconnected ....still misfires. Only slight clue after that test is that plug 6 and 5 look a tiny bit rich and the other 4 look like they just came out of the packet.
Injectors arrive end of next week.
Running out of tests now.

Sagi Badger

589 posts

192 months

Sunday 24th June 2018
quotequote all
Pete,

What makes you think this is fuel?

Reading the symptoms you list again I notice you have a problem at idle as well. I would suspect the memory chip, still can’t think what it is called and I can’t find the one I have in the “stores”. What is interesting is the period of hot running you describe as preceding the problem.

So if getting hot has something to do with it what could have changed in that hot snap. Throttle body seal? Do you have the gaskets fitted, if so check the bolts are tight as an air leak could cause what you describe and could be triggered by a heat cycle, if your throttles are just sealed then less likely but I notice the gaskets do yield after a while so a tighten up is needed. An earth? Check the engine earth is up tight and clean, I have three earths on mine. This could scuttle a lamba readiing. Check all earths while your in low ohm mode. Pressure regulator? Not smashed one apart yet, I have a few, and frankly I wonder what it really is doing here as the closed loop seems to be pretty quick to react to fuelling, but anyway I don’t exactly know what the inner workings are but diaphragm is a good guess and this is actuated by the plenum vacuum from the air box. If it were dumping fuel the injector duration would increase as the pressure would be low, maybe not enough pressure to give a spray pattern but what puzzles me is the full throttle condition you describe.

Be aware over 4 k or thereabouts and, I think, on full throtttle the lambdas are bypassed. From what I understand, prepare for those in the know to correct and please do if you do know, the adaptive map is projected by the ecu in this instance so 4 k onwards with a minus whatever adaptive will be projected or interpolated forwards so if the case it means my memory chip theory is flawed as this relies on that chip...

So my money, this time, is regulator or throttle body seal.

Keep going mate, you’ll find it

J

PetrolHeadPete

Original Poster:

743 posts

188 months

Monday 25th June 2018
quotequote all
Hi John
Really appreciate your thoughts on this...always good to get your input.
The non-volatile RAM chip is new as of around 2 years.
The fuel regulator...I dont think it does much so doubtful on that. Seems hard to fathom that it would block or restrict flow.
The throttle body seal had occurred to me too but they've been in place for about 2 years now with rtv...which has always worked for me. I dont bother with gaskets. There is plenty of vacuum on all cylinders when putting hand over each body mouth. Also in my past experience you always get pops through the bodies when there's a leak...and there is nothing. But it is "on the list"
The only reason I think its fuel is a feeling...you know when you get the throttles out of balance and at first-press of throttle you can get that kind of roughness sensation...it feels a bit like that. Its hard to know if its just one cylinder. Its almost like there is too much fuel and when you open the throttles enough you get back to approximately correct mixture and off it goes.
One other random thought was whether one of the cams has moved relative to its sprocket...i.e. cam timing all to cock...*very* unlikely but also needs to be checked.
The fact it still misfires without lambdas connected seems to rule out the lambdas...obviously wink Well, I suppose its obvious, may surprise me yet!
Gotta luv 'em

twinreal

300 posts

154 months

Monday 25th June 2018
quotequote all
The nvram is a dallas DS1230. It holds the adaptive maps and has a durability of approx. 10 years after initially connected.
When you reset the adaptives, the ECU is filling the RAM with new learned values.

Maybe it is worth to have a closer look to the bulk head connector. The little pins can loosen and slip backwards.