Low rev misfire

Low rev misfire

Author
Discussion

Jhonno

Original Poster:

3,459 posts

82 months

Tuesday 30th April
quotequote all
CerbWill said:
What are the adaptive doing at idle (whatever idle speed you can currently get)?
4.2 chip tried, same symptoms.

When I left it parked 2 days ago it was idling about 1200rpm, now it is trying to idle at 1400rpm..

Adaptives are showing about 0 give or take..

gruffalo

5,937 posts

167 months

Tuesday 30th April
quotequote all
Jhonno said:
Byker28i said:
Misfire at low revs isn't coils, that tends to be over 3k revs
Well, potentially rules one thing out then!
Not really.

If a coil pack is breaking down then the cars will often feel OK at higher revs but struggle under load at lower revs as the weak spark gets blown out.

It still happens at higher revs you just can't feel or hear it because of the increased engine speed.

spitfire4v8

2,840 posts

122 months

Wednesday 1st May
quotequote all
Curing misfires on cerbs reminds me of the instructions on a brand of paint aerosol you could get in the 80s ..

1) shake can
2) shake can again
3) shake can again...

cerb version
1) replace ignition components
2) replace ignition components again
3) replace ignition components again

then move on to anything else you might think of

but steps 1,2 and 3 is always replace ignition components.

Jhonno

Original Poster:

3,459 posts

82 months

Wednesday 1st May
quotequote all
spitfire4v8 said:
Curing misfires on cerbs reminds me of the instructions on a brand of paint aerosol you could get in the 80s ..

1) shake can
2) shake can again
3) shake can again...

cerb version
1) replace ignition components
2) replace ignition components again
3) replace ignition components again

then move on to anything else you might think of

but steps 1,2 and 3 is always replace ignition components.
laugh

So true! Misfire roulette!

Byker28i

17,110 posts

158 months

Wednesday 1st May
quotequote all
gruffalo said:
Jhonno said:
Byker28i said:
Misfire at low revs isn't coils, that tends to be over 3k revs
Well, potentially rules one thing out then!
Not really.

If a coil pack is breaking down then the cars will often feel OK at higher revs but struggle under load at lower revs as the weak spark gets blown out.

It still happens at higher revs you just can't feel or hear it because of the increased engine speed.
I've had two coils go. First time was a misfire at over 3K revs, ran fine below just pootling around but wouldn't run properly when given some throttle and got over 3k revs. That was a whole coil pack breaking down under load, could provide enough spark at low revs/load.

Second time was only one spark plug was affected so the car didn't feel right but still made it home ok from a very wet Wales. One part of the coil had failed leaving me no spark to that cylinder.


Advertisement

ukkid35

4,742 posts

114 months

Sunday 5th May
quotequote all
Failure mode for coil pack that I experienced was simply to lose one cylinder - intermittent failure at first, then gone completely

Since you've been measuring flow rates, you must also have been able to view injector operation

One thing I found was that the wiring to one injector connector had frayed so that only one or two strands remained in place - this was obviously affecting injector performance

Might be worth checking wiring if you are running out of other checks to make

Jhonno

Original Poster:

3,459 posts

82 months

Monday 6th May
quotequote all
Having spent some more time with the throttle setup today, whilst it was set evenly at idle, as the throttles opened the even bank went lean. More time tweaking and evening up this initial opening point has all but eliminated the miss/stutter..

ukkid35

4,742 posts

114 months

Monday 6th May
quotequote all
Jhonno said:
Having spent some more time with the throttle setup today, whilst it was set evenly at idle, as the throttles opened the even bank went lean. More time tweaking and evening up this initial opening point has all but eliminated the miss/stutter..
Ideally the throttle cable would be attached to the link bar, rather than the even bank

That way the forces/play on each bank should be equal

Jhonno

Original Poster:

3,459 posts

82 months

Tuesday 7th May
quotequote all
Well, despite being much improved the symptoms keep making cheeky reappearances.. Almost as if something keeps changing as I have it set right, drive it for a day and it comes back when I next drive it!

Thom

1,673 posts

188 months

Tuesday 7th May
quotequote all
Swap throttle pots to see if the problem moves from one bank to another?

Jhonno

Original Poster:

3,459 posts

82 months

Wednesday 8th May
quotequote all
Tried unplugging a TPS (odd bank) as I believe it should run on 1 and it wouldn't run..

I had it balanced and idling strong at 1k, turned it off, took it out later and it started hunting at idle, but holding (ie not dying), stutter is below 2.5k, although seems better sometimes than others. Adaptives are generally doing roughly the same, but occasionally one will show rich the other lean.

I don't want to just throw money at it, but I need it right before the 28th..

The hunting seems to indicate lambdas? But at £150 a pair, I don't want to do it for fun, plus mine are relatively new.

Jhonno

Original Poster:

3,459 posts

82 months

Sunday 12th May
quotequote all
Right..

Took it for a longer run and part way it started to really lack power and miss. So I reset the adaptives to see if it made a difference and boom. Ran perfectly again, pulsing idle became smooth etc. Lambdas are the correct way round. So I am thinking lambdas need replacing? Even though they seem to read as they should? The do flick back and forward. Generally together, although sometimes one will go lean, the other rich.

ukkid35

4,742 posts

114 months

Sunday 12th May
quotequote all
Jhonno said:
Right..

Took it for a longer run and part way it started to really lack power and miss. So I reset the adaptives to see if it made a difference and boom. Ran perfectly again, pulsing idle became smooth etc. Lambdas are the correct way round. So I am thinking lambdas need replacing? Even though they seem to read as they should? The do flick back and forward. Generally together, although sometimes one will go lean, the other rich.
I discovered this weekend that the MIL light can actually come on - in seven years I'd never seen it before

After a day on the rollers one of the Lambdas started playing up and throwing a MIL warning, with the accompanying beep, much of the time

Joolz assured me that this was due to me being a cheapskate and not using proper NTK sensors

Do your Lambdas need replacing? I don't know, but if they go completely wrong the MIL light will come on

Jhonno

Original Poster:

3,459 posts

82 months

Monday 13th May
quotequote all
Hmm.. Can the lambdas give incorrect readings? I mean they are still showing as active, but they seem to learn it a running issue?

Do the adaptives only work off the lambdas?

spitfire4v8

2,840 posts

122 months

Monday 13th May
quotequote all
We could do with seeing a trace of the live graphing screen showing the lambdas spiking

start diagnostics, at the top press window : real time logging

on the grey part of the window right click and select the lambdas option

take a screen shot of the screen with the engine hot and 30 seconds of running say 10secs idle, 10 secs held steady at 2000rpm, then 10 at 3000 rpm and post it here. You can do that by running the test and then quickly press printscreen and paste that image into a paint program and save it. You can record the run and save it as a file but it's just easier to save a screenshot.


The adaptives trim is based on the lambda feedback .. some lambda outputs look ok on the main diags screen, but when you view it in the realtime logging they don't spike sharply like they should do, the traces might be gentle meanderings which is not correct.

It's ok to see them showing weak one bank rich the other, they're rarely in-sync and you wouldn't expect them to be really anyway.

These engines don't like weak mixtures, but are forced to run that way for catalyst protection.
It's likely that when you reset your adaptives the mixture goes rich when running from the main map, and once trimming back to lamabda=1 happens your rough running reappears.

This does NOT however explain why your car varies from one run to another, unless your NVRAM chip internal battery is on its way out and forgetting the stored trim valves sometimes. To check this, make sure your ecu has some adaptive values when read on the diags. Unplug and take out the ecu for a day. Next day plug it back in and read the daptive tables again. If they're full of zeros your NVRAM needs replacing. If it's full of numbers you're probably ok (for a day of battery back up anyway, try it for longer if you want)

Also try:
On the next outing drive the car until it runs rough, then without stopping the engine reset the adaptives and then drive away immediately .. does the engine behave smoothly initially (until it starts learning values again ..)

if it does drive smoothly then you now know it's the trimming of the mixture that's causing the issue.

You've then got to work out why.

Is there one or two cylinders which are naturally slightly weaker anyway (higher airflows, air leaks, poorer injection / spray rates) and the trimming then puts those weaker than lambda=1 which would be very bad for smoothness.

This is a long drawn out diagnosis period and you really need to know what you're looking at.


As an aside is it ok under load, ie larger throttle openings.

report back your findings smile


Jhonno

Original Poster:

3,459 posts

82 months

Monday 13th May
quotequote all
spitfire4v8 said:
Also try:
On the next outing drive the car until it runs rough, then without stopping the engine reset the adaptives and then drive away immediately .. does the engine behave smoothly initially (until it starts learning values again ..)

if it does drive smoothly then you now know it's the trimming of the mixture that's causing the issue.

You've then got to work out why.

Is there one or two cylinders which are naturally slightly weaker anyway (higher airflows, air leaks, poorer injection / spray rates) and the trimming then puts those weaker than lambda=1 which would be very bad for smoothness.

This is a long drawn out diagnosis period and you really need to know what you're looking at.


As an aside is it ok under load, ie larger throttle openings.

report back your findings smile
Thank you for taking time to respond Joolz!

This I've done.. And it runs perfectly for a short while after resetting them.

Parking up after a 130 mile round trip yesterday, I noticed the even bank was now reading rich on idle, when I had it setup running even across the banks (roughly 0 adaptives +/-3) previously.

My plan is to recheck airflows this evening. I will also try the other things you have mentioned.

Jhonno

Original Poster:

3,459 posts

82 months

Monday 13th May
quotequote all
  • also, over 3k and on larger throttle openings it runs perfectly, and goes very well, I've not taken it above about 4.5k though.

spitfire4v8

2,840 posts

122 months

Monday 13th May
quotequote all
Engines naturally smooth out with higher revs anyway, any poor combustion events are harder to feel as revs increase. You might still have an issue higher up, but its effects may be unobtrusive by then.

You say it drives smoothly just after resetting the adaptives, then its 99 percent certainly a weak mixture misfire that you're experiencing. This might be due to one or two cylinders flowing slightly more air, therefore running weaker than the others, or one or more injectors flowing slightly less, or poor spray patterns (not so much an issue on the 4.5 as it sprays onto the back of the butterfly, but if the droplets are really poor that might affect things, I've never had a car that bad so don't know for sure)

We can move forward a bit I think based on the lambda traces you put up, but if it really does look like a weak mixture (ie the lambdas are spiking sharply, you've totally eliminated all intake air leaks, had your injectors cleaned or changed, and have been really anal about your airflow balancing of individual butterflies) then you'd have really only a couple of options :

1 put in on a mappable ecu so you can clamp the mixture on light duties to be richer than it is right now.

2 remap the existing ecu and disable lambda control in the poor running areas so you can map it richer, but there would be no mixture trimming in those areas so you'd have to be pretty much perfect when resetting your throttle positions versus airflows in the future to what they were during the remap.

Jhonno

Original Poster:

3,459 posts

82 months

Monday 13th May
quotequote all
Well my injectors were refurbished recently, my heat barriers and o-rings are new..

Why might a cylinder be flowing enough to cause a weak mixture?

I shall report back with lambda traces and having checked the airflow balance again this evening.

Edited by Jhonno on Monday 13th May 12:16

spitfire4v8

2,840 posts

122 months

Monday 13th May
quotequote all
Sounds like youve done pretty much everything reasonably possible so far