S8 D3 truly random misfire
 

Hi folks,
A truly useful and informative forum.

I have been plagued by a misfire issue that is truly random. One day it’s cylinder 1 and 9 having misfire counts between 9 and 96 with other eight cylinders on 0. Other times these 1 and 9 cylinder will both be at 0 with other cylinders 4 and 6 having misfires with the others all at 0 - very strange indeed!

I’ve since completed the following:
Replaced coil packs and plugs.
Replaced all connector shells on coil packs
Cleaned earthing connections on cylinder heads for coils.
Replaced MAF connector shells
Unplugged MAF’s with engine idling, resulting in erratic idling as it should be.
Carbon cleaned with walnut shell blasting.
Visual inspected inlet manifold for broken plastic flaps - ok.
VCDS scan resulting misfire codes only for engine.

It shouldn’t be compression as the misfire cylinders can change.

Can anyone advise of any other checks I can run?

Should I power off the battery and unplug ECU connectors and check/clean?
Is there an earth connection that can be corroded or something that can cause this?

Thanks guys,

Is the battery in good health? Have you checked the vacuum lines for a possible leak? I don't know if VCDS can monitor fuel pressure, but what about a possible clogged fuel filter?

Thanks steamship,
Battery good - less than a year old and no VCDS codes.
Vacuum lines mostly replaced - i’ll look into these again.
Fuel rail pressure checked on VCDS, so have not replaced fuel filter - I’ll replace.

one thing when I reset all codes with engine at idle - it sounds like the engine smooths out for a few seconds before pulling misfire codes again.

I willl report back with findings.

Thanks.

Looks like you've covered everything that I can think of.

I just re-read your original post and noticed this line:

Should I power off the battery and unplug ECU connectors and check/clean?


I only recently discovered that the ECUs on the D3 are just lying exposed in the plenum chamber, and now wondering if that area might be the cause, as I'm thinking that since the symptoms are random in nature, maybe water has got into the ECUs, or at least into the connectors.

Since you replaced most of the vacuum lines, and they're probably worse to get at than on a 3.7 D2, I'd opt to investigate the ECU connector first, and at least rule them out before checking the vacuum lines.

Thanks Steamship,

I’ll get straight on that. I’m also hoping to find and clean all earth connections I can find.

Regards,

A bit strange and I do not have any experience with D3 S8.

Considering this I would say ECU detects a problem and cuts fuel to the problematic cylinders to save cats from unburned fuel going into them and damaging them and that results in more misfires. So I would monitor any odd behaviour after you reset and see if there are any problems/reports of misfire in that time to see if it can pinpoint offending cylinder, if that changes over time that it's got to be fuel supply or air metering issue.

Update; ok so I’ve disconnected battery and disconnected connectors to both ECU’s, all looked clean inside no signs of corrosion or water ingress. I sprayed these with Deoxit and reconnected.
Disconnected and checked all electrical harness connectors along with cleaning up all earth connections (coilpack earth on cyl heads and inside wheel-arches) that I could access.

Reconnected battery.
Start-up ok engine idles smooth with 0 misfire counts across all cylinders.
Test drive with VCDS still connected. No misfires detected.

So far so good. I’ll do a cold start and longer drive for the readiness tests to be completed.

Thanks again guys, I’ll keep updating.

Sounds promising, fingers crossed for you fella +++

I'd love to hear your S8 with that Miltek fitted, might have a drive down in fact :cool:

Thanks Noretail,

You're most welcome to drop in and hear it - it sounds great...

there is some drone during motorway driving that takes some getting used to though...

Regards,

I had the middle resonators removed on mine and it sounds great, I'm sure the Miltek system exceeds the current sound though +++

Update

So got wrapped at work for a few days so finally test drove the car on a few trips and the following are results:
1. Misfire is not present with engine warming up. When engine temperature is mid scale / normal. The misfires are now:
- 000768 random multiple cylinder misfire detected P0300-008
- 000769 Cylinder 1 P0301-008 Misfire detected.
- 000774 Cylinder 6 P0306-008 Misfire detected.
- 000777 Cylinder 9 P0309-008nMisfire detected.
I also have the MIL back on and the New fault;
004229 - Mixture regulation bank 4, range 1. P1085-004 Lean limit exceeded. MIL on.

I checked VCDS Readyness and the secondary air has not passed.. can this be connected?
this seems odd as replaced both vacuum valves and blower a year ago and this was ok... I’ll have to remove wheel liner and check all pipes intact etc...

Based on the above I’m thinking a possible fuelling issue or vacuum leak (?) or possible dreaded inlet manifold flaps issue. Thinking about looking into fuel trims to hopefully shed some light...

Really odd as the engine has no misfires, running smooth when warming up.....

Any thoughts or guidance would be greatly appreciated.

Thanks

Bank 4?, it's not W12, you only have 2 banks on V10. But it does sound like not enough fuel, possibly cylinder 6 or 9 not getting enough fuel so (check, swap, replace those 2 injectors). Or maybe not enough fuel in general ( fuel fiter, pump)

I noticed that 'bank 4' too. I wonder if, from an electronic point of view, its treating it as two V5s, or a V4 and V6? It probably has two ECUs like most of the larger FSI engines so that could be where it comes from. Bank 4 in theory could be 7,8,9 or just 8,9 depending on how it's arranged.

I would go with injector too. I did read on another forum that a member on Pistonheads bought a S8 V10 with a misfire and it turned out to be a faulty injector

post #4

https://audisrs.com/v10-misfire-t63354.html

Thanks guys for your suggestions. The v10 has 2x ECU’s and 4 banks.

I’m hoping to figure out how to read the fuel trim on VCDS hopefully understanding this will point the way to diagnosing the issue - it’s really frustrating when considering I did the carbon clean only weeks ago, the car then passed its MOT just fine then deteriorated ....

I think the miltek system contributed greatly to the three points on my licence - the noise of great and makes it hard to have your thoughts elsewhere😊
My only regret-ish is that it does drone a bit at motorway speed 1000rpm-1200rpm...

Hi guys,

Update - connected VCDS and monitored fuel trim so bank 4 (cyl9 and cyl10) is measuring way out with long term fuel trim up to 25 fluctuating between 23 and 25... I believe this should be below 10 as on the other banks...

I did clamp off vacuum pipes and doused a whole can of carb spray around the inlet manifold, breather and various vac pipes - no change in fuel trim readings.

Interestingly I pilled dipstick and change in readings as expected. Couldn’t even remove the oil filler cap - hadn’t realised the vacuum was so extreme at idle...

With the other 3x banks fine the problem looks localised so I’m guessing not a vacuum leak as that would likely affect more banks since this engine although two throttle bodies has a common inlet manifold chamber.

Figuring this means a mixture problem so either an ignition / injection / compression problem on 9 and 10.

Anyone have deeper understanding of VCDS as I’m sure there must be other parameters that will help focus on the actual fault?

Thanks guys for any ideas.... time for a cuppa.

Trim should be within max 12.5% (really within 3-4%) and somewhat equal on all banks, looks like cylinder 9 has a problem, injector most likely. You could swap it around with another cylinder to confirm.

Thanks Ainarssems,
I’m dreading taking off the inlet manifold again... so just want to diagnose as best I can before taking things off again.
In thinking about the high vacuum I’m getting at idle where I cannot even remove the oil cover - any chance you can check your oil cap and let me know if it is easy to remove at idle?

I’m considering that I may have a defective PCV/Crankcase breather-oil separator valve. This would actually make some sense as it is located at the back of the engine near bank-4.

Thanks,

To check this better I am considering the propane test, adding propane to the inlet manifold to richen the mixture. If the fuel trim is driven rich on bank 4 then this should more conclusively point to an injector issue (surely). If not and bank 4 still reports a lean fuel trim then it should be that defective 02 sensor.. any thoughts?

Thanks,

If you remove the oil filler cap whilst the engine is running and the variation in engine speed / rough running improves significantly then it is likely it could be the PCV valve for sure +++

Thanks Architex,
I’m still trying to sort out my misfire issue and +25 fuel trim on bank-4 so looking out for anything abnormal.
I’m prepared to replace Lambda sensor and clean injectors but i understand the inlet manifold flaps can break off and cause the mixture to be off...

I’m taking it all apart again on Monday but VCDS activating the intake manifold test first and will inspect with a camera. Otherwise I’ll continue on to remove and service injectors and replace Lambda S1 Bank-4 while i’m At it. Fingers crossed...

Thanks again...

Update

Ok so all apart again.
Completed intake manifold activation test with VCDS and flaps look perfect.

I have removed the injectors sending to service and test. I had one stuck injector and after waiting on the eBay sourced, FSI removal/installation kit - injector was removed fine.

Research on eBay I found: Injectortune in Wiltshire, can do these injectors so fingers crossed all returned by next week. The challenge now is replacement seal/fitting kits. So far China has these with 2-week lead time. Waiting on Audi to quote...

I’ll update as this develops (this is taking rather long)

Cheers

James at injectortune runs a D2 S8. Or at least did in the past. He did 10 injectors on mine. All good...

The sage continues lean codes and V10 rough idle
 

Hi all, so now all injectors serviced by Injectortune, and refitted with new installation kits, new Lambda sensors bank 2 and bank 4, new Oil separator fitted, and still having this intermittent rough idle. After spirited driving the CEL is on again with the following codes by VCDS:
Bank 1. System too lean at idle. P2187 - 004
Mixture regulation Bank 3 range 1. P1081 - 004 - lean limit exceeded - MIL on.
Mixture regulation Bank 4, range 1. P1085 - 004 - lean limit exceeded.

Looking at the above can anyone suggest additional checks... it looks to be a vacuum leak... possibly?

I can not remove the oil filler cap with the engine idling due to tremendous vacuum - is this normal? On other web threads this points at a faulty oil separator - please let me know what is normal.


I’m at my wits end on this car - so much work and No cigar...

Thanks guys,

From memory the oil seperator / pcv valve isn't hugely expensive on the D2 so perhaps its an idea to change it and see if it resolves things and gets you back to smooth running

Thanks Architex,

I have ordered a new breatherAGAIN but the more costly one. The £38 one was only recently fitted and must be a dud.....

I’ll report again on progress.

Cheers,

Hi, today a fitted another new PCV oil separator and also vacuum tested the fuel tank breather solenoid valve N80 (tested ok). Started the engine and I am now able to remove the oil filler cap with engine idling (well I couldn’t before). So it just goes to prove that cheaper parts on EBay £32 compared to the better £80 Bosch part makes one hell of a difference in quality and function.

Next step - check VCDS for Lean operation after completing my recovery from retinal detachment surgery, hopefully early next week.

Fingers crossed,
Thanks again folks!

+++

Update, well a few hundred miles later and all is well with the car. Idling smoothly, starting and running very well and no engine fault codes - great.

Thanks for all your comments and support��

It’s been a costly and time consuming project to get this car up to the level it’s at. Thinking on a new project now so may consider selling the S8 next.

Cheers

UPDATE
To close out this issue i sorted this issue a while back - sorry for the late post - the ‘new’ Oil Separator that I fitted previously was defective! On recommendation from a USA Audi Forums chap I purchased another new oil separator (Febi) and no more misfire! Lesson on low cost aftermarket parts learned...

Thanks for the update and closing this thread off. I really appreciate it when someone gives us the conclusion. It helps us all with future issues that may be similar.

Well done!