Mysterious Occasional Misfire - VVC 160

[rj]

For what it's worth I don't think 6 seconds is off the mark.

[revilla]

Regin, do you understand what that "rich time" is about then? Is it just the time it takes the ECU to settle back to correct idle fuelling or does it go rich for a period on return to idle for a purpose?

To be honest I think I agree now that there's not much wrong with the lambda and fuelling, certainly when it is running properly and not misfiring. It's just odd that the ECU always seems to log a DTC indicating "Misfire with Low Fuel" (i.e. lean mixture) when the misfire occurs.

I think my next move will be to get the decat pipe made up, move the new lambda down to where it is reading more than one cylinder if possible and then just drive it for a while to see if it still does it. Otherwise I'll be in danger of changing too many things at once as you correctly pointed out earlier.

If it does still do it I'll then look at swapping in some fresh ASNU'd injectors.

[rj]

I don't know what's behind, but any car will only run closed loop when speed has been steady for a while. Some of it will be because there is a distance and hence a delay from the combustion chamber to the lambda sensor.

[p.mole1]

I would try the TPS, had a problem with a Ford years ago that would misfire at a certain speed and we checked the TPS with a Ohm meter but found nothing. There was a tiny break in the carbon track that was causing the misfire.

[kenneth.mula]

Maybe a long shot, and you have probably tried it. But have you checked your earthing? I had similar problems, the car misfired. When I checked, and corrected my earthing, everything was ok afterwards.

Kenneth

[bjw]

Since a lot of plausible possibilities are being eliminated, a (very) long shot - "Misfire with Low Fuel" is equally "Misfire with Too Much Air".

As I read your post#1 it happens at higher vacuum, so possibly an inlet manifold leak ? (Although how that could be intermittent isn't obvious...)

If it were a leak, it would presumably be at cyl #4 for it to register as low fuel.

Cheers

[revilla]

Thanks for the further suggestions.

I have already checked and proved the earthing by looking for earth voltages in the engine loom, but I've double checked that everything was secure tonight anyway.

This afternoon I made up an extension lead to extend the lambda sensor down to the boss on the catalyst rather than the one on No.4 primary. I had the oscilloscope on it again and the signal did look a maybe little cleaner, or maybe I'm kidding myself!

I also scoped the throttle position sensor signal and swept it through a few fully open / fully closed cycles and a few movements around the closed position and the signal was clean, sweeping up and down with the throttle position as expected and always returning to the same idle baseline, so I don't think there's much wrong with the TPS. It is almost brand new too.

I couldn't find any signs of a manifold leak. I had good listen around with a piece of rubber tubing to my ear and there were no sucking or whistling noises.

What I did find was that the idle stop screw on the throttle body was open quite a bit. They don't usually need to be opened up unless the cam timing has been played with and there's a lot of overlap. I had in a previous incarnation of this engine been playing with the cam timing and it was suffering from a diving idle before the IACV caught it, so I had opened it up a bit to compensate. When open too far, they tend to give a big step transition on and off idle (the engine is still producing significant power when the ECU switches to idle strategy) and my car had been suffering from that a bit so this was on my list to check anyway. I've pretty much closed it off now, return to idle is fine with no dive and the idle transition step has gone. Initial throttle response feels sharper so it's generally happier.

It's possible that the increased air supply when the TPS was telling the ECU that it was at idle could have been giving a lean idle mixture under some conditions. It's also possible that the ECU was having to manage the idle speed with some fairly aggressive ignition retard which might have been upsetting it too.

For now I've cleared the DTCs, put it all back together and we'll just have to see if it is any better or whether it misfires again. At least with a cleared DTC list I'll know that anything the ECU logs is fresh and corresponds to future misfire events.

I did take it out for a good blat round this evening and it felt awesome, but then after a few days of driving a worn out Astra diesel it always does!

[revilla]

Car went in for MOT today. They don't usually bother with emissions (they should) but this time they decided to do it properly. I didn't argue, thought it might give me some clues if anything was wrong. It absolutely walked through the emissions and it wasn't even properly hot as it had been standing for half an hour or so. Lambda 1.02, HC 35ppm (limit 200).

[revilla]

An update. After the above changes I did have another occurrence. Having convinced myself that the misfire was only happening when the plugs would be running coolest (low throttle particularly after long coast down with the fuel cut off and cool air being pumped through the cylinders) and noting that the original Rover NLP100290s were the equivalent of an NGK 6 grade, I swapped from NGK BCPR7ES to BCPR6ES gapped down to 0.7mm. It's not missed a beat since. You never know with these intermittent issues whether you've really got them but it's been quite a while now and I've been deliberately trying to provoke the problem and all looks fine. Plugs still look healthy. ECU has logged nothing.

[Nick Bassett]

Fingers crossed for you Andrew - wouldn't it be funny after all that, it was just the spark plugs!

[StevehS3]

Doesn’t it [nearly] always turn out to be something simple, even on a K :)

[revilla]

It does rather look like it!