Slightly lumpy idle

Hi,

I'm after some help (and I'm new to the forum so be gentle!)
I have a 2008 V8 L322 which has an annoying problem:
When at idle you can notice a slight shimmy from the engine, I guess you would call it a slightly lumpy idle. It is a duel fuel car and only does this on petrol.
Also over the past couple of months I have had the check engine light on twice, fault codes indicated "system too lean left bank"
The codes clear themselves after a couple of days,
There are no other noticeable problems.
I have checked for vacuum leaks (including pcv operation)and replaced the plugs and MAF sensor with a genuine item. I have also run some BG44K injector cleaner through the tank. The other thing I did was to remove the throttle body and give it a clean although it wasn't dirty in any way.
Would a bad or sluggish O2 sensor on the left bank cause both these issues i.e. If one bank is running lean then the mismatch between the sides would cause a shimmy.
My next move is to have a look at the O2 sensors on the diagnostics and also the long term fuel trims. Just to reiterate, it does not have the problem when on gas.
Any pointers or shared experience would be much appreciated. It's not that bad at all but I'm fussy about these sought of things.
By the way it's a prins vsi lpg so most of the usual suspects should affect on both gas and petrol.
Thanks in advance

As you say sounds like o2 sensor could be at fault.

First thing would be to reset the fuel trims and see if the same bank then starts moving towards the trim limit again, if it does, swap the O2 sensors over from one bank to the other and reset again...if the fault follows, then it is the sensors, if it remains, then you may need to look at induction air leaks feeding that particular bank making it lean out...but surely that would happen on LPG too....hmmmm

First thing is to reset the fuel trims and see if it continues, then swap the sensors....

That's great, I'll have a look at that today. The only thing that I can think of that would make an induction leak only affect petrol is the amount of trim authority that the lpg ecu has. I.e. Even with the lpg ecu at the limit of fuel trim, it's not enough to cause the lumpy idle. I'm sure I've read somewhere that the amount of trim that the lpg ecu can achieve is less than the petrol ecu?
Thanks for the information

Simonelvery said:

That's great, I'll have a look at that today. The only thing that I can think of that would make an induction leak only affect petrol is the amount of trim authority that the lpg ecu has. I.e. Even with the lpg ecu at the limit of fuel trim, it's not enough to cause the lumpy idle. I'm sure I've read somewhere that the amount of trim that the lpg ecu can achieve is less than the petrol ecu?
Thanks for the information

Click to expand...

Best of luck and let us know how you get on....it is posting answers, help and fixes that makes a forum grow and help others!

Ok so here is an update on this issue:
Had the car on the tester today and it is showing that both banks are running lean. When you look at the graphs on the diagnostic kit they are both showing a very lean condition when at idle (O2 sensor readings) When you rev the engine, they both react but it takes a good boot of the throttle to get the readings to drop towards the rich. When on LPG it is still running lean, but not quite as bad so this maybe is the reason why I only get the lumpy idle on petrol.
So this takes me back to a vacuum leak:
I have systematically removed all the vacuum pipes and blocked them off with no change to the lumpy idle, the only thing I can't see is if the mixture is changing because I haven't got access to the kit at home. I have smoke tested the intake and there is no sign of a manifold leak. I filled the intake with smoke and even after 5 mins there was plenty of smoke still in there as it came out when one of the vacuum pipes was removed. I'm assuming from this that if there was a leak I would have seen it.
Like I said before, it's had a MAF sensor so I have to assume it's not that.
Could it really be both pre cat oxygen sensors are causing the issue? In all my years fixing aircraft, this in my point of view would be highly unlikely.
Is there anything else that anyone can think of that would cause this?
By the way long term fuel trims are showing +11 to +15 at idle.

Cheers

One other thing:
I have read that the O2 sensors should be seen to flip flop from rich to lean and back at idle , is this correct because they're definitely not doing that.

Simonelvery said:

Ok so here is an update on this issue:
Had the car on the tester today and it is showing that both banks are running lean. When you look at the graphs on the diagnostic kit they are both showing a very lean condition when at idle (O2 sensor readings) When you rev the engine, they both react but it takes a good boot of the throttle to get the readings to drop towards the rich. When on LPG it is still running lean, but not quite as bad so this maybe is the reason why I only get the lumpy idle on petrol.
So this takes me back to a vacuum leak:
I have systematically removed all the vacuum pipes and blocked them off with no change to the lumpy idle, the only thing I can't see is if the mixture is changing because I haven't got access to the kit at home. I have smoke tested the intake and there is no sign of a manifold leak. I filled the intake with smoke and even after 5 mins there was plenty of smoke still in there as it came out when one of the vacuum pipes was removed. I'm assuming from this that if there was a leak I would have seen it.
Like I said before, it's had a MAF sensor so I have to assume it's not that.
Could it really be both pre cat oxygen sensors are causing the issue? In all my years fixing aircraft, this in my point of view would be highly unlikely.
Is there anything else that anyone can think of that would cause this?
By the way long term fuel trims are showing +11 to +15 at idle.

Cheers

Click to expand...

Was the replacement MAF a cheapy? If so it could well be that, the cheap ones often do not work well.

No, original oem denso unit

Ok so just in case anyone is interested, here is what the problem was:

After much head scratching and fault diagnosis it turns out that it was a dirty injector.
I narrowed it down by doing the following:
With the diagnostic kit on the car, I once again checked for a vacuum leak which would have given the lean codes. I changed the pcv valve to be sure and used both the smoke method and the propane method to check for air leaks around the intake manifold and both sets of injectors.
I then moved onto the O2 sensors, one thing of note is the upstream O2 sensors on the AJV8 are wideband sensors so you are looking for a change in current, not voltage as the mixture changes. This however is a maximum of 2 milliamps either side of zero which is incidentally the correct value for the correct mixture. As the mixture changes between rich and lean the wideband sensor makes very small changes in the magnitude of about .1 milliamps so it's very hard to check. By introducing propane into the intake you can however see a change to full rich and then lean as the propane is removed.
I then drove the car with the kit connected and both long term and short term trims were doing exactly what they should, again indicating that the lambda probes were operating.
However, at idle the long term fuel trims kept creeping up when idling. Since I had ruled out a vacuum leak there had to be something causing this.
I then checked every sensor I could to check correct operation, including Maf, throttle and pedal position sensors, map and inlet temp, coolant temp etc etc. All were to spec
I then did an injector balance check and found that one of the injectors on the left bank was causing less of a reaction of the fuel trims than the others. That was the problem, the poor injector was squirting in less fuel to one cylinder, causing the O2 sensor on that bank to see a lean condition and start ramping up the fuel trims. When driving, the injector behaved itself and the trims came down. By the way 3 of the injectors were none to clever which explained why there was a slightly high trim on the right bank but not bad enough to cause the lean code.
Simples ....I don't think!

Clever stuff
Well done for sorting it !

Woo hoo, success....

Well done chap, and many thanks for posting the result...I am sure that will help others in fault finding... top man!!

I'm currently chasing a similar issue.
Can i ask how you did the balance test?
I have GAP IID but i don't see any option to do a balance test.

Thanks in advance.