Understanding Fuel Trims

[Whodafunk]

I am trying to get my head around short and long term fuel trim (STFT / LTFT) information to help diagnose a fault.

I have code P2196 which is Bank 1 Sensor 1 Rich.

LTFT for Bank 1 is +22% (Bank 2 is normal, around the zero mark)

I understand that the engine is compensating for a LEAN condition by increasing fuel to try and achieve stoichiometry (lambda) of 14.7 air to 1 part fuel. Therefore creating a RICH condition as far as the O2 Sensor reads.

So basically with all codes reset the O2 Sensor initially detects a lean condition. It therefore tells the engine to increase fuel to restore stoichiometry. This sets a new baseline for the LTFT of around +22% (in my case). It maintains this and the engine operates perfectly, bar the P2196 code and EML.

So, the million dollar question is, what is causing my lean condition to begin with?

Now Im pretty certain it's not a fuel delivery problem as the problem is unique to bank 1.

I'm fairly confident it's not a vacuum leak as when I increase RPM the percentages remain roughly the same. When in the case of a vac leak increased RPM makes it less prominent therefore LTFT should come down, but it stays the same.

Any ideas what else I can look for? Perhaps an exhaust leak around/prior to the B1 O2 Sensor that causes the initial lean reading? Would LTFT percentages increase with RPM as the exhaust would draw in more air?

Whatever it is I appear to have a lean condition that is equal at idle and at load.

 

[Whodafunk]

Don't all shout at once, lol

 

[audismart]

Or could be a faulty lambda sensor?

 

[Whodafunk]

Or could be a faulty lambda sensor?

Nah, I've ruled that out that stupid way by changing it

 

[S3dave1990]

Faulty pcv?

 

[Whodafunk]

Faulty pcv?

I thought about this but I'm not getting any sort of symptoms and idles perfectly

 

[audismart]

Could be the maf, they dont like newly cleaned and re oiled air filters.

 

[Whodafunk]

Could be the maf, they dont like newly cleaned and re oiled air filters.

Nah, I've changed that too

 

[george_seamons]

The "bank 1" part is a bit of a red herring. On a 4 cylinder engine, the whole engine is bank 1, you don't get a bank 2 until you go up to the "V" engines (v6, v8, v10 etc).

 

[Whodafunk]

The "bank 1" part is a bit of a red herring. On a 4 cylinder engine, the whole engine is bank 1, you don't get a bank 2 until you go up to the "V" engines (v6, v8, v10 etc).

It is a V6, it's a 3.2 V6

My problem is only on Bank 1. Looking at some live fuel trim info this morning my LTFT is at 20% at load and still around 0% at idle. Clearly the issue is only occurring under load which rules out a vacuum leak.

 

[Marek196c]

so it's not maf, lambda on bank1 nor vacum hoses. So did you have exhaust leaks checked? Fuel pump, fuel lines? Maybe not creating enough compression at beginning? Do you got flexi before the sensors by the chance?
P.S. got same results from the computer after 1 minute of running? (till 1 minute shouldn't read info from lambda sensors as they need to warm up)

 

[Whodafunk]

I've waited until the sensors are upto temperature and in closed loop before taking readings. The flexi is after the sensors but only just. I've done a smoke test on Bank 1 down pipe and found no exhaust leak. It's not a failsafe test though and I'm still not 100% it's not an exhaust leak, particularly as the fault seems to only really manifest itself under load. I'm also suspecting the MAP sensor as a possible cause too.

 

[Marek196c]

you got 1 map for both sides or 1 per side? if you got 1 for both sides should inflict also bank 2... What about initial fuel pressure?

 

[Whodafunk]

you got 1 map for both sides or 1 per side? if you got 1 for both sides should inflict also bank 2... What about initial fuel pressure?

That's a good point actually. Again fuel pressure would affect both banks. I'm down to injector problem or exhaust leak. Although I'm still not convinced a leaking injector would cause a lean condition and then solve it by injecting more fuel. Mpg would be well down if that was the case, and it's only down by 2/3.

 

[Whodafunk]

The fact that this problem arose immediately after work on the exhaust I'm still leaning towards an exhaust leak. It's too much of a coincidence for it to be anything else

 

[Marek196c]

That's a good point actually. Again fuel pressure would affect both banks. I'm down to injector problem or exhaust leak. Although I'm still not convinced a leaking injector would cause a lean condition and then solve it by injecting more fuel. Mpg would be well down if that was the case, and it's only down by 2/3.

not if it's on rail or hose (dirty) on some hose or copuler, T connector... or injector itself (maybe you can see on vcds the current voltage or pressure)

The fact that this problem arose immediately after work on the exhaust I'm still leaning towards an exhaust leak. It's too much of a coincidence for it to be anything else

so it's good if you can get proper inspection of it, like if you go to kwikfit they will inspect it for you free of charge without obligation... (better vision from bottom)

 

[audismart]

I`m inclined to say an injector, might be an o ring seal gone or blocked injector, have you tried an injection cleaner in the fuel?

 

[Marek196c]

I`m inclined to say an injector, might be an o ring seal gone or blocked injector, have you tried an injection cleaner in the fuel?

but really concentrated if not wouldn't got any results like on reserve put 1 liter of it, shake well the car and drive it for 20 miles then top up (I ve done that on my ex. 1.4 fiesta worked pretty well).

 

[Whodafunk]

but really concentrated if not wouldn't got any results like on reserve put 1 liter of it, shake well the car and drive it for 20 miles then top up (I ve done that on my ex. 1.4 fiesta worked pretty well).

Yeah that's my next step. Nothing to lose really

 

[Marek196c]

Yeah that's my next step. Nothing to lose really

meanwhile you can change fuel filter cost some quids (4-ish?) but can be 1 possible cause less.

 

[Whodafunk]

meanwhile you can change fuel filter cost some quids (4-ish?) but can be 1 possible cause less.

Changed that too as it had the pressure regulator built in.

 

[Marek196c]

some progress? If you suspect the injectors, check voltage on them eventually take them off and check on bench.

 

[Whodafunk]

I've put some injector cleaner in and will see what difference that makes.

I checked voltage but I wasn't really sure what it should be. At idle cylinders 1, 2 and 3 were around 0.4 and cylinders 4, 5 and 6 were around 0.9. Any reason why they would be different? When I increase RPM all of the voltages shoot up to 3.5 and above.

I did try to check for injection quantity (bank 13 I believe), but for some reason it wouldn't bring this measuring block up??

With regards taking them off to inspect/clean there's quite a bit of work involved on the 3.2 to get to them. The bumper has to come off in order to get to the lower bolts on the intake manifold

 

[Marek196c]

Did your car been remaped? Because they shouldn't differ, the scale in voltage is basically how much the injectors are open, that mean on 0 or nearly 0voltage they are closed on 3.5v (more or less, sepend on model of injector) should be fully opened, this was important info. If wasn't the fact of overfuelling I guess for wiring, but for me is something with mapping into ecu itself, because as initial voltage putting 0.4V and should 0,9 as on other ones. As on idling is ok (don't need too much force to rotate the engine) but its struggling under load(need a lot of force to rotate engine + move the car that is over a tone). So yeah logical will be check mapping of ecu.

 

[Whodafunk]

It's not been remapped, no.

 

[Marek196c]

Ok so if you can not read bank 13 that should be reflecting the injector values, better checking the ecu and wiring to them, some recalls from audi for ecu upgrade?

 

[Whodafunk]

How do I know if mine was one of those recalled?

 

[Marek196c]

Calling audi and asking them, you will need your number plate, anyways they offer also software upgrade to newest one but dunno costs of it.

 

[Whodafunk]

I'll have another go with VCDS as sometimes it needs restarting as not all options available in measuring values

 

[Marek196c]

you got original vcds or ebay one?

 

[Whodafunk]

you got original vcds or ebay one?

It's a cheap one from Amazon. That was my next thought as to why bank 13 isn't showing up. I've never had an issue with any of the other measuring blocks but might need to try a proper Ross tech one and see if that makes a difference

 

[Marek196c]

yes, if you can get some1 with it it's better, meanwhile could be good idea to check electric diagram of injectors, if they are all controlled directly from ecu, or using some rely or amp to drive them?

 

[Whodafunk]

Any idea where I can get the diagram from mate? Good shout

 

[Marek196c]

I will check on my Haynes later (like 2 or 3 hours from now) if it's also suitable for 3.2 petrol, or may be on ETKA like explode view of parts or on forum or google. However it's always good to get one (haynes) due to torque wrench settings as well wiring diagrams etc... and it's relatively cheap (less than 20£).

 

[Marek196c]

just checked, my haynes not covering 3.2 engine version.