2005 A8 D3 4.2 (BFM) - Low MPG

[MatthewH]

Hi,

I purchased a 2005 A8 around 12 months ago and have gradually been fixing its problems.

The final issue with the car is the low MPG. I realise the MPG is never going to be great with a 4.2L V8 petrol engine but it's currently averaging ~13mpg with careful driving. This is on open NSL roads - around town can easily be 8mpg. It feels fine to drive - no rough running or misfires.

I've spoken to a few other owners of this car with the same engine who all seem to be averaging at least 20mpg in mixed driving. The figures mine returns seems far too low in comparison with that.

The only fault codes are P1128 and P1130:
P1128 - Fuel Trim; Bank 1 (Mult): System too Lean
P1130 - Fuel Trim; Bank 2 (Mult): System too Lean

I've cleared these codes but they return approximately every 100 miles.

These codes are the opposite of what I'd expect - with such high fuel consumption I would expect rich codes but these have never occured.

I've already looked at the following areas:
- MAF sensor (this seems to be the primary cause for these error codes - I've replaced it twice with new genuine parts but it hasn't improved things)
- Pre-cat O2 sensors (replaced with new Bosch sensors)
- Replaced fuel pressure regulator
- Replaced intake manifold flap links
- Replaced throttle body gasket
- Replaced intake manifold gaskets
- Checked all hoses for vacuum leaks (none found)
- Replaced all vacuum hoses
- Checked fuel pressure (3.5bar with FPR connected, 4.0 bar without FPR)
- Replaced fuel filter
- Replaced air filter
- Replaced spark plugs
- Ignition coils were replaced under recall a few years ago
- Replaced G62 coolant temperature sensor
- Replaced N80 EVAP valve
- (probably more that I've forgotten)

When starting the car and driving off, the average MPG reading usually shows a steady 15mpg, but 20 seconds later, it immediately drops to 5mpg with no change in driving style. This makes me think it may be related to the change-over from open loop to closed loop fuelling.

Both cooling fans run at full speed at all times while the engine running for unknown reasons but I don't think this would make such a difference to the MPG. All temperatures seem normal and it heats up as expected (up to 90 degrees within a couple of minutes)

Has anyone had a similar problem? I've run out of ideas. Is it possible that one of the fuel pumps may be failing? I've read that someone else had similar symptoms and it turned out to be a faulty fuel pump, although I don't fully understand why this would cause excessive fuel consumption.

Thanks.

[Mechcanico Lee]

The continuous running cooling fans are a clue here , this is a sign of the ecm not seeing the correct coolant temp reading and running fans as safety default

Coolant Temp sensors ..... is the ecm data block showing correct temp ?
Does it have two temp sensors ...one in engine block somewhere near cyl heads usually ,and maybe radiator outlet temp aswell ?

The lamda codes of lean mixture is strange like you say , are the pre cat lamdas 4 wire or 5 wire wide band with a red or pink wire on the 5 wire type ?


In data blocks for lamdas does it have a block called 'lamda equivalence ratio ' if it's five wire it will have , if it's 4 wire it will have a cycling voltage between 0 volts and 1 volt or 0 and 4 volts ...the cycling voltage is correct ( lamdas switching from rich to lean ) if lamda Volta sticks one way or the other it's over rich or to lean ....drive the car up the road at a steady pace and watch what these values .

Also have a look in the data that it says it's in closed loop mode , the running cooling fan default mode may keep it in a open loop situation
The running fans has to be your first port of call here , sometimes the fans control module can go faulty and put the fans on also ....you normally get codes for this though

But you donot always get codes for engine temps being massively out of range but ecm runs fans just because it does not know what the true temp is .

[steamship]

About the only other possibility to consider is a clogged catalytic converter due to collapse, or some other blockage in the exhaust system.

From what else I've read, if the o2 sensor(s) are on the way out, they may not throw up an error code, but could be providing false information that the engine is running lean, and hence feeding more fuel through it.

[MatthewH]

Originally Posted by Mechcanico Lee

The continuous running cooling fans are a clue here , this is a sign of the ecm not seeing the correct coolant temp reading and running fans as safety default

Coolant Temp sensors ..... is the ecm data block showing correct temp ?
Does it have two temp sensors ...one in engine block somewhere near cyl heads usually ,and maybe radiator outlet temp aswell ?

The lamda codes of lean mixture is strange like you say , are the pre cat lamdas 4 wire or 5 wire wide band with a red or pink wire on the 5 wire type ?


In data blocks for lamdas does it have a block called 'lamda equivalence ratio ' if it's five wire it will have , if it's 4 wire it will have a cycling voltage between 0 volts and 1 volt or 0 and 4 volts ...the cycling voltage is correct ( lamdas switching from rich to lean ) if lamda Volta sticks one way or the other it's over rich or to lean ....drive the car up the road at a steady pace and watch what these values .

Also have a look in the data that it says it's in closed loop mode , the running cooling fan default mode may keep it in a open loop situation
The running fans has to be your first port of call here , sometimes the fans control module can go faulty and put the fans on also ....you normally get codes for this though

But you donot always get codes for engine temps being massively out of range but ecm runs fans just because it does not know what the true temp is .

Thanks for the response.

The O2 sensors are 5 wire wideband sensors - the colours are: white/pink/yellow/grey/black.

I do believe the MPG dropped around the same time as the fans became faulty so I think you may be correct. I've just been unable to find any particular fault which would make the fans run constantly so far.

It does have 2 coolant temperature sensors. G62 near the cylinder head (I replaced this one recently) and G83 sensor somewhere near the radiator outlet (although I've never actually been able to find this). My understanding is the G62 is used for fuelling purposes and the G83 is just used for the gauges. If the G83 is also used for fuelling this may be contributing to the fault, though I've not had any fault codes relating to this sensor (assuming it would show up in the engine module).

I've watched the data blocks for the G62 temperature sensor and it appears to be working correctly. It climbs up to 90 degrees and stays there, which reflects what the gauge is showing. I've never seen the gauge go above half-way.

I've heard that a faulty air-con pressure sensor may also cause one or more fans to run constantly (not sure if this would also cause the main fan to spin). The fans spin constantly whether air-con is turned on or off so I'm not sure if this is the case.

I think it does go into closed-loop mode as expected but I will double check.

I assume the post-cat O2 sensors are irrelevant in terms of fuelling? I know on most cars they just check catalytic converter efficiency but apparently some also take them into account for fuel correction.

I will perform the tests you mentioned in the morning and post an update with my findings.

Thanks

[MatthewH]

Originally Posted by steamship

About the only other possibility to consider is a clogged catalytic converter due to collapse, or some other blockage in the exhaust system.

From what else I've read, if the o2 sensor(s) are on the way out, they may not throw up an error code, but could be providing false information that the engine is running lean, and hence feeding more fuel through it.

Hi, the catalytic converter is something I'd considered. I'm not sure how that could be tested though. It passed MOT with no emissions problems last time around if that makes any difference. It did originally fail on emissions but after replacing both leaking flex pipes it passed (didn't know a leaking flex pipe could cause an emissions failure at the time as it's after the cat)

The pre-cat O2 sensors have just been changed (finished today) as that was the only thing left that I could of, but they don't seem to have shown any improvement. Post-cat sensors are still originals as far as I know.

Thanks

[Mechcanico Lee]

Ok , so you can see the temp reading in ecm block , the other temp sensor is usually in radiator outlet hose or near the rad outlet .... so the lower colder radiator hose , it's used for diagnosis but can phase ecm if giving a mad reading

Go into temp blocks it may not be listed under rad outlet but just a temp reading with no description ....its normally 20 /30 degrees lower that engine actual temp

Do you have any codes in air con hvac system for ac pressure switch ?

So five wire wide bands , like said make sure ecm in closed loop , go into lamda equivalence ratio for both banks , idling and steady driving on the road they should be close to lamda 1 .

Look at the after cat lamdas on steady driving they should be pretty steady at about 0.6 ... 0.7 volts ( low oxygen ) this means cats are working and burning any excess oxygen , if the readings are lower volts or cycling shows cats in efficient .
After cat lamdas are there for diagnostic purposes but be careful because they do get used to trim the fuel readings .

Do the petrols have any type of exhaust manifold air injection control , pulse air and the likes , on some cars you hear this in a morning on warm up ...sounds like a Hoover running , if this was to run whilst in closed loop it would make the exhaust high in oxygen so the lamdas will detect it as lean and the ecm will respond by making the mixture richer

Your emissions failure due leaking down pipes ......as the co testing machine is taken from the tail pipe any leaks in the exhaust draws air in so the lamda value will go out of range as seen by the co tester .... lamda reading will go high , so over the 1.030 lean threshold ....higher than lamda 1 = leaner , lower than lamda 1 = richer

[steamship]

Originally Posted by MatthewH

Hi, the catalytic converter is something I'd considered. I'm not sure how that could be tested though. It passed MOT with no emissions problems last time around if that makes any difference. It did originally fail on emissions but after replacing both leaking flex pipes it passed (didn't know a leaking flex pipe could cause an emissions failure at the time as it's after the cat)

The pre-cat O2 sensors have just been changed (finished today) as that was the only thing left that I could of, but they don't seem to have shown any improvement. Post-cat sensors are still originals as far as I know.

Thanks

I came across this site about catalytic converters and some checks you can do, including exhaust back pressure. However, from what you've posted so far and what it says on this site, this may not be your problem. The site is as follows:

http://www.aa1car.com/library/converter.htm

[Mechcanico Lee]

Another thought

Oil temp readings ok ??

[IT]

Originally Posted by MatthewH

I've spoken to a few other owners of this car with the same engine who all seem to be averaging at least 20mpg in mixed driving. The figures mine returns seems far too low in comparison with that.

Yes, with my BFM 20mpg was easily achievable, and 30+ on a longer run - drop a couple of MPG for LPG maybe, but still sounds like you're sub 50% of where you should be.

Fans running constantly is definitely a sign of something not right. Would be looking there first for sure.

Hope you find a solution, this thread sounds quite interesting...

[MatthewH]

Thanks for the responses.

I have done some further checks today.

The car does have "secondary air injection". The pump is currently broken but all vacuum hoses to the combi-valves have been replaced as I know these are common ones to leak. It does give the occasional "insufficient flow" error code but this is to be expected with no pump. I did think about this and thought maybe the combi-valves were stuck open and causing the lean condition, but I tried to blow through the main secondary air injection hose while the car was turned off but I couldn't.

Here are the values I recorded after the car had warmed up (idle only, air-con off):
- Post-cat sensor voltage => Both around 0.710V steady
- Pre-cat sensor voltage => Both around 1.540V steady
- Fan 1 Duty Cycle => 56.1%
- Fan 2 Duty Cycle => 56.1%
- Fan request from A/C => 100%
- A/C Readiness => A/C-Low
- A/C Pressure => 51.0 bar

Would it help if I went for a short drive before testing these values? I mentioned before the average MPG starts off at 15mpg then drops to 5mpg after a few seconds of driving (even downhill with no throttle).

I checked the HVAC module for errors and it does give the following error:
- 00819 - High pressure sensor G65 (short to ground)
This comes back immediately after clearing codes.

I also had a look for the oil pressure sensor reading but couldn't find it. I found the oil level value but no oil pressure. Do you know where I'd look for this on this engine? Oil level is correct and it doesn't seem to burn any oil at all. It's also still a fairy clean colour after nearly 10k miles (doing another oil and filter change next weekend). On my Fords it goes black after 1k miles!

Edit: Just noticed you asked about oil temperature rather than oil pressure! I'll have a look.

I also wasn't able to find the radiator outlet reading. There was a value called "Coolant temperature (G65)" and one just called "Coolant temperature" but they seemed to show the same value. There is definitely a G83 sensor aswell according to all cooling system diagrams I've found for this engine.

Thanks