gems 4.0 v8 bad fuel mpg .techical possible lambdas faulty
Moderator: phpBB2 - Administrators
gems 4.0 v8 bad fuel mpg .techical possible lambdas faulty
Hi .all my p38 rangerover is really bad on fuel i see about 15mileon 20pound of fuel ( steady driving .after buying a hawkeye reader i found bank b had a faulty lambda .was moving from cold to hot .so i replaced it and reset ecu adaptations it starts idiolsa lot better but mpg has got worse .im woundering if anyone can advise on wot reading i should see cold and warm .so i can check and report back ...
What's that.. about 5mpg?
Have you checked the other sensors as well? I remember a major fault with my TVR when it was in the garage. We tried to start it and it wouldn't fire up at all.
We scratched our heads for a while and checked the obvious stuff (fuel delivery, pipes connected, etc.). It all seemed OK so we did some diagnostics and discovered that the temp sensor had gone bad. The ECU thought it was like minus 100 degrees or something. When we were trying to start the car, the system was trying to enrich the mixture so much that it couldn't actually fire up at all. This excess fuel then seeped down the bore and into the sump. We had to give the car a full oil change (annoying as it had only just had one).
One new sensor and an oil change later, it all fired up fine.
Have you checked the other sensors as well? I remember a major fault with my TVR when it was in the garage. We tried to start it and it wouldn't fire up at all.
We scratched our heads for a while and checked the obvious stuff (fuel delivery, pipes connected, etc.). It all seemed OK so we did some diagnostics and discovered that the temp sensor had gone bad. The ECU thought it was like minus 100 degrees or something. When we were trying to start the car, the system was trying to enrich the mixture so much that it couldn't actually fire up at all. This excess fuel then seeped down the bore and into the sump. We had to give the car a full oil change (annoying as it had only just had one).
One new sensor and an oil change later, it all fired up fine.
-
Ian Anderson
- Forum Contributor
- Posts: 2453
- Joined: Sun Nov 19, 2006 9:46 pm
- Location: Edinburgh
What does it smell like?
When mine was overfuelling (about 6 mpg on limp home mode) when I started it in the back garden on a still summer day after about5 minutes my eyes would burn and boy she did smell rich!
Ian
When mine was overfuelling (about 6 mpg on limp home mode) when I started it in the back garden on a still summer day after about5 minutes my eyes would burn and boy she did smell rich!
Ian
Owner of an "On the Road" GT40 Replica by DAX powered by 3.9Hotwre Efi, worked over by DJ Motors. EFi Working but still does some kangaroo at low revs (Damn the speed limits) In to paint shop 18/03/08.
Might also be an air leak on the intake side off the system some where might not actually be a lamba sensor fault too much unmetered air going in would also play havoc with the fuel system. By dumping too much fuel into the cylinder black plugs suggest over fueling.
I would change the plugs and leads for decent quality leads and a set off new plugs check the inlet system with a tin off carb cleaner or easy start and listen for the engine note drop. Which will indicate a air leak oh another thing is That a bad throttle position sensor can also cut out the lamba sensors if it reads upto 5 volts on the live data from the ecu should give you values off what the engine sensors are doing. Mine was over fueling multi meter showed the lambas switching voltage. Plugged a snap on scanner on to it and read the values the throttle position sensor was showing 5 volts no matter if it was tick over fully open throttle. So the engine was reading full throttle and lamba sensors wasn't working as the ecu thought it was at full throttle and a v8 will switch them off around 3100rpm to allow for a richer fuel map for more power.
I would change the plugs and leads for decent quality leads and a set off new plugs check the inlet system with a tin off carb cleaner or easy start and listen for the engine note drop. Which will indicate a air leak oh another thing is That a bad throttle position sensor can also cut out the lamba sensors if it reads upto 5 volts on the live data from the ecu should give you values off what the engine sensors are doing. Mine was over fueling multi meter showed the lambas switching voltage. Plugged a snap on scanner on to it and read the values the throttle position sensor was showing 5 volts no matter if it was tick over fully open throttle. So the engine was reading full throttle and lamba sensors wasn't working as the ecu thought it was at full throttle and a v8 will switch them off around 3100rpm to allow for a richer fuel map for more power.
Land rover discovery 3.9 Hotwire v8 nick named Christine.
You don't need a special reader to check a TPS. It is just a potentiometer (variable resistor) which sends about 0-5 volts to the ECU according to position. I'm not an expert on the Gems system, but I'd guess all the other sensors are analogue too, so fairly easy to test with a DVM. Of course an ECU with decent diagnostics can make things easier - but they're not always infallible.
Dave
London SW
Rover SD1 VDP EFI
MegaSquirt2 V3
EDIS8
Tech Edge 2Y
London SW
Rover SD1 VDP EFI
MegaSquirt2 V3
EDIS8
Tech Edge 2Y
On mine I checked with a volt meter moved it that was showing it was working. Plugged I'n my snap on scanner and It was showing what the ecu was seeing and a tps at 4.98 volts with the ignition on engine running no change. Changed the tps and it worked and the lamba sensors began working and I had wasted £90 on a New lamba sensor and didn't need it
Land rover discovery 3.9 Hotwire v8 nick named Christine.
I'm no expert on EFI but it seems to me that it's one of three options:
The ECU thinks more air is coming in than actually is, and is adding extra fuel to enrich the mixture,
It is actually putting more fuel into the system than it thinks it is, this could be caused by higher than expected fuel pressure delivering more fuel per injector pulse than is expected,
or it thinks that the amount of fuel being added for the amount of air being seen is correct, when in fact it is wrong (i.e. bad fuel map).
The ECU adding excess fuel could be caused by a lot of things, for example:
Air leaks in the inlet manifold. I would expect this to normally cause a lean mixture as a leak would allow extra air to be sucked in, bypassing the MAF and causing the ECU to see less air going in than there actually is, and so inject less fuel than is actually required. A rich mixture would only be created if positive manifold pressure was forcing previously metered air to escape to atmosphere, maybe in a forced induction setup, for example.
A bad MAF showing more airflow than is actually passing through
A bad TPS suggesting that the throttle is open more than it is
A bad MAP sensor saying that the manifold is experiencing more pressure than it is
A bad temp sensor telling the ECU that the air is very cold (or, if there's a fuel temp sensor, maybe the fuel is very hot?) and therefore requires more fuel to be delivered
Perhaps a bad crank sensor or something saying the engine is revving at a different speed than it actually is
An o2 sensor reporting lean exhaust gasses causing the ECU to add fuel to compensate. This could be a faulty lambda or unburnt air getting in via a cracked manifold, bad gasket, loose manifold, etc.
Other sensors playing up like Barometric or EGT
And probably a load of other things I haven't thought about
Theoretically the car should adjust the fuel injection according to the lambda feedback, so a faulty sensor elsewhere could potentially be compensated for... however, any compensation for lambda feedback will only happen when the the ECU is operating in closed loop mode.
If the closed loop system is compensating for a duff sensor elsewhere, it may appear ok at idle as the ECU will be in closed loop then, but when you go for a drive and accelerate, the ECU will switch to open loop and add fuel only according to the other sensors, then any duff reading may cause uncompensated overfuelling issues. Also, if the ECU is staying in open loop (because the TPS is showing WOT, for example, and so it thinks you are trying to accelerate) the car may never go closed loop and therefore the ECU will ignore the Lambda altogether...
It all depends on what sensors you have, what they are telling the ECU and what decisions the ECU is making based upon that info.
If you think all the sensors are working correctly, check for air leaks in the inlet and exhaust system, missing, leaking, detached or perished pipes, bad gaskets, seals, etc. Also, don't just check the sensors, but check all the wiring back to the ECU for any shorts and/or breaks, and failing that... maybe give up, assume that everything is working correctly and the fuel maps are somehow b0rked? lol.
The ECU thinks more air is coming in than actually is, and is adding extra fuel to enrich the mixture,
It is actually putting more fuel into the system than it thinks it is, this could be caused by higher than expected fuel pressure delivering more fuel per injector pulse than is expected,
or it thinks that the amount of fuel being added for the amount of air being seen is correct, when in fact it is wrong (i.e. bad fuel map).
The ECU adding excess fuel could be caused by a lot of things, for example:
Air leaks in the inlet manifold. I would expect this to normally cause a lean mixture as a leak would allow extra air to be sucked in, bypassing the MAF and causing the ECU to see less air going in than there actually is, and so inject less fuel than is actually required. A rich mixture would only be created if positive manifold pressure was forcing previously metered air to escape to atmosphere, maybe in a forced induction setup, for example.
A bad MAF showing more airflow than is actually passing through
A bad TPS suggesting that the throttle is open more than it is
A bad MAP sensor saying that the manifold is experiencing more pressure than it is
A bad temp sensor telling the ECU that the air is very cold (or, if there's a fuel temp sensor, maybe the fuel is very hot?) and therefore requires more fuel to be delivered
Perhaps a bad crank sensor or something saying the engine is revving at a different speed than it actually is
An o2 sensor reporting lean exhaust gasses causing the ECU to add fuel to compensate. This could be a faulty lambda or unburnt air getting in via a cracked manifold, bad gasket, loose manifold, etc.
Other sensors playing up like Barometric or EGT
And probably a load of other things I haven't thought about
Theoretically the car should adjust the fuel injection according to the lambda feedback, so a faulty sensor elsewhere could potentially be compensated for... however, any compensation for lambda feedback will only happen when the the ECU is operating in closed loop mode.
If the closed loop system is compensating for a duff sensor elsewhere, it may appear ok at idle as the ECU will be in closed loop then, but when you go for a drive and accelerate, the ECU will switch to open loop and add fuel only according to the other sensors, then any duff reading may cause uncompensated overfuelling issues. Also, if the ECU is staying in open loop (because the TPS is showing WOT, for example, and so it thinks you are trying to accelerate) the car may never go closed loop and therefore the ECU will ignore the Lambda altogether...
It all depends on what sensors you have, what they are telling the ECU and what decisions the ECU is making based upon that info.
If you think all the sensors are working correctly, check for air leaks in the inlet and exhaust system, missing, leaking, detached or perished pipes, bad gaskets, seals, etc. Also, don't just check the sensors, but check all the wiring back to the ECU for any shorts and/or breaks, and failing that... maybe give up, assume that everything is working correctly and the fuel maps are somehow b0rked? lol.
