Lambda Sensors gone haywire
Discussion
There seem to be a few officiandos of engine management systems on this forum, so I have a question for you, please....
After having new cats and lambda sensors fitted recently my V8 ran well for a couple of hundred miles (in one journey) then, just as I was pulling into home I got an ECU fault and the engine started to run poorly.
On investigating I found the ECU had logged a Lambda sensor fault (though I don't know if the workshop cleared the logs after fitting the new ones which had, apparently, burnt out). Running the logger showed 0v from 1 lambda and a normal range from the other and the adaptive maps were all over the place - +30's on one bank and -30's on the other.
Reset the adaptive maps and the engine ran smoothly for a few miles before loosing the plot again, coughing and spluttering. This happened about 10 days ago and I haven't had chance to test the sensors or check the connections but I ran the car yesterday and got 0v and 1.2v constant from the 2 sensors respectively.
Now it's my guess that - on top of a failed sensor (the 0v one) - the workshop may have fitted the sensors the wrong way round meaning that one bank is running lean and being leaned off even further whilst the other is running rich and being constantly increased.
My question is - if that's the case, why didn't I notice poor running for the first couple of hundred miles?
After having new cats and lambda sensors fitted recently my V8 ran well for a couple of hundred miles (in one journey) then, just as I was pulling into home I got an ECU fault and the engine started to run poorly.
On investigating I found the ECU had logged a Lambda sensor fault (though I don't know if the workshop cleared the logs after fitting the new ones which had, apparently, burnt out). Running the logger showed 0v from 1 lambda and a normal range from the other and the adaptive maps were all over the place - +30's on one bank and -30's on the other.
Reset the adaptive maps and the engine ran smoothly for a few miles before loosing the plot again, coughing and spluttering. This happened about 10 days ago and I haven't had chance to test the sensors or check the connections but I ran the car yesterday and got 0v and 1.2v constant from the 2 sensors respectively.
Now it's my guess that - on top of a failed sensor (the 0v one) - the workshop may have fitted the sensors the wrong way round meaning that one bank is running lean and being leaned off even further whilst the other is running rich and being constantly increased.
My question is - if that's the case, why didn't I notice poor running for the first couple of hundred miles?
Don't know how your ECU works but from what little I know, the lambda sensor will be a narrow band heated sensor which should output one of two values depending whether the lambda ratio is less than or more than 1. In normal operation, the lambda ratio should oscillate around 1 so you will normally see the output cycling rapidly between the two states. This oscillation makes it hard to measure the output directly using a conventional multimeter. In my setup (Rover V8 / 14 CUX) readings from both lambda sensors are combined to control fuelling for the whole engine. In others, each bank of the engine is controlled by its own lambda sensor. If one sensor fundamentally isn't working, it may be a wiring fault, heater feed failed, a genuine fault on one bank (why did the previous sensor fail?) or something silly like the wiring for the two sensors crossed over so the engine isn't really running closed loop. If I was faced with a problem like this I'd check for wiring faults and then hand it over to the experts.
My thoughts exactly.
Generally you can check to see that you have allthe correct voltages to and from the O2 sensors, feed is obviously 12v to the heater, with an earth return for it..make sure you have an earth!!
The signal should constantly vary between 0 and 1 volt, it will alternate rapidly if the sensor is at working temperature and operating correctly.
You may have a wiring fault thats allowing the signal to be interrupted before it gets to the ecu for measurement, hence it reads as a faulty O2 sensor.
Generally you can check to see that you have allthe correct voltages to and from the O2 sensors, feed is obviously 12v to the heater, with an earth return for it..make sure you have an earth!!
The signal should constantly vary between 0 and 1 volt, it will alternate rapidly if the sensor is at working temperature and operating correctly.
You may have a wiring fault thats allowing the signal to be interrupted before it gets to the ecu for measurement, hence it reads as a faulty O2 sensor.
My cerb is also running rough, fitted new throttle pots & set up with laptop, runs fine for a day then like a bag of shite, with huge hesitation between 2-3000 rpm.
I've now set the motor up for the th time but am rapidly losing my rag.
checked plugs & today fitted new KV85 leads ( thanks to Tim at ACT- great service) & car is better but not 100%, its almost as if the setup has been lost from the ecu.
The only fault code mine throws up is a battery voltage one- not enough to cause these problems?
I think it could be throttle cable related ( just too loose- will investigate tomorrow).
Can any one else shed any light on any other areas worth checking.
Darren.
I've now set the motor up for the th time but am rapidly losing my rag.
checked plugs & today fitted new KV85 leads ( thanks to Tim at ACT- great service) & car is better but not 100%, its almost as if the setup has been lost from the ecu.
The only fault code mine throws up is a battery voltage one- not enough to cause these problems?
I think it could be throttle cable related ( just too loose- will investigate tomorrow).
Can any one else shed any light on any other areas worth checking.
Darren.
Darren, apart from the initial lambda fault, I also have a battery voltage fault logged. Must ask Joolz about that one.
What are the lambda sensor readings? I have one side flat-lining at 0v which I suspect as either a dead sensor or cable fault (the other seems ok) I was wondering if that could be the root cause - not a dead sensor but duff wiring.
Still haven't had time to check it though (planned to do it at the weekend but that went pear shaped when our production suite in France fell over).
What are the lambda sensor readings? I have one side flat-lining at 0v which I suspect as either a dead sensor or cable fault (the other seems ok) I was wondering if that could be the root cause - not a dead sensor but duff wiring.
Still haven't had time to check it though (planned to do it at the weekend but that went pear shaped when our production suite in France fell over).
Make sure the sensors are wired in correctly.
If you shine a torch down the back of the engine, you'll see the lambada plugs resting on top of the gearbox ( they are marked odd & even ) make sure the wiring marked odd feeds the lambada for the passenger side bank, they may have been put back together wrong.
Check the wiring first, otherwise a new lambada maybe?
Darren.
If you shine a torch down the back of the engine, you'll see the lambada plugs resting on top of the gearbox ( they are marked odd & even ) make sure the wiring marked odd feeds the lambada for the passenger side bank, they may have been put back together wrong.
Check the wiring first, otherwise a new lambada maybe?
Darren.
this sounds just like the lambdas are connected to the wrong side .. the weak bank gives out a weak signal 'cos it's actually running the lambda that should be connected to the rich running bank, so the wek bank runs weaker and the rich bank runs richer till the running limits are reached (about +/- 35%).
change the plugs over on the gearbox and you'll hopefully be ok again.
Joolz.
PS if it turns out to be that, go back to your dealer and give them a slap!
change the plugs over on the gearbox and you'll hopefully be ok again.
Joolz.
PS if it turns out to be that, go back to your dealer and give them a slap!
Finally got the car on the axle stands and got to the wiring / sensors. Marks on both the connectors and the wiring are aligned (O-O E-E) so I checked the impedance of the heating elements: 6-9ohms one side, nothing on the other. Called the dealer and they are shipping a replacement (remember Joolz - I'm in Amsterdam now, can't just pop round the corner anymore
) (AND you recommended the "dealer" - so who do I slap? 
)
Sh1t happens and parts do sometimes fail even after only a couple hundred miles - I can't blame anyone for this other than SAGEM who I believe manufacture the said lambda sensor....
... YES I can, I can blame the French
..but that's another thread...
) (AND you recommended the "dealer" - so who do I slap? ) Sh1t happens and parts do sometimes fail even after only a couple hundred miles - I can't blame anyone for this other than SAGEM who I believe manufacture the said lambda sensor....
... YES I can, I can blame the French
..but that's another thread...SAGEM!!! Say no more! pity you cant get a Bosch or Nippon Denso replacement,(you should actually) theyre unlikely to cause you any problems.
Sagem ecus and electrics are CRAP!!!! Renault Megane coil packs...Sagem...burn out regular as clockwork, eventually they started to replace em with Jap Nippon Denso parts and no more problems!
Sagem Ecus...man talk about problems?
If you can possibly replace em with the two i mentioned, i would.
Sagem ecus and electrics are CRAP!!!! Renault Megane coil packs...Sagem...burn out regular as clockwork, eventually they started to replace em with Jap Nippon Denso parts and no more problems!
Sagem Ecus...man talk about problems?
If you can possibly replace em with the two i mentioned, i would.
octane junkie said: Finally got the car on the axle stands and got to the wiring / sensors. Marks on both the connectors and the wiring are aligned (O-O E-E) so I checked the impedance of the heating elements: 6-9ohms one side, nothing on the other. Called the dealer and they are shipping a replacement (remember Joolz - I'm in Amsterdam now, can't just pop round the corner anymore
Sh1t happens and parts do sometimes fail even after only a couple hundred miles - I can't blame anyone for this other than SAGEM who I believe manufacture the said lambda sensor....
... YES I can, I can blame the French
ah-ha .. lol!!
the lambdas I use are distributed through intermotor but seem to actually come from various sources .. most common is NTK from Japan (so it says on the lambdas in stock at the mo) .. are the sagem ones cheaper? .. might explain why they're used sometimes .. seem to remember factory cars having these as std though. Funny thing is though that the NTK ones are half the price of the std factory lambdas ..
Spare parts deliveries are like the proverbial bus. You wait for ages then along come 3 together.... the huge pile of lambda sensors in my garden is now so visible that the European Commission are offering french manufacturers subsidies to cut back on production and local air traffic control have issued NOTAMs.
Replaced the offending sensor and the car runs like a dream (again). Managed to find the clicking noise coming from the back axle, too: offside anti-roll bar tie rod was snapped and the handbrake cable was catching on the half shaft.
wonder if I can make a new tie rod from all those O2 sensors......
Replaced the offending sensor and the car runs like a dream (again). Managed to find the clicking noise coming from the back axle, too: offside anti-roll bar tie rod was snapped and the handbrake cable was catching on the half shaft.
wonder if I can make a new tie rod from all those O2 sensors......Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff