37 replies to this topic

[Phil-R]

The lambda reading is likely a symptom and not the cause. Try to concentrate on the other sensors.

You checked lots, but you didn't mention the TPS. This is a spring loaded potentiometer and over time it gets stiff enough that it will open fine and you'll see the voltage change on your Sykes / volt meter, BUT when you release the throttle, the potentiometer's spring won't pull the sensors arm all the way back. This may upset the ECU telling it the throttle is open a touch more than it is, perhaps making it think you're ready to accelerate or perhaps causing the fueling to go out of closed loop control, so ignoring the lambda sensor.

It's easy to check. Remove the the 2 screw holding it on, and actuate it manually. Check your Sykes and note the voltage when pushed to the closed position, and also when you let it close slowly with its own spring. It should be obvious at this point whether it's returning to the closed position by itself. It's near imposimble to check for this particular fault on the car.

Edited by Phil-R, 07 January 2016 - 07:35 PM.

[minisilverbullet]

The lambda reading is likely a symptom and not the cause. Try to concentrate on the other sensors.

You checked lots, but you didn't mention the TPS. This is a spring loaded potentiometer and over time it gets stiff enough that it will open fine and you'll see the voltage change on your Sykes / volt meter, BUT when you release the throttle, the potentiometer's spring won't pull the sensors arm all the way back. This may upset the ECU telling it the throttle is open a touch more than it is, perhaps making it think you're ready to accelerate or perhaps causing the fueling to go out of closed loop control, so ignoring the lambda sensor.

It's easy to check. Remove the the 2 screw holding it on, and actuate it manually. Check your Sykes and note the voltage when pushed to the closed position, and also when you let it close slowly with its own spring. It should be obvious at this point whether it's returning to the closed position by itself. It's near imposimble to check for this particular fault on the car.

 

I thought this too, I think the lambda value you see is some symptom of "limp home mode"

[Phil-R]

Agreed, it could well be; Though I suspected there are two possibilities...

When ever mine tripped into Limp Home Mode the carbon monoxide reading went skyward (6 to 8% after cat). My gut feeling is that would actually cause a voltage higher than 0.8V though.

Another possibility is that the ECU hasn't gone into Limp, but is simple operating open loop. It knows it should only use a narrowband lambda sensor only around idle, but one of the sensors (maybe TPS) is making it think it's just outside of an idle condition (like slow moving traffic). It would be normal for fuel map areas around idle to be a touch richer than required as this would give a smoother transition when closed loop kicks in.

Edited by Phil-R, 07 January 2016 - 10:23 PM.

[FlyingScot]

You should be able to read the at "idle" voltage on the TPS with the ACR it should be close to 0.4V. If it doesn't return correctly this voltage will be higher and in fact if this occurs several times MEMS will record a fault in the log.
The reason for asking the OP to measure the lambda voltage at the plug is exactly what is described above - that the 0.8V is what MEMS is saying (and being read by the ACR) as opposed to what is really being generated by the lambda itself.

FS

[Phil-R]

Good points. Reading the Lambda at the plug is a must especially if it's suspicipusly static. Take a reading both plugged in and unpluged - The current generated by the sensor is tiny, so if there was a wiring fault that put 0.8V on the sensor +ve wire this would overwhelm any voltage the sensor could generate. Ideally, if you can leave the heater wires attached while disconecting the signal wires, then this would be better than just unplugging it completely.

Adding to what FS says about MEMS taking several voltage readings before flagging a fault, its quite interesting timing how long it takes to go into limp mode. I don't know how obvious it is with various diagnostics readers, but it can be spotted by listening to the exhaust note (gets steadier as it sudenly goes rich) and also listen to the injector - it starts firing at twice the rate and half the pulse width, so sounds faster and softer. I have no idea why it does that, but timing how long it takes from start up to limp a few times, some times gives another small clue!

I'd still recommend that the TPS is physically removed to test for a couple of reasons. The stepper motor will cause small changes in TPS voltage, so may mask any error caused by the TPS sticking. Secondly, you may not be able see any change if the TPS gets stuck in the same place most times. Thirdly, the throttle plate can only push the TPS open, and not pull of closed; With the TPS in your hands, its pretty clear when it's sticking.

It's late and I'm starting to ramble. I've probably given OP a headache and starting to reconsider carbs! Good luck OP!

[96minimig]

Thanks for the replies. Not rambling that s very helpful will try that later tonight and test the signal wires with it running too.
Although I only have a standard multimeter so it may not pick up the low voltages

[David Wayne]

I run an MPI so I do not know about the SPI's ECU, but wouldn't there be a check engine indication on the dash when in limp mode? 

[96minimig]

Hi got round to getting back to this today and fitted another brand new lambda sensor from minispares and still have a reading on my Skykes Pickavant of 0.8 that's it.

Re checked plugs they are correct colour and the car runs fine. Since changing to the new sensor it has a very slight popping misfire note when cold from the exhaust but when warm it disappears still drives fine on cold though.

I also checked the voltage from the TPS and its at 0.56 volts from idle and 3.98 when full throttle and its voltage goes up smoothly when you press the throttle linkage with no snags or rough spots.

[FlyingScot]

I run an MPI so I do not know about the SPI's ECU, but wouldn't there be a check engine indication on the dash when in limp mode?

Nope, nor on MPi, there is no visual indication that MEMS is in limp home mode.

FS

[FlyingScot]

Hi got round to getting back to this today and fitted another brand new lambda sensor from minispares and still have a reading on my Skykes Pickavant of 0.8 that's it.
Re checked plugs they are correct colour and the car runs fine. Since changing to the new sensor it has a very slight popping misfire note when cold from the exhaust but when warm it disappears still drives fine on cold though.
I also checked the voltage from the TPS and its at 0.56 volts from idle and 3.98 when full throttle and its voltage goes up smoothly when you press the throttle linkage with no snags or rough spots.

You need to follow the advice above, TPS voltage seems right but worth checking off the SPi unit. Measure voltage at the plug with a multimeter not the ACR as it seems more likely you have a wiring issue.

FS

[96minimig]

How would this throw out voltage to a lambda sensor as that's what's confusing me and sure.y the car would run rough.
I also had a friend who's got an old mot emissions tester probe machine and the HC were well within the limit but lambda and CO were fluctuating right on the limits epescially the lambda so surely this shows the lambda is working.
I am aiming to change the cat too as I think the cheapo one I bought doesn't work at its best unless you take it down th road to get it hot then the CO passes easily but soon cools and goes into failing.

[96minimig]

I am going to try and check the TPS tomorrow but just curious to the above

[Phil-R]

How would this throw out voltage to a lambda sensor as that's what's confusing me and sure.y the car would run rough.

 

Narrowband lambda sensors only give readings for air fuel ratios of roughly 14:0 - 15:1. It's not good for emissions or economy, but Mini engines prefer to run slightly rich, and will sound VERY healthy anywhere from 11:1 to 14:1. An AFR of 14:1 would be rich enough to give you a rich lambda reading but not blacken the plugs.

 

Since changing to the new sensor it has a very slight popping misfire

 

When you slow down the ECU cuts all fuel until you get down close to idle speed. This is why modern cars don't pop and crackle on over-run. If the TPS incorrectly says you've got your foot slightly on the accelerator, it thinks you want to maintain that speed and starts putting fuel back in. This is where your exhaust popping is coming from - unburnt fuel in the exhaust. There are so many variables that could affect why, when, where this happens, the new sensor may or may not have triggered it, but probably didn't cause it.

 

 

I also checked the voltage from the TPS and its at 0.56 volts from idle and 3.98 when full throttle and its voltage goes up smoothly when you press the throttle linkage with no snags or rough spots.

 

Yep, my sticking TPS looked just like this on the car... I did the same tests and couldn't believe mine was at fault; When playing with it off the car for a few minutes, it was suddenly very clear  . Press and very slowly release the TPS. When the arm stops moving, see if you can pull it back closed any further.

Edited by Phil-R, 09 February 2016 - 01:19 AM.

[96minimig]

Thanks for getting back to me again guys. Hope your patience are still intact lol sorry .

Ok thanks Phil-R will remove the potentiometer tonight and report back if it moves back as you said.
my MOT is booked for March and I pretty sure everything is ok just this emissions business. if I can get this sorted I just need to fit a better catalytic converter and cross my fingers

Will report back soon

[Phil-R]

Whatever you find, you can at least rule things out and you'll be 1 step closer to the answer. Good luck.