Jump to content

Lambda Sensor Problem


alanrichey

Recommended Posts

I know I'm supposed be the expert on the Rover MEMS system but this one has me stumped. So appealing for anyone who understands the basic principles of ECUs :)

 

AndyW has recently purchased a Superspec and joined the Club, so the first thing we did was run my MEMSAnalyser software on the car.

 

This is what we see on my car:

 

Me_zpssntlwkxv.jpg

 

All working as expected, the ECU waits patiently (green line shows Status is flagged as Good) while the sensor warms up and then after 1:30 min it all settles down, engine runs in Closed Loop for the rest of the trip with the mixture being controlled.

 

But this is what we are seeing on Andy's car

 

Andy_zpsmtlxtkhk.jpg

 

His sensor seems to be doing the same as mine, although maybe not as efficiently, but you can see from the Green Line that after about a minute the ECU decides the sensor is not working properly and switches the Status flag to OFF for the rest of the trip so he is running Open Loop in the default 'Limp Home' mode.

 

We have done all the obvious wiring checks and rebooted the ECU, and obviously the next step will be to try a different sensor, but meanwhile does anyone know what the ECU is checking for ? And why it would switch the sensor out after 1 minute before it has a decent chance to warm up.?

 

Al

Link to comment
Share on other sites

I don't know the MEMS ECU, but most lambda controllers have a certain voltage setting that can be used to indicate a fault. For example, they might read 1v to 4v for actual lambda readings, then 0v for warming up and 5v for error. It depends on the controller and also the way the user sets it up (LC1 for example is quite flexible in this sense). Now I'm guessing you don't have an external lambda controller, this is all just built into the MEMS ECU?

 

In that case, is there any setting in the MEMS ECU that is maybe set to flag an error when voltage exceeds 0.44v? Or is there a separate wire or module that might handle the warm-up state? Basically how does it get an error state from the sensor - either it's indicated via an out-of-range voltage value, or there's a separate wire (which would mean some external/separate lambda controller I would think?).

 

If it's all contained within the MEMS ECU and there's no config settings in the software that might be to blame for this, then it might be crack-open-the-ECU time.... after you've tried swapping in your sensor, that would of course be the first thing to do...

Link to comment
Share on other sites

it's a narrow band sensor that generates 0.1V for too lean up to 0.9V for too rich and yes everything is built into the ECU and not user-adjustable. And no external controllers. And the only wires are the heater, signal and ground.

 

I'm beginning to think it is just that Andy's sensor is just a bit lazy. Maybe it has been contaminated with silicone in the past ?

 

Shan't be opening the ECU, as we think it is all on EEPROM and no-one knows anything about it. But I do have a spare ECU from jez, so if a new sensor doesn't help we can try that.

Link to comment
Share on other sites

it's a narrow band sensor that generates 0.1V for too lean up to 0.9V for too rich and yes everything is built into the ECU and not user-adjustable. And no external controllers. And the only wires are the heater, signal and ground.

 

I'm beginning to think it is just that Andy's sensor is just a bit lazy. Maybe it has been contaminated with silicone in the past ?

 

Shan't be opening the ECU, as we think it is all on EEPROM and no-one knows anything about it. But I do have a spare ECU from jez, so if a new sensor doesn't help we can try that.

 

Cool, ok, I think then it's either faulty sensor or faulty ECU. Let's hope it's the former, but given the voltage readings are being measured correctly I fail to see why the ECU is considering the status as faulty.... I mean, all the ECU can see from this thing is a voltage, and your graph is reporting nothing untoward (other than the slightly slower warm-up). Maybe when the voltage peaked above 0.9v or dropped below 0.1v I could understand it flagging an error, but it seems to go into that state rather early with no obvious indication on the voltage readings as to why it thinks there is a fault...

 

I have one idea... probably the only one explainable by the graph...

 

Maybe the MEMS ECU has a 1m hard-coded timeout within which it expects to see some increased voltage from the lambda, and if it doesn't get it, it flags the lambda in an error state indefinitely until the ignition is cycled?

Link to comment
Share on other sites

 

I have one idea... probably the only one explainable by the graph...

 

Maybe the MEMS ECU has a 1m hard-coded timeout within which it expects to see some increased voltage from the lambda, and if it doesn't get it, it flags the lambda in an error state indefinitely until the ignition is cycled?

 

We thought that as well, as the status flipped at 48 seconds on his first two runs, but on the 3rd run it waited until 1:06 before it flipped. And when I look at mine, although my sensor tends to start 'stirring' after 40-50 seconds on one occasion it took 1:30 mins before anything happened and my ECU happily sat waiting.

 

What fun :)

Link to comment
Share on other sites

What time is the start period of the graph? Could it be is sees the slow warm up very quickly but allows the 1:30 min before going into limp mode irrespective of fault logged? Still, the easiest diagnosis is the sensor swap.

Link to comment
Share on other sites

Could it be the heater element in the Sensor? if this keeps going open circuit or is just not drawing enough current, then maybe the ECU is then failing the device. May also be why the Sensor is looking lazy at switching..

Link to comment
Share on other sites

What time is the start period of the graph? Could it be is sees the slow warm up very quickly but allows the 1:30 min before going into limp mode irrespective of fault logged? Still, the easiest diagnosis is the sensor swap.

 

VERY good point. The start point is when the ignition is turned on and the lambda sensor doesn't start until the engine starts up. I need to revisit the graphs and see what the gap is from engine start to going Open Loop.

 

Later: Well picked up. When I checked all the runs for time between engine start and the ECU giving up it is pretty constant at about 40 seconds. So apologies to Brumster, you were right. The ECU must have a hard-wired time while it looks for activity from the sensor.

Edited by alanrichey
  • Like 1
Link to comment
Share on other sites

Could it be the heater element in the Sensor? if this keeps going open circuit or is just not drawing enough current, then maybe the ECU is then failing the device. May also be why the Sensor is looking lazy at switching..

 

Well done Nigel, I think you win the prize :) Andy did 2 runs today, the first from cold and the second after a 5 min stop so the engine was hot. Standard behaviour on the first run but on the second run everything was perfect

 

image_zpsnrux0uap.jpg

 

So the problem must be that the lambda heater is not working properly. So all we need to do now is find out if the sensor is at fault or the relay. And at least we can eliminate the ECU as a culprit, which is a load off our minds.

  • Like 1
Link to comment
Share on other sites

And even more good progress, turns out the sensor heater is only getting a 2V feed rather than 12V. So now we know the sensor is OK and the fault is probably in the relay.

 

The only problem is that we cant find the relay :) And I can't help as I have been looking for mine for 2 years and haven't managed to track it down. I suspect it is hidden deep inside the back of the dash.

 

I did wonder about wiring up another relay in parallel with a decent 12V supply but unfortunately the 'trigger' 12V is sent by the ECU when the engine starts. And I can't thing of any other system that receives a 12V trigger when the engine starts that we could tap into.

 

However we have at least identified the problem and I am very grateful for all the help offered here. What could have taken us days was sorted in hours.

 

I love this Club !!

Link to comment
Share on other sites

the heater relay is most probably in the ecu mounted on the circuit board

I have a load of ecu's at my workshop I'll open one up tomorrow and have a look. might even have a rover one.

Link to comment
Share on other sites

Thanks for the offer but please don't, you might ruin it. We know for a fact the lambda sensor heater relay is a separate yellow relay that comes together with the fuel pump relay and the ECU relay. It is on the wiring diagram and I have actually seen it on 2 part-built Superspecs so I know it exists. We just cannot find where our builders have put it,

 

However, we might have a solution. Now we can communicate with the ECU we can instruct it to cycle the relay. Hopefully we can follow the clicks ... :)

Link to comment
Share on other sites

their only scrap one's anyway, good luck.

by the way

after reading one of your earlier post's about blocking one of your exhaust pipes I realised that my mates transit's( 2.3 petrol/lpg) poor running and unstable tick over was being caused by the hole in his exhaust so thanks for that.

Edited by ivorbigun
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...