55 replies to this topic

[madazv8]

Hi,

 

I know this was discussed in depth in a previous thread,  as I scanned over it one evening - but can I find it again... no.  Sorry if this is duplicating.

 

Story goes,  just got the car and it is obviously running rich.  I have addressed a filthy fuel trap and cleaned out the MAP as best I can.  Today I managed to get the memsfcr program working on the laptop and checked the status of everything.  Vacuum is ok-ish and the map drops to about 40-50 from 100 when started.  The O2 sensor starts from mid way and slowly rises without any fluctuation until 840mv and sits there solid - this is flagged as a fault. All other sensors reading as they should. 

 

Previous owner says he replaced the O2 already, although I don't know with what.  Can anybody point me in the direction of what causes this (blocked O2?),  or even the long thread where this was thrashed out.   Appreciate any help.  Thanks in advance.

Attached Files

•  o2.png   217.94K   6 downloads

Edited by madazv8, 16 February 2023 - 12:05 PM.

[MatthewsDad]

Hi, I'm working through a similar problem at the moment. High lambda reading that flat lines at around 1000mV. Rich emissions.

These links may be of help. Let us know how you get on.

https://www.theminif...ambda-readings/

https://www.theminif...-oxygen-sensor/

[humph]

The error some make is rushing to replace the sensor without checking the wiring to it, one of the larger specialists did exactly this with mine.  On mine the wiring had rubbed through and broken behind the throttle body, I believe that this is quite common. The break meant that the supply to the sensor was intermittent.

 

The second time I had a similar issue the sensor was damaged by a failed coolant temp sensor.  Again this is mentioned in the threads Mattewsdad posted.

[madazv8]

I am having a different take on this now.  I was sure it was running rich as it was blowing wet soot out of the exhaust during startup.   I gave the car a good run and that has now cleared.  Maybe it was related to the blocked fuel trap and dirty MAP sensor?   A good run didn't improve anything - can always hope!

 

Anyway I pulled the spark plugs and they are saying the mixture is LEAN (flaky white deposits on outside electrode).  So,  I checked the inlet manifold for any leaks, and didn't find anything.  memsfcr still indicates the O2 sensor is reading steady high.  Everything else within range.

 

Does a high reading mean it is detecting (true or false) that the mixture is fat? Or is this indicitive of a lean mixture ? 

[genpop]

"  Everything else within range."

 

Closed loop is zero!

 

What about lambda status?

[genpop]

According to this:

"Engine running, with the crypton ACT, the reading is in millivolts ( 0.001v), the true voltage the lambda sensor outputs. with the sensor disconnected the reading was fixed at 403mv (0.403v). connecting a sensor that was not in the exhaust and sitting freely on the plastic oil filler cap, the sensor tip started to warm up as would be expected (careful, very very hot, burny burny fingers!), but the volt reading remained 403mv. simply placing the body of the sensor onto something bare metal with a ground path the volt reading instantaniously changed to 200mv, a reading consistant with high oxygen content (it is in the air of course), suggesting the grounding of the sensor body is an impotant factor. not sure if this is just a cheapo after market sensor thing, but i would always recommend a little extra money is spent in buying a Bosch part as i've rarely had issues with them other than damge or old age (bit like me )"

 

800mV = "running lean"

<403mV = "running rich"

[madazv8]

According to this:

"Engine running, with the crypton ACT, the reading is in millivolts ( 0.001v), the true voltage the lambda sensor outputs. with the sensor disconnected the reading was fixed at 403mv (0.403v). connecting a sensor that was not in the exhaust and sitting freely on the plastic oil filler cap, the sensor tip started to warm up as would be expected (careful, very very hot, burny burny fingers!), but the volt reading remained 403mv. simply placing the body of the sensor onto something bare metal with a ground path the volt reading instantaniously changed to 200mv, a reading consistant with high oxygen content (it is in the air of course), suggesting the grounding of the sensor body is an impotant factor. not sure if this is just a cheapo after market sensor thing, but i would always recommend a little extra money is spent in buying a Bosch part as i've rarely had issues with them other than damge or old age (bit like me )"

 

800mV = "running lean"

<403mV = "running rich"

 

Thanks for taking the time.  I may be a bit slow here so apologies...  The quote says 200mv with sensor in air =  High Oxygen = lean running.  so I don't follow the final summary that lean is a high mv reading... ?

 

Additionally..

If the voltage is stuck high, using a hose pipe clamp, gently pinch the fuel supply 'flexi' line (orange), momentaraly, until the engine can be herd to alter tone, too much and the engine will stall. This gives a lean condition and the sensor should read low volts. The lambda sensor is working correctly and the problem lies elsewhere. If the voltage stays high, the sensor may be faulty, or there may be another problem, such as a faulty lambda sensor heater relay.

If the voltage is stuck low, using a hose pipe clamp, gently pinch the fuel return 'flexi' line (green), momentaraly until the engine note can be heard to alter tone, too much and the engine will stall. This gives a rich condition and the the sensor should read high volts. The lambda sensor is working correctly and the problem lies eslewhere. If the sensor voltage stays low the sensor may be faulty.

 

Edited by madazv8, 20 February 2023 - 10:42 AM.

[genpop]

What about a logfile? memsfcr can do that.

[genpop]

i am sorry,

of course this

800mV = "running lean"

<403mV = "running rich"

is wrong.

lean is < 403 < rich (800)

or look to that picture.

Attached Files

•  lambda curves.JPG   26.73K   21 downloads

Edited by genpop, 21 February 2023 - 01:20 PM.

[madazv8]

OK,  so as per the great advice on this forum I checked over the wiring tonight.  The O2 sensor relay (and the fan relay beside it) has no continuity on the earth feeds.  I hooked up a temporary earth and the relay started working.  Gave her a start and plugged in the laptop,  lambda behaving the same.  Didn't run it for long as I was in an enclosed garage,  and couldn't (didn't want to) take it out for a spin to warm up as it pouring rain. 

 

All other wiring to relay and O2 plug seems to be in order.  It is not a genuine O2 that has been fitted a year ago (potentially all this time with no relay).  May go ahead and order up the Bosch unit, unless there is something else I should be checking?

 

Finally,  anybody got suggestions on this wire with no continuity?  According to the service manual,  pin 35 of the relay (earth) is a black &green wire (confirmed) and is connected to pin 36 of the ECU,  why is this?  Is it important? i.e. can I just make a localised ground?

 

Thanks folks

Edited by madazv8, 21 February 2023 - 10:57 AM.

[sonscar]

Generally the ECU supplies earth to turn relays on.Is it earthed when ignition is on,this may revert to open circuit if the ECU does not see the engine running so may not be earthed for more than a second or two.Steve..

[madazv8]

Will check when engine is running tomorrow.  I did check with ignition on and there was still no continuity.  Hopefully it is that,  as I don't want to be bodging the wiring.  Cheers.

[genpop]

What about a logfile? memsfcr can do that.

On the page settings you may start the logging.Log from cold engine until it runs steady warmed up.What year is the car?

[madazv8]

Generally the ECU supplies earth to turn relays on.Is it earthed when ignition is on,this may revert to open circuit if the ECU does not see the engine running so may not be earthed for more than a second or two.Steve..

Bang on.  Thanks for that,  saved me going down a rabbit hole there!  Checked and confirmed,  ground is normal when engine is running.

 

All sensors changed,  and seem to behave as expected when monitoring memsfcr.  Seem to be running out of options other than a dead lambda sensor or faulty ECU.   I'll grab a log file when I get a chance.   Cheers.

[R32Egor]

 

According to this:

"Engine running, with the crypton ACT, the reading is in millivolts ( 0.001v), the true voltage the lambda sensor outputs. with the sensor disconnected the reading was fixed at 403mv (0.403v). connecting a sensor that was not in the exhaust and sitting freely on the plastic oil filler cap, the sensor tip started to warm up as would be expected (careful, very very hot, burny burny fingers!), but the volt reading remained 403mv. simply placing the body of the sensor onto something bare metal with a ground path the volt reading instantaniously changed to 200mv, a reading consistant with high oxygen content (it is in the air of course), suggesting the grounding of the sensor body is an impotant factor. not sure if this is just a cheapo after market sensor thing, but i would always recommend a little extra money is spent in buying a Bosch part as i've rarely had issues with them other than damge or old age (bit like me )"

 

800mV = "running lean"

<403mV = "running rich"

 

Thanks for taking the time.  I may be a bit slow here so apologies...  The quote says 200mv with sensor in air =  High Oxygen = lean running.  so I don't follow the final summary that lean is a high mv reading... ?

 

Additionally..

If the voltage is stuck high, using a hose pipe clamp, gently pinch the fuel supply 'flexi' line (orange), momentaraly, until the engine can be herd to alter tone, too much and the engine will stall. This gives a lean condition and the sensor should read low volts. The lambda sensor is working correctly and the problem lies elsewhere. If the voltage stays high, the sensor may be faulty, or there may be another problem, such as a faulty lambda sensor heater relay.

If the voltage is stuck low, using a hose pipe clamp, gently pinch the fuel return 'flexi' line (green), momentaraly until the engine note can be heard to alter tone, too much and the engine will stall. This gives a rich condition and the the sensor should read high volts. The lambda sensor is working correctly and the problem lies eslewhere. If the sensor voltage stays low the sensor may be faulty.

 

HI There - I had similar problems last year but only with the Lambda sensor.

 

Have you actually done the tests described above when the engine is at temperature?

If not then do it - this should see if the sensor is actually working - You will be able to see the fluctuation of the voltage in the Software.

 

If it does react the correct way then you can say the Lambda sensor (be it a cheap one or the correct bosch one) is working and then if its not the wiring then it maybe the ECU. 

 

In the end I had to get my ECU refurbished as the Lambda sensor and the heating of the sensor was working correctly.

 

Just my input