High CO levels, how can I fix this?

[dann2707]

Hello.

I'm so not in the know when it comes to petrols and their sensors

I had the MOT today and the CO levels were really really high according to the tester.

I had a quick google and loads of stuff kept coming up about lambda sensors being dodgy before the CAT however that's related to O2 isn't it so changing that wouldn't affect performance would it?

Could the MAP sensor be on its way out?

I've got an OBD scanner and Torque app on my phone but it returned no stored faults.

Thanks for any replies, would love to get this sorted asap especially if it's killing fuel.

 

[KevF182]

Pre cat Lambda (O2) sensor is absolutely related to CO emmissions, yes. Do you have the print out of the test results? You should have been given this by the tester, and if you post it up you might get a some opinions on whats up. But the best thing to do really will be to plug it in to something that can read live data (the signals the ecu is recieving as the engine is running). Most garages have these. Sensors can be at partial fault, enough to miss-inform the ecu, without actually triggering a fault code.

Going by statistics its probably either the pre-cat O2 or an injector at fault. But post up the results and it might give a better idea.

 

[dann2707]

I'll go get it off him tomorrow. Was in a bit of a rush today and must have forgotten to get it.

Are you talking about like a picoscope machine?

Would I be able to see if it's functioning correctly with a bog multi-tester?

Thanks for the reply Kev.

btw, the cars had all of its injectors changed a few months or so ago.

 

[dann2707]

Had the precat lambda sensor and it looked OK on inspection. Not that dirty and didn't smell of fuel or anything.

Would it be worth just replacing it anyway? Really unsure on what to do in terms of testing it.

 

[Dan@SJM]

You can't tell by looking at them.

 

[Steven103]

Had the precat lambda sensor and it looked OK on inspection. Not that dirty and didn't smell of fuel or anything.

Would it be worth just replacing it anyway? Really unsure on what to do in terms of testing it.

Two things you can easily do at home.

1) Swap the pre and post cat lambda around. The pre cat manages the air fuel mixture so if this sensor is dodgy then the ECU is fed the wrong info and puts the wrong amount of fuel in (hence poor performance & MPG). The post cat lambda sensor is only there to check that the catalyst is working and throw up a fault if not so that you know to go and have the cat replaced and aren't single handedly killing all of the polar bears. So let's say that the pre-cat is faulty and the post-cat is ok, when you swap them around it should fix the fuelling but might throw up an orange emissions warning light (bottom right of dash). If this happens then you know you just need to buy one lambda sensor.

2) Use an oscilloscope or live data to look at the voltage that the pre-cat lambda is putting out. There is an ISO standard that defines the voltage range of a "healthy" lambda. I can't remember exactly what it is, but the reading should constantly alternate very quickly between about 0.2 and 0.8v and IIRC the ISO standard says it should alternate no less than 3 times in a 10 second period. With the engine running you can use Torque to produce a graph of the "pre cat o2 sensor" (it might be called "lambda 0 bank 1" now that I think about it) and check that it does alternate. Torque/mobile phones do have a pretty slow refresh rate so they're not ideal, but they can be good enough to spot a half-dead lambda. E.g. my pre cat lambda wasn't dead enough to cause any warning lights, but on Torque you could see it only alternated between 0.4 and 0.6v then would go dead and just remain at 0.4v for 7 seconds etc. before alternating again.

 

[dann2707]

Two things you can easily do at home.

Thank you so much for this

I got torque up and running and did two logs. One of the first sensor 1. There wasn't an option for lambda 0 only 1. This is what came up for the sensor 1.

I'm guessing that looks pretty much spot on and healthy?

The second sensor however returned this graph when doing the same log. This is sensor 2

Absolutely nothing, however that's sensor 2 presumably the post-cat sensor? But if it was completely dead surely it would bring up an emission light on my dash or some kind of error?

Thanks for any replies in advance.

 

[Steven103]

This is all well and good (the first graph looks sort of as you'd expect for a healthy lambda on a s*** piece of data logging equipment that does not have a fast enough refresh rate), but I'm not convinced we're/you're looking at the right lambda sensor.

The second screenshot for example begs two questions: 1) are you sure the lambda is plugged in and 2) are you sure you're telling Torque to log the correct data (there are about 10 different options for Lambdas in the app IIRC to cope with different makes and models of cars)


Have you swapped the lambdas around and repeated the test to see if they appear to perform any different? I think this is the best course of action.

 

[dann2707]

I was thinking I'd done something wrong but I'm literally just sat in my car about to fire it up after swapping them over. Let's see what happens now!

 

[dann2707]

So after the sensor swap it delivered the exact same results. Sensor 1 was heartbeaty and the other was dead.

 

[dann2707]

This is all well and good (the first graph looks sort of as you'd expect for a healthy lambda on a s*** piece of data logging equipment that does not have a fast enough refresh rate), but I'm not convinced we're/you're looking at the right lambda sensor.

The second screenshot for example begs two questions: 1) are you sure the lambda is plugged in and 2) are you sure you're telling Torque to log the correct data (there are about 10 different options for Lambdas in the app IIRC to cope with different makes and models of cars)


Have you swapped the lambdas around and repeated the test to see if they appear to perform any different? I think this is the best course of action.

I think with the torque app on my phone the options highlighted in green are those that can only be monitored on my car. And for the o2 sensor there are only two options so I think it has to be the correct ones.

So strange how the second sensor constantly says nothing is happening even after the swap. Logged the map sensor too and that looks normal.

This is so strange!

 

[ben lee]

normaly high co with lambda and hc ok would be a faulty cat.