CO too high - MOT fail

[Studio263]

The car is an early Mk4 Golf, the original version of the 1.6L engine (AKL).

At 200k miles the catalyst started rattling and eventually the middle fell apart completely. It was mentioned on last year's MOT but it just about got through on the emissions.

To get it through this year I've fitted a new one (Klarius, you can't get the originals any more). This was two weeks ago. I've done it carefully, no paste before the cat, handled with care throughout etc. The car runs well, no smoke, no hunting, 50mpg (as it did when it was new), no oil used.

However, it has now failed the MOT on too much CO. The numbers are:

RPM: 2684
CO(%): 0.75 (0.3 is the max)
HC: 187 ppm (200 is the max)
Lambda: 1.017

What could be wrong? My first thought was the lambda probe but the test suggests otherwise. HC looks high too. Otherwise I'm a bit lost, it drives absolutely perfectly otherwise. Could the new catalyst be duff?

[tegwin]

I had similar on a 2001 polo 1.6 16v gti.

Lots of head scratching and eventually found a tiny vacuum leak on a hose going up to what I assume was a fuel pressure regulating diaphragm on the cylinder head fuel rail. New hose and instant compliance with the mot numbers.


It could be several things. I’m not sure on the specifics of your engine though. E.g. faulty outside air temp probe, faulty MaF if fitted… etc

[Belle427]

Cheap cats need to be practically glowing to pass an Mot, lambda reading suggests it's running lean, I would double check the seal of the cat to make sure its not pulling air in first.

[stevieturbo]

Studio263 said:

The car is an early Mk4 Golf, the original version of the 1.6L engine (AKL).

At 200k miles the catalyst started rattling and eventually the middle fell apart completely. It was mentioned on last year's MOT but it just about got through on the emissions.

To get it through this year I've fitted a new one (Klarius, you can't get the originals any more). This was two weeks ago. I've done it carefully, no paste before the cat, handled with care throughout etc. The car runs well, no smoke, no hunting, 50mpg (as it did when it was new), no oil used.

However, it has now failed the MOT on too much CO. The numbers are:

RPM: 2684
CO(%): 0.75 (0.3 is the max)
HC: 187 ppm (200 is the max)
Lambda: 1.017

What could be wrong? My first thought was the lambda probe but the test suggests otherwise. HC looks high too. Otherwise I'm a bit lost, it drives absolutely perfectly otherwise. Could the new catalyst be duff?

CO suggests slightly rich, Lambda suggests lean.

Usually exhaust leak

[paul_c123]

Another interpretation could be running lean and ineffective cat.

[Studio263]

stevieturbo said:

CO suggests slightly rich, Lambda suggests lean.

Usually exhaust leak

Where would the leak be, downstream from the catalyst? The other end of it goes straight to the manifold and I used a new gasket there, also it runs quietly. Could a leak elsewhere in the system cause these problems?

Could a defective lambda probe make the engine think that its running lean and therefore richen itself up to try to compensate?

[GreenV8S]

Studio263 said:

Where would the leak be

Upstream of the lambda sensor, or very close downstream of it. It could also be a faulty lambda sensor, as you mentioned. And it could also be a misfire or air/fuel distribution problem combined with an ineffective cat. But the leak stevieturbo suggested seems the most likely cause and is also easiest to check.

Are you in a position to monitor the lambda signal yourself? If so, try obstructing the tailpipe to produce a slight backpressure and see if that removes the symptoms.

[Studio263]

The lambda sensor is in the exhaust manifold so I can't see how there could be an air leak before it. The catalyst is only two weeks old so it should be working. The lambda sensor is the original one, I've had the car for over 20 years and I've never replaced it.

I've got access to a VAG COM test set so I can see the lambda voltage in the measurement blocks function. I'll measure it and see how it compares with the MOT test values.

[paul_c123]

Is the cat type approved?

[GreenV8S]

Studio263 said:

The lambda sensor is in the exhaust manifold

I assume you'd have noticed a blowing manifold.

IIRC the problems started when the cat physically disintegrated and there were no signs of emissions problems before that, and no other signs of an engine management problem except for the MOT test results. So either some new problem has started, which seems unlikely, or the new cat is simply not working. How do you know it has got hot enough?

[stevieturbo]

Studio263 said:

Where would the leak be, downstream from the catalyst? The other end of it goes straight to the manifold and I used a new gasket there, also it runs quietly. Could a leak elsewhere in the system cause these problems?

Could a defective lambda probe make the engine think that its running lean and therefore richen itself up to try to compensate?

literally anywhere from the engine to the test probe

[stevieturbo]

Studio263 said:

The lambda sensor is in the exhaust manifold so I can't see how there could be an air leak before it. The catalyst is only two weeks old so it should be working. The lambda sensor is the original one, I've had the car for over 20 years and I've never replaced it.

I've got access to a VAG COM test set so I can see the lambda voltage in the measurement blocks function. I'll measure it and see how it compares with the MOT test values.

lambda sensor has sweet FA to do with the MOT test probe in the tailpipe.

Although yes, a leak before a lambda sensor will skew the readings it sees, and will then affect the data it reports to the ecu.

[Studio263]

paul_c123 said:

Is the cat type approved?

It says it is for vehicles made before 2000, this is a 1998 model. It came from Euro Car Parts, nothing fancy.

I plugged it in yesterday, the values from the sensor are:

Idle (800 RPM):

Lambda = 0.04 > 0.8

3000RPM:

Lambda = 0.1 > 0.78

I also disconnected the battery for 10 minutes to clear the learned values in case the ECU had adapted itself to the previous failed catalyst. I don't have an exhaust analyser but exhaust did not smell rich; at idle there was almost no smell at all, just that slight chemical odour that catalysts seem to make when they are working.

I checked the engine ECU for stored fault codes, there were none. The ambient air measurement was about right and the engine temperature read 88C when the gauge on the dash showed 90C - there is a separate sensor for each. The value from the air mass sensor changed when the throttle was opened and all the other measurements seemed to be making sense.

I also removed one of the spark plugs (not easy with this engine!) and it had a light carbon deposit around the end of the metal part which brushed away easily; the electrodes and the ceramic core nose were clean.

Edited by Studio263 on Friday 8th August 07:46

[paul_c123]

Studio263 said:

It says it is for vehicles made before 2000, this is a 1998 model. It came from Euro Car Parts, nothing fancy.

But.........is it type approved? Being new, and being from Euro Car Parts, and being a Klarius don't mean much, it could still be faulty. The only way to know is with a gas analyser, although post-cat lambda should give a good idea for complete failure.

[Studio263]

What would be the point of a major distributor to the trade selling parts that could not pass an MOT?

When I worked in a garage we used to fit Klarius catalysts all the time, they were our go-to brand if an original one was too expensive or not available.

[the-norseman]

Put some decent fuel in the car (V-Power etc), add some fuel additive and take it for a decent hard drive on the way to the MOT centre.

[paul_c123]

Studio263 said:

What would be the point of a major distributor to the trade selling parts that could not pass an MOT?

When I worked in a garage we used to fit Klarius catalysts all the time, they were our go-to brand if an original one was too expensive or not available.

Let's rewind a bit here. You say you've checked the engine and its running right. The only part you've replaced is the cat. Its failed the MoT on emissions:

- Engine isn't right, your checks are wrong
- New cat is faulty
- Something else

[stevieturbo]

Studio263 said:

What would be the point of a major distributor to the trade selling parts that could not pass an MOT?

When I worked in a garage we used to fit Klarius catalysts all the time, they were our go-to brand if an original one was too expensive or not available.

Who said any part has caused it not to pass ?

[stevieturbo]

paul_c123 said:

Let's rewind a bit here. You say you've checked the engine and its running right. The only part you've replaced is the cat. Its failed the MoT on emissions:

- Engine isn't right, your checks are wrong
- New cat is faulty
- Something else

Or it never needed a cat in the first place.

Why was it replaced ? History is always important

[richhead]

My money would be on an exhaust leak somewhere, probably before the cat, you can get catalist friendly exhaust paste, it only takes a pin hole.