Tips on NCT Emssions Failure

gline

I just put my 98 legacy through the NCT, it failed on emissions, only slightly.

Lambda was 1.035 and pass is between 0.98 and 1.02 so i failed by 0.015, is that a lot for emssions?

the car has the original Cat, but modified back-box (does this make a difference?)

Does this show a large problem like I need a new Cat or lambda sensor?

Or would cleaning the sensor work?

I was also thinking of some of that injector/engine cleaning stuff that you add to your fuel?? would that help?

Also which fuel should I use when going in for the test? like the stuff with ethanol etc? Which garage has the best fuel for use in nct e.g. texaco/maxoll etc?

Any other tips for getting a car through the emssions?

Carawaystick

That's a lot, it's 15 parts out of the 40 range. There is a lambda sensor in the exhaust, could be gone.

I think the fuel additives only improve CO or particles.

But I'm not sure, a mechanic here should know more- especially balpark price for new sensor and replacement

BlackWizard

Was the car given a good drive before you sent it in?

BrianD3

If you had lambda 1.035 and everything else was OK you probably have a leak in your exhaust downstream of the lambda sensor

gline

BlackWizard wrote: »

Was the car given a good drive before you sent it in?

no, literally started and went there (20min drive), hasnt been driven in 2 days.

I had the exhaust repaired a few days ago, there was a hole in the exhaust, but i got a new section welded in and removed the bit with the hole.

gline

i can get a new lambda for €40, any mechanics on here? You think that will fix it?

BrianD3

The Lambda result of 1.035 does not mean the lambda probe is faulty, in fact there is no evidence here to suggest that the car needs a new lambda probe. There is also no evidence to suggest that injector cleaner would do any good or that the result was due to the engine temperature.

A car failing the NCT solely for Lambda >1.02 suggests diluted gases coming out of the tailpipe with the leak being located downstream of the lambda probe. The modified back box could be affecting it. If the leak were upstream of the probe then it gets more complex.

The best way to diagnose this problem for certain is to bring the car to a decent garage with an exhaust gas analyzer.

gline

BrianD3 wrote: »

The best way to diagnose this problem for certain is to bring the car to a decent garage with an exhaust gas analyzer.

Most garages have one of them?

So it isnt any easy fix for this issue?

Beta Ray Bill

Hi

Is the Car Jap import?
Japan uses different Fuel in their cars compared to over here.
could possibly be affecting it?

gline

grahambo wrote: »

Hi

Is the Car Jap import?
Japan uses different Fuel in their cars compared to over here.
could possibly be affecting it?

dont think its a jap import. originally UK car.

Pkiernan

I got around a similar SMOG check issue in California.

I drove the car until there was about 1 gallon petrol in it.
Added 1 gallon of pure ethanol (from a hardware store).

Passed emissions no bother, but C02 readings were obviously higher.

Then filled up tank with normal petrol.

Tha car was a 1996 model, so I wasn't too concerned about any effect the temporary high ethanol level would have on how it ran.

Cosmo K

BrianD3 is spot on, nothing wrong with your lambda sensor or cat, small leak in the exhaust is causing the problem.

gline

Cosmo K wrote: »

BrianD3 is spot on, nothing wrong with your lambda sensor or cat, small leak in the exhaust is causing the problem.

cool, il get that checked
Any garage ok for that or do u need a specialist garage?

Darragh29

BrianD3 wrote: »

If you had lambda 1.035 and everything else was OK you probably have a leak in your exhaust downstream of the lambda sensor

This is probably the issue OP. Check if you fitted a modified back box, that the coupling between the box and the exhaust pipe is sealed. You can do this by getting under the car, running the engine and putting your hand over the exhaust for a second and listening for air leaks downstream from the cat & lambda sensor.

drunkmonkey

I've a jap impreza sti, i've spent roughly in the region of 4,500 over 5 tests + retests, I can not get it to pass the test on emissions, i've gone though 3 garages and 1 main dealer.....last time we had it perfect in the garage, passed emissions, took the 20 minute drive to the test center, failed again, i'm at the end of my patience with the NCT center....i'm sorry but if you rev the bolix out of my car your going to get a blip in the emissions.....3 years of ownership now and I still don't have an NCT after import, also had to replace the engine after one test as a piston went while they were testing it...:mad:

i've just resigned myself to the fact I will never have an nct, at this stage i'm just going to book it in every few months for a test, let it fail, at least i'll have a cert to say i tried if a garda ever gets funny about it....

last time I was told replace the cat, oh ya no problem, 1300 from Subaru, special order only.....they really must think subaru owners have money trees growing out the back...:mad:

Darragh29

drunkmonkey wrote: »

I've a jap impreza sti, i've spent roughly in the region of 4,500 over 5 tests + retests, I can not get it to pass the test on emissions, i've gone though 3 garages and 1 main dealer.....last time we had it perfect in the garage, passed emissions, took the 20 minute drive to the test center, failed again, i'm at the end of my patience with the NCT center....i'm sorry but if you rev the bolix out of my car your going to get a blip in the emissions.....3 years of ownership now and I still don't have an NCT after import, also had to replace the engine after one test as a piston went while they were testing it...:mad:

i've just resigned myself to the fact I will never have an nct, at this stage i'm just going to book it in every few months for a test, let it fail, at least i'll have a cert to say i tried if a garda ever gets funny about it....

last time I was told replace the cat, oh ya no problem, 1300 from Subaru, special order only.....they really must think subaru owners have money trees growing out the back...:mad:

This sounds to me like a classic intermittant vacuum leak issue, just one that is not very obvious.

drunkmonkey

Darragh29 wrote: »

This sounds to me like a classic intermittant vacuum leak issue, just one that is not very obvious.

and the cure would be......

gline

was in with an exhaust place, no hole upto cat, possible hole beween cat and manifold, but he said manifold and cat and all piping between will have to be taken out and heatshield removed top be checked.

He said though that there mightent even be a hole, just that their might be, he also said, maybe change the lambda.

What do you guys think?

Anyone know someone who would remove manifold+ cat etc pretty cheap? Or will I just put in a new lambda?

Darragh29

gline wrote: »

was in with an exhaust place, no hole upto cat, possible hole beween cat and manifold, but he said manifold and cat and all piping between will have to be taken out and heatshield removed top be checked.

He said though that there mightent even be a hole, just that their might be, he also said, maybe change the lambda.

What do you guys think?

Anyone know someone who would remove manifold+ cat etc pretty cheap? Or will I just put in a new lambda?

Also OP, check the engine RPM that the lambda failed at. The NCT don't test for Lambda at low idle so on the high idle test, check the form to see what the engine RPM was when the lambda failed and post the RPM back up here...

Darragh29

drunkmonkey wrote: »

and the cure would be......

I've seen this problem a few times. What happens is usually a perished or cracked brake servo pipe or a MAP sensor rubber hose is cracked, (conveniently always at the back of the pipe where you cannot see it, Murphy's Law being what it is!!!), anyway, on your inlet manifold you have a vacuum and this vacuum is used for the brake servo assist and other functions as well. If there is a problem with air getting into the manifold, what this does is cause more air to pass through the engine than should be passing through. The upstream lambda sensor will detect the excess oxygen in this excess air, (air contains approx 23% oxygen), and will start making serious adjustments to your air:fuel ratio, through a closed loop feedback control instruction that is running on your ECU.

The ECU is programmed under conditions where there is too much oxygen in the inlet tract, to adjust this misture by adding more fuel, so more fuel is horsed into the combustion chamber by the ECU on the basis of air leaking into the inlet tract from outside the engine.

So you have a condition now where the engine is running rich and this causes as NCT fail on emissions. What can happen is that if you have a partially cracked pipe in a brake servo or on a MAP sensor that is half intact for the want of a better description, it can be intact for 2 minutes, then the engine gets a rev and the pipe twists slightly due to the engine moving and you get a load of air leaking in and you get all of what I've described above.

Post your emissions fail data up here for all the tests you did and I'll see if what I've said above can be fully reconciled with your particular fail problem.

Darragh29

gline wrote: »

was in with an exhaust place, no hole upto cat, possible hole beween cat and manifold, but he said manifold and cat and all piping between will have to be taken out and heatshield removed top be checked.

He said though that there mightent even be a hole, just that their might be, he also said, maybe change the lambda.

What do you guys think?

Anyone know someone who would remove manifold+ cat etc pretty cheap? Or will I just put in a new lambda?

Either there is a leak or there isn't. If you do the quick check I suggested above, you'll know straight away if you have a leak or not.

Don't change a lambda sensor unless you know there is a problem with it!!! They are not cheap and it is unlikely you have a problem with it as the item you are failing for is almost always due to a small leak into the exhaust system due to a hole or a bad joint between a back box and the pipe, or a small leak somewhere.

gline

the mechanic jacked it up and tested the whole exhaust system (the parts he could reach) and there were no leaks. He just said the only other place it could be is up beween the cat and manifold, but there is all heat shielding around it so he cant access it to check, he said get the manifold and cat and any piping in-between out and come back and he can check for leaks, so there is a POSSIBLE leak, but he cant tell, il post up exact NCT results in a minute

gline

Darragh29 wrote: »

Also OP, check the engine RPM that the lambda failed at. The NCT don't test for Lambda at low idle so on the high idle test, check the form to see what the engine RPM was when the lambda failed and post the RPM back up here...

it was at 2670rpm, result was 1.035 and pass is between 0.98-1.02

Darragh29

Just on a gut instinct, I'd be paying real close attention to that modified back box that was fitted, in particular the coupling between the back box and the exhaust pipe.

There could be a cracked manifold up at engine but I'd be inclined to make sure that every other possibility has been eliminated before getting into that, particularly given that your lambda fail is a borderline one.

BrianD3

Has the back box an oversized outlet? I'm wondering if that could have something to do with it.

Also, if you could post the CO and HC readings and the engine oil temperature that might give some clues.

Darragh29

gline wrote: »

no, literally started and went there (20min drive), hasnt been driven in 2 days.

I had the exhaust repaired a few days ago, there was a hole in the exhaust, but i got a new section welded in and removed the bit with the hole.

Another thing OP is I'd have driven the car well before going near an NCT centre with any car. One thing that the catalythic converter does is it recombines gases...

It performs oxidation, using O2 (oxygen) of CO (Carbon Monoxide), to give CO2, a harmless gas.

The thing is, it does nothing really until it has reached its operating temperature. The same is true for the lambda sensors, they all operate in open circuit mode until they warm up and start modulating your air:fuel ratio...

Given that your car was lying up for 2 days and only driven to the NCT centre, I'd bring it on a 30 minute hard drive and arrange with your mechanic beforehand to do an emissions test immediately afterwards when the engine is hot. Just make sure you have a legitimate fail here before you go paying someone to take out the manifold, etc.

gline

BrianD3 wrote: »

Has the back box an oversized outlet? I'm wondering if that could have something to do with it.

Also, if you could post the CO and HC readings and the engine oil temperature that might give some clues.

CO was 0.00% (below 0.3% is pass)
HC was 14ppm (below 200 is pass)
at 2670rpm

The backbox is larger than standard, but no leaks around it or at the joints

I was told to take out the air filter aswell before the test and that can help.

BrianD3

Your HC and CO are way below the limits, this supports the theory that fresh air getting into the exhaust and diluting the emissions. If your CO and HC had been on the high side with a lambda of 1.035 that would have been...puzzling.

Taking out the air filter is just adding possible variable to the mix and is not necessary. I'd put it back in and get the garage to go ahead with stripping down the exhaust system. If they think they've solved the problem then make sure that they do an exhaust gas check to confirm it.

Cosmo K

@Op

Don't get the cat or the lambda sensor changed, they are not the problem. I'd say, its your modified back box, thats causing all the trouble. Do you still have the original box?

gline

Cosmo K wrote: »

@Op

Don't get the cat or the lambda sensor changed, they are not the problem. I'd say, its your modified back box, thats causing all the trouble. Do you still have the original box?

no i dont, i bought it with it on.

how would that effect the lamda reading?

Bearcat

i've an old 88 911...no cat....and it walked the nct. I think after 20 yrs old the tolerances are much easier. ???