Failing emissions Test even with a replaced Cat? Diagnose?

[Mr2Owner]

My Mr2 is failing the emissions test for the Nct. (I'm in Ireland and my 94 Mr2 needs to pass emissions here).

The original failed readings were
Low idle = CO 3.76 vol% Above 0.50% is a fail
High idle = CO 4.92 vol% Above 0.30% is a fail
HC 193 ppm Above 200 is a Fail. (So this figure passed)
Lambda = 0.96 Not between 0.97 and 1.03 is a fail (So this only failed by 0.01)

I removed the Cat and it was broken, the element was loose and broken. I fitted a 2nd hand cat which looked in good condition and now the Low idle is passing but the high idle is still failing up at 3.30 and even as high as 4.0.

Can anyone advise on what my next move is, maybe the 2nd hand cat is also broken?
The car is driving fine and doesn't seem smokey or to be burning oil. Could it be a sensors causing the fault?
How do I check if the car is running too rich?
The previous owner replaced the Lambda sensor in the tiurbo Downpipe elbow. Theres only 1 Lambda sensor, correct? And only 1 Cat, correct?

Any tips would be appreciated here before it turns into a money pit.

[synXero]

The most often given piece of advice is find an MOT station who are willing to be helpful, by allowing you to get it tested the minute you turn up. Take the car out and get it hot. With both of my MR2s I always needed this in order to get them through. TBH I was always struggling but you get there in the end

[bobhatton]

The Lambda sensor needs to be a Toyota one, not a Chinese one from ebay

[MartinF]

The Lambda sensor needs to be a Toyota one, not a Chinese one from ebay

As above, it looks like your Lambda sensor is on it's way out. But don't tell Bob they're all manufactured in China even the Toyota ones.

[jon]

It's worth checking the lambda sensor first, http://mr2.ie/mr2/bgb/mk2/mechanical/34.html

If you have to replace the lambda sensor, get a Denso one, they're the OEM for Toyota. Same part as Toyota would sell but cheaper!

[Mr2Owner]

I think the replacement Lambda was a decent one, I'll have another look at it and see what brand it is. But why does the lambda look suspect from the figures above, its only out of range by 0.01 Does this not suggest the Lambda is sensor itself is fine?

Also, I got the car tested the 2nd time in a friendly garage I know, the mechanic plugged his emissions tester into my car the minute I arrived, didnt even switch the engine off so it was nice and hot and still failed a mile over.

Does a high idle figure point towards the cat?

[jon]

Your HC, CO and Lambda exhaust readings all point to a rich mixture.

Normally that's due to a lambda sensor, air or coolant temperature sensor fault.

Even without a catalytic converter fitted, CO exhaust readings should be much lower, around 0.5%.

[Mr2Owner]

Your HC, CO and Lambda exhaust readings all point to a rich mixture.

Normally that's due to a lambda sensor, air or coolant temperature sensor fault.

Even without a catalytic converter fitted, CO exhaust readings should be much lower, around 0.5%.

So you reckon its running rich from the readings, cheers, I'll start with that basis then so.

I also printed out that link above from "Jon" on how to test the Lambda so I'll try that too.

One thing I noticed which I doubt has any bearing on this problem, but I noticed the fan beside the intercooler is on constantly, even with the engine off and just the ignition on full. Should it be like this?

Heres a pic of the Lambda, I can't make out what make it is, its a 4 wire one and looks new so looks like the previous owner did replace it as he told me. Just said i'd throw up a pic in case someone can identify it.

Edit: The forum wont let me post url links yet, how many posts do you need before I can post pictures?


Edit: Looking at the procedure linked to above for testing the Lambda sensor, is it correct to say I back-probe the terminals to get the readings leaving the connector plugs connected??

[jon]

Looking at the procedure linked to above for testing the Lambda sensor, is it correct to say I back-probe the terminals to get the readings leaving the connector plugs connected??

E1 & VF1 mentioned in the test procedure are terminals in the diagnostic connector in the engine bay.

[Kiddo]

Firstly, is your car standard or modified? Also, when the engine's warm, what temp is the temperature gauge indicating?

You say it's a '94 so that should be a rev3 and therefore have a MAP sensor instead of a MAF just to rule that out.

Lambda is all about the amount of oxygen required for a perfect burn and the actual measured amount of oxygen, so a reading of 0.96 would suggest the car is running slightly rich but you mention the cooling fan is constantly on...? If the coolant temp sensor was goosed then that would explain it running rich (thinks the engine's cold) but then why would the fan be on...? If you can measure the resistance across the coolant temp sensor when the engine's cold and up to temperature then someone on here might be able to tell if the figures look ok.

[Mr2Owner]

Firstly, is your car standard or modified? Also, when the engine's warm, what temp is the temperature gauge indicating?

You say it's a '94 so that should be a rev3 and therefore have a MAP sensor instead of a MAF just to rule that out.

Lambda is all about the amount of oxygen required for a perfect burn and the actual measured amount of oxygen, so a reading of 0.96 would suggest the car is running slightly rich but you mention the cooling fan is constantly on...? If the coolant temp sensor was goosed then that would explain it running rich (thinks the engine's cold) but then why would the fan be on...? If you can measure the resistance across the coolant temp sensor when the engine's cold and up to temperature then someone on here might be able to tell if the figures look ok.

The car is slightly modified from previous owner, but I was actually just thinking of putting it back standard before tackling this emissions problem.

It has an aftermarket air intake exposed filter and a blow off valve and some hoses of some description. Not sure if the blow off valve is vented to atmosphere or not but I think I'll just look for the standard parts and put it the way it should be.

Its got 3" twin backboxes and the exhaust is standard apart from the back boxes I think. But I'm leaving the exhausts the way they are.

The fan thats on constantly is the intercooler fan, not the rad fan, so would a faulty coolant temp sensor make that fan run or more likely be the radiator fan in the frunk run? The car temp gauge sits in the middle once warmed up.

[Nic]

i'd also check the gasket that seals the oxygen sensor to the down pipe, if it's leaking (or there are any leaks in the area) it could be effecting the sensor reading, causing it to run rich.

[abovetherim]

The engine bay fan will run constantly if the temp sensor attached to the engine lid is either unplugged or goosed.

When hot the temp sensor for the ecu should read 1.05 - 1.35 KOhms i think.

[Mr2Owner]

I got a chance to have another look at the car today, I checked around the Lambda sensor as suggested for a leak and it seems ok.
Also as suggested, I checked the air temp sensor at the engine bay cover and its connected but the wiring may be a little damaged/exposed at the connector plug, I might patch it up and see.

I also removed and fitted new spark plugs, the old plugs were Denso K16R-U11. The new ones I got from Toyota were Denso PK20R8. These new ones are correct yeah?

Heres a pic of the old ones, I don't know how long they were in the car though.

[Mr2Owner]

I'm going to test the Lambda sensor, from reading the procedure linked to earlier in my thread, it states to count the number of tims the multimeter needle fluctuates.

Do I need an old style multimeter to do this, or can my digital multimeter be used to count the fluctuations, if yes, what will I be looking for on the digital meter? Will it fluctuate from like 12v to 4v a few times or such? Note: I just picked 12v and 4v as an example, does anyone know the exact volts Ill be looking for to fluctuate between?

Here's a pic of the current Lambda, looks fairly newish but I'm unsure of its brand? It doesn't seem to be leaking at its gasket either.

[MartinF]

You will need an analogue multimeter with a needle.

[Mr2Owner]

A little update on this, it failed the NCT (Irish version of MOT) again.

I got the car tested in another exhaust centre place and it passed the readings in that place. So I brought it to the official NCT test centre today and it failed again.

Here's the original NCT readings:
The original failed readings were
Low idle = CO 3.76 vol% Above 0.50% is a fail
High idle = CO 4.92 vol% Above 0.30% is a fail
HC 193 ppm Above 200 is a Fail. (So this figure passed)
Lambda = 0.96 Not between 0.97 and 1.03 is a fail

Here's todays Nct's readings: (Only difference to car was new 2nd hand CAT fitted and new spark plugs and an oil service.)
Low idle = CO 0.08 Above 0.50% is a fail (So this low idle passed)
High idle =CO 1.51 Above 0.30% is a fail
HC 63ppm Above 200 is a fail (so this figure passed again)
Lambda 1.19 (Not between 0.97 and 1.03 is a fail)

So basically, the figures are well down compared to the original test and it now passes the low idle and only fails the high idle and the lambda.

Is it strange that the lambda reading is now failing on the other end of the scale? Is the lambda reading suggesting the car is now running lean?

Also, I've never had the engine management light on since I bought the car 2 months ago, but the NCT tested said it was on when he was doing the emissions test. I had the car fully warmed up before the test and he done the test straight away whilst it was warm.

I havn't done the Lambda test you folks suggested above as I don't have an anolougue multileter but I plan to get 1 the weekend.

Are the above readings still pointing towards a faulty Lambda sensor?

Any help appreciated as its starting to become very annoying

[Mr2Owner]

Anyone got an opinion on the 2nd set of emissions readings?

[Mr2Owner]

I got the car looked at today in a mechanics garage, the exhaust system doesn't seem to be blowing and the Lambda sensor seems to be working, was tested by being disconnected and watching it on the emissions tester machine.

Not sure what to do next, I think ill fit a genuine Denso lambda sensor and see if that corrects things. I'm trying to put the car back to standard too regards the air box and blow off valve.

What does the readings below suggest is wrong,, ie: Cat / Lambda sensor / exhaust leaks, etc...?

Low idle = CO 0.08 Above 0.50% is a fail (So this low idle passed)
High idle =CO 1.51 Above 0.30% is a fail
HC 63ppm Above 200 is a fail (so this figure passed again)
Lambda 1.19 (Not between 0.97 and 1.03 is a fail)

[SonicSW20]

High lambda reading means that the gas mixture is lean. 1.19 would be *very* lean. To me, this suggests you've got a leak somewhere - extra air in the exhaust system is causing the gas mixture in the exhaust to read lean. This does *not* mean the engine itself is running lean. If the leak is upstream of the lambda sensor, then this will throw off the ECU's lambda reading, and therefore throw off the fuelling. So, this could be linked to your high CO problem.

Find the leak!