Cav 1.8 Emissions Nightmare
Discussion
GrannieTwoEight said:
To be honest I have only asked them to test the emissions and haven't asked them to do any other fault finding, so really all they can do is cast a second eye over it and see if they can spot anything. What they would do is more or less do what I have already done except stuff like vac tests. What he said today was that he can't plug this car up to a computer to test anything like a more modern car.
Can't comment on specific value for lambda but i'm not going to dispute 2 MOT testers both telling me it is too high.
The lambda figure you have quoted is not high. It is impossible. If the car failed, surely they have given you a sheet listing the failure along with the test results ?Can't comment on specific value for lambda but i'm not going to dispute 2 MOT testers both telling me it is too high.
So some facts are getting thrown out the window between testing, and this forum.
And there is no point taking the car to a garage and asking them to test emissions, but undertake no form of diagnostic work whatsoever.
That's like going into a hospital with a broken arm and asking them to look at it, but not actually do anything about it.
As for plugging it in etc...Obviously this mechanic you have taken the car has no clue whatsoever. And that's being polite. Stop to think what mechanics did before cars had some form of self diagnostics ?
Yes, they tested things manually. Computers dont fix cars, competent mechanics do.
But without some proper factual information relating to the problem, nobody anywhere will ever be able to fix it. So either that is relayed correctly so you might have a slim chance of a remote fix, or you find a competent mechanic that can conduct tests and diagnose the car properly. It really shouldnt be rocket science for any decent mechanic.
Or just sell the car.
what stevieturbo said.
Post a pic of the emissions sheet.
The only way "plugging it in" is going to help is if you find someone with tech2 and experience of the engine, they can read live data from the car and use their experience to tell you what the data means and hopefully diagnose something.
Have you put new plugs in it yet ? even cleaned and gapped the old ones ?
does it use any oil ?
Post a pic of the emissions sheet.
The only way "plugging it in" is going to help is if you find someone with tech2 and experience of the engine, they can read live data from the car and use their experience to tell you what the data means and hopefully diagnose something.
Have you put new plugs in it yet ? even cleaned and gapped the old ones ?
does it use any oil ?
stevieturbo said:
The lambda figure you have quoted is not high. It is impossible. If the car failed, surely they have given you a sheet listing the failure along with the test results ?
So some facts are getting thrown out the window between testing, and this forum.
And there is no point taking the car to a garage and asking them to test emissions, but undertake no form of diagnostic work whatsoever.
That's like going into a hospital with a broken arm and asking them to look at it, but not actually do anything about it.
As for plugging it in etc...Obviously this mechanic you have taken the car has no clue whatsoever. And that's being polite. Stop to think what mechanics did before cars had some form of self diagnostics ?
Yes, they tested things manually. Computers dont fix cars, competent mechanics do.
But without some proper factual information relating to the problem, nobody anywhere will ever be able to fix it. So either that is relayed correctly so you might have a slim chance of a remote fix, or you find a competent mechanic that can conduct tests and diagnose the car properly. It really shouldnt be rocket science for any decent mechanic.
Or just sell the car.
Well it looks like I missed a decimal point in the values I quoted for lambda. The sheet I have from the original fail sheet is:So some facts are getting thrown out the window between testing, and this forum.
And there is no point taking the car to a garage and asking them to test emissions, but undertake no form of diagnostic work whatsoever.
That's like going into a hospital with a broken arm and asking them to look at it, but not actually do anything about it.
As for plugging it in etc...Obviously this mechanic you have taken the car has no clue whatsoever. And that's being polite. Stop to think what mechanics did before cars had some form of self diagnostics ?
Yes, they tested things manually. Computers dont fix cars, competent mechanics do.
But without some proper factual information relating to the problem, nobody anywhere will ever be able to fix it. So either that is relayed correctly so you might have a slim chance of a remote fix, or you find a competent mechanic that can conduct tests and diagnose the car properly. It really shouldnt be rocket science for any decent mechanic.
Or just sell the car.
natural idle values:
CO = 0.13 - Pass
Fast idle test:
CO = 0.99 - fail
HC = 99 - pass
Lambda = 1.035 - fail
2nd fast idle:
CO = 0.89 - fail
HC = 90 - pass
Lambda = 1.035 - fail
On this point Stevie:
And there is no point taking the car to a garage and asking them to test emissions, but undertake no form of diagnostic work whatsoever.
I wanted a 2nd opinion - as I said - because I have had this problem before with another garage, their gas tester wasn't working and I was cleaning plugs and changing oil and filters only for him to ring me up and say they found a fault with it and to come back in for testing.
Edited by GrannieTwoEight on Thursday 29th November 22:48
Edited by GrannieTwoEight on Thursday 29th November 22:50
Crafty_ said:
what stevieturbo said.
Post a pic of the emissions sheet.
The only way "plugging it in" is going to help is if you find someone with tech2 and experience of the engine, they can read live data from the car and use their experience to tell you what the data means and hopefully diagnose something.
Have you put new plugs in it yet ? even cleaned and gapped the old ones ?
does it use any oil ?
Uses no oil and plugs are newPost a pic of the emissions sheet.
The only way "plugging it in" is going to help is if you find someone with tech2 and experience of the engine, they can read live data from the car and use their experience to tell you what the data means and hopefully diagnose something.
Have you put new plugs in it yet ? even cleaned and gapped the old ones ?
does it use any oil ?
So the C18NZ (I'm presuming its that engine) has a simple management system, no MAF, it uses manifold pressure and lambda to set the fuelling.
Usually if the map sensor is faulty the car would be undriveable - poor power delivery, hesitation etc.
Have you had a good look for air leaks ? pipes, gaskets and so on ?
It might be worth first of all unplugging the lambda sensor and see what the ecu does in default mode. Unplugging will throw a light, but we aren't too interested in that. If it starts behaving I'd suggest there is an air leak somewhere (thus the readings from lambda are inaccurate and causing overfuelling).
It would be worth checking the operation of the map sensor, you should find out which pins to check and the expected values. I don't have the data to hand but I googled and found "0.5v at idle to 3.7v full chat across the outer pair."
Apparently the same map sensor is used in volvos and renaults too.
I'd do the same with TPS, again this would usually result in poor running, but I'd want to figure out whats good and whats bad.
I don't think things like catclean are worth messing with.
Problem with all of this is you need a gas analyser to fault find. It might be worth finding a local expert and have them diagnose. Found this, just for info
Usually if the map sensor is faulty the car would be undriveable - poor power delivery, hesitation etc.
Have you had a good look for air leaks ? pipes, gaskets and so on ?
It might be worth first of all unplugging the lambda sensor and see what the ecu does in default mode. Unplugging will throw a light, but we aren't too interested in that. If it starts behaving I'd suggest there is an air leak somewhere (thus the readings from lambda are inaccurate and causing overfuelling).
It would be worth checking the operation of the map sensor, you should find out which pins to check and the expected values. I don't have the data to hand but I googled and found "0.5v at idle to 3.7v full chat across the outer pair."
Apparently the same map sensor is used in volvos and renaults too.
I'd do the same with TPS, again this would usually result in poor running, but I'd want to figure out whats good and whats bad.
I don't think things like catclean are worth messing with.
Problem with all of this is you need a gas analyser to fault find. It might be worth finding a local expert and have them diagnose. Found this, just for info
It really is a borderline fail ( from what sounded like a huge fail at the start of the thread )
Again, whether you have replaced the lambda sensor or not. It's operation needs tested properly.
The high lambda value, which in some respects does indicate lean...although contradicted by the CO reading, there could well be a small air leak in the exhaust.
Again, any half decent mechanic should see this on a gas analyser. o2 readings would be higher than expected if there is an air leak.
But as Ive been saying, all that is needed is some basic diagnostic work and tests to determine if things are working as they should, or if there may be a problem.
It doesnt need any super fancy diagnostic equipment, although an oscilloscope would be very helpful.
Again, whether you have replaced the lambda sensor or not. It's operation needs tested properly.
The high lambda value, which in some respects does indicate lean...although contradicted by the CO reading, there could well be a small air leak in the exhaust.
Again, any half decent mechanic should see this on a gas analyser. o2 readings would be higher than expected if there is an air leak.
But as Ive been saying, all that is needed is some basic diagnostic work and tests to determine if things are working as they should, or if there may be a problem.
It doesnt need any super fancy diagnostic equipment, although an oscilloscope would be very helpful.
stevieturbo said:
It really is a borderline fail ( from what sounded like a huge fail at the start of the thread )
Again, whether you have replaced the lambda sensor or not. It's operation needs tested properly.
The high lambda value, which in some respects does indicate lean...although contradicted by the CO reading, there could well be a small air leak in the exhaust.
Again, any half decent mechanic should see this on a gas analyser. o2 readings would be higher than expected if there is an air leak.
But as Ive been saying, all that is needed is some basic diagnostic work and tests to determine if things are working as they should, or if there may be a problem.
It doesn't need any super fancy diagnostic equipment, although an oscilloscope would be very helpful.
The values posted were from first test, The CO values were triple what they should have been on the 2nd fast idle, although since I done all the engine work it is down to about 0.4 so it's really borderline now.Again, whether you have replaced the lambda sensor or not. It's operation needs tested properly.
The high lambda value, which in some respects does indicate lean...although contradicted by the CO reading, there could well be a small air leak in the exhaust.
Again, any half decent mechanic should see this on a gas analyser. o2 readings would be higher than expected if there is an air leak.
But as Ive been saying, all that is needed is some basic diagnostic work and tests to determine if things are working as they should, or if there may be a problem.
It doesn't need any super fancy diagnostic equipment, although an oscilloscope would be very helpful.
I was reading that the lambda switch on/off should be measured, which as you mentioned, I would need a scope. It's not something I can put my hands on and will see if I can borrow one, or ask the local car electrician if he has one.
Apologies the reading below was incorrect when I posted above:
CO = 0.13 - Pass
Fast idle test:
CO = 0.72 - not 0.99
HC = 99 - pass
Lambda = 1.035 - fail
2nd fast idle:
CO = 0.89 - fail
HC = 90 - pass
Lambda = 1.035 - fail
Edited by GrannieTwoEight on Friday 30th November 15:25
Edited by GrannieTwoEight on Friday 30th November 15:26
stevieturbo said:
At the minute you're pissing into the wind and just changing parts at random and for no good reason, and with no ability to see if they make a change.
Find a competent mechanic.
But as Puma says, everything is still pointing to a small air leak in the exhaust
Sound advice.Find a competent mechanic.
But as Puma says, everything is still pointing to a small air leak in the exhaust
GrannieTwoEight said:
I've spoke to 4 mechanics and not one has said anything about a leak in the exhaust but if that is what you think it is then I will ask the guy tomorrow to specifically look for an exhaust leak when it's up in the air.
Unfortunately, the term mechanic covers a wide range of people.From someone who reads about cars in a magazine and then thinks they can repair them, to those who actually can repair them. Filtering out the competent ones from the incompetent is the hard part.
But maybe the crap mechanics call those who can diagnose and repair something different too.
You need to find a competent one, but then it may also cost more to pay this person. And if they dont have a proper 4/5 gas analyser, you're wasting your time taking it to them, find somewhere that can carry out testing and diagnosis properly.
GrannieTwoEight said:
I've spoke to 4 mechanics and not one has said anything about a leak in the exhaust but if that is what you think it is then I will ask the guy tomorrow to specifically look for an exhaust leak when it's up in the air.
LOL @ mechanics. I wouldn't trust one of those to diagnose his way out of a paper bag.The CO value indicates whether the air/fuel mixture that IS ACTUALLY BEING BURNED is rich or weak. Ideally it's close to zero when the engine is running at the correct stoichiometric 14.7:1 A/F ratio.
The lambda value is a calculated value from the measurements of CO, CO2 and O2 in the exhaust according to the Brettschneider equation developed by Bosch and indicates whether the exhaust gas contents are weaker than stoichiometric (value greater than 1.0) or richer than stoichiometric (value less than 1.0). An engine running rich should have high CO, low CO2 and low O2 in the exhaust and a lambda value of less than 1. An engine running weak should have low CO, high CO2 and high O2 and a lambda value of greater than 1.
You have a rich CO reading but a weak lambda value which is impossible if there are no mechanical faults. The two values should match. Rich CO should = lambda less than zero and vice versa. That means you MUST have high O2 in the exhaust gases to generate such a high calculated lambda value from the given CO measurement.
The reason is that the exhaust is leaking air in and the O2 content in this is tricking the O2 sensor into thinking the engine is running weak. It therefore injects more fuel than really needed which raises the actual burned A/F mixture CO value above zero. Meanwhile the excess oxygen from the exhaust leak which just passes through the exhaust system unchanged leads to a calculated lambda value above 1.0 even though the high CO reading without any leaks should generate a lambda value below 1.0.
Simples. A complete no-brainer for anyone other than - well the average garage mechanic.
MGJohn said:
Very informative thread and a good read.
There's still a loose end to be tied though.
How did the car finally fare with those MoT emissions tests ?
Ha, 6 odd years later and this car is still going strong would you believe. I just got pissed off and ended up taking it to a mate who said he would "pass it" the last 6 years, but then this year I've had a bit of a rethink, I did want to get to the bottom of what was causing this and had another look at it. Firstly I discovered a huge rip in the air intake pipe that goes to the carb, I should think that likely wouldn't make a difference to un-metered air getting into the engine as it does on likes of the 2.0 cavy which uses the motronic and air-flow sensor, but it seems to have made a difference to the running of the car. There's still a loose end to be tied though.
How did the car finally fare with those MoT emissions tests ?
I also changed the idle control stepper valve as the revs seemed to hunt up and down. This has more or less cured that. Then I realised that I'd never changed the rotor arm, possibly never been changed since new, and I might have changed the dizzy cap way back in 2008, so plummed and got a new bosch one of each. this has made the actual running of the car much better. Also have changed the 02 sensor for a genuine new one.
I also fixed the back box and clamp which was poor fitting and the box was rusty; i'd patched welded it before but it was swinging like the dogs danglies in the wind. I also noticed that some of the back boxes I had got over the years has a little drain at the back, some didn't, and they were all VX parts. I reckoned maybe the ones that had a drain were pre-cat and so if I fitted one with a drain it would maybe have upped the Lambda. Although in saying that my Omega has a drain hole that I put in my self on all 3 silencer boxes and it still passed with flying colours.
The car is going through a bit of a refurb at the moment so it will be going for MOT soon and we will find out what happens. I do have a good spare cat that I can utilise as a last ditch attempt if it fails again, the cat is quite old now and was aftermarket so it might now be goosed.
Edited by GrannieTwoEight on Wednesday 30th May 19:05
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff