Failed CO emissions, 1.4 Petrol!

[cmcar08]

Hi all, I’m hoping to gain some knowledge from you guys!

 

My 2007 1.4 (Bud) petrol octavia failed 3 emissions tests the latest being CO at both idle and high revs. Lambda is just about in range at 0.98. HC reading well within limits now with the fuel cleaner through the system.

 

Here’s the annoying bit, it’s driving 100% and no fault codes present. It is smoking quite noticeablely when up to temperature though especially in this cold weather. Fuel consumption appears normal. service done, sparks, air filter , oil filter, oil, fuel filter too.

 

I have a small diagnostic tool and it shows preCAT O2 sensor at oscillating between 0.085-0.85V so I think that’s normal? Not sure what the average voltage though, I don’t have means to see it in graph form.

MAP sensor is showing around 28 hg in. With engine off and around 8-9hg in at idle, which I also think is normal? Not sure what it’s supposed to be at higher revs?

 

LTFT% is high at mostly at +33%idle but it intermittently changes , might be 26% today, 30 tomorrow. Then increasing revs it does usually reduce to less than 10%. Im sure whatever is causing the high idle % is a big part of the overall fault.

hard to Judge the STFT% as the values change so quickly. Anywhere between +\-25% at anytime at idle.

 

ETC reading appear fine and sensor

was Checked with multimeter. Fuel rail pressure is 4 bar so I think that’s normal and injectors are working fine also

 

Something I’m struggling to confirm, what is the normal range of the post CAT sensor supposed to be? Currently it generally limits around 0.55-0.77V. I would say 0.66V would be the average.

sometimes it did oscillate, from 0.1-0.8V, but that may be to me messing with other parts trying to find an issue elsewhere.

 

I tried my best to check for air leaks ( but I wouldn’t have the best tools to 100% confirm), we didn’t cure a very small leak between exhaust and CAT but didn’t make any difference for the last test. Plenty of fuel cleaner in the tank also now, so I’d say that’s as good as that going to be. We took off both CATs and both appears to be as clean as any second hand car would be.

Both O2 sensors and MAP sensors were removed and cleaned.

 

I have advise that it may be the ECU, but I find it hard to believe the ECU can resist throwing a fault code and keep the engine running well, while also sabotaging the fuel/ air ratio and emissions? Surely something one of the above with be off.

 

My personal thinking is it’s something wrong with the engine such as an air leak, oil or oil vapours getting back into the engine through a faulty PCB or Oil separator, burning oil or coolant (does need occasional topping up, just discovered a small radiator leak though also), that the sensors aren’t picking up properly, the ECU is just trying to compensate the problem by changing the fuel trim, is this possible?Problem is I don’t know how to test this without expensive and possibly needless trial and error part replacing?

[Warrior193]

Hello, welcome to the forum. You say there is noticeable smoke when up to temperature - does this look like oil burning ? (blue smoke) What mileage and how much oil lost in, say, 1000 miles?

[TheClient]

Yeah. as above I'm hoping and surely the noticeable smoke is....  condensation? 

[TheClient]

It could be as simple as a faulty lambda sensor, or worse poisoned or ineffective cat. 

[cmcar08]

16 hours ago, TheClient said:

Yeah. as above I'm hoping and surely the noticeable smoke is....  condensation? 

This was my initial thoughts when in the shed, it just appeared to be excessive white smoke, but during the NCT when reved, it looks a bit dark alright. Maybe it’s the difference in lighting.

 

id say I need to top up oil 4-5 times a year, so maybe 200ml over 10,000km maybe. Top up amounts are small volumes

The engine itself has only an oil capacity of 3.2l

[cmcar08]

15 hours ago, TheClient said:

It could be as simple as a faulty lambda sensor, or worse poisoned or ineffective cat. 

Can anyone confirm what voltage the lambda sensor is supposed to be? Is it be be really steady or slightly oscillating?

My understanding is that the postCAT 02 sensor (lambda) is a warning only to the dash of an ineffective CAT, but no warnings so far, and that this has no bearing on the ECU operation so something else is causing the high fuel trim?

 

I did visually inspect both CATs and no clogging or oil residue on either of them.

[TheClient]

On 17/12/2022 at 12:17, cmcar08 said:

I have a small diagnostic tool and it shows preCAT O2 sensor at oscillating between 0.085-0.85V so I think that’s normal? Not sure what the average voltage though, I don’t have means to see it in graph form

They do oscillate. 

 

https://www.hella.com/techworld/uk/Technical/Sensors-and-actuators/Test-lambda-sensor-4379/#:~:text=If the operating temperature of,0.1 V and 0.9 V.

 

I don't have enough experience to know if yours could still be the problem. Something is not right with Lamda staying over rich state as that could be the cause of the fail on CO and HC.

 

On 17/12/2022 at 12:17, cmcar08 said:

LTFT% is high at mostly at +33%idle but it intermittently changes , might be 26% today, 30 tomorrow. Then increasing revs it does usually reduce to less than 10%.

+33% seems really high. Adding a lot of fuel to compensate for vacuum leak as you suggest is one logical interpretation but experience could link elsewhere.

 

1 hour ago, cmcar08 said:

This was my initial thoughts when in the shed, it just appeared to be excessive white smoke, but during the NCT when reved, it looks a bit dark alright. Maybe it’s the difference in lighting.

 

id say I need to top up oil 4-5 times a year, so maybe 200ml over 10,000km maybe. Top up amounts are small volumes

The engine itself has only an oil capacity of 3.2l

200ml over 10k km is tiny.  

 

I'd say your dark smoke is unburnt fuel mixture (carbon) as that is what everything else is pointing towards.

 

How many miles has the car done.

 

Have you considered taking it to a trust worthy shop to diagnose?

If you want me to guess I'd probably try a new respected brand lanbda sensor as they don't last forever and it is a logical place to start but it is not based on experience with your other circumstances..

 

https://www.autosuccessonline.com/understanding-fuel-trim-theory-and-operation/#:~:text=Long-Term Fuel Trim (LTFT,10% to %2B10%.

 

https://www.motor.com/magazine-summary/fuel-trimming-diagnostic-time/#:~:text=Short-term fuel trim (STFT,for a perceived lean condition.

[J.R.]

There must be some confusion on units here, 4-5 top ups totalling 200ml means 40-50ml each time, you would need a very small graduated jug to measure that and you certainly would not be able to make out the rise in level on the dipstick from the addition of 40-50ml.

 

2 litres per year perhaps?

[cmcar08]

On 19/12/2022 at 13:45, J.R. said:

There must be some confusion on units here, 4-5 top ups totalling 200ml means 40-50ml each time, you would need a very small graduated jug to measure that and you certainly would not be able to make out the rise in level on the dipstick from the addition of 40-50ml.

 

2 litres per year perhaps?

You are correct maybe less than 1.5 lites (I just finished the 5litre can, 3.2 of which was used during oil change, I also used some topping up other cars too over the year.

[cmcar08]

On 19/12/2022 at 13:04, TheClient said:

They do oscillate. 

 

https://www.hella.com/techworld/uk/Technical/Sensors-and-actuators/Test-lambda-sensor-4379/#:~:text=If the operating temperature of,0.1 V and 0.9 V.

 

I don't have enough experience to know if yours could still be the problem. Something is not right with Lamda staying over rich state as that could be the cause of the fail on CO and HC.

 

+33% seems really high. Adding a lot of fuel to compensate for vacuum leak as you suggest is one logical interpretation but experience could link elsewhere.

 

200ml over 10k km is tiny.  

 

I'd say your dark smoke is unburnt fuel mixture (carbon) as that is what everything else is pointing towards.

 

How many miles has the car done.

 

Have you considered taking it to a trust worthy shop to diagnose?

If you want me to guess I'd probably try a new respected brand lanbda sensor as they don't last forever and it is a logical place to start but it is not based on experience with your other circumstances..

 

https://www.autosuccessonline.com/understanding-fuel-trim-theory-and-operation/#:~:text=Long-Term Fuel Trim (LTFT,10% to %2B10%.

 

https://www.motor.com/magazine-summary/fuel-trimming-diagnostic-time/#:~:text=Short-term fuel trim (STFT,for a perceived lean condition.

Hi, the cars supposedly at 230,000km (I recently Realised Even cheap scanners can alter the mileage though).

Im trying my best to avoid going to a specialist as my mechanic is a relative and will get in for if I go elsewhere at this point (prefer not to have an earful over the next 20 years or so, 🙄), not many  locally with a great rep either that won’t charge an arm and a leg.

hopefully I’ll get access to a higher grade scanner that I can share on the thread and see if anyone can make sense of it.

 

one question if someone can clarify, what is the voltage readings of the preCat O2 sensor compared to post CAT supposed to be?

If PreCAT is supposed to cycle 0.1-0.9v which it is, but my postCat O2 appears to be mirroring it with the same

range, would this not indicate the Cats doing noting to the emissions? But again no warning or error code present?

 

I’ve recently blocked of the PCV pipe to see if it was leaking air, but doesn’t appear to have changed anything regarding the smoke emissions.

 

 

[TheClient]

Downstream should be stable mid range voltage I believe.

 

The downstream isn't the main input to the ECU for fuelling trim / adjustments though. So may simply show the fault is elsewhere.

 

Some basics; is the car warming up ok?  Have you monitored the reported engine coolant temp with diagnostics (not the temp gauage). You've already replaced and checked air filter, spark plugs, what is their condition now?

 

It certainly wouldn't be completely unusual to have to replace the CC inside 230k Kms.

 

Even at 1L to 1.5l oil per 10k Kms, that would not be a concern for cat poisoning I don't think. Or visible burning oil in exhaust smoke.

 

Probably not the place to comment publicly on altered odometers but lets say it has at least 230k KMs. I have no experience, but didn't think it would be that easy, unless you've made a career out of such things.... I'm sure it is possible with the "right" equipment and software...

 

• https://www.autoserviceprofessional.com/articles/7599-easy-air-fuel-and-oxygen-sensor-diagnosis#:~:text=Post-cat oxygen sensors%2C when,some effect on fuel control.

"

For post-catalytic converter oxygen sensors used for fuel control:

Post-cat oxygen sensors, when good, feature a steady voltage usually between 500 to 700 mV. If it zigzags, the catalytic converter is highly suspect.

On some vehicles the rear sensor does have some effect on fuel control. For our purposes, it’s just good to know that when testing the sensor, the voltage should go up when the fuel mixture is rich and should go down when it is lean. Sadly there is no way of generically knowing what is an optimal post-cat oxygen sensor voltage. It differs by the manufacturer."

[cmcar08]

On 22/12/2022 at 09:40, TheClient said:

Downstream should be stable mid range voltage I believe.

 

The downstream isn't the main input to the ECU for fuelling trim / adjustments though. So may simply show the fault is elsewhere.

 

Some basics; is the car warming up ok?  Have you monitored the reported engine coolant temp with diagnostics (not the temp gauage). You've already replaced and checked air filter, spark plugs, what is their condition now?

 

It certainly wouldn't be completely unusual to have to replace the CC inside 230k Kms.

 

Even at 1L to 1.5l oil per 10k Kms, that would not be a concern for cat poisoning I don't think. Or visible burning oil in exhaust smoke.

 

Probably not the place to comment publicly on altered odometers but lets say it has at least 230k KMs. I have no experience, but didn't think it would be that easy, unless you've made a career out of such things.... I'm sure it is possible with the "right" equipment and software...

 

• https://www.autoserviceprofessional.com/articles/7599-easy-air-fuel-and-oxygen-sensor-diagnosis#:~:text=Post-cat oxygen sensors%2C when,some effect on fuel control.

"

For post-catalytic converter oxygen sensors used for fuel control:

Post-cat oxygen sensors, when good, feature a steady voltage usually between 500 to 700 mV. If it zigzags, the catalytic converter is highly suspect.

On some vehicles the rear sensor does have some effect on fuel control. For our purposes, it’s just good to know that when testing the sensor, the voltage should go up when the fuel mixture is rich and should go down when it is lean. Sadly there is no way of generically knowing what is an optimal post-cat oxygen sensor voltage. It differs by the manufacturer."

Just getting the motivation to get back to this now over the next few days. 

The car is warming up fine at idle/low gears, but drops off to 60-75degC in 4-5th gear driving (I think the gearbox is not the original and may have a higher gear ratio, as speed cameras always clock a higher speed the the dash gauge, so there must be more air cooling of the engine than it anticipates).

Engine Coolant temp will reach 90degC as seen on the scanner, and as reported by the NCT test sheet also. I do notice however the dash gauge will reach 90degC faster though (ETC may be around 60-70degC while the gauge shows 90degC, but ETC will continue to rise to 90degC and start the fan when needed.

 

The is oil over spilling small amounts into the air filter from the top pipe off the engine, this was noticed when the pipe was off and the engine reved up, not sure if that's normal. Will need to check sparks again, but I would expect its wet with petrol as the LTFT is +38%, I will check later.

 

I attach some live data of the O2 readings. So O2S1 is alternating as expected approx 0.1-0.85V which seems fine and looks to be well centered around the 0.45V mark, so I cant see this requesting +37.5% Long term fuel trim (LTFT). There's a separate "Lambda Control" figure that I suspect is the lambda instruction to fuel trim, and at idle doesn't go above +/-4%, which if true would suggest the O2S1 is doing its job.

 

(Just as an overall update, the LTFT is approx 37-39% at idle, and drops to 9% at 1600rpm, down to 0% at 2500rpm currently, theirs no fault codes present, which is surprising as I though 25% is the limit for LTFM, in terms of driving, its going 100% with no signs of an fault apart from the obvious volume of smoke from the exhaust, the PCV to Oil separator broke over the holidays is now blanked of temporarily, new battery installed also)

 

What could alter the LTFM at idle apart from the O2S1 itself on a MAP only engine? I've read that a MAP only engine's LTFM doesn't get affected by air leak on the intake side, besides a high idle (which isn't occurring as the idle is at it usual level), Fuel pressure is checked to be within normal range. 

I can check the MAP readings which is returning 25kPA at idle, but I don't know what the target would be for 1500rpm/2500rpm?

 

02S2 is between 0.55-0.75V approx, average is likely 0.68V. Its zig-zagging a bit but that probably due to the sample rate of the scanner? Not sure what the Skoda spec for this is? Would anyone know if this indicates a good or bad CC1 or CC2?

For CC1/CC2 I can see 2no temp readings of ranging from 484-750degC (ish) depending on the RPM and run time (PDF attached). The second CC seems to be around 200degC less at any given time, I assume the higher temp is the CC closest the engine, its not well labelled. Would this be an indicator of CC condition?

CC Temp.pdf

[TheClient]

I'll read your data and questions a few more times. Maybe I'll come up with someting....

 

But you are asking in great depth for my experience on this. In my past experience I've resolved with obvious checklist things.    An obvious observation from your latest update is that temperature, in higher gears, will surely cause over fuelling...  But I note your comments that it is 90C at idle and lower gears and presumably when running emissions tests?.  Even if the gearbox has been changed and ratios altered, the thermostat should still be regulating the temp properly. I suspect the thermostat is faulty and may be unreliable.  I'm not sure on the style of thermostat in the BUD.  Is it an easy change?

 

What is your outside ambient temperature at the moment when you see temp drop to 60-75C while driving?  

 

A workshop manual (aftermarket or VW Erwin you can pay for 1 hour access online) "may" help with exactly what voltages to expect on the diff O2 sensors. it may give some other testing procedures, maybe...

 

The dash temp gauage in most VW group cars is usually Bravo Sierra. So I'm not surprised it reports 90C even when temp is lower. As long as the EMC is seeing the actual coolant temp.

 

Presumably the diagnostic temp sensor you are seeing with the scanner is the reading being presented to the EMC. So unless it is too low, like when dropping in high gears, that should not be the cause of the over fuelling when engine is at 90C...

 

 

 

 

[cmcar08]

On 11/01/2023 at 16:59, TheClient said:

I'll read your data and questions a few more times. Maybe I'll come up with someting....

 

But you are asking in great depth for my experience on this. In my past experience I've resolved with obvious checklist things.    An obvious observation from your latest update is that temperature, in higher gears, will surely cause over fuelling...  But I note your comments that it is 90C at idle and lower gears and presumably when running emissions tests?.  Even if the gearbox has been changed and ratios altered, the thermostat should still be regulating the temp properly. I suspect the thermostat is faulty and may be unreliable.  I'm not sure on the style of thermostat in the BUD.  Is it an easy change?

 

What is your outside ambient temperature at the moment when you see temp drop to 60-75C while driving?  

 

A workshop manual (aftermarket or VW Erwin you can pay for 1 hour access online) "may" help with exactly what voltages to expect on the diff O2 sensors. it may give some other testing procedures, maybe...

 

The dash temp gauage in most VW group cars is usually Bravo Sierra. So I'm not surprised it reports 90C even when temp is lower. As long as the EMC is seeing the actual coolant temp.

 

Presumably the diagnostic temp sensor you are seeing with the scanner is the reading being presented to the EMC. So unless it is too low, like when dropping in high gears, that should not be the cause of the over fuelling when engine is at 90C...

 

 

 

 

I did a a couple of flight recordings while driving to work;

intake temperature was 8degC at turn on so that would be the ambient outside air temp.

during driving it reaches 20degC Intake air temp.

With ETC at 84degC (stationary idle)

LTFT was 39%, MAP was 24kPa.

O2 was oscillating similar to the previous graph upload. Does the above look correct apart from the LTFT figure itself? Are these the only factors that affect fuel trim possibly resulting in overfuelling/ high CO at idle or am I missing something?
 

You are correct NCT emissions test takes the ETC reading and can only be carried out at above 80degc.

 

regarding the 60ish degC running temp at higher speeds, I assume the additional airflow is what’s cooling the engine rather than the coolant being sent through the radiator by the stat? Again assuming the gear ratios are higher than programmed to the ECU.

I’m not overly concerned with this for now as it’s not at factor while stationary testing during NCT.

 

Not sure if it helps troubleshoot this but increasing revs at stationary through to 2600rpm , the MAP only drops to a low of 21kPa, however fuel trims reduce down to acceptable levels by this point so I assume Map pressure isn’t actually much of a factor until driving the car on under load with respect to LTFT? I am aware that the map also takes the IAT reading, but if the above looks reasonable for IAT + Map at idle can I rule that out as a potential remedy?

[cmcar08]

Car has since been fixed and tested, there was a combonation of clogged Cat and worn piston rings.

 

I changed to a 2nd hand cat and the pretty much fixed the fuel trims but the car still failed emissions, im guessing thisnl was also caused by the worn piston rings. Fixed the piston rings and emissions are now almost 0.01% vol.

 

Fyi, a tip to diagnose worn piston rings was to start measuring emissions at a low engine temp, and if the figures start out ok, and worsen as the temp rises thats a sign that increasing volumes of oil is innthe exhaust as the oil temp increase. Mine was passing NCT limits below 75degC but above got worse, at 90degC it was well beyond the test limits.

[T2000]

2 hours ago, cmcar08 said:

Car has since been fixed and tested, there was a combonation of clogged Cat and worn piston rings.

 

I changed to a 2nd hand cat and the pretty much fixed the fuel trims but the car still failed emissions, im guessing thisnl was also caused by the worn piston rings. Fixed the piston rings and emissions are now almost 0.01% vol.

 

Fyi, a tip to diagnose worn piston rings was to start measuring emissions at a low engine temp, and if the figures start out ok, and worsen as the temp rises thats a sign that increasing volumes of oil is innthe exhaust as the oil temp increase. Mine was passing NCT limits below 75degC but above got worse, at 90degC it was well beyond the test limits.

When you first mentioned smoke I thought of valve stem seals, a compression test would have possibly picked up the rings.