Hi Screwloose,
Let me start again.
At the MoT yesterday the car failed for - 'Exhaust emissions carbon monoxide content after 2nd fast idle excessive (7.3.2b)' - the garage man suggested changing the Lambda sensor!
Thinking about it on the way home I wondered how this sensor would be able to change the reading! What do you think, what should I be thinking of doing to bring the reading down?
|
Doing some digging I found this online:
'If a vehicle fails for high CO at idle and /or fast idle then.... if Lambda is too low, the mixture is too rich. This can be caused either by a misfire or by a faulty Lambda sensor. If it's running very rich, the 02 and HC will be higher than normal.'
The car does have a slight flat-spot, would this affect the reading?
The readings on the Crypton Gas Analysis are: CO 0.67% vol (the max should be .20%); HC 76ppm; Co2 13.93% vol; O2 1.01% vol; Lambda 1.02; Oil temp 17C.
Perhaps it just needs a good tune-up, I had a new exhaust/cat fitted a fortnight ago by the way.
|
OK; got some good 4-gas data there; CO2 is a little low and the O2 is a little high. Was it taken at natural idle or fast idle?
Let's start with the engine code [there not being a 1.2 Saxo.] HDX? KFW? Something like that. [Should be in the VIN number.]
Check the plug on the oxygen sensor for corrosion and, if you've a meter, read the voltage on the black wire of the sensor - plug connected at idle. [Stick a pin in the back of the plug.]
|
It was taken at fast idle (7.3.2b).
Vin number VF7S0HDZF followed by eight numbers.
Plug looks pretty clean and the voltage measures 14.3 with the engine running at idle.
|
HDZ is fine; 1.1 unit with single-point injection.
If that was fast idle; then the O2 is too high by quite a margin. Are there any air leaks into the exhaust manifold?
That battery voltage reading must have been off one of the heater circuit wires; try sensor pin 4 - the black wire from the sensor.
|
I've just listened to the exhaust manifold and couldn't hear any leaks.
The sensor has three wires coming out of the back, two white and one dark grey, I took the reading from the dark grey wire. (If I follow the wires around to the connector it reveals four wires going into the sleeving and only three coming out. The colour of the four wires at the connector are two white, one dark grey and one black).
I was just talking to a mechanic about something else and mentioned this failure problem and he queried how hot the car was during the test, as I had only driven it for five minutes to the test station it couldn't have been that hot.
The oil temp reading was 17C, would this temperature be regarded as hot or would it be worth putting in some emission additive and going for a blast down the dual carriageway to get it nice and hot before retesting?
|
Three wires?? There should indeed be two white, one gray - and one black - going to the sensor too.
The wires in the loom have code numbers printed on them; the signal should be on wire code 130.
I'd ignored that temp reading on the assumption that it was reading the ambient temp and wasn't in the oil. Nobody would do an emission test without the car being hot. Oil ideally should be around 80C when tested.
|
"Nobody would do an emission test without the car being hot"
I would add, should, instead of, would, screwloose? Have known some kwikfit places test when engine virtually cold. Nearly always fails :) as you know.
|
yorkie
How silly of me!
..."nobody would do a test cold - unless they were under pressure to sell lots of unnecessary replacement cats."
Edited by Screwloose on 21/10/2008 at 20:24
|
|
|
screwloose they may have bypassed the temp bit on the mot and looking for work, how ever if its had a new exhaust/cat recently was/is there an underlying problem which has destroyed the cat Has it in fact had a new cat!!!!!!
|
|
|
Ok, found the fourth (black wire) it was hiding under the other three.
The reading is 0.8v varying to 0.1v.
I've just spoken to the MoT tester and he said he gave the car chance to warm up as he didn't want the car to fail either, he seems an honest fellow with a good local reputation so I'm inclined to believe him.
I asked him could the oil temp make that much of a difference to the outcome and he seemed to think it didn't! I also asked him about putting an additive in and he said nine times out of ten it didn't work and that I would need to change the sensor.
I changed the exhaust and cat by the way because it had rusted out.
|
That's more like it. If the sensor is sweeping between 0.1 and 0.8 volts, then it's capable of doing everything it's supposed to.
As it only failed the emissions at 2750 revs; what is the sensor doing at those revs?
|
...I suppose that this could be rare but my Fiesta failed on emissions yesterday and today I discovered that my lambda sensor has a "CSD" failure which manifests itself so that the sensor is giving -1 to 0 volts instead of 0 to 1 (no, I didn't reverse the polarity:). And it was working OK on 5 measurements in 3 hours but failed on the 6th!
You can ignore this but you could consider this possibility if your car shows some other problems and shows them only sometimes (stalls, looses power etc)
But you should measure it at high revs as Screwloose suggested. If it's not switching (stays high) maybe the engine isn't getting enough air. If it does then I suppose that the CO must be ok because the PCM is regulating but I'm not an expert.
|
At higher revs the reading is between 0.7 and 0.1v, when it drops to tick-over it returns to between 0.8 and 0.1v, is this what I should expect or is it misbehaving?
|
As a general rule; if the oxygen sensor is sweeping up and down, then the mixture isn't too far out.
So then we turn to that new cat...? Have you any access to a laser thermometer?
|
|
Unfortunately not, but are you suggesting that the sensor seems to be working ok and the cat could be faulty? Is this likely/common? Thanx for your interest so far.
|
It's not an unbreakable law; but if an oxygen sensor has entered closed loop with the ECU, then the mixture is fully under control and is unlikely to be that far wrong.
As your excess CO reading is within the "cat window" - the amount that a functioning cat can correct - then I'm wondering about that new cat.
Cheap and nasty replacement cats are the current bane of my life.
If the temperature of the pipe coming out of the cat isn't 50C hotter than that going in - then it isn't working.
Edited by Screwloose on 23/10/2008 at 21:44
|
hmm... But isn't it worrying that CO an O2 are to high?? I mean, it seems logical that if CO is too high than there wouldn't be any O2 because there is no oxygen for complete burning. Maybe the air is leaking directly into the exhaust before the sensor somehow? I am guessing that the O2 sensor senses oxygen as the name says and if it gets oxygen from some leak or misfiring cylinder it would send the information that the mixture is too lean (too much o2) and the ecu will make it richer producing more CO.
In my case O2 was very low (even zero at idle).
|
hmm... But isn't it worrying that CO an O2 are to high?? I mean it seems logical that if CO is too high than there wouldn't be any O2 because there is no oxygen for complete burning.
Complete burning cannot feasibly occur within any engine - it's the job of the cat to complete the combustion process and (among other things) transfer oxygen left in the exhaust onto the carbon monoxide molecules.
If the cat isn't working, there will be higher than usual levels of all gasses (except CO2) left in the exhaust.
|
Cliff,
I was just wondering what exactly you have recived as a gas print out sheet.
At the start of an MOT test, a BET (Basic emmisions test) is carried out. This test does NOT involve having to put the oil probe into the engine. If the car fails this, you then carry out a vehicle specific test, with data supplied by the manufacturer. This test can only be done after the oil temprature is measured. If the first "fast idle" test fails, the car is held at fast idle for a further 3 min, and then tested again. Your gas sheet should therefore show the two fast idle "Fails".
One further question, how is the car normally driven? Mainly for short runs by any chance?
|
|
Sorry, my theory is illogical even based on my assumptions: if there was too much O2 and O2 measures it then it would stay low...
|
|
|
|
