Hello - need some help wrote:
MOT failed today. Emissions. Failed on CO and Lambda.
First test CO 1.76% (target less than 0.3%) Lambda 0.954 (target 0.97 to 1.03)
Second test CO 2.45% Lambda 0.934
HC was 9 and 14ppm (target less than 200ppm). Indicates that fuel is being burned well?
New coils plugs and leads fitted in December. No noise/rattles heard. 29K miles 05 plate.
Does this indicate the CAT has gone? If yes does this indicate another underlying problem - or do they just go?
Thanks for any advise.
Using some Cataclean may well help though there is an issue with the way the test is run that can create a problem for our cars.
That target of 0.97-1.03 is for the fast idle test, where the engine revs have to be held between 2,500-3,000 rpm for 2 minutes.
The problem is that our ECU knows when the engine is being rev'd on no load and will automatically drop the revs back to idle after about 90 seconds.
So, for the first 89 seconds the measurement is likely to be well within limits, then the guy in the driver's seat has to blip the throttle to get the revs back up, at which point there's a spike that takes it outside the limit, before it settles back again.
Since the machine registers the min and max figures, this results in a fail.
It is possible to avoid this by being very gentle catching the revs as they drop, to limit the size of the blip.
My local tester has me sit in the drivers seat and we didn't have any issue, despite his lead-footed mate managing to fail it three times in a row.
It also helps if you switch off the climate control first.
Have a word with the guys doing the test and have them watch the lambda in real-time to see this effect. I have known some testers acknowledge that they're seeing a false spike and repeat the test at idle for the recorded data.
Or, better still, take the car to a tester than knows about RX-8's.
Of course, 2,500-3,000 revs is half-throttle for a conventional engine but barely a third for us and well below where we should be reving while driving.
Note: the 2 minutes and 90 second figures are from memory. It's about those times anyway.