engine tuning gas analyser Q

[steveo3002]

i have use of a wideband tuner and seperate co tester , they dont agree with each other

are the circumstances with a bad tune /engine fault that they wont match this chart or is it cast in stone that xx afr = xyz co

charts like so
https://www.mgexp.com/article/co-afr.html

[GreenV8S]

steveo3002 said:

i have use of a wideband tuner and seperate co tester , they dont agree with each other

are the circumstances with a bad tune /engine fault that they wont match this chart or is it cast in stone that xx afr = xyz co

charts like so
https://www.mgexp.com/article/co-afr.html

The wideband sensor is measuring the oxygen concentration in the exhaust. The CO sensor is measuring CO concentration in the exhaust. Both types of sensor are fallible and can give inaccurate readings if they are not working properly.

Both these concentrations will vary with the air/fuel ratio. If the combustion process is perfectly uniform and consistent then they will vary in a predictable way and the AFR estimated from the two readings should agree. If they disagree significantly, either one / both of the sensors is giving an inaccurate reading, or the combustion process is not perfectly uniform and consistent.

What sensors are you using? How have they been calibrated? The more expensive they are and the more recently they have been calibrated, the more I'd be inclined to trust them.

I have an ordinary cheap hobby grade CO sensor which gives broadly sensible readings but it is very sensitive to pressure and temperature changes and sometimes give erratic readings - I use it as a guide but don't assume it's accurate. I have a variety of wideband controllers, some of which need calibrating (i.e. in fresh air) to produce accurate readings. Commercial systems might need calibrating in a controlled gas mixture, but they're outside my budget. Sometimes a wideband sensor will go 'bad' such that they produce a very inaccurate reading even after calibration. If I see an unexpected reading I can cross-check these different sensors and controllers against each other as a sanity check. I wouldn't blindly trust any of them, especially if they disagree.

[steveo3002]

innovate wideband...self calibrates , snap on co meter , you dial it into 0 in free air (also goes back to 0 after removal from the tailpipe)

afr readings are somewhat believable when driving and under wot

ive been having issues getting it idle at a sensible mixture according the afr gauge, misfires /pops in the tailpipe ...with it on the co meter its very weak and smooths out if i richen to approx 1% co

i dunno i guess i cant trust either of them ..may have to wait till something gets mot'd then compare the co meter

[Mignon]

The relationship between AFR and CO is fixed and I agree the numbers in the chart you posted. You don't say what engine you are trying to tune or what CO reading you think it ought to idle nicely at but there might be mechanical issues that mean it prefers a richer mixture.

[GreenV8S]

steveo3002 said:

innovate wideband...self calibrates , snap on co meter , you dial it into 0 in free air (also goes back to 0 after removal from the tailpipe)

afr readings are somewhat believable when driving and under wot

ive been having issues getting it idle at a sensible mixture according the afr gauge, misfires /pops in the tailpipe ...with it on the co meter its very weak and smooths out if i richen to approx 1% co

i dunno i guess i cant trust either of them ..may have to wait till something gets mot'd then compare the co meter

What actual readings are you getting?
Where are the sensors reading from - tailpipe?
Does the exhaust include a catalyst?

I've had a couple of the LC things. They aren't my first choice, but they're so much cheaper than the alternatives that I keep coming back to them. I've twice had sensors which after calibration produced a reading which was off by about 1 AFR, which is roughly the error you're seeing here. I don't know what caused it because the affected sensors calibrated correctly with a different controller, but just seemed to consistently calibrate 'wrong' with these specific units. My guess was it was a heater performance issue and not a total failure. The fact your engine misfires does support the theory that it is actually lean despite what the AFR reading says.

Edited by GreenV8S on Friday 23 November 15:49

Edited by GreenV8S on Friday 23 November 15:50

[steveo3002]

its a 1988 mpi polo....doesnt have a cat

wideband is in the downpipe

snap on tester up the tailpipe

right ok well long story its a rebuilt engine that ran fine and passed a mot on the original carb , i then converted it to mpi from a slightly later polo and its never idled well

been using the wideband , like i say okay figures on the road by those google charts its needed to be 3% or above or has what i think is a lean missfire at idle...the tailpipe instead of a constant purrr is purr-pop-pop purr , unless at 3% or more on the wideband

now today ive got the snap on up it....fiddle and richen to 1% or so and it sounds ok yet the wideband says its rich

mechanicly ive been over everything i can , compression is good , vac checks out like it should , all service parts new etc

[GreenV8S]

steveo3002 said:

unless at 3% or more on the wideband

3% what?

[steveo3002]

needs to idle at 13.4 afr (3% ish on chart) read from the wideband

or

happy at 1% co on the snap on meter (which should read 14.1 via the chart)

im guessing a healthy engine should happily idle well below 1% , vw set up suggests 1% for set up

[Mignon]

Sounds to me more like there's an air leak on one cylinder so the whole thing needs to be rich to get that one running ok.

[steveo3002]

Mignon said:

Sounds to me more like there's an air leak on one cylinder so the whole thing needs to be rich to get that one running ok.

yeah i dont disagree...had it apart twice and new gaskets /checked with straight edge blah blah

[Mignon]

Then I suggest you pull each spark plug lead in turn and see if one makes less difference to the idle.

[steveo3002]

its not a completly dead cly ...just the odd pop pop out the tailpipe

[Mignon]

Could also be an exhaust manifold leak.

[steveo3002]

GreenV8S said:

I've had a couple of the LC things. They aren't my first choice, but they're so much cheaper than the alternatives that I keep coming back to them. I've twice had sensors which after calibration produced a reading which was off by about 1 AFR, which is roughly the error you're seeing here. I don't know what caused it because the affected sensors calibrated correctly with a different controller, but just seemed to consistently calibrate 'wrong' with these specific units. My guess was it was a heater performance issue and not a total failure. The fact your engine misfires does support the theory that it is actually lean despite what the AFR reading says.

Edited by GreenV8S on Friday 23 November 15:49

Edited by GreenV8S on Friday 23 November 15:50

yeah might set it up on the co meter ignoring wideband for now , see how drives and what the plugs look like etc

cant do alot more as a home user

[steveo3002]

Mignon said:

Could also be an exhaust manifold leak.

seems it was a injector o ring leak,had them in /out a few times , new seals etc , carb cleaner sprayed on didnt show anything

happened to wiggle em while it was running and its okay now ! can only think one hadnt seated right