This is the second time I have noticed strange behaviour with my car in relation to the AFR. The car is UK MY00, AE802 ECU, Dawes and SS DP.

Yesterday coming home along the M25 I noticed that the AFR meter was only lighting the first two segments of the display (720/740mV) pulsing them on and off very quickly at an 80MPH ‘Cruise’. Normally I would be seeing a ‘knight-rider’ effect up to 840/860mV (5-6% CO). WOT produced a maximum of 6% only!! Normally I am well into 9%+ CO

This stayed all the way home, a 40 mile or so drive. I re-checked the calibration of the AFR as soon as I got home and it was spot on.

Using the car today and the AFR is displaying mixtures as ‘normal’ cruise is OK and WOT is back to 9%+

The last time this happened was when the weather was much colder so I don’t think it is that.

The only thing I can think of is that I had to fill up with Esso 97ROM fuel and then on the return journey I managed to fill up with Optimax as I normally do. The car was left overnight with Optimax in the tank before I noted the above behaviour.

Any body got any ideas/theories?

Harry

 

No idea. At least it is not my dodgy AFR design then

 

The lambda reads lean when hot...you weren't looning around before you got it stationary were you?

/J

 

John, no problems with your design

No launch tests before hand! Straight out the car park, warm her up get on to the M25, look at AFR...................

Harry

 

How are you guys coupling the ground for your lambda signals?

It really should be directly coupled to the sensor ground for accurate results that are not distorted by electrical currents in the car's chassis / earth wires.

For example, if you take your earth reference as some random negative wire near the ECU, then simply turning your lights on may cause a few tens of milivolts change in the detected lambda voltage (which can be quite significant to the AFR reading).

Apologies if this issue has already been discussed, but it could cause inaccuracies. Perhaps not in the above case, though.