Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Misterious AFR Swings

General Tuning Discussion

Forum Posts



Tech Articles

Discuss all things tuning in this section. News, products, problems and results. 

= Resolved threads


Hello all,

RB25 DET; GT35R ; 1000cc EV14 ; 2X 044 surge tank; Haltech PS1000

This car is driving me nuts. AFR swings wildly lean when I turn on the car on the dyno, after some running and a couple of dyno runs all gets back to normal. This is not the initial lean out because of fuel vapor (or so I think). This lasts for minutes and the car which was once idling at 14.7 idles at 19, 20 after startup. All the map is affected but more so at idle.

Nothing changes, no corrections active and it is quite consistent in this behavior. Have you ever found anything like this?

Fuel temperature? Injector weird behaviour? Ran out of ideas really

It is very hot here (35ºC)


have you checked the plugs if all of them are firing? I've had a car on the dyno, which did exactly what you described and the coils were broken ... sometimes they were firing the plugs, sometimes one coil didn't.

It was an inline 6 with cams all sorts of mods, so it wasn't easy to distinguish.

It sounds like you are getting heat soak, your IAT compensation should be able to compensate for the extra heat, unless you look into the IAT compensation for a warm start only

I've had the same issue on my rb25 running similar mods. turned out to be an injector that was dirty and not firing right at low pulsewidth. check the car is running on all cylinders when it happens.

also you could try logging the event the checking to make sure the ECU isn't trimming fuel in or out of the map due to a faulty sensor.

and obviously check your wideband is calibrated and you have no exhaust leaks giving you a false lean reading.

Thanks all great insights

Wideband is calibrated and agrees with narrowband.

Coils are new yellow jackets. Car pulls cleanly to redline and only runs rought at idle when lean. Car not misfiring ignition 100% it appears

Funny you say injectors... I felt one clicking faintly and mixture was rich all over the map. Sent to clean they said all was fine and within 5% of each other which I find odd. Installed them and all seem to click the same. Looked at plugs at idle, cruise and Wot and all look the same in all tests. The symptons do point to them... May do some more plug checks to confirm...

No compensations going on. Fuel swings up to 40% at idle.

Yesterday tested again, having retuned the fuel map to the conditional the day before. Checked on dyno and it was out of each again, very lean to the point of not running at idle and consitently richer at Wot by around 15%. Afr graph was paralelo to the one before but a full point richer...

Curved ball here. Will report with more tests done tomorrow.

Sorry for the testament haha. Enjoy your Sunday


I'd start by taking note of inputs such as IAT and ECT when the AFR is on track and as expected, then compare what these parameters are doing when you're seeing the AFR massively lean. This is a good guide as to where you should be looking for the problem.

One point I would make that is often misunderstood is that if you are running the Haltech in VE mode then the air temp correction is handled automatically in the fuel model. If you've also tuned the IAT fuel trim table then you will be overcompensating for temperature fluctuations. At the same time the automatic IAT comp tends to over compensate at idle and light load so often you can get better results by reducing the amount of trim being applied in these areas of the IAT trim table.

I've also had Iat sensor dramas with my rb25. It turned out to be a case of my ECU (adaptronic) was a multi model ECU to do rb25, rb26 and vg30 engines. problem was the rb25 didn't have an Iat sensor in stock form so the Ecu was picking up some sort of signal saying intake temps were 125centigrade then randomly trimming 18% of the fuel in and out of the map.

Regarding diagnosing the injectors/ plugs the easiest way ive found is to wait till the problem is happening then disconnect the injectors/ plugs 1 by 1 and listen to the engine change/look at the wideband.

Hello and thanks again.

Andre, did that already - ECT, IAT, FP, Map, Battery Voltage, RPM, TPS are normal and stable.

What happens is when cold or after standing for a while but still hot (ETC>50º) the engine wants more fuel - aprox 1,64ms pulsewidth as reported by the haltech.

After some running it needs less and less fuel until it stabilizes at around 1,56ms.

This is only a 0,1ms difference but causes all this drama on idle and light load. I cannot, for the life of me know what is causing this. I suspect maybe of injector dead time, fuel temperature or electric interference?? But it is unclear why it developed this behaviour over time, since I tuned it 1000kms ago

This behavior is consistent.

Thanks for everything


We usually reply within 12hrs (often sooner)

Need Help?

Need help choosing a course?

Experiencing website difficulties?

Or need to contact us for any other reason?