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

Ends in --- --- ---

Lambda Correction Map

Understanding AFR

Forum Posts



Tech Articles

Discussion and questions related to the course Understanding AFR

= Resolved threads



I'm really new to tuning and have just completed my first pass of the standalone course, so bear with me if any of my questions are a bit dumb. But I have a VW VR6 Turbo setup for my drag car I purchased complete with KMS MP25 Standalone. I got basically everything except the car and have fitted it all to mine and am about ready to fire it up. The original map was done by a tuner on a RR but my aspirations are to to be able to map the car myself at the track as and when required, or as I upgrade turbo/injectors etc.

But I hit a hiccup sooner than expected when the seller got in touch and mentioned it was mapped on 106 race fuel (Stoich 14.0:1). I have no intention of running it on 106 and would like to adjust the map to pump fuel, as safely as possible. I figured that I can achieve this by pulling some fuel from the map to compensate for the difference in stoich value, which I calculate at roughly 5%. And although both fuels will still be knock limited I assume I would be safer to pull a little timing too and the 106 will be less susceptible to knock than the 99 pump fuel?

I've spent some time recently getting to know my ECU well and have found that there is a lambda correction map feature, and wondered if there was a general consensus on how good these are and if people actually use them. Its powered in the case of my ECU by a lambda target table. Again this was originally populated for the 106 fuel but I can add 5% to the whole table and the numbers look to match the 14.7:1 stoich figures I'm looking for, and i wondered if this could be a good way to get the map close, safely before fine tuning with some more data logging? Or if there's horror stories out there and I'm best to stay away?

Sorry for a long winded post. Hope it makes sense and appreciate any feedback received.



You have a lot going on here, and are likely over thinking this process. Lambda is Lambda, it doesn't care what kind of fuel you are using. That is why lambda is superior to AFR when tuning. Be careful adding or removing any amount to lambda targets universally, especially in percentages, as it does not do what I think you think it does. Meaning a target lambda of 1.0 with 93 octane is 14.7. A target lambda of 1.0 with this 106 might be 14.0, but as you can see the target lambda would not change, just how the characteristics of the fuel itself react to that target. A target lambda of .8 would equate in 93 octane to 11.8 AFR (best rich torque), but that same target lambda on e85 would be an AFR of 7.8. I have attached two pictures of lambda to AFR tables to help with understanding this more.

You do not compensate stoich changes in target lambda in my experience. You adjust fuel characteristic changes by altering the calculations of how MUCH fuel the engine is going to use to reach the target lambda you've set. Unfortunately I am not familiar with your standalone system, but in Bosch-Motronic terms it would be called KRKTE. After this is adjusted, using the same target lambdas, you would use the lambda correction table (and log the axis hysteresis) to adjust the amount of fuel that needs added or removed to meet said target lambda goals by adding or removing whatever percentage is added or removed at that point of data. In Bosch-Motronic again, this map (or at least the one I prefer to use) this is called FKKVS, which has an axis of RPM by Injector On Time. Very large, high resolution adjustments.

As a point of reference, on any small, medium, or large frame turbo I have ran in VW 1.8t and 2.0t applications my target lambda at full load is between .78-.82 depending on how the engine reacts to any given amount of boost or available fueling abilities of the system. Regardless of fuel type, that is my target period.

After you have made the correct adjustments to your injector data in the ECU, and if so desired the target lambda values depending, then yes definitely look at pulling timing as that is going to have a large impact on your cars performance when switching between fuel octanes as well. Likely it will need just a little trim, but its important to get logs and see for yourself. I do not know what your setup is like, how much boost, or even how your engine was built (factory pistons with compression dropping head gasket is what most 12v guys do, and works well) so those changes are going to be very specific to how your engine reacts to the new fuel. As a general rule, I would start your high load/rpm which I am assuming is currently set to around 16-18 degrees back to 12-14 and go from there. Again, those are assumptions, and I have no idea what you are working with.

Good luck, hopefully this helps.

Attached Files

Hello Ed,

Regarding ignition timing, to fully optimise this you will need to take it to a load control dyno and calibrate to MBT/knock limit using a standalone knock detection system from the likes of Phormula or Plex as the KMS doesn't have any knock control functionality. Once this is set on the dyno then fine adjustments can be made on the road/track. Assuming its been calibrated correctly previously on 106 you will have to pull timing to run 98 as it will have a lower knock limit. The amount is entirely application specific but around -4 to -6 degrees should be enough to keep the engine safe until you can get it calibrated correctly.

For the fuelling; as Anthony suggested it is easy to over think things. Depending on the ECU functionality the simplest way is to run the engine with closed loop fuelling enabled and the clamp values set wide (+-20%) then perform an RPM sweep at various load sites 100%, 75%, 50% load etc and correct the base map to suit your specific lambda target based on how much the controller is having to trim to achieve target. You could do this with the base map either as it is or with -5% fuel it wont matter so much as the 40% controller window should cover the 5% delta. To define the lambda target using a dyno start at stoic (lambda 1.0) and increase fuel until you reach MBT, usually around 0.90-0.95 lambda as combustion efficiency will be sub 100%. You will also have to consider thermal limits for exhaust valves, turbine wheels etc and add a little extra to reduce combustion temperature hence why turbo engines can be as low 0.82 however there is a common misconception that richer is better when in reality running too rich causes bore wash resulting in piston ring failure, ring lands collapsing, bore concentricity issues etc so it is best practice to be precise as possible with the fuelling requirements.

Thanks a lot for the reply's its really appreciated! I suspect you're right and I am overthinking it! I think first things first I'm going to go and sit the understanding AFR course again to refresh my memory.

Anthony - The reason i have myself twisted up about adjusting the correctly table is because I can see that the AFR of the table starts at 14.0 whereas the KMS documentation actually gives a default starting at 14.7 so it looks like this was amended to suit in my head. And since the rest of the table looks far from default it looked to have had some work put in, it seemed a waste. Adding 5% brings the first entry's in the table up to 14.7 which is where the 5% figure came from originally. I've put some screen shots in to try and illustrate what I'm saying a little better as i'm not sure how well im explaining it. FYI you're correct its a stock 12v with spacer gasket. Originally with the old owner it was mapped to 330bhp but they had issues with the boost control so it was left at wastegate pressure. I have a larger BorgWarner turbo & injectors to go in the future but want to cut my teeth on this setup first, getting all the features working as i want them and getting some experience/confidence.

Scott - Thanks for the reply that makes a lot of sense. I think fuelling wise that's what I was hoping for. That info's really helpful though and definitely what I needed to see. Regarding knock, in the absence of knock control is there any mileage in external knock sensors to tune with on the road or are they snake oil/not reliable enough? I came across this https://phormula.com/product/phormula-ks-3-knock-detector/ which on the face of it this device or similar in my head would help bridge the gap and realistically cost wise is going to be the same investment as a few hours on the dyno, plus I get to hang onto it for future use.

To be clear, I'm not against the dyno at all I'm just exploring the options and if possible I'd rather invest in something I could potentially get continual use out of that suits my application if that were an option.



Attached Files

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?