I just noticed this. This is exactly why so many people have issues with the VAFC. IF they have the VAFC setup so that it will make fuel adjustments during closed loop, then the ECU will tune the changes out by adjusting the LTFT. Then when the car goes WOT, the LTFT is still applied, and their tune is messed up. The solution is to make sure the VAFC does NOT apply fuel corrections when in closed loop, so that the LTFT won't be impacted by the fuel corrections.

 

Here is a thread on an STI forum about a guy who has to keep reseting his ECU to clear the LTFT because it is causing his engine to run very rich at WOT. Basically, his part throttle tune is too lean, so the ECU is applying large positive STFT values which slowly forces the LTFT to go higher and higher, which pushes his WOT fueling very rich.

 

Does it make any sense that the ECU would apply an adjustment to a parameter it isn't monitoring? By your own admission, you acknowledge that in open loop, o2 feedback is ignored. So the ECU is just hoping the AFR is ok based on adjustments determined at partial throttle?

Futhermore, how does this concept work with the AP1's equipped with a narrowband?

 

Spectacle, consider this. If you are an average joe (not a car guy), and have a totally stock car with 150K miles on it, would you expect the fuel injectors to flow like new? I wouldn't. I would expect that the ECU would need to be applying a pretty significant fuel trim to keep AFR on target. I would also expect that WOT would run pretty damn lean if the ECU wasn't applying some extra trim while in open loop too. I wouldn't want my motor to detonate to death just because the ECU assumes that my obviously clogged injectors magically flow like new when the pedal is pressed to the floor.

That's what the LTFT is for! It's just a running average of STFT that is applied ALL THE TIME! Nobody is saying that LTFT is being adjusted during open loop. It's only being APPLIED during open loop. The ECU has to assume that if it needs to consistently add 10% at part throttle (on average), then it will also need to add 10% at full throttle to be safe.

Why would it be designed any other way? Even with a narrowband, if the ECU is constantly having to add 10%, it will assume that it needs to add it all the time in all cases!

 

BTW, you mentioned that you run you 400+ HP car with closed loop disabled. I imagine you either have a great tune, or you're stupid. However, even if you have a great tune, you either need to check that tune from time to time to make sure it still "works", or you need to turn closed loop back on, so it can keep things in check with LTFT.

 

ECU's don't assume, they interpolate. If there's nothing to interpolate, they don't do anything. You're crazy if you think just because a fuel adjustment needs to be made @ 4k it also needs to be done at 8k. I don't know what else to tell you.

 

I have a great tune and no seasons here in FL. It's either hot or very hot. All those other factors you named about wear and tear and all that jazz don't play a factor.

 

Spectacle, I know you're a smart guy, but I just think you are making assumptions that you are right, and not even listening.

I'm a hardware guy turned software guy. I understand that computers don't "assume", but it's a good word for laymen. We are talking about an ECU that is designed mostly for people that aren't going to modify their cars AT ALL, and aren't really even designed for performance. Most ECU's are designed with reliability in mind.

I'm not saying that just because an engine needs 10% more fuel at 4K, then it must also need 10% at 8K. What I'm saying is that a "smart" car engineer is going to assume that if ON AVERAGE an engine needs 10% more fuel at part throttle to be on target, then it's probably a good idea to give it 10% more fuel at full throttle too.

BTW, are you even using an OBDII ECU? Do you even have any tuning experience with OBDII ECU's or piggy backs? I've got a couple thousand hours working with them, since I've developed 2 self-tuning piggy backs of my own.

 

So you think that your fuel maps will be just as good 2 years / 50K miles from now with no "touch ups"? You don't think your fuel injectors will flow slightly differently, and your MAP sensor will still read exactly the same?

 

Isn't this stating that applying a LTFT during open loop would be a good "assumption" for an ECU to make?