Tune learning with headers

[ManBearPig]

I have had a Lund 93 tune for my stock 16 GT for a few years now. I just finished putting a set of off-road longtube headers on the car and am wondering if it really is true that the car will “learn” to adapt to the headers, and how I can tell. I am now monitoring LTFT and STFT on the nGauge and the STFT seems a lot more active than I’d think it would be. Of course I never thought to look at these values previously so I really don’t have anything to compare to. At WOT both LTFT and STFT usually hover around zero. However under part throttle I see STFT as much as +/- 5 and when I come to a stop the STFT will be off as much as +20 until it slowly settles down to a more reasonable number. LTFT is usually 2-5 under these scenarios. Is the STFT supposed to be near zero at all times, or is it only crucial under heavy load?

Sponsored

 

[Jay-rod427]

Short terms bounce around quite a bit. It's a constant striving for perfect with the long terms. Every situation has it's own long terms, rpm, , atmospheric conditions, throttle, load, etc. Within 5% is not bad at all.

 

[ManBearPig]

Good to know the STFT can move around without concern. What about LTFT? I have noticed that they agree at WOT but part throttle bank 1 is usually +2 and bank 2 is -2. Is it acceptable to be off 4% from each other?

 

[Jay-rod427]

Within 2% of zero is pretty good. Any slight variations could cause the difference between each bank. Intake air flow, injector balance differences, O2 sensor readings, etc. The blessing and curse of modern day computers. Sooooo much accurate data coming in can be overwhelming/worrisome when it's not. You should see the cylinder and/or bank to bank discrepancy on old stuff. It'd scare ya. A bbc carb base intake manifold with widebands/egt on every cylinder is quite alarming how much each cylinder differs, but if you didn't have that data you wouldn't really know otherwise.

 

[ManBearPig]

I guess the car is still learning. I’ve just now driven 100 miles since the header install and the STFT is much less active at part throttle, but the long term trims are getting farther off. I’m seein +3 to +5 on bank 1, +5 to +8 on bank 2. They are only this far off when I’m very light on the throttle. The harder I roll into it the closer to zero they get. WOT is very close to zero on bolts banks. I know the car is learning like it should be, but at what point is the learned error deemed to be too much? Anything
over 5%? I know Lund says you do not need time revisions with headers, is that to say that whatever error the long term trims learn is good enough?

 

[ManBearPig]

I guess the car is still learning. I’ve just now driven 100 miles since the header install and the STFT is much less active at part throttle, but the long term trims are getting farther off. I’m seein +3 to +5 on bank 1, +5 to +8 on bank 2. They are only this far off when I’m very light on the throttle. The harder I roll into it the closer to zero they get. WOT is very close to zero on bolts banks. I know the car is learning like it should be, but at what point is the learned error deemed to be too much? Anything
over 5%? I know Lund says you do not need time revisions with headers, is that to say that whatever error the long term trims learn is good enough?

Edit, I asked my tuner about updates and they will only review a data log for a $50 charge. I’m fine with the charge if they end up making a revision to the tune, but would like some insight on the review. Have others found that’s a datalog after header install actually did require an update? I hate to spend 50 bucks only to find out everything is fine, but I suppose that would be the safe thing to do....

 

[Bluemustang]

I guess the car is still learning. I’ve just now driven 100 miles since the header install and the STFT is much less active at part throttle, but the long term trims are getting farther off. I’m seein +3 to +5 on bank 1, +5 to +8 on bank 2. They are only this far off when I’m very light on the throttle. The harder I roll into it the closer to zero they get. WOT is very close to zero on bolts banks. I know the car is learning like it should be, but at what point is the learned error deemed to be too much? Anything
over 5%? I know Lund says you do not need time revisions with headers, is that to say that whatever error the long term trims learn is good enough?

100 miles should be enough for the car to have fully learned, from what I have heard. 50-100 miles is what I've been told by my tuner.

 

[Zelek]

I do believe there is some adjustment to be had in the tune when long tube headers are added. I would not rely on the car to learn it. I don't know exactly what it is but the entire goal of adding a header is to increase the scavenging effect which allows exhaust to be pulled out faster so more air could go in. I could be completely wrong though but that is my best educated guess without referring to someone who tunes for a living.

 

[Bluemustang]

I would do one or more of the following: Do a KAM reset. Get a datalog and email that to your tuner to have it checked.

My guess is the extra flow is not being accounted for as well now in the tune. I mean it is being accounted for, perhaps just not adjusting quickly enough. Could the car trying to reconcile the new values it's seeing vs. how it used to be before the header install. The LTFTs take a lot of things into account. I'm on stock headers and every once in awhile I do a KAM reset with my FF tune if I feel like it's acting a little wonky. After that it starts the re-learning process again and before long everything is peachy.

The above is my educated guess but I bet a KAM reset will set it right. Also, don't pay look at the STFTs all the time. They're always adjusting and bouncing around. You could drive yourself nuts doing that.

 

[Bluemustang]

I do believe there is some adjustment to be had in the tune when long tube headers are added. I would not rely on the car to learn it. I don't know exactly what it is but the entire goal of adding a header is to increase the scavenging effect which allows exhaust to be pulled out faster so more air could go in. I could be completely wrong though but that is my best educated guess without referring to someone who tunes for a living.

Yeah. Well his MAF lb/min should have increased post header install.

 

[ManBearPig]

I had the battery unplugged for 3 days while I was getting the headers put on. NGauge doesn’t have a KAM reset but I think having the battery out for that long should’ve done it. I may try it again just to see if the long term trims learn the same thing. I guess my question now...is 5-8% off at low throttle enough to warrant a $50 tune review and potential slight revision?

 

[Zelek]

I had the battery unplugged for 3 days while I was getting the headers put on. NGauge doesn’t have a KAM reset but I think having the battery out for that long should’ve done it. I may try it again just to see if the long term trims learn the same thing. I guess my question now...is 5-8% off at low throttle enough to warrant a $50 tune review and potential slight revision?

That's not the same as a crank relearn is it?

That $50 can get you several revisions with datalogs if you really want to dial it in so it's really up to you on if you want to do it. It's good piece of mind knowing that your engine is running well.

 

[Bluemustang]

That's not the same as a crank relearn is it?

That $50 can get you several revisions with datalogs if you really want to dial it in so it's really up to you on if you want to do it. It's good piece of mind knowing that your engine is running well.

No crank re-learn is entirely different. KAM reset stands for "Kill Active Memory" I believe. It resets the PCM adaptive learning. Long term fuel trims start over from the beginning as if you loaded a fresh tune into the PCM.

 

[Zelek]

No crank re-learn is entirely different. KAM reset stands for "Kill Active Memory" I believe. It resets the PCM adaptive learning. Long term fuel trims start over from the beginning as if you loaded a fresh tune into the PCM.

Ah, I gotcha. Thanks! Might explain how I got my O2's to go green on my Lund 93 tune by flashing to the stock tune, driving a bit, then flashing back to the 93 tune. That or I have some magical O2 sensor spacers.

 

[AZ18yote]

No crank re-learn is entirely different. KAM reset stands for "Kill Active Memory" I believe. It resets the PCM adaptive learning. Long term fuel trims start over from the beginning as if you loaded a fresh tune into the PCM.

Pretty sure its keep alive memory. But resetting it does kill it

Sponsored