How To: Use Learning View to verify closed-loop MAF scaling

nsfw · **Joined:** Wed Nov 22, 2006 10:23 pm **Posts:** 2565

Please start a thread in the tuning discussion subforum for that question.

madcorona · **Joined:** Wed Sep 19, 2007 1:01 am **Posts:** 104

NSFW wrote:

You're mostly correct, but there is a little bit more to it.

The D trim (or E trim, if your car has 5 of them) is effective even in open loop. If you do all of your fuel tuning without resetting the ECU, that doesn't matter. But if you reset the ECU when you reflash, then there's a good chance that you're logging and tuning with a 0% D trim. The problem is that if the D trim moves to +5% after a couple days' driving (it can take a while) then you're now running about 0.5 richer than what you saw during your tune/flash/log iterations. If the D trim moves to -5%, then you're 0.5 leaner, which is asking for trouble.

For the other trims, it's mostly just about the warm fuzzy feeling that comes from getting the balance right without the ECU having to make up the difference.

I was wondering will this LV D trim reset on its own? I have this -3% after a few days of driving. I did a check yesterday and it was 0!!!

and when D trim shows negative 3% thought this means we are running rich by 3%?

nsfw · **Joined:** Wed Nov 22, 2006 10:23 pm **Posts:** 2565

They all move around over time. My A trim in particular can move a few percent while I sit idling at a stoplight.

The trim percentage is the amount of fuel that the ECU added or removed (from its initial calculations, based on MAF scaling and injector settings) to achieve the AFR that it wanted (usually around 14.7). I think it's a mistake to say that you are or were rich or lean - it's probably not true and it implies some other things that may or may not be true.

For closed-loop, as long as your trims are between +10% and -10% you're almost certainly getting the AFR that the ECU wants. If you're between -5% and +5%, thats great.

For open-loop, you really need a wideband to know what's going on. You can speculate based on what your AF Learning values are doing, but I don't think they tell you enough to actually make any good guesses.

Aex1 · **Posted:** Mon Mar 12, 2012 6:07 pm

Hi, NSFW, thanks for your great topic!

I have a question, please. After a winter i decided to check my closed-loop MAF scaling and was surprised that airflow ranges were all negatives, about 5-6%, and the same persentage for D+.

My MAF is stock. Ok, i used Dweeb method and substruct some persents from my MAF scale in closed loop. Then i drove some days, checked LV and... A-D+ were all negatives again! Ok, i used the same method and reduced some values even more from 1.20 to 2.60, so the difference from stock values achieved already - 8-10% somewhere!! But guess when i recently checked LV again, the negatives airflow ranges were present as the first time. :shock:

Now what? Subtract even more?? But it seems enough...

Is it a leak? I checked everything, nothing... My last LV is enclosed. Thank you.

nsfw · **Joined:** Wed Nov 22, 2006 10:23 pm **Posts:** 2565

Have you checked for a vacuum leak? That might explain the negative "A" values.
If that's not it, I have no idea, sorry.

Minty · **Joined:** Wed May 30, 2012 9:43 pm **Posts:** 16

NSFW.. As far as I know, my ADM 04/05 Liberty GT has dual O2 sensors. It's a 2.0L twinscroll 5EAT similar to the JDM spec. We run 98RON down here too.

Just an fyi.

throttlehappy · **Posted:** Tue Jun 12, 2012 2:58 am

Base Ignition timing has a major influence in the A and B range. Too much timing and you end up with blue values, not enough and you get red values.

You want to find ignition timing values that give a stable smooth idle and give optimal LV values as well BUT also give a smooth constant g/sec of airflow

joeleodee · **Joined:** Sat Nov 17, 2012 7:51 pm **Posts:** 22

Good information here. I'm currently tuning my vehicle and I'm new to this. I've done a ton of reading and maybe I'm not understanding this. This issue has me perplexed. I adjusted some cells in the idle range and got the immediate trims to stabilize rather tightly. I reset with those values and started the car up. After the reset, corrections were well into the + range, but within a few minutes, those corrections stabilized back down. However, in the meantime, the learned values had steadily crept up and maxed out. After idling for 10+ minutes of reasonable corrections, the learned value didn't drop back down at all. This just seems to be an issue at idle. However, I don't have enough driving logged to determine if this is just isolated to the idle ranges. Should the learned values drop back down in that time range or is there something else I need to be aware of. Injectors are DW1100cc sidefeeds and intake is an AEM CAI.

joeleodee · **Joined:** Sat Nov 17, 2012 7:51 pm **Posts:** 22

I never figured out what the problem was, so I basically started from my previous stage 2 tune, made a few changes in unaffected areas to account for dditional parts since, and I'm building it from there. I've got my learning view under control now as that map had maf and injector scaling and I haven't changed those since.

joeleodee · **Joined:** Sat Nov 17, 2012 7:51 pm **Posts:** 22

Quote:

Do not confuse the "AF Learning" parameters with "AF Correction." If you're looking at a number that fluctuates widely and rapidly, you're looking at the Correction value, which is not helpful. The ECU itself observes this value and uses it to make fueling adjustments, and over time it will adjust the Learning values based on trends in the Correction value. If you're familiar with signal processing, the Learning value is just a low-pass-filtered view of the Correction value. I have watched the A value climb to +15% over the course of several seconds while I had a vacuum leak, but under normal circumstances the Learning values change so slowly they're boring to watch.

I'm having an issue with this. I noticed my a/f learning values were way out of wack. The first 3 ranges were nearly maxed out (+12 to +14%). However, my correction values while logging in real time were very tight (-2 to +2). I reset the Ecu so I could observe. After everything calmed down, my corrections came back to like they were before, but my learning started climbing quickly. At idle, they jumped to almost +15 all while the corrections remained very tight. Same thing happened throughout the ranges. If what is quoted above is true, how could my learned values get like that if the immediate corrections are where they need to be? Could a vacuum leak cause only an issue in the learned a/f but not the a/f correction?

It's an 05 LGT, stg3, e85, DW 1100 sidefeed, AEM CAI.

nsfw · **Joined:** Wed Nov 22, 2006 10:23 pm **Posts:** 2565

Just ignore the Correction values.

High Learning values for the low-airflow ranges usually indicate a vacuum leak.

05Pzev · **Joined:** Sun Jan 03, 2016 10:05 am **Posts:** 217

On a car with aftermarket forced induction, where should I set the upper limit of the "C" range?

Should it be the highest g/s I would see during closed loop... or slightly higher than that?

RomRaider

Forum rules

How To: Use Learning View to verify closed-loop MAF scaling

Who is online