STFT will typically try and pull fuel a littlebit after you hit the throttle due to 'tip in' (i.e. ecu anticipates you hammering the throttle so it over compensates a bit with fuel at first). That is why it is really important to get flat land, and set cruise control on an rpm. Wait 5 seconds, then start the logging for about 15 seconds.

If your LTFT is -2 at idle ... and your STFT is averaging 0 .. then you shouldn't really see the LTFT change.

Few things could contribute to the ltft:
a. the tune always had -2 lt at idle
b. environment/temperature change
c. you modified the intake a bit
d. air filter needs some cleaning.

(florida) lol its pretty flat, but i very well have logged right away when getting to desired RPMS for fuel injector scaling. im going to back track the maps to see where it was "0". Kane was big help during this, and the video makes it 100% better.

Now it may be the porting, but i have a glitch at 4800ish rpms, doesnt matter what load its there. the throttle hicupps at the rpm range. cant see it in the logs though

I would leave it alone and see if LTFT's build from it.

2 things to remember

1. the maf curve is broken into 3 learning zones for LTFT
0-9 g/s
10-22 g/s
23-max g/s

if there is LTFT in the 23-max g/s that trim carries into WOT/Openloop

2. injector staging

if you look at your logs I bet those negative STFT are from the injectors opening, I would not worry much about them unless you got LTFT from it

Also guys, remember - 3% is 0% for all practical purposes. If you get too anal you WILL end up chasing your tail a lot.

An engine is not an isolated, physics driven ideal model - so no math will ever be 100% accurate.

1. That is neat... i didn't know that. Very good information to have. So far only that slight LT at idle, but my ST seems to balance it out

2. I am using MazdaEdit so I dont know if there is a way I can log that ... looks like there is a pulse width param

sorry I should have been more specific, for both the cobb logs and mazdaedit logs,
look at the pulse width and or the duty cycle, it will go down then back up,

this is because those parameters are taking the average from the 2 or 3 different sets of injectors and omitting the ones that are off, I would love it if we could log them individually

so, for example, say you had 30% duty on P1 while accelerating and the secondary kick on at say 10% you would see this in the log as 30% followed by a drop to 20%

its not that cut and dry but its the basic idea of whats happening

I think the P1 actually back off some while the secondary kick on then they ramp up together, then back off for the P2 to kick on then they ramp all together. that's why sometimes the duty cycles or PW, fall for a few cells then start going back up

So here's a question what do you do when you have your maf scaled properly but when you try to get a WOT log you notice that around 4000 to 5000 the AFR drops to about 13-14 when it is commanded to go to 11.8. I figure it is not an injector scaling issue because my low end low load logs have been perfect on the trims and my top end wot is running a bit rich which is a secondary. my fear is to scale the secondary without first understanding why I have a lean blip at wot

Posted From RX8Club.com Android App

Post the log

Extension renamed to .txt ... it is a .csv file though. Its a 2nd gear pull

Throttle enrichment.

But also remember the AFR lags about 1/2 second behind what the car is actually doing - so it' really not as bad as you think.

aha .. makes sense. in your opinion worth the effort to zero out the TE, and do a slow wot to get the afrs ... or just continue on and scale my top end (secondary)

If the car isn't acting weird I'd leave it. It's not a lean spot really - it's just the delay from WOT to getting down into the 12's.

However if it is bucking or whatnot, I'd add more fuel to the throttle maps.

good to know. nope no bucking, rather smooth actually. I'll try again tomorrow just to confirm. I need to find out how to get the injector duty cycle / pulse from mazda edit. They have a Pulse Width parameter but it never captuers any data.

It's possible. During my last class the Mazdaedit guy got it for me.

Off the top of my head I can't remember what it was called.

I don't like how it takes forever for it to go into open loop, something else funny is happening there. and your log doesn't have throttle position data.

those fuel tables remind me of the stock closed loop tables, you could edit those to smooth things out, or get your car to go into open loop sooner.

Sweet. I'll send him an email and report back here so others have it. Thanks.

Getting close ..... scaled the secondaries up by 5%. Still running richer than i am targeting up top (richest I have the afrs to is 11.80).

Sent a note over to the ME guy to see if he can tell me how to get the Injector data from ME. Would be nice to have so I can actually see where the secondaries are on before i continue mucking with the scaling.

edit: nix that ... looks like my maf scale is off a tad @ ~4.06v+

That log is definitely showing rich. And it is across the board at the higher RPMs, I'd look at the Inj 2 situation.

The lean blips might be your transition; scale the Inj2 and see if those stay leaner up top. If you're still not getting on target across the board then double tap your MAF settings.

will do. thanks for the tips.

Looks like the afr curve is looking better. Scaled up 2ndaries 7% from stock. Looks like going in the right direction. Maybe just a tad more and im there.

+10% on InjectorB scale from Stock. Getting better (fyi my intent with these pics are for the guys getting their feet wet to get an idea of how scaling can affect their results)

And lastly, here is injector B scaled 12.5% up from the base map. Unless i am blind to something else that is alarming, i think this is pretty much right on the money while allowing a 'safety' factor.

My targets are 11.87

Last week I was messing with my injector bank 1 size to hone in on some better ST and LTFT values in the 1500 to 4000 RPM range.

I got some pretty good values, but the problem I am seeing is similar to Carbon, at idle, my LTFT is -6. Any sort of rpm increase sends it to near 0. I monitored it live data as I was driving. LT and ST are between 0 and 2 +/-

See my attached log. (rename to .csv)

I also attached an image of the compiled logs I have been doing. As you can see, all my fuel trims are doing decently (first section of the screen capture)

Since I got these pretty damn close, I went on to the 4500 to 6000 logs, thats where my idle issue showed up. I also wanted to know if this is a a time where I would adjust the VE map. You can see that I am 12% lean at 5500rpm.

I haven't changed anything between the first compiled table and the second aside from drive the car.

Any suggestions?

because you changed the P1 injector sizing so it affects *everything* P1 injector related, which not only includes idle but also WOT AFR across the board and a few other things too

I don't make suggestions that people "calibrate MAF first" or "don't touch the VE table unless you really know what you're doing" just to see myself post ...

Is there really ever a case where your primaries would be scaled far off? I can understand secondaries as they may get affected by increased airflow due to an intake or something... but primaries shouldn't need scaling unless you have old injectors with flow issues .. or new injectors that are a different size.... would that be a fair assumption?