tuning frs maf/injector

[QTR FMS]

i tried to scale manually and spent 12 hours on the street using vgi and my afr error wont get better than this graph, i tried different front 02 scale that helped a bit but still,


i was wondering if engine load compensation is what cause this issue i read @Tor old thread and he mentions the same issue around 3500 rpm, did anyone came with better engine load compensation or can i just set the value to 1

[tomm.brz]

do you have stock engine load limits?
it forces a lean condition before 3k even if you try to make it richer and then maybe you make richer even the maf voltage around 3200-3400 and then it goes too rich

also you have ecutek or am i wrong? you could set the racerom v10 and enable the Closed loop fueling , it fixes the afr for you

[QTR FMS]

Originally Posted by tomm.brz

do you have stock engine load limits?
it forces a lean condition before 3k even if you try to make it richer and then maybe you make richer even the maf voltage around 3200-3400 and then it goes too rich

also you have ecutek or am i wrong? you could set the racerom v10 and enable the Closed loop fueling , it fixes the afr for you

im using ecutek but i dont have racerom, and i would rather fix the issue than work around it

[tomm.brz]

well rr10 CL works extremely good i have racerom license, i could set it up for you for a small fee if you are interested

anyway you haven y answered about engine load limits, and if at 3400 rpm is that rich, maf scaling at that voltage is too high

[QTR FMS]

Originally Posted by tomm.brz

well rr10 CL works extremely good i have racerom license, i could set it up for you for a small fee if you are interested

anyway you haven y answered about engine load limits, and if at 3400 rpm is that rich, maf scaling at that voltage is too high

yeah im using stock engine load compensation and i dont really know what im supposed to do with it, i tried to work with maf scale but i have been chasing my tail for the last 5 months

[tomm.brz]

i asked you about "engine load limits" not compensation !

[QTR FMS]

Originally Posted by tomm.brz

i asked you about "engine load limits" not compensation !

yeah that stock too i havent touch lots of tables on this tune

[Tor]

Originally Posted by QTR FMS

i tried to scale manually and spent 12 hours on the street using vgi and my afr error wont get better than this graph, i tried different front 02 scale that helped a bit but still,


i was wondering if engine load compensation is what cause this issue i read @Tor old thread and he mentions the same issue around 3500 rpm, did anyone came with better engine load compensation or can i just set the value to 1

I set all values above 0.4 bar to 1 before starting MAF scaling (in the examples 0 as RomRaider use %). I then start out with a good smooth scale, make 2-3 WOT pulls and check where it has peaks and dips. Example from one of the last I did:



I then I roughly correct the dip/peaks from 0.9 bar and above:



With those corrections let the owner do full logging. Depending on the fuel errors I either do MAF scaling/Load limit corrections (emphasis to differentiate from ELC) and then repeat the above steps (without zeroing out).

Or if the fuel errors are small in both OL and CL (as they were in this example), I only do load limit correction, OL MAF scaling and otherwise apply the error corrections in the ELC table (CL corrections below 0.9, manual corrections above to control WOT curve). In this case, it ended up looking like this:



Which produced this result:



Then I let the owner drive for 150 miles and then do a check on the LTFT. In this case, the WOT LTFT was 0.78 after 150 miles.

I this case it was lucky the initial MAF was pretty close in CL, and I had it done in 2 revisions and could move on to ignition timing. Otherwise it takes a few more revisions to get the MAF scale done in between.

This is the method that works for me. Others may have other procedures. Not that many even bother with the ELC table. For my way of doing it, it is essential.

The ELC table is going to be as individual for each car as the MAF scaling. But they always have the same pattern for a given header. It depends on where your MAF scales average the differences out. I don't like applying more than 6-7% corrections max in ELC (except below 0.3 bar as stock). If the corrections needed are higher than that I rescale the MAF again. Or if you need to apply large negative or positive corrections over the entire table, your MAF is not averaging the differences out (either too high or low).

[QTR FMS]

Originally Posted by Tor

I set all values above 0.4 bar to 1 before starting MAF scaling (in the examples 0 as RomRaider use %). I then start out with a good smooth scale, make 2-3 WOT pulls and check where it has peaks and dips. Example from one of the last I did:



I then I roughly correct the dip/peaks from 0.9 bar and above:



With those corrections let the owner do full logging. Depending on the fuel errors I either do MAF scaling/Load limit corrections (emphasis to differentiate from ELC) and then repeat the above steps (without zeroing out).

Or if the fuel errors are small in both OL and CL (as they were in this example), I only do load limit correction, OL MAF scaling and otherwise apply the error corrections in the ELC table (CL corrections below 0.9, manual corrections above to control WOT curve). In this case, it ended up looking like this:



Which produced this result:



Then I let the owner drive for 150 miles and then do a check on the LTFT. In this case, the WOT LTFT was 0.78 after 150 miles.

I this case it was lucky the initial MAF was pretty close in CL, and I had it done in 2 revisions and could move on to ignition timing.

This is the method that works for me. Others may have other procedures. Not that many even bother with the ELC table. For my way of doing it, it is essential.

The ELC table is going to be as individual for each car as the MAF scaling. But they always have the same pattern for a given header.

thats really interesting i wish i knew about it long time ago, so much time i wasted datalogging because of this table, how low can i start with, can i set 1 below 0.4 bar

[Tor]

Originally Posted by QTR FMS

thats really interesting i wish i knew about it long time ago, so much time i wasted datalogging because of this table, how low can i start with, can i set 1 below 0.4 bar

I don't bother for several reasons:

1) How much time does the engine actually spend below 0.4 bar? The answer is not a lot. Below 0.3 it's mainly coasting, or very very low accelerator pedal (like 1%, foot just resting at the pedal, driving down down a hill and wanting to avoid engine braking. Most driving is done between 0.4 and 0.8, 2000 rpm to 5000 rpm.

2) Because of the above, you can adjust it forever. You pretty much get a different result from each set of logs.

3) In any case, I think you are better off starting out from the stock values since they will be pretty high in any case.

4) If it's running below 0.3 bar, it will with high likelihood run in CL with STFT correcting any errors.

5) WOT LTFT is never set in this area. But in the area mentioned in 1 where most driving takes place. Hence that area is super important to get right to be sure to get a low LTFT at WOT.


Basically, I mostly care about LTFT for WOT. All other fuel trims will be applied in CL only. It's nice that they are low, but it doesn't really matter that they are super low from a fueling standpoint (large error would indicate a poorly scaled MAF which could have an influence on load calculations).

Idle LTFT you can forget completely. Just let the engine idle for 2-3 mins and look how it's going to vary with IAT. It idles at 0.3 and 0.33 bar. If you wanted to change the idle fuel trim you can theoretically just add whatever value you want in those two fields for your idle rpm.

[QTR FMS]

i ended up buying racerom, and the first thing i noticed is how inaccurate the front o2 sensor is compared to aem wideband, this is stock scaling in the graph i tried different scaling i found online but there werent any better so i went back to stock o2 scaling,


this was one of the main issues i had i trusted the front o2 too much and i was wondering why i was having issue getting my afr close, right now my highest afr is 12.7, lowest is 12.1 but with afr smoothing the graph its between 12.58 and 12.33 while the target is 12.46 i might try to get it closer and smoother if possible before going to dyno



i found out about standard deviation in log stats in vgi tool and its way better for me than using openloop and closed loop it was really helpful i wish it was already built into ecutek


using the standard deviation i got my fuel trim average +-1% but on my ecutek its -7% +27% so im not sure if those are just tip in error or something should i ignore it or can i get it closer?


i tried using engine load compensation i have set 3600 to 0.95 and i think my afr looks the same as before maybe it did improve only at 3600 but not around 3590 and 3900


from what i read about elc i think it doesnt effect the actual afr just the target right?


edit: just remembered i have scaled the front o2 to read 0.72 lambda at -0.76 and it seems the actual afr is off by about 1 afr