SPREADSHEET: Scale Closed Loop MAF Table and Injectors

nj1266 · **Joined:** Thu May 03, 2007 2:02 pm **Posts:** 70

Should I be concerened that a stock STi has AFR Learning #1 between -6.25 and -9.38 at idle? Or is that normal? I attached the log

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 3040

Is the car completely bone stock? With intake silencer in place?

AFR Learning A between 0% and -10% is really not a concern.

However, if you are planning to use the injector scaling function in the future, I would correct the MAF scale so that it is between 0% and -5%. I do this by using constant throttle walking speed logs in the parking lot in 1st gear.

rodney757 · **Joined:** Wed Feb 18, 2009 5:06 pm **Posts:** 37

I'm using this spreadsheet to scale my MAF and from 2.62 - 2.72v I have over 80 pieces of data but there is no numbers in the S column. I thought that there only had to be over 20 pieces of data and it would give you a number. Should I just use the number in the N column since its the same as the number in the S column if I have enough data for that particular voltage? Thanks

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 3040

Winzip the file and pm or post it.

rodney757 · **Joined:** Wed Feb 18, 2009 5:06 pm **Posts:** 37

Heres the file.

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 3040

It's because you have more data than I expected.

Copy the formula from S76 and fill the cells below it.

rodney757 · **Joined:** Wed Feb 18, 2009 5:06 pm **Posts:** 37

Okay its working now. Thanks for this very helpful spreadsheet.

pwrx · **Posted:** Mon Mar 23, 2009 1:16 pm

Mickey,

I gave up on this stuff for a long time, but I finally decided I should finish up and get my injector scalar/latency/maf all dialed in.

I was running a scalar of 659.x, and after running a bunch of highway cruise data through your spreadsheet it came up with a latency -.07ms, and bumping up the scalar by about 10cc. I applied these changes, and logged on the way to work today. I know I don't have enough data yet, but its still starting to take shape, so I wanted to solicit a little feedback.

The spreadsheet now tells me that the latency is spot on. however it says I need to raise my scalar even higher to 692cc. HOwever when I look at the graphs, it looks like I'm pretty close on the upper end of the closed loop data. And my learning view says the same on the A/F Learning D field.

Code:

CAL ID:A4TC401L
Description: 03 USDM Impreza WRX MT
ECU ID:3614486405

IAM    16

A/F Learning #1:
0 - <5.60    5.60 - <10.00    10.00 - <50.00    50.00+
4.20         1.17             -0.59             -0.68 

FLKC Table       0 - <0.50    0.50 - <0.70    0.70 - <0.90    0.90 - <1.10    1.10 - <1.30    1.30 - <1.50    1.50 - <4.00    4.00+
0 - <1400        0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
1400 - <1800     0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
1800 - <2600     0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
2600 - <3400     0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
3400 - <4200     0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
4200 - <5000     0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
5000 - <12800    0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 
12800+           0.00         0.00            0.00            0.00            0.00            0.00            0.00            0.00 

Wouldn't shifting my scalar even higher cause me to have more positive corrections?

Should I leave the injector data as is and just finish off the fine tuning with MAF corrections, or do you think the injectors are still off? (assuming the data stays the same after adding another commute worth of data to the xls)

Attached is the spreadsheet, feedback greatly appreciated.

Thanks man,
pwrx

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 3040

i would just leave it. Your AFR Learning looks really close.

pwrx · **Posted:** Mon Mar 23, 2009 2:24 pm

mickeyd2005 wrote:

i would just leave it. Your AFR Learning looks really close.

Leave the injector scalar/latency and scale the MAF in the bottom end?
Or leave everything?

Thanks,
pwrx

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 3040

I wouldn't touch anything below 60 g/s. Your AFR Learning A is close enough.

pwrx · **Posted:** Mon Mar 23, 2009 2:45 pm

mickeyd2005 wrote:

I wouldn't touch anything below 60 g/s. Your AFR Learning A is close enough.

Learning A is +4.2%. Isn't this still a bit high?

Looking at "MAF Check 1", I'm steadily around +5% below 1.6v. Also, your spreadsheet notes that idle trims are usually -3 to -6%, however mine are about +4. Does this not deserve some adjustments? I'm still getting some crazy correction swings on Tip In/Out, and figured cleaning this range up could help. Am I just being too anal?

Thanks again,
pwrx

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 3040

Really... I think everything is fine.

I used to try to get AFR A within +0 to -6% but now I just let it fall withing +-6%.

pwrx · **Posted:** Mon Mar 23, 2009 7:10 pm

Ok, thanks for the feedback.

So, now to scale in OPEN loop above 60g/s.

I have made a new copy of my ROM similar to what you have in post 1 (but a bit richer in the top, just in case). Is there a good thread on how to best do this scaling (important parameters to log, steps to repeat, etc)?

Thanks,
pwrx

wrxsti-l · **Posted:** Wed Apr 22, 2009 11:52 pm

Quick question Mickey.

What does it mean if in learningview A/F learning are pretty much under 1%, but Airboy's spreadsheet reports fairly large (13%) A/F errors?

Learningview results - apart from 0-4.2, all are under 1% (please ignore knock):

Airboy's spreadsheet - AFR Interpolate results:

What causes this and how can I correct it? It makes it all but impossible to hit target AFR.

RomRaider

Forum rules

SPREADSHEET: Scale Closed Loop MAF Table and Injectors

Who is online