MAF Tuning Getting Started

DCMoney

Goblin Guru
Need some help getting started tuning the MAF.

I've been watching Goat Rope Garages videos on tuning, started with the first video and got to where he goes through talking about tuning the MAF. I've watched this video more times than I can count. What I'm trying to log is the EQ error ratio like detailed in his video. Link to where he starts talking about the math needed to find the EQ error

Im using Chads templates to set the values to tune the MAF. I have a AEM Wideband hooked up through a serial to usb connector and it seems to be working. Just cant get anything to graph with the MAF Error graph.

Am I going about it the wrong way? Has anyone else done this?
1.JPG
 

DCMoney

Goblin Guru
Think I figured out why I can get that graph to log anything. Doesn't look like we have the equivalence ratio commanded sensor. Is there another sensor I can use?

2.JPG
 

DCMoney

Goblin Guru
Thanks for everyone's help! LOL

Got it to work, used the following to take the Commanded AFR divided by 14.7 and got data. Not sure if this is the right way to do it but it worked.

24419



24418
 

Ross

Goblin Guru
Looks like you got it all figured out.
I calculated the error percentage from the perspective of the commanded AFR. That is why my divisor is the commanded.
You calculated it from the sensor AFR perspective. Probably both will work fine.
Good work!

24421
 
Last edited:

DCMoney

Goblin Guru
Looks like you got it all figured out.
I calculated the error percentage from the perspective of the commanded AFR. That is why my divisor is the commanded.
You calculated it from the sensor AFR perspective. Probably both will work fine.
Good work!

View attachment 24421
Thanks, good to know I'm moving in the right direction.
 

DCMoney

Goblin Guru
Noticed while tuning my MAF there are a few points where adjusting the values do very little to change their actual value. Any ideas on why its only at these two points 2700 Hz and 3000Hz? Also noticed in my other logs the values below 3000Hz seem to be all over the place, where above 3000Hz I've gotten them to stay closer to +- 1.5%

Run 1.JPG


Maf set 6.JPG
View attachment 24741
 

Attachments

ctuinstra

Goblin Guru
Can you save your scanner layout and upload it? Hopefully we can attach XML files. My layout and math is completely different than yours and the values I get seem to fall in line.
 

ctuinstra

Goblin Guru
Have you complete the VE tuning? How did that turn out?

When doing the MAF, the throttle should be very smooth and slow. I see in the scan there is a lot of jitter and fairly abrupt movements in the throttle.
 

DCMoney

Goblin Guru
Have you complete the VE tuning? How did that turn out?

When doing the MAF, the throttle should be very smooth and slow. I see in the scan there is a lot of jitter and fairly abrupt movements in the throttle.
I haven't started the VE turning yet. I rewatched a few of the Goat Road Garage videos and he mentioned smooth throttle application, yeah... I've been getting data driving to work not exactly being easy on the throttle. Next time I go out I'll do a run specifically for collecting smooth throttle applications.
 

ctuinstra

Goblin Guru
Thanks! Zipped them, see if that works.
That works but I'm still missing data or something. All the values I see are much higher, but still show some differences.
24759


The values you have aren't really very linear to those around it.

24760


Maybe try something like this:

24762


Something still a bit off, usually as you get closer to the commanded ARF, the O2 oscillates a lot between rich and lean, that's how you know you are riding the fence. But in your case, it's still running rich most of the time other than the dips around the area that you are questioning (2400-3000hz).

24763


Here is an example of my O2 sensor at the final (and no longer in tuning mode):

24764
 

Attachments

ctuinstra

Goblin Guru
I haven't started the VE turning yet. I rewatched a few of the Goat Road Garage videos and he mentioned smooth throttle application, yeah... I've been getting data driving to work not exactly being easy on the throttle. Next time I go out I'll do a run specifically for collecting smooth throttle applications.
I personally did the VE first. I know there is conflicting information out there about which is better to do first. Your VE tables show a peak that I would smooth out. But you could spend all day tweaking the table.

24765


24766
 

Ross

Goblin Guru
Chad, did you build a table for his data? I can't see it.
Andrew, Can I get your graph.table.xml for your MAF error table? Or any other tables you want me to see.
 

DCMoney

Goblin Guru
That works but I'm still missing data or something. All the values I see are much higher, but still show some differences.
View attachment 24759

The values you have aren't really very linear to those around it.

View attachment 24760

Maybe try something like this:

View attachment 24762

Something still a bit off, usually as you get closer to the commanded ARF, the O2 oscillates a lot between rich and lean, that's how you know you are riding the fence. But in your case, it's still running rich most of the time other than the dips around the area that you are questioning (2400-3000hz).

View attachment 24763

Here is an example of my O2 sensor at the final (and no longer in tuning mode):

View attachment 24764
Ah, could be missing a divide by 10 somewhere in the math, I'm logging in lambda.

I turned off all the MAF only tuning the O2 chart logged like this at idle.

Attached my MAF Error table.

O2 Graph.PNG
 

Attachments

Ross

Goblin Guru
Dang, VCM Scanner sucks...still can't see your data. Now I think I need your Math.Parmameter1.xml
 

OptimizePrime

Goblin Guru
Do you have your evap plug plugged in when you're taking these scans? If so, unplug to lock the LTFT's while you're dialing in STFT's

Are these scans long? I see one is 3mins. Usually I go for a quick spin before I start logging to get everything (not just coolant temps) up to operating temps and then go for a long scan as you can have outliers like this show up.
 

Ross

Goblin Guru
Yes, I did. If your Math1 parameter uses the same parameters as his logs, then it will work.
My math parameters didn't fill the table... but I do see his data logs, just need to reconstruct his table.
I get the red dots in the corner of the MAF Error table - which indicates a missing parameter, or an error getting the data.
24770
 
Last edited:

DCMoney

Goblin Guru
Do you have your evap plug plugged in when you're taking these scans? If so, unplug to lock the LTFT's while you're dialing in STFT's

Are these scans long? I see one is 3mins. Usually I go for a quick spin before I start logging to get everything (not just coolant temps) up to operating temps and then go for a long scan as you can have outliers like this show up.
That 3 min run was just running in the garage to see what the car would do after turning everything that was off to MAF tune back on. I thought I did away with the evap during the build? Believe the template file Chad shared disabled LTFTs.

Yes, I did. If your Math1 parameter uses the same parameters as his logs, then it will work.
My math parameters didn't fill the table... but I do see his data logs, just need to reconstruct his table.
I get the red dots in the corner of the MAF Error table - which indicates a missing parameter, or an error getting the data.
View attachment 24770
Well **** that means something in my math/graph is wrong as well, but I've gotten data from it.

Another variable that I'm not sure how much of an affect it will have but some of my morning runs to work the temps would be in the 60's then the drive home would be closer to 100.

Appreciate everyone's help with this.
 

ctuinstra

Goblin Guru
Another variable that I'm not sure how much of an affect it will have but some of my morning runs to work the temps would be in the 60's then the drive home would be closer to 100.

Appreciate everyone's help with this.
Air density will be different for the two runs. I would try to stick to doing all the tuning at one time and the same air. My original tune was done in the summer in the hotter air and earlier this year in the cooler air, it was wanting to add a lot of fuel. I re-tune in the cooler, dense air so it was would remove fuel in the hotter summer air and not have to add in the the cooler, denser air. Not that it really matters much because the ECM will correct for the air, but it was just my preference.
 
Top