MAF Scaling - Open/Closed Loop Fueling info

[troek]

heres what i got so far, the corrections % prett much mirror the fuel trims im seeing, makes sense to me i guess.

[steve99]

Originally Posted by troek

heres what i got so far, the corrections % prett much mirror the fuel trims im seeing, makes sense to me i guess.

The fuel trims at idle tend to vary a bit as thats where the iat can climb when your stuck in traffic, its also where you get the most variation in iat especially on a hot day and you have been going hard.

looks fine to me

[troek]

the ltft is usualy -3 to -5% now, and ive noticed what you mean about the IAT affecting it, right around 90 i start to see the changes it seems. while cruising with steady throttle the fuel trims rarely go over 2-3% in either direction. so ill maybe do another round of scaling on closed loop after i put some miles on, then start on the open loop. during open loop fueling i never see above 3-4% so i would say its not realy necessary, but its fun learning how to do this. this program is pretty nifty and this is a great thread. i started yesterday with aboslutely no idea what i was doing, and after reading everything and everyones help, i feel i got a solid basic understanding of this. im constantly amazed at the amount of help and expertise you find on the internet.

[Kodename47]

Originally Posted by troek

the ltft is usualy -3 to -5% now, and ive noticed what you mean about the IAT affecting it, right around 90 i start to see the changes it seems. while cruising with steady throttle the fuel trims rarely go over 2-3% in either direction. so ill maybe do another round of scaling on closed loop after i put some miles on, then start on the open loop. during open loop fueling i never see above 3-4% so i would say its not realy necessary, but its fun learning how to do this. this program is pretty nifty and this is a great thread. i started yesterday with aboslutely no idea what i was doing, and after reading everything and everyones help, i feel i got a solid basic understanding of this. im constantly amazed at the amount of help and expertise you find on the internet.

You don't have to wait as the tool uses both LTFT and STFT. What I would say is that set the maximum IAT filter to keep your temperatures within a decent range. Once you have a good scale, don't worry about constantly chasing 0% error as it will change day to day. The next area to check is the OL to CL transition. Using the spreadsheet tool as a checker/smoothing can be useful. I've also started checking non-WOT OL scaling which made a big improvement on my car. Not sure if it will be as big a change for NA but I'll throw up a how to anyway.

[Kodename47]

Here's my look at non-WOT OL MAF changes. I was seeing up to 3v in my CL scaling but I noticed that I could be going into OL as low as around 2.5v and I was actually going lean, which could be due to the vac/boost change on my S/C. The lowest MAFv I could get by using the main OL scaling screen was over 3v so this was of no use in this instance. So, to do this you need to add a couple of columns into the spreadsheet. Here are the formulas, ensure that once you've calculated these you then copy and paste the actual cell values into the spreadsheet in place of the formulas. Otherwise the tool can't use them. If you paste the below into the 1st row of excel then it will work once you use the correct cells.

1 - dV/dt: =IFERROR(ABS(((V2-V1)/(t2-t1))*1000),0) - (V = MAFv, t = time in ms)
2 - Error %: =IF(Fuel Sys Status=4,((AFR/Comm AFR)-1)*100,LTFT+STFT)

Then load up the log stats screen.
-Create custom Axis: X Axis: 1,2,4 / Y Axis: MAFv Range
Then do as below, note the data columns and filters:


So in essence you have all OL and CL error, which is also pretty much what the OL and CL tools do.

I use the count function to determine where my OL and CL transition areas are, I've highlighted that in red. This is easily done as the cell count in OL starts to increase. You can then apply the % change of these cells to your MAF scale. You can see in my picture that between 2.75v and 3v the error goes crazy.

I highlighted in black the areas where the normal OL scaling would take over so this can be discounted.


It's also worth noting to be wary of areas where you tip in the throttle regularly. As the AFRs catch up with the commanded AFR if the swing is large enough, the displayed error % could be quite large when in reality it's not an issue. This will only be temporary but can be quite influential on the averaged error figures. Once you know that the error % is fairly consistent, you can reduce the maximum error % filter to give you better results. It's also why in the one above I would discount the 2.5v and possibly the 2.57v values.

[AL008]

Does anyone get the following error in the log when you try to open the .jar file?

2014-06-29 11:11:27.141 [ERROR] Config.java:342 - java.io.FileNotFoundException: config.xml (The system cannot find the file specified)

I double click and it does nothing.

[solidONE]

I found that when taking logs of the fuel corrections it's a good idea to run in the low rpm and low maf voltage range at the beginning before the intake plumbing soaks up heat then run it in the middle then higher rpm maf voltage range as the engine gets warmer and warmer. This will make sure your IAT's are nice and tight within a similar range throughout the entire log.

[steve99]

Originally Posted by AL008

Does anyone get the following error in the log when you try to open the .jar file?

2014-06-29 11:11:27.141 [ERROR] Config.java:342 - java.io.FileNotFoundException: config.xml (The system cannot find the file specified)

I double click and it does nothing.

Ok I just downloaded the latest version from here
https://github.com/vimsh/mafscaling

Works fine for me :-)

You need use the "run.bat" file to start the program not the .jar file.
if that does not work you might need to update your java runtime and try running as administrator

[AL008]

I did my closed loop run and I think it's a bit screwed up, I'm getting up to 11% st fuel trim at around 1800 rpms...

[vgi]

Originally Posted by AL008

Does anyone get the following error in the log when you try to open the .jar file?

2014-06-29 11:11:27.141 [ERROR] Config.java:342 - java.io.FileNotFoundException: config.xml (The system cannot find the file specified)

I double click and it does nothing.

you only get this error (in the log file) the first time you run the program as the file doesn't exist yet - it's created/updated when you close the program.

[AL008]

I have done over 4 runs so far to correct my MAF Scaling for CLOSED loop and I don't think it is working as expected, maybe I am doing something wrong here but here is what I have done.

Log before any application of VGI scaling seemed to indicate STFT ranging from -9.38% to 10.94%, highway driving and no stops.

This log was Highway driving, no stops, constant throttle at about 50-60 degrees farenheit. processed it through the VGI program and loaded onto car.

STFT's seemed to hit around -12/15% up to +15% at times throughout the ranges.

My latest log (after the 4th time of doing these scales) probably the same run as my original (highway, no stops, constant throttle, 60-70 degrees farenheit) resulted in
STFT -10.16% to 12.5%..

I am kind of lost as to what I am doing incorrectly. Any help would be greatly appreciated.

[steve99]

Originally Posted by AL008

I have done over 4 runs so far to correct my MAF Scaling for CLOSED loop and I don't think it is working as expected, maybe I am doing something wrong here but here is what I have done.

Log before any application of VGI scaling seemed to indicate STFT ranging from -9.38% to 10.94%, highway driving and no stops.

This log was Highway driving, no stops, constant throttle at about 50-60 degrees farenheit. processed it through the VGI program and loaded onto car.

STFT's seemed to hit around -12/15% up to +15% at times throughout the ranges.

My latest log (after the 4th time of doing these scales) probably the same run as my original (highway, no stops, constant throttle, 60-70 degrees farenheit) resulted in
STFT -10.16% to 12.5%..

I am kind of lost as to what I am doing incorrectly. Any help would be greatly appreciated.

Short term fuel trims will always vary widly as throttle applied load changes ect. Its the long term fuel trims you need to look at during cruise and idle

long term fuel trim is the average over time of the short term fuel trim in the four bands.

as long as the long term fuel trims are low all is good

[AL008]

Originally Posted by steve99

Short term fuel trims will always vary widly as throttle applied load changes ect. Its the long term fuel trims you need to look at during cruise and idle

long term fuel trim is the average over time of the short term fuel trim in the four bands.

as long as the long term fuel trims are low all is good

I just monitored before.

Idle LTFT: 3.12
During cruising it hovers around 0% most of time
Backing off acceleration hit 7% briefly in some spots.

Does this seem about right?

Thanks

[steve99]

Originally Posted by AL008

I just monitored before.

Idle LTFT: 3.12
During cruising it hovers around 0% most of time
Backing off acceleration hit 7% briefly in some spots.

Does this seem about right?

Thanks

Yep sounds pretty good , the idle LTFT seems to vary a bit as that where you get the most iat variation ie when stopped heats up heaps and throws trims off. Some jokers play around with iat/maf compensations but unless your rearly keen best leave it alone.

post up a log if you want.