Welcome to the North American Subaru Impreza Owners Club Wednesday January 28, 2015
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Engine Management & Tuning > Open Source Reflashes

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 02-23-2008, 03:43 AM   #101
mickeyd2005
Scooby Newbie
 
Member#: 122595
Join Date: Aug 2006
Chapter/Region: NWIC
Vehicle:
2005 LGT Wagon
ABP

Default

^^^ I added a newer version of the spreadsheet to it... but truth be told... the spreadsheet isn't really needed.

I think it does calculate latency better than using fast idle logs. Fast idle corrections seems to vary depending upon heat soak so using the full data range for latency seems to be more accurate.
* Registered users of the site do not see these ads.
mickeyd2005 is offline   Reply With Quote
Old 02-24-2008, 01:05 AM   #102
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

Guys, is it a bad idea to tune for AFR at the same time as I am scaling my MAF? If so, how do I know that I've achieved my correct (or perfect) MAF scaling and can proceed to tune AFR and other parameters?
savageblitz is offline   Reply With Quote
Old 02-24-2008, 01:12 AM   #103
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

Tune closed loop maf scaling with this method. Tune open loop maf scaling by setting your fuel targets to something easy, like filling the entire OL portion of your fuel map with 12 or something (so you have an easy target to hit) adjust the upper end of the MAF scaling until your WBO2 readings come close to those targets over as much of the map as you can. Switch back to whatever fueling you actually want to be running and then finesse the fuel targets until you're hitting exactly what you mean to be running.

Note that if you don't tune the CL portion of your MAF scaling pretty damned well, AF Learning #1 D will drift enough that you'll never manage to make your OL fueling stay consistent.
williaty is offline   Reply With Quote
Old 02-24-2008, 10:48 AM   #104
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

Quote:
Originally Posted by williaty View Post
Tune closed loop maf scaling with this method. Tune open loop maf scaling by setting your fuel targets to something easy, like filling the entire OL portion of your fuel map with 12 or something (so you have an easy target to hit).
Do you mean filling up the parameters "Primary Open Loop Fueling A" and "Primary Open Loop Fueling A" ? Why is there a parameter A and B?

Quote:
Originally Posted by williaty View Post
adjust the upper end of the MAF scaling until your WBO2 readings come close to those targets over as much of the map as you can. .
So basically I will definite need to use a WBO2 for this right? Why can't I use the AFR readings from the ECU?

Quote:
Originally Posted by williaty View Post
Switch back to whatever fueling you actually want to be running and then finesse the fuel targets until you're hitting exactly what you mean to be running.
Sorry, I'm still a little confused here. Am I still doing MAF Scaling (for OL) now or am I tuning for AFR?

Quote:
Originally Posted by williaty View Post
Note that if you don't tune the CL portion of your MAF scaling pretty damned well, AF Learning #1 D will drift enough that you'll never manage to make your OL fueling stay consistent.
How will I know if my CL portion of my MAF scaling is good enough? If I were to follow your MAF scaling method, is it when there are no more mean and mode values to change in my MAF table?

Thanks for your patience in answering my questions.
savageblitz is offline   Reply With Quote
Old 02-24-2008, 01:18 PM   #105
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

Quote:
Originally Posted by savageblitz View Post
Do you mean filling up the parameters "Primary Open Loop Fueling A" and "Primary Open Loop Fueling A" ? Why is there a parameter A and B?
You'll have to ask someone else why there's an A and B. Yes, I mean filling that table with a single value for everything above, say, 1.3g/rev or something like that.



Quote:
So basically I will definite need to use a WBO2 for this right? Why can't I use the AFR readings from the ECU?
You will have to have a WBO2 to tune the OL range of MAF Scaling and to dial in the fueling on the top end. The stock O2 sensor is only any good right near stoich.



Quote:
Sorry, I'm still a little confused here. Am I still doing MAF Scaling (for OL) now or am I tuning for AFR?
Honestly, it's both. You tune you MAF scaling until you're coming as close to your Primary OL Fueling map as possible. Then you finesse those last little errors out of existence by changing the Fueling map. For instance, being NA, my final target AFR is 12.5. I tuned my CL MAF Scaling until my modal correction was zero and all my AF Learnings were <3%. Then I tuned my OL MAF scaling until it was hitting as close to 12.5 all the way up the rev range at WOT a I could get it. However, MAF scaling alone still let AFR wobble up and down between about 12.2 and 12.8 at WOT as the revs changed. So I very slightly changed the Primary OL Fuel map in those cells where I wasn't hitting exactly what I wanted to be until I WAS hitting exactly what I wanted to be.

MAF scaling is a chainsaw, Primary OL Fueling is a scalpel. Get it as close as possible with the chainsaw before you start on the detail work with the scalpel.



Quote:
How will I know if my CL portion of my MAF scaling is good enough? If I were to follow your MAF scaling method, is it when there are no more mean and mode values to change in my MAF table?
I stopped when almost all of my modal corrections were 0 and my averages were less than 1% and my AF Learnings were <3%. It's the AF Learnings that really matter.
williaty is offline   Reply With Quote
Old 02-26-2008, 10:28 AM   #106
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

ty,

thanks so much for your replies.

I've got several questions regarding your method of MAF scaling.

1. When logging, you only used AF Correction #1 and AF Learning #1. Why aren't you using #2 and #3? Wouldn't the scaling be more accurate if #2 and #3 were included as well?

2. In the first filtering of data, you exclude the spikes for dMAFv/dt. Is the reason for doing this is that when dMAFv/dt is very high, the ECU is not measuring air flow accurately (maybe due to latency) and thus the data would be useless. Also, is this the reason why when logging, we should move the throttle slowly?

3. When logging I noticed I have very little data for MAFv > 2.66. How do get more data at higher values so that I can do my MAF scaling at those higher regions?

4. What would be the implications if I do not do proper scaling at higher MAFv? Does that mean that at higher boost, my ecu will not calculate the AFR accurately?

5. BTW, just to confirm: -ve values in correction/learning means the ECU is leaning out AFR - ie. feeding less fuel right?

6. You mentioned that MAF scaling for idle is freakish.. how important is MAF scaling for idle? is it just for fuel consumption efficiency?

thanks so much for your replies.. I'm learning so much from your thread.
savageblitz is offline   Reply With Quote
Old 02-26-2008, 01:05 PM   #107
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

Quote:
Originally Posted by savageblitz View Post
1. When logging, you only used AF Correction #1 and AF Learning #1. Why aren't you using #2 and #3? Wouldn't the scaling be more accurate if #2 and #3 were included as well?
No, #2 and #3 aren't fully understood and aren't first-order effects.

Quote:
2. In the first filtering of data, you exclude the spikes for dMAFv/dt. Is the reason for doing this is that when dMAFv/dt is very high, the ECU is not measuring air flow accurately (maybe due to latency) and thus the data would be useless. Also, is this the reason why when logging, we should move the throttle slowly?
Yes, if the amount of air is changing too rapidly, it'll take a moment for the ECU to catch back up via the feedback loop.

Quote:
3. When logging I noticed I have very little data for MAFv > 2.66. How do get more data at higher values so that I can do my MAF scaling at those higher regions?
Well, you can make it more difficult for the car to get out of Closed Loop, but in a boosted application, I'm not sure that's wise. The real answer is to get a wideband.

Quote:
4. What would be the implications if I do not do proper scaling at higher MAFv? Does that mean that at higher boost, my ecu will not calculate the AFR accurately?
Correct. Without correct MAF scaling up top, the fueling won't be right, the timing won't be right, the... etc. Anything that references airflow or load requires proper MAF scaling all the way up.

Quote:
5. BTW, just to confirm: -ve values in correction/learning means the ECU is leaning out AFR - ie. feeding less fuel right?
Yes, the correction is what the ECU is doing to the fuel + is more fuel, - is less fuel.

Quote:
6. You mentioned that MAF scaling for idle is freakish.. how important is MAF scaling for idle? is it just for fuel consumption efficiency?
Closed Loop will eventually force idle to stoich. However, the O2 sensor is not active for the better part of a minute after you turn the car on. If you get your idle scaling off too far, the car won't want to idle before the O2 sensor is ready. It would also fail to idle if the O2 sensor failed.
williaty is offline   Reply With Quote
Old 02-27-2008, 03:36 PM   #108
SubiSam
Scooby Newbie
 
Member#: 156401
Join Date: Aug 2007
Chapter/Region: South East
Location: Greensboro NC
Vehicle:
'03 WRB WAGON
2.1 stroker GTX3071R

Default

Subscribed
SubiSam is offline   Reply With Quote
Old 03-01-2008, 10:58 AM   #109
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

ty,

on average, how many iterations does it take to get CL MAF Scaling done? I'm into my fifth iteration now and there's always some small corrections to make..

my wideband just arrived today (Innovate LM1). Sorry, I'm kind of new to this but how
do I use this to tune my OL MAF Scaling?

You've mentioned that I should set my Primary Open Loop Fueling maps to a single value say 12. What should I do next? Do WOT and do logging using my wideband and then adjust the higher values of MAF scaling? Which values of MAFv should I normally start for OL MAF scaling?
savageblitz is offline   Reply With Quote
Old 03-01-2008, 04:56 PM   #110
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

If you don't do it all in a very narrow temperature range, you'll never get the whole thing settled. The best thing is to take a day where you can spend the whole day working on it. Do 3-4 iterations in a single day at a pretty constant temp and it'll fall in line.

Yeah, basically to sale OL, you do this, but instead of logging AFL, you log WBO2. So you take the actual AFR, use the AFC to figure out what it would have been hitting had AFC not been applied, then compare that to what you wanted to hit. The difference is your MAF scaling correction.
williaty is offline   Reply With Quote
Old 03-19-2008, 11:03 AM   #111
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

Quote:
Originally Posted by williaty View Post
Yes, I mean filling that table with a single value for everything above, say, 1.3g/rev or something like that.
ty,

I think I've almost got my CL MAF Scaling dialed in using your method. I've just got my LC-1 running and want to tune my OL MAF Scaling. Am I right to say that with OL MAF Scaling I will still be changing just the MAF Scaling table?

Using the method you described above, say I filled up everything above 1.3g/rev with a single value, what should I be logging?

Is AFR readings from my LC1 and MAFv enough? How do I know which values in the MAF Scaling table I should change? (I'm a little lost here in linking the values to change with what I am logging).
savageblitz is offline   Reply With Quote
Old 03-19-2008, 11:14 AM   #112
Rene2.5RS
Scooby Specialist
 
Member#: 4574
Join Date: Mar 2001
Chapter/Region: NESIC
Location: Bristol, CT
Vehicle:
2014 Outback 2.5i L
VRP

Default

Quote:
Originally Posted by savageblitz View Post
ty,

I think I've almost got my CL MAF Scaling dialed in using your method. I've just got my LC-1 running and want to tune my OL MAF Scaling. Am I right to say that with OL MAF Scaling I will still be changing just the MAF Scaling table?

Using the method you described above, say I filled up everything above 1.3g/rev with a single value, what should I be logging?

Is AFR readings from my LC1 and MAFv enough? How do I know which values in the MAF Scaling table I should change? (I'm a little lost here in linking the values to change with what I am logging).
When I am tuning for OL fueling I am referencing:
RPM, Engine Load, MAF v, LC-1, Map AFR

I use Airboy's interpolation spreedsheet to generate the map AFRs and then use the spreadsheet I made to reference what the MAF scaling was the error in Mapped values and logged values, and generates new values for the MAF scaling table. Change the map, reflash and then go out and do it again.

Like Williaty said this needs to be done roughly all at constant conditions.

-Rene
Rene2.5RS is offline   Reply With Quote
Old 03-19-2008, 12:18 PM   #113
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

Thanks Rene for your reply. When you mentioned Airboy's interpolation spreadsheet, are you refering the Excel spreadsheet in "Log file Interpolation ver3p7c Jan08.zip"?

Its quite complex and has many functions - so how do you use it to (I quote from your post):
1. generate the map AFRs
2. use the spreadsheet you made to reference what the MAF scaling was the error in Mapped values and logged values,
3. generate new values for the MAF scaling table

Sorry, I'm really lost. Do you mind elaborating abit further? Thanks.

(So now you can see why I used williaty's method coz he detailed his method step by step) :-)
savageblitz is offline   Reply With Quote
Old 03-19-2008, 02:01 PM   #114
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

If you like how Airboy's works, and you can make it work for you, use it, it's probably the simplest.

The way to do it without airboy's is to set the OL Fueling table so that all the areas above where you've currently got the MAF scaled are one value. I guess for a turbo you'd want to pick something rich like 11 but you should probably confirm that with someone who actually has a turbo.

Then go out and log MAFv, AFL, LC-1 AFR, and CL/OL Status for some driving that keeps you up on OL most of the time. When you get home, filter out the stuff with a high d/MAFv/dt just like the original method. Then filter out all of the CL stuff (just like you filtered out the OL stuff for the other method). Sort the whole thing by MAFv. Create a new column called "Target" and fill it with whatever your target AFR was. Create a new column called "Error". Fill the Error column with the formula ((((Actual-Target)/(Target))*100)+AFL). Then just do the modal correction thing like you did in the original method.


Basically the only difference is that since you're in OL, there's no AFC, so you have to calculate it yourself (AFC is just (Actual-Target)/Target expressed as a percentage).
williaty is offline   Reply With Quote
Old 03-19-2008, 02:56 PM   #115
Rene2.5RS
Scooby Specialist
 
Member#: 4574
Join Date: Mar 2001
Chapter/Region: NESIC
Location: Bristol, CT
Vehicle:
2014 Outback 2.5i L
VRP

Default

Quote:
Originally Posted by savageblitz View Post
Thanks Rene for your reply. When you mentioned Airboy's interpolation spreadsheet, are you refering the Excel spreadsheet in "Log file Interpolation ver3p7c Jan08.zip"?

Its quite complex and has many functions - so how do you use it to (I quote from your post):
1. generate the map AFRs
2. use the spreadsheet you made to reference what the MAF scaling was the error in Mapped values and logged values,
3. generate new values for the MAF scaling table

Sorry, I'm really lost. Do you mind elaborating abit further? Thanks.

(So now you can see why I used williaty's method coz he detailed his method step by step) :-)
Ok here we go:

(Make sure the Airboy spreadsheet is filled with the proper data from your current ECU map, and my spreadsheet would need your current MAF scaling table)

So I start out by making a log with the parameter I want, open Airboy's interpolation spreadsheet and open the log and you get this:


Next I open the AFR Interpolation tab, you can see every thing you need is highlighted red


Next I copy the data generated from interpolation over to my spreadsheet in this order:


I then go back to the Data tab of Airboy's sheet to grab the MAF v column


And then copy it over to my sheet and clean up the extra data lines:


And then I click on the MAF Table tab on my sheet and this is a screen shot of the only changes (in yellow)


I usually get about 700-1000 data points when doing this and then once I change it I go back out and get about another 700-1000 and check it again.

-Rene
Rene2.5RS is offline   Reply With Quote
Old 03-20-2008, 12:04 PM   #116
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

williaty/Rene,

thanks so much. you guys are awesome. I'll try both methods out during this long weekend.
savageblitz is offline   Reply With Quote
Old 03-20-2008, 12:25 PM   #117
Rene2.5RS
Scooby Specialist
 
Member#: 4574
Join Date: Mar 2001
Chapter/Region: NESIC
Location: Bristol, CT
Vehicle:
2014 Outback 2.5i L
VRP

Default

Quote:
Originally Posted by savageblitz View Post
williaty/Rene,

thanks so much. you guys are awesome. I'll try both methods out during this long weekend.
PM me your email address and I will send you my spreadsheet so you won't have to make one yourself

-Rene
Rene2.5RS is offline   Reply With Quote
Old 03-21-2008, 12:49 PM   #118
savageblitz
Scooby Newbie
 
Member#: 78394
Join Date: Dec 2004
Location: Earth
Vehicle:
03 Euro WRX MY03
Dirty White

Default

Thanks so much Rene, I've received your spreadsheet.

One question: Should the data logging be done under WOT (coz your spreadsheet is named "WOT"). If WOT is not needed, do I need to minimize throttle movement, etc when logging?
savageblitz is offline   Reply With Quote
Old 03-21-2008, 07:06 PM   #119
Rene2.5RS
Scooby Specialist
 
Member#: 4574
Join Date: Mar 2001
Chapter/Region: NESIC
Location: Bristol, CT
Vehicle:
2014 Outback 2.5i L
VRP

Default

The sheet is really only designed for WOT scaling. I didn't put any logic into it for tip-in. the only thing it disregards is any time the map afrs call for 14.7.

When I was doing my scaling the CL spreadsheet I used scaled everything up to the OL/CL change over and the spreadsheet I made covered the rest of the MAF v range. If you wanted to tune out side of WOT then yes you would have to make very little to no throttle movement. Best advice would be find a long step hill and hold throttle going up it while building boost.

PM or email me if you have any other questions.

-Rene
Rene2.5RS is offline   Reply With Quote
Old 04-09-2008, 05:41 PM   #120
pignoseWRX
Scooby Specialist
 
Member#: 93152
Join Date: Aug 2005
Chapter/Region: BAIC
Location: San Frandisco, CA
Vehicle:
2006 Impreza WRX
SGM

Default

have you guys compared this to the ROMRaider test build MAF scaling tool? should just be the same results, right?
pignoseWRX is offline   Reply With Quote
Old 04-09-2008, 05:59 PM   #121
Rogan
Scooby Specialist
 
Member#: 142133
Join Date: Feb 2007
Chapter/Region: MAIC
Location: Virginia - hate it
Vehicle:
3 Jeeps
And a Saab

Default

Quote:
Originally Posted by pignoseWRX View Post
have you guys compared this to the ROMRaider test build MAF scaling tool? should just be the same results, right?
I havent gotten the new rev MAF/INJ tabs to work yet, on either an 04 STI or 02 WRX..
Rogan is offline   Reply With Quote
Old 04-09-2008, 06:47 PM   #122
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

Quote:
Originally Posted by pignoseWRX View Post
have you guys compared this to the ROMRaider test build MAF scaling tool? should just be the same results, right?
Yes, extensively. The MAF tool is pretty good for getting the first couple of iterations done to get the MAF scaling close. It's biggest limitation is that it misses narrow peak/valleys almost completely.

Right now, my current workflow is to run the MAF scaling tool 3-4 times of an hour each to get the MAF scaling's general shape roughed in. Then I do 1-2 iterations of the method described here to get it nailed dead to rights and to account for any narrow errors.

Once I have the closed loops scaling dead-nuts on, I do the OL scaling using the same method as listed above.
williaty is offline   Reply With Quote
Old 04-09-2008, 09:00 PM   #123
Bullwinkle1983
Scooby Newbie
 
Member#: 172332
Join Date: Feb 2008
Location: St Paul, MN
Vehicle:
2006 WRX

Default

How about the newest test release?

http://www.romraider.com/forum/topic3029.html
Bullwinkle1983 is offline   Reply With Quote
Old 04-09-2008, 11:12 PM   #124
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

I'll test it tomorrow
williaty is offline   Reply With Quote
Old 04-19-2008, 11:12 PM   #125
williaty
Scooby Guru
 
Member#: 71092
Join Date: Sep 2004
Chapter/Region: MWSOC
Location: Delaware County, Ohio
Vehicle:
2005 2.5RS Wagon
Regal Blue Pearl

Default

There have been no additional definitions released since the initial release for the NAs. The one person working on Subaru defs is also the person trying to get real time tuning working. RT comes out before he resumes work on the Subaru defs.
williaty is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
need help with scaling the MAF 07wrx4life Open Source Reflashes 11 05-21-2008 03:21 PM
One man's trek across 5 dealerships in the Bay Area... fluxwl Dealership Reviews 1 07-14-2004 07:45 PM
One man's junk is another man's needed subaru part mynew02 Private 'For Sale' Classifieds 44 05-01-2004 09:40 PM
One man's junk.... chkltcow Off-Topic 15 04-12-2004 05:41 PM
One man's trash is another man's treasure... Jarvis Texas Impreza Club Forum -- TXIC 13 04-23-2002 12:07 AM


All times are GMT -4. The time now is 03:17 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2015, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2015 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.