Pages: 1 [2] 3 4 ... 6
Author Topic: Mbox KFMIOP Interpolator  (Read 21587 times)
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #15 on: January 10, 2012, 05:19:15 PM »

hey Berttos - Could you share your KFMIRL with us?  Or at least a methodology for arriving at a good KFMIRL.  I'm not sure the "tuning wizard" generates torque curves relevant to a 2.7tt, maybe I'm wrong?  Please correct me if so.

i have observed that it is best to keep the factory curve and simply italian tune the entire table rather than messing with individual columns.  that said - try multiplying all data points by 1.25 - 1.30.

here is what i've been using in my stg 3 file with excellent results ---



i'll post my .bin once i'm satisfied with my LAMFA fueling in all situations.  it's tough to find a safe place to do 3rd gear logs  Sad
Logged
nyet
Administrator
Hero Member
*****

Karma: +443/-62
Offline Offline

Posts: 10104


WWW
« Reply #16 on: January 10, 2012, 05:21:24 PM »

I never understood the flat top at the max of pedal position

Why not leave the ENTIRE IRL table stock, and just extend the last 2 columns into the max req load you want?
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
s5fourdoor
Hero Member
*****

Karma: +33/-1
Offline Offline

Posts: 617


« Reply #17 on: January 10, 2012, 05:24:57 PM »

italian tuning generally yields functional solid results.  however, i'm more a fan of aston martin tuning myself.  lol, jokes aside:  i took the rs4 KFMIRL and here's what i've generated for everyone.  this spreadsheet has an input for your "max power level" and it scales everything appropriately using your italian tuning technique.  i'm going to try it tonight but i feel confident enough that i'll attach the spreadsheet for others to use.  i can't wait to try this out.  i miscalculated this stuff yesterday and got a dreaded L2 intervention, but man was it fun up until that!  hard to believe its the same car honestly...
Logged
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #18 on: January 10, 2012, 05:35:08 PM »

I never understood the flat top at the max of pedal position

Why not leave the ENTIRE IRL table stock, and just extend the last 2 columns into the max req load you want?

that's where the wonky timing comes from. strangely and inexplicably - ME7 seems to take the entire table into account when doing load calculations, not just the columns that contain values that would apply to a certain % requested torque in a given situation.  this behavior means that when one greatly increases requested load from one column to another - weird things happen.  i've found that the Torque Monitoring Level 2 will intervene with a throttle cut until a certain load level - then it will intervene via N75 limitation (10% WG duty) - and finally, at the top end, if you make it that far without intervention - it will intervene via timing.

also - i love increased load earlier.  the car is so much more enjoyable this way.  i've tried altering KFPED and a couple of others in place of KFMIRL (light-mid load columns) and found that retaining the stock pedal curve with a completely raised KFMIRL results in a file that is exciting but not unrefined.  for me - it has to pass the wife test - she has to be able to drive it in stop and go traffic.
Logged
nyet
Administrator
Hero Member
*****

Karma: +443/-62
Offline Offline

Posts: 10104


WWW
« Reply #19 on: January 10, 2012, 06:27:36 PM »

that's where the wonky timing comes from. strangely and inexplicably - ME7 seems to take the entire table into account when doing load calculations, not just the columns that contain values that would apply to a certain % requested torque in a given situation.  this behavior means that when one greatly increases requested load from one column to another - weird things happen.

But if you have IOP set up properly, how does this still happen?
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #20 on: January 10, 2012, 06:48:21 PM »

But if you have IOP set up properly, how does this still happen?

i wish i knew...

perhaps ME7 uses a polynomial interpolation for load calcs rather than a linear?

i do know that i have yet to see a 'professionally tuned' K04 file that properly sets up KFMIOP.  they all seem to do some sort of weird underscaling of MAF values, either via a physically tapered MAF housing or via a tricked out MLHFM in order to trick ME7 into thinking there is a lower load - then they use some fueling tricks to meet actual air intake.  am i crazy? or has no one figured out the relationship between IRL and IOP in all these years?
Logged
amd is the best
Sr. Member
****

Karma: +9/-5
Offline Offline

Posts: 255



« Reply #21 on: January 10, 2012, 08:53:18 PM »

Sub'ed.
Logged

2018 Audi RS3
2001 Audi A4 2.8 30v Supercharged
1991 Audi 200 20v
s5fourdoor
Hero Member
*****

Karma: +33/-1
Offline Offline

Posts: 617


« Reply #22 on: January 10, 2012, 08:57:01 PM »

i wish i knew...

perhaps ME7 uses a polynomial interpolation for load calcs rather than a linear?

i do know that i have yet to see a 'professionally tuned' K04 file that properly sets up KFMIOP.  they all seem to do some sort of weird underscaling of MAF values, either via a physically tapered MAF housing or via a tricked out MLHFM in order to trick ME7 into thinking there is a lower load - then they use some fueling tricks to meet actual air intake.  am i crazy? or has no one figured out the relationship between IRL and IOP in all these years?

this.  i literally have an email thread with a fellow nefmoto poster on almost this exact topic.  hopefully in the next few days i'll present results on MLHFM flow normalization based on an idea we've discussed.
Logged
nyet
Administrator
Hero Member
*****

Karma: +443/-62
Offline Offline

Posts: 10104


WWW
« Reply #23 on: January 10, 2012, 08:59:53 PM »

or has no one figured out the relationship between IRL and IOP in all these years?

Yup. So much for the god like properties of "pro" tuners.
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
amd is the best
Sr. Member
****

Karma: +9/-5
Offline Offline

Posts: 255



« Reply #24 on: January 11, 2012, 01:12:29 AM »

i have observed that it is best to keep the factory curve and simply italian tune the entire table rather than messing with individual columns.  that said - try multiplying all data points by 1.25 - 1.30.

here is what i've been using in my stg 3 file with excellent results ---

https://lh4.googleusercontent.com/-Zme1TzzH8t8/TwzUYkzlb6I/AAAAAAAADOc/FlRFW12sAJE/s800/irl_op.GIF

i'll post my .bin once i'm satisfied with my LAMFA fueling in all situations.  it's tough to find a safe place to do 3rd gear logs  Sad


Something didn't seem to be adding up when I ran your spreadsheet so for the hell of it I took the time and entered in all of your KFMIRL values and your KFMIOP header information and everything matched up perfect except for the last column (attached below).

Am I doing something wrong?

No idea why half of it is purple, lol.

« Last Edit: January 11, 2012, 01:20:39 AM by amd is the best » Logged

2018 Audi RS3
2001 Audi A4 2.8 30v Supercharged
1991 Audi 200 20v
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #25 on: January 11, 2012, 04:26:40 AM »

Something didn't seem to be adding up when I ran your spreadsheet so for the hell of it I took the time and entered in all of your KFMIRL values and your KFMIOP header information and everything matched up perfect except for the last column (attached below).

Am I doing something wrong?

No idea why half of it is purple, lol.



no - you're right.  i've been playing with that last column and trying to understand it's effects on the load calcs throughout the table.  those values are just what i happen to have in there at the moment.  the calculated 'proper' values work fine.  i've yet to determine if they are the absolute best to use.
Logged
TTQS
Guest
« Reply #26 on: January 11, 2012, 12:03:10 PM »

Cool stuff, has 'sparked' a renewed interest in this issue.

But what is 'Italian tuning'?  Grin

I'm off-message on this streetwise lexicon.

TTQS
Logged
ejg3855
Full Member
***

Karma: +6/-0
Offline Offline

Posts: 107


« Reply #27 on: January 11, 2012, 12:24:27 PM »

I think Italian tuning refers to tuning for WOT conditions
Logged
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #28 on: January 11, 2012, 12:42:46 PM »

Cool stuff, has 'sparked' a renewed interest in this issue.

But what is 'Italian tuning'?  Grin

I'm off-message on this streetwise lexicon.

TTQS

it means simply turning up the wick for the entire table - all data points * 1.30 for example, without regard for massaging certain load/rpm areas to alter the overall requested load curve.  though, it also refers to FFing an entire table Smiley

imo the factory got the load curve right for the B5 and it would be hard to improve upon without more resolution in the table (more columns).

 
Logged
TTQS
Guest
« Reply #29 on: January 11, 2012, 02:19:20 PM »

Ah right, Italian tuning methodology; just applying the same multiplication factor throughout all addresses without affecting the OEM 'shape'

I suppose you would have to bank on the factory getting it right because they can hold one parameter fixed, e.g. nmot and go right through the rl_w or mrfa_w range on the engine dyno so can tweak response at will.

Thanks for an interesting thread guys. I need to go back and do KFMIOP & KFMIRL 1.01.   Roll Eyes

TTQS
« Last Edit: January 12, 2012, 03:23:09 AM by TTQS » Logged
Pages: 1 [2] 3 4 ... 6
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.027 seconds with 18 queries. (Pretty URLs adds 0s, 0q)