Pages: 1 ... 5 6 [7] 8 9 ... 52
Author Topic: Nefmoto community project: Stage1 1.8t ME7.5 A4 (8E0909518AK-0003)  (Read 502388 times)
SB_GLI
Hero Member
*****

Karma: +115/-10
Offline Offline

Posts: 1022


« Reply #90 on: October 12, 2014, 07:48:17 AM »

I found the correct locations for the KFZKLAMFAW maps.  The locations previously mentioned in this thread were incorrect.
26B04 and 26B64, which are the two maps right directly before LAMFA.

It seems that ZKLAMFAW is at 26C3C, I am unsure if this is a map or a single value, 8 or 16bit.

http://nefariousmotorsports.com/forum/index.php?topic=5407.11

I think we just FF the KFZKLAMFAW tables and go from there.  Once we know this works, we can slightly decrease the values to smooth out the request, but not enough to delay it this much.
« Last Edit: October 12, 2014, 07:54:05 AM by SB_GLI » Logged
SB_GLI
Hero Member
*****

Karma: +115/-10
Offline Offline

Posts: 1022


« Reply #91 on: October 12, 2014, 06:18:42 PM »

Here's an updated xdf with correct locations for kftlafa and kfzklamfaw maps.
Logged
thelastleroy
Sr. Member
****

Karma: +19/-2
Offline Offline

Posts: 260



« Reply #92 on: October 13, 2014, 12:22:19 PM »

What software version is this updated xdf for?
Logged
SB_GLI
Hero Member
*****

Karma: +115/-10
Offline Offline

Posts: 1022


« Reply #93 on: October 13, 2014, 12:33:12 PM »

What software version is this updated xdf for?

The only one that should be in this thread.  Smiley  368072
Logged
nyet
Administrator
Hero Member
*****

Karma: +604/-166
Offline Offline

Posts: 12232


WWW
« Reply #94 on: October 13, 2014, 12:37:52 PM »

The only one that should be in this thread.  Smiley  368072

that is 0003... should we be using 369307-0004?

Also, can you post the 0003 ori, I don't seem to have it, only the 0004 ori

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.
ddillenger
Hero Member
*****

Karma: +637/-21
Offline Offline

Posts: 5640


« Reply #95 on: October 13, 2014, 12:48:03 PM »

My addresses were wrong because everyone that asks for these has always done so for the version I don't use (004). I assumed that you guys were also using that shit software.

Sorry about that.

Nye:

Please find the software version referenced in this thread attached below.
Logged

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!

Email/Google chat:
DDillenger84(at)gmail(dot)com

Email>PM
SB_GLI
Hero Member
*****

Karma: +115/-10
Offline Offline

Posts: 1022


« Reply #96 on: October 13, 2014, 01:15:42 PM »

My addresses were wrong because everyone that asks for these has always done so for the version I don't use (004). I assumed that you guys were also using that shit software.

DD, can you explain your experience as to why 0004 is shit?  Smiley  Just so we know why we are going back to 0003.  I think the reason why this thread began using 0003 was because you advised people to use 0003 over 0004.

I had issues in the past with 0004, but I can't tell you for sure if switching to 0003 made a difference or not as I don't have new logs yet.  My issue was that after the kickdown happened, and it dropped back down a gear, the load/boost request was all over the place and oscillated up and down until the driver let off the accelerator.  It is for this reason, I have particular interest in this thread.
Logged
ddillenger
Hero Member
*****

Karma: +637/-21
Offline Offline

Posts: 5640


« Reply #97 on: October 13, 2014, 01:25:24 PM »

Fueling problems. I ported over to 004 after having a c'sum error with custom code (before I had ecufix). The file was perfect on 003, but on 004 actual AFR didn't follow requested at all. I double, triple, quadruple checked EVERYTHING. I ended up going back to 003 and all issues were gone.
Logged

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!

Email/Google chat:
DDillenger84(at)gmail(dot)com

Email>PM
thelastleroy
Sr. Member
****

Karma: +19/-2
Offline Offline

Posts: 260



« Reply #98 on: October 13, 2014, 03:17:29 PM »

Ok, back from an awesome Canadian Thanksgiving road trip to Parry Sound. Beautiful drive once the traffic opened up. I ended up leaving the tuned file on the car, and it was fine for ordinary tame driving, and reasonable fuel economy on the highway.

Anyways, now that we have kftlafa and kfzklamfaw in the latest xdf (thanks SB_GLI), what can be done to improve the LAMFA situation based on the last log? Someone want to explain how these work?

Logged
SB_GLI
Hero Member
*****

Karma: +115/-10
Offline Offline

Posts: 1022


« Reply #99 on: October 14, 2014, 04:53:12 AM »

Anyways, now that we have kftlafa and kfzklamfaw in the latest xdf (thanks SB_GLI), what can be done to improve the LAMFA situation based on the last log? Someone want to explain how these work?

Here's what I would use for my first pass.  Attached are screenshots of the stock values, and modified values.  I set kftlafa (time delay) to .2 seconds across the board and raised the kfzklamfa (smoothing) maps to higher values, not so much that there is no smoothing, but hopefully not too little so that the smoothing doesn't delay the desired lamdba that we want to achieve.
Logged
aef
Hero Member
*****

Karma: +69/-46
Offline Offline

Posts: 1566


« Reply #100 on: October 14, 2014, 05:38:16 AM »

Not sure if i understood the last posts:

It is necessary to do the changes to force the ecu to follow only lamfa map?

Are kftlafa and kfzklamfaw available in very me7.5? I searched the board and found only two posts for BFB/BEX A4 1.8t

Logged
phila_dot
Hero Member
*****

Karma: +170/-11
Offline Offline

Posts: 1709


« Reply #101 on: October 14, 2014, 08:22:49 AM »

In ME7, they are TLAFA and ZKLAMFAW.

Why use the time delay?

I have posted on here a conversion factor to display ZKLAMFAW as a gain percentage.

Logged
SB_GLI
Hero Member
*****

Karma: +115/-10
Offline Offline

Posts: 1022


« Reply #102 on: October 14, 2014, 08:45:09 AM »


Why use the time delay?


I assume you are referring to why I chose .2 seconds vs 0 seconds?  I guess these aren't final values that would be used, but more just to see what a constant value across the board would do.  I'd like to see a .2 second delay when you step on the throttle before lamfa kicks in so we get an understanding of how this table works.


I have posted on here a conversion factor to display ZKLAMFAW as a gain percentage.


I've seen that before.  I will look for that conversion again and apply it to the definition file.
Logged
nyet
Administrator
Hero Member
*****

Karma: +604/-166
Offline Offline

Posts: 12232


WWW
« Reply #103 on: October 14, 2014, 09:27:40 AM »

I guess these aren't final values that would be used, but more just to see what a constant value across the board would do.

Fine for experimenting, but the smoothing that ZK provides is a happy side effect, the more the better, but adding non-zero TLA means less smoothing for a given (total) delay.

Quote
I've seen that before.  I will look for that conversion again and apply it to the definition file.

I'm using 0.001526 in mbox (for 0-100%)

If you use 0.000015 you run out of precision. Stupid winols Smiley
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.
thelastleroy
Sr. Member
****

Karma: +19/-2
Offline Offline

Posts: 260



« Reply #104 on: October 14, 2014, 07:45:49 PM »

Fine for experimenting, but the smoothing that ZK provides is a happy side effect, the more the better, but adding non-zero TLA means less smoothing for a given (total) delay.

I'm using 0.001526 in mbox (for 0-100%)


I've got the conversion for ZKLAMFAW set up according to this factor (x*0.001526), which is now displaying the ZKLAMFAW values as 2-digit numbers. What does it mean "gain percentage" like phila_dot said? Will Higher number produce more delay before LAMFA enrichment? or other way around? In your opinion, where should we leave the delay/smoothing effect to allow for daily-drivability? This is a stage 1 project after all.

If we can get this sorted, I'll flash and do another log to see where we're at.
« Last Edit: October 14, 2014, 07:47:48 PM by thelastleroy » Logged
Pages: 1 ... 5 6 [7] 8 9 ... 52
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.025 seconds with 19 queries. (Pretty URLs adds 0s, 0q)