|
ddillenger
|
 |
« Reply #300 on: November 28, 2014, 10:29:13 PM »
|
|
|
Thanks D. I'm really happy with the project. Mostly now I just want to make sure it's bullet-proof. Then it's on to bigger and faster things I guess. This train doesn't stop does it?
Never. |
|
|
|
|
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
|
|
|
|
nyet
|
|
« Reply #301 on: November 29, 2014, 12:07:11 AM »
|
|
|
FanTASTIC job Leroy! You should be proud. Job well done. Now you need to get a 2.7t so you can go save the completely stalled 2.7t community project  |
|
|
|
|
Logged
|
|
|
|
|
littco
|
|
« Reply #302 on: November 29, 2014, 01:35:56 AM »
|
|
|
Logs really good.
It compares well to the stage 1 I have on my BFB running around 150maf grams as well, altough I'm guessing its quite cold where you are at the moment? During the summer when I was logging mine I was only ending up with 12 degrees timing at the top end compared to your 20 and that was on 99ron fuel as well, IATS were up to 40c though on the run...
|
|
|
|
|
Logged
|
|
|
|
|
thelastleroy
|
|
« Reply #303 on: November 29, 2014, 06:28:03 AM »
|
|
|
I'm guessing its quite cold where you are at the moment? During the summer when I was logging mine I was only ending up with 12 degrees timing at the top end compared to your 20 and that was on 99ron fuel as well, IATS were up to 40c though on the run...
Yes It was -2*C ambient last night. I didn't consider the effect of air temp on timing advance. Is there a correct way to tune timing for temperature variance? I know the ECU will pull timing if there is knock, but it would be cool to prevent pull if possible. |
|
|
|
|
Logged
|
|
|
|
|
SB_GLI
|
|
« Reply #304 on: November 29, 2014, 10:20:05 AM »
|
|
|
FanTASTIC job Leroy! You should be proud. Job well done. Now you need to get a 2.7t so you can go save the completely stalled 2.7t community project I think I'll pop my head in there and do some work on my allroad. |
|
|
|
|
Logged
|
|
|
|
|
thelastleroy
|
|
« Reply #305 on: November 29, 2014, 06:50:48 PM »
|
|
|
Here's the *finished* file. I'm sure there will be 25 more revisions after this one, but it's pretty much all I can do with stock equipment. I'm going to re-name the original file in the OP "stock" and put the tuned file here, so anyone who wants it will hopefully read through the thread.
This is the 24th version of this file on my car, and that's only counting the versions that were different enough to warrant keeping the old one for comparison. It's been a labour of love, and I'm very greatful to all the great people on this forum who have helped me to learn.
I'm going to continue to "refine" the file and if I see small improvements, I'll post them.
|
|
|
|
|
Logged
|
|
|
|
|
littco
|
|
« Reply #306 on: November 30, 2014, 09:28:15 AM »
|
|
|
Yes It was -2*C ambient last night. I didn't consider the effect of air temp on timing advance. Is there a correct way to tune timing for temperature variance? I know the ECU will pull timing if there is knock, but it would be cool to prevent pull if possible.
Kftarx is about it which is load based. So it pulls load when temps rise. |
|
|
|
|
Logged
|
|
|
|
A4Rich
Full Member
Karma: +10/-0
 Offline
Posts: 116
|
|
« Reply #307 on: November 30, 2014, 05:17:36 PM »
|
|
|
Here's the *finished* file. I'm sure there will be 25 more revisions after this one, but it's pretty much all I can do with stock equipment. I'm going to re-name the original file in the OP "stock" and put the tuned file here, so anyone who wants it will hopefully read through the thread.
This is the 24th version of this file on my car, and that's only counting the versions that were different enough to warrant keeping the old one for comparison. It's been a labour of love, and I'm very greatful to all the great people on this forum who have helped me to learn.
I'm going to continue to "refine" the file and if I see small improvements, I'll post them.
Leroy et al., thank you. This thread is great; fantastic tool for learning! I have a couple of questions based on the latest bin you posted, anyone please feel free to jump in. 1. Is it necessary to adjust KFMDS, KFZWOP, and KFZWOP2 based on the change made to KFMIOP load axis as it is shared? 2. The bin doesn't include the P1681 Work around, is the consensus still that it should be included for all 2004+ cars? |
|
|
|
« Last Edit: November 30, 2014, 05:32:15 PM by A4Rich »
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #308 on: November 30, 2014, 05:21:42 PM »
|
|
|
1. Is it necessary to adjust KFMDS, KFZWOP, and KFZWOP2 based on the change made to KFMIOP load axis as it is shared?
ABSOLUTELY! The easiest way by far is to just move columns/rows around, so you don't have to re-interpolate stuff. This is generally the case whenever you change axis data that is shared.. trouble is, an incomplete mappack might not have all the maps that share a given axis, so it can be a bit of guessing game, unless you are good with IDA. 2. The bin doesn't include the P1681 Work around, is the consensus still that it should be included for all 2004+ cars?
Maybe? hopefully others can chime in on this, I have no personal experience on this topic  BTW I can't express how happy I am about this thread. Really. Truly. Thanks everybody! |
|
|
|
« Last Edit: November 30, 2014, 06:41:02 PM by nyet »
|
Logged
|
|
|
|
|
ddillenger
|
|
« Reply #309 on: November 30, 2014, 05:28:42 PM »
|
|
|
Anything >03 should include the P1681 workaround IMO. It doesn't ALWAYS happen, but when it does, it can ruin your day.
In sw 368072, it is 0x6B3AE
in sw 369307, it is 0x6B598
2D--->0D to kill the code.
I too am pleased at the direction of this thread. I think this is 100x better than the alternative. I'm going to simplify it and clean it up at some point once we have a concrete conclusion. I'd really like to see some logs of another car or two with the same file before we call it conclusive.
|
|
|
|
|
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
|
|
« Reply #310 on: November 30, 2014, 06:39:15 PM »
|
|
|
Leroy et al., thank you. This thread is great; fantastic tool for learning!
I have a couple of questions based on the latest bin you posted, anyone please feel free to jump in.
1. Is it necessary to adjust KFMDS, KFZWOP, and KFZWOP2 based on the change made to KFMIOP load axis as it is shared?
2. The bin doesn't include the P1681 Work around, is the consensus still that it should be included for all 2004+ cars?
I didn't adjust any of the other maps that share the KFMIOP load axis. Oops. Like I said, 25 more revisions ahead. Also, the P1681 work around... this will avoid no start condition after reading back the file with nef?? I guess it's a good idea to do the map adjustments for the KF group and re-log just to check for any other ill effects. I've already started "tweaking" a few things here and there on the file anyways.....What does KFMDS do? I don't have it in xdf. Where might I find it?
I too am pleased at the direction of this thread. I think this is 100x better than the alternative. I'm going to simplify it and clean it up at some point once we have a concrete conclusion. I'd really like to see some logs of another car or two with the same file before we call it conclusive.
I totally agree about cleaning this up. I was actually thinking of making a "walk-through" including all steps and links to theory on how to make the Stage 1 file. Then it could just be a closed thread "intro to tuning me7.5", leaving this thread for discussion and future optimizing of the file. This would be a large undertaking though, so it might take me a while. And yes, we need some logs on other cars, with different conditions than mine. I've got the advantage of cool ambient temps and decent gas at the pump. Would be good to troubleshoot and show solutions for other people's cars. Cheers everyone, lets make this great. |
|
|
|
« Last Edit: November 30, 2014, 08:13:10 PM by thelastleroy »
|
Logged
|
|
|
|
em.Euro.R18
Full Member
Karma: +4/-5
Offline
Posts: 219
|
|
« Reply #311 on: December 01, 2014, 10:16:04 AM »
|
|
|
I think you will end up needing something like this in order to get lamdba where you want it at .85 with those time delays active
1k .80
3k .76
4k .79
5k .83
6k .84
No need to lean it up top. You'll find little gain by requesting .84 lambda. Keeping it at .82 will keep things cool and safe with little to no sacrifice in the power curve. |
|
|
|
|
Logged
|
|
|
|
MadCow
Full Member
Karma: +2/-1
Offline
Posts: 108
|
|
« Reply #312 on: December 01, 2014, 01:29:17 PM »
|
|
|
I've ported your maps to an 018CH bin and looking to do some logs, what variables should I log? I've taken it for a quick drive and even not going above 3/4 throttle it feels great, but I wanna know for sure everything's safe before I really push it.
As an aside, I edited the KFZWOPs to reflect the changed axis, wanna know if I did the right thing. I moved each of the bottom 3 rows up by 1 and kept the bottom row the same. Is that fine or should I try to somehow extrapolate the bottom row values based on the previous 2-3 rows?
Also, what's the logic behind the LAMFA axis changes? Why compress the RPM axis? More mid-RPM resolution? Why stretch out the requested torque axis in the areas where it's all 1 (under 90)? Wouldn't it be better to have more resolution in the upper load range where the values aren't 1?
I apologize if these questions are stupid or have been answered in the thread, I'm just starting to learn this stuff.
|
|
|
|
« Last Edit: December 01, 2014, 01:41:09 PM by MadCow »
|
Logged
|
|
|
|
|
SB_GLI
|
|
« Reply #313 on: December 01, 2014, 01:42:59 PM »
|
|
|
No need to lean it up top. You'll find little gain by requesting .84 lambda. Keeping it at .82 will keep things cool and safe with little to no sacrifice in the power curve.
Yeah, this was a bad suggestion on my part. |
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #314 on: December 01, 2014, 01:48:05 PM »
|
|
|
I've ported your maps to an 018CH bin and looking to do some logs, what variables should I log?
dwkrz_0
dwkrz_1
dwkrz_2
dwkrz_3
dwkrz_4
dwkrz_5
etazwbm
fldrrx_w
fr_w
frm_w
frxt
frxta_w
gangi
lambts_w
lamfa_w
lamfaw_w
lamsbg_w
lamsoni_w
ldimn_w
ldimx_w
ldimxak_w
ldrlts_w
ldtvm
ldtvr_w
mdverl_w
miasrl_w
mifa_w
misol_w
miszul_w
msdk_w
mshfm_w
nmot_w
perffilt_w
plsol_w
ps_w
pu_w
pvdkds_w
rl
rl_w
rlmax_w
rlmx_w
rlroh_w
rlsol_w
tabgm_w
tanslin
te_w
ti_b1
tmotlin
uhfm_w
wdkba
wkrm
wped_w
zwgru
zwist
zwopt
zwout
zwsol
I moved each of the bottom 3 rows up by 1 and kept the bottom row the same. Is that fine or should I try to somehow extrapolate the bottom row values based on the previous 2-3 rows?
should be fine Also, what's the logic behind the LAMFA axis changes? Why compress the RPM axis? More mid-RPM resolution? Why stretch out the requested torque axis in the areas where it's all 1 (under 90)? Wouldn't it be better to have more resolution in the upper load range where the values aren't 1?
The idea is to get lamfa to be as smooth as possible in areas where it will be changing a lot. I haven't looked carefully at the lamfa for this project lately, so I can't comment if it is done correctly, but from leroy's logs, the results seem good, compared to other implementations of LAMFA that i've seen (including some i've done myself lol) I apologize if these questions are stupid or have been answered in the thread, I'm just starting to learn this stuff.
They are all EXCELLENT questions, no apologies needed! |
|
|
|
|
Logged
|
|
|
|
|
