[hassmaschine]

Ho doesnt - I'm hoping i can reconstruct a binary from the F-series data, but its way different from the E-series BMWs.

I can already generate an MSD87 binary, of course I'm not sure if it can be flashed using something like an MSD85.prg. There's likely no such thing as an MSD87.prg, although if there was theoretically it could be flashed with an ICOM. Same with the GKE2xx.

[GTZ]

Two days ago I plugged a jB+ unit to the manifold sensor.I have started with the minimum setting.The results were good.Yesterday I tried one step upper to the settings of the JB+.
Max boost was 0.95bar, it seems the problem is 90% solved.I will remain to this setting and waiting for ecu to be adapted and report again!

[GTZ]

Update...
1 hour ago I put JB+ one more step to the right.Problem was totally solved.Max boost is always 0,95 to 1 bar with or without full throttle.I remind you that without JB+ connected to manifold sensor,there was boost reduction at WOT except in some cases of 75% of the right pedal press.
Of course acceleration looks insane now.
I have also noticed that transmission behavior on "D" position and "Normal" drivetrain selection looks too sporty.I mean that reaction in right pedal is much faster and kick down less sensitive.You have to be too gently with your right foot in order to have smooth rolling from 0 km/h.Maybe transmission slippage is less now.
AFR has not any affection as it has to do with boost level.Ignition advance behavior had also no difference comparing with recent log runs.
Of course I have to wait for some adaptations.
The point is that I have no limp mode,no codes...
To be continued...

[rjahl]

That's great news.

With the Automatics, it's very important to have the DME report good torque values to the TCU. Too low and the transmission will shift early and produce sloppy , slow shifts. Too high and you will get the opposite plus the potential to reach the Torque limit "nannies". That's not even discussing the TCC lock ups values, those maps are crazy.

The early N54 tuners ran into these issues on the N54s with Automatics and developed ways to report the "desired" toque values to the TCU. These are "work arounds" and are inferior to a real TCU tune.

Their is a high degree of probability that if you and Bimmerlabs find a way to read and Flash the TCU, I can map it and build a tune. Just for the fun of it.

[GTZ]

Till now only alone DME MSD87.2 is flashable and of course away from the engine hood...
I have no idea what kind of tools are essential to retrieve TCU program from an F01 N54.
Can tools like WINKFP,tools32 etc do anything in this case?

For the time being this liitle solution named JB+ works properly.
I will drive for some more km and make my conclusions safely...

[Dsycks]

Sorry that I am late to the party.

Any help needed here or is this issue solved with the results you want in hand?

[Dsycks]

There are commercially available software modifications for this application.

How much like or not like this are they?

Is one wise to simply purchase that software or for more serious use should one spend the time doing as some have done here?

[blackout_ssri]

Hi guys,

Long time lurker, first time poster here.

First, I want to say that your work is amazing and thank you for what you've built already.

Based on information that I gathered on this forum and others, I'm in the process of building an XDF for the GKE211.

The file is not complete at the moment and lacks some of the rows and column information as it was not of the most importance (I've only been on it for a few days).

I'd like to share this file with you in the hope that it'll help others and maybe help me understand quicker what are some of the maps.

The files is offset to work with a binary of the 'oda' file, no full dump is required (or supported). It is based on the 'public' xdf for the Alpina tunes (can't remember what DME it's using) and only contains the already figured parameters + the scalars that are not mapped yet.

What gave me some thought was that I found 5 sets of clutch pressure maps, whereas in the original xdf there were only 4 (including the Alpina M map).

Disclaimer: the categories for the parameters are not updated, use this file only for educational purposes.

rjahl Thank you for posting the TCC maps, they were the starting point for me.

In the zip you can find the XDF and a stock binary to use with it.

[rjahl]

Originally Posted by blackout_ssri

Hi guys,

Long time lurker, first time poster here.

First, I want to say that your work is amazing and thank you for what you've built already.

Based on information that I gathered on this forum and others, I'm in the process of building an XDF for the GKE211.

The file is not complete at the moment and lacks some of the rows and column information as it was not of the most importance (I've only been on it for a few days).

I'd like to share this file with you in the hope that it'll help others and maybe help me understand quicker what are some of the maps.

The files is offset to work with a binary of the 'oda' file, no full dump is required (or supported). It is based on the 'public' xdf for the Alpina tunes (can't remember what DME it's using) and only contains the already figured parameters + the scalars that are not mapped yet.

What gave me some thought was that I found 5 sets of clutch pressure maps, whereas in the original xdf there were only 4 (including the Alpina M map).

Disclaimer: the categories for the parameters are not updated, use this file only for educational purposes.

rjahl Thank you for posting the TCC maps, they were the starting point for me.

In the zip you can find the XDF and a stock binary to use with it.

Here is a link that will provide you with more clues, It's my OLS file for the GKE211 program in the E90. It's an early OLS file so even the Demo versions should be able to open the file. Almost everything of value is found and mapped. It's really is the "keys to the city"

I'll can only provided limited help these days, just too busy to contribute much to the community.




https://mega.nz/#F!VtgDCAbD!ZsHEind-O7Q_jOTM7xCnsQ



Just a reminder, Bimmerlabs has the ability to process a custom GKE 211 tunes into a fully signed and check sum corrected 0Da file. I've done about 50 different tunes so I'm pretty sure it's safe.

My only request is you keep your work public so others can benefit.

[blackout_ssri]

Thank you very much rjahl , I'll continue working on this and post my progress.
I'll probably create a GitHub repo for easier version control and public access.

I know about the bimmerlabs (hats off to you guys), I've actually been using it for a while now, wrote about 20-25 0da's, mind me there were shots in the dark until last night when I managed to get some quicker shifts and early TCC lockup. It's safe for sure!

We'll keep you guys posted, it's my primary side project and I do have free time to work on it .

[rjahl]

Originally Posted by blackout_ssri

Thank you very much rjahl , I'll continue working on this and post my progress.
I'll probably create a GitHub repo for easier version control and public access.

I know about the bimmerlabs (hats off to you guys), I've actually been using it for a while now, wrote about 20-25 0da's, mind me there were shots in the dark until last night when I managed to get some quicker shifts and early TCC lockup. It's safe for sure!

We'll keep you guys posted, it's my primary side project and I do have free time to work on it .

One thing to keep in mind with shift speeds. It will take adjustment to both the shift time targets and the shift pressures to achieve the results you are look for.

In the stock maps there are areas the are more controlled by shift pressure and in other area more control is derived by shift times. If your balance is off, the TCU will adapt away from your changes. The adaptation is slow so sometime it took a week or 1,000 miles before I lost what I thought was a good setting.

I good place to start is by adding 20% to the shift pressures and pulling 15% from the shift times. Drive it for a week and then make more adjustments. I never found any values with the off going clutch pressure.

I had added as much as 40% to the shift pressures at high RPM and load but don't try that at lower loads or speeds, gets bumpy.

I believe your Kombi has a programmed delay or dampening built into the tach. This will make you feel like the shifts are slower then they really are. The dampening can be coded but I never missed with it.

Share your experiences, I was up to tune #70 before I moved to a different car

[blackout_ssri]

Originally Posted by rjahl

One thing to keep in mind with shift speeds. It will take adjustment to both the shift time targets and the shift pressures to achieve the results you are look for.

In the stock maps there are areas the are more controlled by shift pressure and in other area more control is derived by shift times. If your balance is off, the TCU will adapt away from your changes. The adaptation is slow so sometime it took a week or 1,000 miles before I lost what I thought was a good setting.

I good place to start is by adding 20% to the shift pressures and pulling 15% from the shift times. Drive it for a week and then make more adjustments. I never found any values with the off going clutch pressure.

That's a good point that I completely missed, that the TCU will adapt away. I haven't got much time to transpose a lot of maps, also the OLS file confused me a bit as some of the maps don't seem to be in the same place.

Managed to add the gear indicator in all driving modes to show in the kombi + the 'show selected gear vs current gear' and some torque reduction (downshift) maps.

I went 'all in' with some of the shift times and pressures about 50% difference from normal just to see which map's which, finally managed to get the gearbox to enter limp mode, seems I have a 3->2 discrepancy where either the off going clutch has too much pressure or the oncoming one is too fast and they're both engaged.

Cool thing that in ISTA+ it actually says what's wrong, I didn't expect it to be that specific, in any case you could clearly feel that downshift was badly off.

Tonight I'll have more time to spend on the XDF and next on the list is to figure which maps are for downshifts and which are for upshifts. I'll create the repo as well, but thinking about it (how the xHP one just disappeared) I've decided to periodically post it in here as well, just in case, here it will live forever I guess.

Found a sweet spot for the 1->2 shift which was absolutely horrible before and 3->4 is so quick it feels like a completely different car!

With the off going clutch pressures, am I mistaken that they should be reduced rather than increased? When I tried to increase them it felt like it was trying to engage both gears at the same time, although I'm not sure if it was actually the right map.

On another note, it feels like limp mode has it's own map or something. The first time I got into it I was blocked in D and the second time I was blocked in D5 with the torque converter doing the whole job (this thing can start from a stop on 5th gear :|, maybe it was just an error)

I need some data logging tools, trying out 'Testo' but can't seem to find the 'Status Eingaenge AEGS' job too useful, it's good for TCC lockup but not much more, I should probably look through all of them.

Anyway, I'll come back tomorrow with some updates.

[rjahl]

Originally Posted by blackout_ssri

That's a good point that I completely missed, that the TCU will adapt away. I haven't got much time to transpose a lot of maps, also the OLS file confused me a bit as some of the maps don't seem to be in the same place.

Managed to add the gear indicator in all driving modes to show in the kombi + the 'show selected gear vs current gear' and some torque reduction (downshift) maps.

I went 'all in' with some of the shift times and pressures about 50% difference from normal just to see which map's which, finally managed to get the gearbox to enter limp mode, seems I have a 3->2 discrepancy where either the off going clutch has too much pressure or the oncoming one is too fast and they're both engaged.

Cool thing that in ISTA+ it actually says what's wrong, I didn't expect it to be that specific, in any case you could clearly feel that downshift was badly off.

Tonight I'll have more time to spend on the XDF and next on the list is to figure which maps are for downshifts and which are for upshifts. I'll create the repo as well, but thinking about it (how the xHP one just disappeared) I've decided to periodically post it in here as well, just in case, here it will live forever I guess.

Found a sweet spot for the 1->2 shift which was absolutely horrible before and 3->4 is so quick it feels like a completely different car!

With the off going clutch pressures, am I mistaken that they should be reduced rather than increased? When I tried to increase them it felt like it was trying to engage both gears at the same time, although I'm not sure if it was actually the right map.

On another note, it feels like limp mode has it's own map or something. The first time I got into it I was blocked in D and the second time I was blocked in D5 with the torque converter doing the whole job (this thing can start from a stop on 5th gear :|, maybe it was just an error)

I need some data logging tools, trying out 'Testo' but can't seem to find the 'Status Eingaenge AEGS' job too useful, it's good for TCC lockup but not much more, I should probably look through all of them.

Anyway, I'll come back tomorrow with some updates.

Status Eingaenge AEGS is the same job I used. It's about the best you'll get.

This gearbox does not report pressures so the best you can do is monitor input and output RPMs. Add a column that tracks the slip between them based on the actual gear. You can watch your clutches through the slip. The gear reported only changes when the shift is complete.

Sorry about any errors you find, the file is pretty old and I did most of my work on the E85 version. Nearly 100 percent the same except for an annoying 28 byte offset in the middle of the file.

Don't worry about the Xhp files, I have them all plus the original ols files they were based on. Even a few more. Use my files, it will be less work. Let me know if you need anything specific this will allow you to keep tuning and testing.

[rjahl]

Originally Posted by rjahl

Status Eingaenge AEGS is the same job I used. It's about the best you'll get.

This gearbox does not report pressures so the best you can do is monitor input and output RPMs. Add a column that tracks the slip between them based on the actual gear. You can watch your clutches through the slip. The gear reported only changes when the shift is complete.

Sorry about any errors you find, the file is pretty old and I did most of my work on the E85 version. Nearly 100 percent the same except for an annoying 28 byte offset in the middle of the file.

Don't worry about the Xhp files, I have them all plus the original ols files they were based on. Even a few more. Use my files, it will be less work. Let me know if you need anything specific this will allow you to keep tuning and testing.

One more note, my files show upshift "HS" and downshifts "RS".

[rjahl]

Originally Posted by rjahl

One more note, my files show upshift "HS" and downshifts "RS".

I promise one last note, try playing with the curve detection settings in sport mode. You can turn off or shorten the upshift delay while cornering in Sport mode. Default setting is just too long and is annoying. I never liked or used sport mode for regular driving until I fixed it.

[blackout_ssri]

Hello, this past week has been hectic at work, barely got to do something for myself in the evenings.
However I did some mapping in the weekends and got some interesting results.

After I transposed all the Shift pressure and shift time maps from the OLS file to XDF I increased all the pressures by 40% all around the map, and reduced the shift time by about 30% all around.

Well, to simply put it, this thing was brutal, especially when downshifting from 3->2 at +50% throttle, amazing!

Still, there was something a bit off, especially that I went all in and increased so much from the first try, it seemed like it was lagging before engaging the next gear, there was a moment where it lost power.

I started to fiddle with the maps that I had from the other XDF (the 5 sets of "HS x->y"), first tried to increase the values (20% all over the 5 sets), it didn't like that, afterwards I reduced the values (from stock) by 30% all over the 5 sets and I could really feel a difference in how it changed gears, that lag, loss of power was not as proeminent as before, it also seemed that the gear changes were not so harsh anymore but still quick and quite brutal, especially on the downshifts.

I can't put my finger on it, and considering this is my first time diving into mapping stuff (TCU / ECU) not entirely sure how to find out what's what, but certainly those older maps did something, I don't know if they are pressures or something else even though they fit perfectly as pressures.

I've been running this tune for around 5 days now, commuting and driving around and it's quite good in all 3 modes (D/S/M), of course not refined at all, but I really like the change, suddenly that small 2.0 got some life in it.

Will try to better understand what's with those 5 sets of maps (if anyone has any idea or theory about them, please let me/us know), these maps are not present at all in the OLS file posted by rjahl which is even stranger to me, especially that I modified them before getting the OLS and I could feel some difference, though not even close to what I got from the actual shift pressure maps (from OLS).

Also received some more parts for my "carPC" project so am in between nodeJS, video connections, CAN-BUS stuff and TunerPro, WinOLS + work.

Here's the latest XDF that I have, as before the categories are totally wrong and not all columns and rows are listed correctly (seemed less important than the actual values). I will update the categories accordingly after I understand more about the maps. Haven't transposed any of the shift point maps as they were of less interest at the moment.

[rjahl]

Originally Posted by blackout_ssri

Hello, this past week has been hectic at work, barely got to do something for myself in the evenings.
However I did some mapping in the weekends and got some interesting results.

After I transposed all the Shift pressure and shift time maps from the OLS file to XDF I increased all the pressures by 40% all around the map, and reduced the shift time by about 30% all around.

Well, to simply put it, this thing was brutal, especially when downshifting from 3->2 at +50% throttle, amazing!

Still, there was something a bit off, especially that I went all in and increased so much from the first try, it seemed like it was lagging before engaging the next gear, there was a moment where it lost power.

I started to fiddle with the maps that I had from the other XDF (the 5 sets of "HS x->y"), first tried to increase the values (20% all over the 5 sets), it didn't like that, afterwards I reduced the values (from stock) by 30% all over the 5 sets and I could really feel a difference in how it changed gears, that lag, loss of power was not as proeminent as before, it also seemed that the gear changes were not so harsh anymore but still quick and quite brutal, especially on the downshifts.

I can't put my finger on it, and considering this is my first time diving into mapping stuff (TCU / ECU) not entirely sure how to find out what's what, but certainly those older maps did something, I don't know if they are pressures or something else even though they fit perfectly as pressures.

I've been running this tune for around 5 days now, commuting and driving around and it's quite good in all 3 modes (D/S/M), of course not refined at all, but I really like the change, suddenly that small 2.0 got some life in it.

Will try to better understand what's with those 5 sets of maps (if anyone has any idea or theory about them, please let me/us know), these maps are not present at all in the OLS file posted by rjahl which is even stranger to me, especially that I modified them before getting the OLS and I could feel some difference, though not even close to what I got from the actual shift pressure maps (from OLS).

Also received some more parts for my "carPC" project so am in between nodeJS, video connections, CAN-BUS stuff and TunerPro, WinOLS + work.

Here's the latest XDF that I have, as before the categories are totally wrong and not all columns and rows are listed correctly (seemed less important than the actual values). I will update the categories accordingly after I understand more about the maps. Haven't transposed any of the shift point maps as they were of less interest at the moment.

Keep up the hard work. The maps you mention are probably the the Pressure maps for the off going clutch. There are 4 of these 8x10 maps for each down shift.

It seems like there are two groups of maps for the offgoing clutch. KAB and KABFOL. One group for D and another for sport or shortened. KABSP and KABFOL21. Learn to use the Map ID rather than any name provided by the Xhp or a translated OLS file.

Note: This screen capture is from my Z4 file and it takes a 28 byte offset somewhere in the middle. I developed the Z4 file much further than the 3 series file that I posted earlier.

I was also able to learn from Patents filed by ZF. Those gave important clues.

[rjahl]

In case you can't sleep tonight and need something to read. LOL

[rjahl]

Check the original Mega download link, I placed another file for your studies. Offsets will be correct or with a 28 byte offset.

[blackout_ssri]

Nice literature ).
The maps that I have found based on the xHP XDF match the ones you are showing in the screenshot, the offset seems correct and the values are the same.

I downloaded the new OLS and will have a look today/tonight this will certainly help me properly categorise the off-going clutches maps and figure their order out, thanks.

[blackout_ssri]

Hi guys,

Sorry for not posting any updates in a while but I haven't got any at the moment, I've been caught up with work and other personal stuff.

Not going to lie again and say tomorrow but by the end of next week (yay, holidays) I should have most of the important stuff migrated to XDF format, categorised and all that.

I'm looking to swap my N46 for an N52/53 3.0, I think the gearbox should handle the torque right? In any case I'm going to buy a donor car for this project, but not going to start it in the near future anyway.

[UCDHIUS]

Originally Posted by blackout_ssri

Hi guys,

Sorry for not posting any updates in a while but I haven't got any at the moment, I've been caught up with work and other personal stuff.

Not going to lie again and say tomorrow but by the end of next week (yay, holidays) I should have most of the important stuff migrated to XDF format, categorised and all that.

Awesome to see, the community still working on this ! Don’t forget to post up whenever possible!