[Mik325tds]

Originally Posted by TDIwyse

How was your heart rate and blood pressure during this? Yikes...

Elevated
I think I'm slowly becoming a flash junkie. Need to find a group soon.

[JP-76]

Originally Posted by Mik325tds

Makes me think of a new mod...

If you're thinking about retrofitting electronic gearshiter, you'll need new mechatronics also. Gearshifting mechanism is totally different in newer gearbox. Just thinking out loud...

[Mik325tds]

Originally Posted by JP-76

If you're thinking about retrofitting electronic gearshiter, you'll need new mechatronics also. Gearshifting mechanism is totally different in newer gearbox. Just thinking out loud...

How do you know it's totally different? The flash files say the HW number is identical.
E90:
;ZB-NR TYP-NR HW-NR IX SW-NR AM PIN S CS
7587867,0000000,7573100,A,7587868DA,0FFFFFFFFFD,00 0,1 C
7610590,0000000,7591972,A,7610591DA,0FFFFFFFFFD,000,1 H
7606273,0000000,7591972,A,7606274DA,0FFFFFFFFFD,00 0,1 J
...

E70:
7507225,0000000,7591972,A,7507226DA,0FFFFFFFFFD,000,1 L
7607058,0000000,7605219,A,7607059DA,0FFFFFFFFFD,00 0,1 7
...

[JP-76]

Originally Posted by Mik325tds

How do you know it's totally different? The flash files say the HW number is identical.
E90:
;ZB-NR TYP-NR HW-NR IX SW-NR AM PIN S CS
7587867,0000000,7573100,A,7587868DA,0FFFFFFFFFD,00 0,1 C
7610590,0000000,7591972,A,7610591DA,0FFFFFFFFFD,000,1 H
7606273,0000000,7591972,A,7606274DA,0FFFFFFFFFD,00 0,1 J
...

E70:
7507225,0000000,7591972,A,7507226DA,0FFFFFFFFFD,000,1 L
7607058,0000000,7605219,A,7607059DA,0FFFFFFFFFD,00 0,1 7
...

[Mik325tds]

Ah. Thanks for the insight!
I remember seeing a DTC about a valve. Is that the park lock actuator or what are we looking at there? I still find it strange how they could have the same HW number for two different mechatronics. Maybe the electronic part is identical?

[JP-76]

Originally Posted by Mik325tds

Ah. Thanks for the insight!
I remember seeing a DTC about a valve. Is that the park lock actuator or what are we looking at there? I still find it strange how they could have the same HW number for two different mechatronics. Maybe the electronic part is identical?

Yes, that is the park sylinder-rod on the picture. There might be possibility ...in theory anyway...to change the mechatronic-unit to newer one and then replace the gearlever to electronic one but I dont know what else it would require...

I believe the electronic control-unit is the same in both.

[Xtremektm]

E-version is for E65 that have electronic parking. All E90 have Regular version

[rspring]

Is there a performance TCU tune available for our cars? Noob with a noob question, but I did not find it using search.

[DWR]

Originally Posted by rspring

Is there a performance TCU tune available for our cars? Noob with a noob question, but I did not find it using search.

That's a good question. Some say yes. Some no. Depends upon your definition of performance. Look here:
http://www.jrtuning.co.uk/articles/b...p-tuning.shtml

[DWR]

Mik325tds and I have been working on understanding the calibration file from our transmission. A while back I reported that we were able to see a hundred or so maps using WinOLS. Unfortunately, the size of the file indicated we must be missing something. Examine the file manually and started to see many more maps the software was not recognizing. So, I was left with nothing to do but create my own map finding tool.

Below a graphic display to explain the structure of 3D maps in the cal file. I imported the cal file hexadecimal dato into Excel, and converted it into decimal numbers. Then created some logic to check for map data that was consistent with the structure illustrated below.

Ended up finding a lot more maps. Hooray and darn, all at the same time ;-) Created a matrix to keep count of the various sizes of 'potential' maps. That is also below. Next the fun begins - see if we can match x and/or y axis values with ones similar to maps we are seeing in the HP Tuners maps for Australian Fords.

[rspring]

Thanks DWR. I have looked them up before, unfortunately they don't offer prices or any performance info on their site for 335d, so I'm less than fully comfortable.

[Mik325tds]

Originally Posted by DWR

Mik325tds and I have been working on understanding the calibration file from our transmission. A while back I reported that we were able to see a hundred or so maps using WinOLS. Unfortunately, the size of the file indicated we must be missing something. Examine the file manually and started to see many more maps the software was not recognizing. So, I was left with nothing to do but create my own map finding tool.

Below a graphic display to explain the structure of 3D maps in the cal file. I imported the cal file hexadecimal dato into Excel, and converted it into decimal numbers. Then created some logic to check for map data that was consistent with the structure illustrated below.

Ended up finding a lot more maps. Hooray and darn, all at the same time ;-) Created a matrix to keep count of the various sizes of 'potential' maps. That is also below. Next the fun begins - see if we can match x and/or y axis values with ones similar to maps we are seeing in the HP Tuners maps for Australian Fords.

Great work Derek!
So you think the 101 6x6 maps are our shift maps?

[mob17]

Originally Posted by rspring

Thanks DWR. I have looked them up before, unfortunately they don't offer prices or any performance info on their site for 335d, so I'm less than fully comfortable.

It's about £400. Si-335d has done a review on it recently.

It is expensive though.

[DWR]

Originally Posted by Mik325tds

Great work Derek!
So you think the 101 6x6 maps are our shift maps?

I did. But I don't now. Lol, turns out I didn't go high enough in the y axis dimension. Pretty sure I found at least a dozen of the shift maps with a 10 X 13 dimension. Below is a Map from a Australian Ford Falcon compared to a map from the cal file you flashed (A7610591). The Ford is a high reving V8, so the difference in the maps seems to make sense. OSS is the transmission output speed. I have not scaled the throttle % on the 335D data, just know 13 is the max on the map.

Now, if we can figure out when each map is used ...
Mik325tds get ready to do a lot of data logging ;-)

[DWR]

BTW, I'm still looking for the torque converter maps. They are amazingly elusive. Really have been concentrating on the torque converter, and found the shifts maps in the process. Should look similar to shift maps but with different x-axis dimensions. Lots of 6 X ? maps, but have not found a grouping that looks right. Also, found some smaller maps that look similar to torque management maps. Have not spent much time on those, just marked their locations.

I don't think we are that far from testing stuff out. Mik325tds, you think you can crack the checksum, etc. so we could flash a custom cal file? - man, if you told me 2 months ago I would be typing this pargraph, I would have shaken my head.

[Mik325tds]

Originally Posted by DWR

I don't think we are that far from testing stuff out. Mik325tds, you think you can crack the checksum, etc. so we could flash a custom cal file? - man, if you told me 2 months ago I would be typing this pargraph, I would have shaken my head.

DWR, you are the man! Phantastic work!
Seems like all the research on the Ford forums was worthwhile. I had a bit of trouble wrapping my head around the notation above, but things are starting to make sense. Maybe it's worthwhile translating OSS into wheel speed in order to get a feel for the maps as you suggested before.
As far as logging and sorting the maps goes, I think we should start at simple coast loggs (0% throttle input from 60mph) and find the downshift points. They should be significantly different between DS and D.
Once we have split these two types of maps we could start copying a DS map into the D area and see when it starts affecting the loggs.

I'll start focussing on the checksums again....

[Chief Orman]

Originally Posted by DWR

That's a good question. Some say yes. Some no. Depends upon your definition of performance. Look here:
http://www.jrtuning.co.uk/articles/b...p-tuning.shtml

I have been in touch with these guys and they are very clear that it is a transmission remap and not just a mechanism to reset adaptations. I also believe they are claiming a reflash of the transmission through the OBD port using their tool (said tool apparently can't be used to reflash ECU due to some new security feature). I am sure this is old news too many.

[DWR]

Originally Posted by Mik325tds

Maybe it's worthwhile translating OSS into wheel speed in order to get a feel for the maps as you suggested before.

Yes, I have an OSS to wheel speed to engine rpm spreadsheet. I believe can do that for both maps.

As far as logging and sorting the maps goes, I think we should start at simple coast loggs (0% throttle input from 60mph) and find the downshift points. They should be significantly different between DS and D.
.

Great idea. I think it would be good to search for the map switching parameter, also. There appears to be more maps than there are modes.

[TDIwyse]

This is great stuff guys. Thanks for your continued efforts and posting the progress.

Curious ... any long term plans to create a way to swap a manual transmission into the D?

[Mik325tds]

Originally Posted by TDIwyse

Curious ... any long term plans to create a way to swap a manual transmission into the D?

I'd be all over trying to get a manual in our D's but from what I read somewhere is that BMW doesn't have a manual transmission that can take the torque. Not sure if that is true though.

[Mik325tds]

Here's is a little math challenge for those of you that like puzzles:
:1000F0000007FF60FFFFFFFFFFFFFFFFFFFFFFFFA6
:10010000FFFFFFFFBBD81D600000000000000000E3
:1001100000000000000000000000000000000000DF
:1001200000000000000000000000000000000000CF
:1001300000000000000000000000000000000000BF
:1001400000000000000000000000000000000000AF

See the last byte in these lines? Those are the first checksums we'll have to get the correct formula for. So far I thought it was adding up all bytes in the line and then inverting (NOT operation) bitwise. But that will always end up 1 off.

Anyone else sees the solution?

[Mik325tds]

Originally Posted by Mik325tds

Here's is a little math challenge for those of you that like puzzles:
:1000F0000007FF60FFFFFFFFFFFFFFFFFFFFFFFFA6
:10010000FFFFFFFFBBD81D600000000000000000E3
:1001100000000000000000000000000000000000DF
:1001200000000000000000000000000000000000CF
:1001300000000000000000000000000000000000BF
:1001400000000000000000000000000000000000AF

See the last byte in these lines? Those are the first checksums we'll have to get the correct formula for. So far I thought it was adding up all bytes in the line and then inverting (NOT operation) bitwise. But that will always end up 1 off.

Anyone else sees the solution?

Duh.
It's Checksum=NOT(sum(allbytes))+1
One down, now the difficult one.