[3lli0t]

Originally Posted by Mik325tds

Yes, that'll probably a single value in the cal file different from the standard cal file. Probably impossible to find unless it is in a field of same values.
What do you mean, with "only one mapping available in the data"?

That sounds not so nice :P.
My description concerning the mapping was a little bit confusing: For my HW-# there are min. 25 different files in the DATA which can be flashed. My actual ZUSB is for the N57 and there is no file for the M57 for the GKE215. The reason is, that the e92 325d from 2009 was one of the last models with the M57TU2UL, the 330d had already the N57 (245 HP), so BMW didn't develop an extra mapping for the M57 with GKE215.
Generally I would try to flash the Alpina Soft for the N54 and test the result. But I don't know if that could harm the transmission, because it's made for an entire different torque and power curve. On the other hand I have to start somewhere and like so see if it's possible to combine the Alpina-Soft with the standard-N57-mapping. The plan: Take the Alpina file and overwrite the N54-tables with the N57-tables. If this works I have a file with the alpina-display (Gears in D and DS) and can start to optimize the mapping. Now my question: Is it possible to isolate only the shift-maps und leave the rest?

[Mik325tds]

Originally Posted by 3lli0t

That sounds not so nice :P.
My description concerning the mapping was a little bit confusing: For my HW-# there are min. 25 different files in the DATA which can be flashed. My actual ZUSB is for the N57 and there is no file for the M57 for the GKE215. The reason is, that the e92 325d from 2009 was one of the last models with the M57TU2UL, the 330d had already the N57 (245 HP), so BMW didn't develop an extra mapping for the M57 with GKE215.
Generally I would try to flash the Alpina Soft for the N54 and test the result. But I don't know if that could harm the transmission, because it's made for an entire different torque and power curve. On the other hand I have to start somewhere and like so see if it's possible to combine the Alpina-Soft with the standard-N57-mapping. The plan: Take the Alpina file and overwrite the N54-tables with the N57-tables. If this works I have a file with the alpina-display (Gears in D and DS) and can start to optimize the mapping. Now my question: Is it possible to isolate only the shift-maps und leave the rest?

First you have to find out if the cal files have the same structure for your HW number and the alpina flash. I use TotalCommander to do a binary file compare for that. If the files are the same length and the sectioning is the same you should see only differences in the data itself but not in the addresses. Best case would be if the HW number is the same for the alpina flash and your HW, but since it didn't flash I assume that is not the case.
As far as identifying the shift maps goes, I guess we'll have to wait until DWR shares his magic on that.

[3lli0t]

Originally Posted by Mik325tds

First you have to find out if the cal files have the same structure for your HW number and the alpina flash. I use TotalCommander to do a binary file compare for that. If the files are the same length and the sectioning is the same you should see only differences in the data itself but not in the addresses. Best case would be if the HW number is the same for the alpina flash and your HW, but since it didn't flash I assume that is not the case.
As far as identifying the shift maps goes, I guess we'll have to wait until DWR shares his magic on that.

I compared the N54-Alpina file and the N57 file. The adresses are the same. If I make the right settings in WinOLS it finds around 100 maps, but I have no idea if the software is right .

[Mik325tds]

Originally Posted by 3lli0t

I compared the N54-Alpina file and the N57 file. The adresses are the same. If I make the right settings in WinOLS it finds around 100 maps, but I have no idea if the software is right .

Wich one is the alpina file?
WinOLS will not help you. As DWR found out, the structure of the maps do not follow the Bosch standard structure, so WinOLS will not correctly identify maps. He used an excel file to read in the cal file and convert it to decimal numbers and then some magic on how to identify the maps. Hopefully, he'll let us in on how to exactly transfer that to other cal files. Success in this case certainly doesn't come without sweat.

[Mik325tds]

Originally Posted by DWR

I'm in the midst of upgrading my map finding tool and have it inoperable right now. But you and the forum will have it shortly after I do . In the meanwhile, this is what I can tell you. After the first block of zeros, use the data value "534C" as the starting point, row1 and column1. The infield data for the shift maps is at row 7839, column 6, starting with value "0122". The maps do not follow Bosch format (that's why I could not find them at first), but rather run infield data consecutively, 10 columns, row 1 to 13, 10 maps, in that sequence.

I'm not quite sure I'm following you. In the file A7610591.0da row 7839 looks like this:
:10E4D00000010303000000140028026C030C03B6C3
Can't find 0122 anywhere close to that. The next 0122 is in row 8008 address :10EF6000. What am I missing?

[DWR]

Originally Posted by 3lli0t

Is it possible to isolate only the shift-maps und leave the rest?

I think you are asking is it possible to change only the shift maps? The answer is yes.

[DWR]

Originally Posted by 3lli0t

I compared the N54-Alpina file and the N57 file. The adresses are the same. If I make the right settings in WinOLS it finds around 100 maps, but I have no idea if the software is right .

It is not. WinOLS doesn't find the shift maps.

[DWR]

Originally Posted by Mik325tds

I'm not quite sure I'm following you. In the file A7610591.0da row 7839 looks like this:
:10E4D00000010303000000140028026C030C03B6C3
Can't find 0122 anywhere close to that. The next 0122 is in row 8008 address :10EF6000. What am I missing?

We are not working with the same file. I'm working with the data in Excel and not the *.0da. That's why I had to give a reference starting point and then the number of rows down from the reference. Right now the spreadsheet is being used as a kind of scratch pad to manipulate the data in ways that make map finding easier. Since I have not cared where I found the maps, I made no attempt to keep the Excel file and the *.0da file in sync - Sorry. I will resolve that shortly.

For anyone wanting to match up patterns, here is the data for the first shift map in A7610591:
XXXX XXXX XXXX XXXX XXXX 0122 0122 0122
0140 017C 019F 01BD 01C2 0262 0262 0262
0262 0320 00A0 00A0 00A0 00FF 00FF 00FF
00FF 015E 015E 015E 01E0 01E0 0258 044C
044C 044C 044C 047E 04B0 04F1 04FB 0622
0622 0622 0622 06F9 0352 0352 0352 0352
0352 0352 0352 03E8 0460 0460 0500 0500
0609 0622 0622 0622 0622 06D6 073A 079E
07B7 09C4 09C4 09C4 09C4 0AC8 04B0 04B0
04B0 04B0 04B0 04B0 04B0 0514 0690 06C2
07C1 07C1 09DD 083E 083E 083E 083E 097E
0A0F 0AAA 0ABE 0D48 0D48 0D48 0D48 0EA6
0672 0672 0672 0672 0672 0672 0672 07D0
0A32 0A87 0A9B 0A9B 0DAC 11F8 11F8 11F8
11F8 11F8 11F8 11F8 11F8 11F8 11F8 11F8
11F8 1388 0BC2 0BC2 0BC2 0BC2 0BC2 0BC2
0BC2 0BD1 0DC5 0E10 0EE2 0EE2 1194 XXXX

[DWR]

So after many hours of trying to make a square peg fit into a round hole, I gave up. I found the Torque Converter Control Maps data a couple of weeks ago, but just could not get it into a map that made sense. Was trying too hard to make it fit the Ford maps in HP Tuners. Well, it doesn't. Here is the OSS data for the first TCC Map. I believe it is an application map. There are 6 more adjacent maps. Release maps come later? 8160 is approximately 200mph, so that means, in this map, first and second gear will not lock up. The multple columns per gear is not fully understood. The Australian Ford maps have 3 columns that some say are for high, med and low OSS - That doesn't make sense as it will always lockup at the low OSS. So, it will take manipulation and datalogging to figure it out.

[Hoooper]

The left column could be apply and the right release, it would make sense when only looking at those numbers, the TCC applies at a higher speed and the release happens at a lower speed. The chart doesn't look right though to me. No variation based on throttle. Does 2nd gear lock while manual shifting in your experience? It definitely locks up in D.

[Mik325tds]

Originally Posted by DWR

We are not working with the same file. I'm working with the data in Excel and not the *.0da. That's why I had to give a reference starting point and then the number of rows down from the reference. Right now the spreadsheet is being used as a kind of scratch pad to manipulate the data in ways that make map finding easier. Since I have not cared where I found the maps, I made no attempt to keep the Excel file and the *.0da file in sync - Sorry. I will resolve that shortly.

For anyone wanting to match up patterns, here is the data for the first shift map in A7610591:
XXXX XXXX XXXX XXXX XXXX 0122 0122 0122
0140 017C 019F 01BD 01C2 0262 0262 0262
0262 0320 00A0 00A0 00A0 00FF 00FF 00FF
00FF 015E 015E 015E 01E0 01E0 0258 044C
044C 044C 044C 047E 04B0 04F1 04FB 0622
0622 0622 0622 06F9 0352 0352 0352 0352
0352 0352 0352 03E8 0460 0460 0500 0500
0609 0622 0622 0622 0622 06D6 073A 079E
07B7 09C4 09C4 09C4 09C4 0AC8 04B0 04B0
04B0 04B0 04B0 04B0 04B0 0514 0690 06C2
07C1 07C1 09DD 083E 083E 083E 083E 097E
0A0F 0AAA 0ABE 0D48 0D48 0D48 0D48 0EA6
0672 0672 0672 0672 0672 0672 0672 07D0
0A32 0A87 0A9B 0A9B 0DAC 11F8 11F8 11F8
11F8 11F8 11F8 11F8 11F8 11F8 11F8 11F8
11F8 1388 0BC2 0BC2 0BC2 0BC2 0BC2 0BC2
0BC2 0BD1 0DC5 0E10 0EE2 0EE2 1194 XXXX

Ok. I understand now. I found 6 maps with this similar signature:
Addresses:
:10EF6000 (line 8008)
:10F78000 (line 8138)
:10F88000 (line 8154)
:10FF1000 (line 8259)
:10012000 (line 8294)
:10022000 (line 8310)
:10032000 (line 8326)
Will start to mess with them soon....

[Mik325tds]

Originally Posted by Hoooper

The left column could be apply and the right release, it would make sense when only looking at those numbers, the TCC applies at a higher speed and the release happens at a lower speed. The chart doesn't look right though to me. No variation based on throttle. Does 2nd gear lock while manual shifting in your experience? It definitely locks up in D.

I agree, doesn't make too much sense yet. Why would the OSS for lock up be lower in 5th gear then in 4th?

[Hoooper]

Originally Posted by Mik325tds

I agree, doesn't make too much sense yet. Why would the OSS for lock up be lower in 5th gear then in 4th?

The only reason I can think of would be to ensure that the TCC is always locked in 5th gear at any speed it will allow you to be in that gear. That doesnt make sense though because that OSS corresponds to about 58 MPH which is way higher than my 5th gear can be engaged at and is also way higher than the TCC is typically engaged in 5th

[DWR]

Originally Posted by Hoooper

The left column could be apply and the right release, it would make sense when only looking at those numbers, the TCC applies at a higher speed and the release happens at a lower speed. The chart doesn't look right though to me. No variation based on throttle. Does 2nd gear lock while manual shifting in your experience? It definitely locks up in D.

Throttle is zero to max, down the column. Sorry, I did not make that clear.

[DWR]

Originally Posted by Hoooper

The only reason I can think of would be to ensure that the TCC is always locked in 5th gear at any speed it will allow you to be in that gear. That doesnt make sense though because that OSS corresponds to about 58 MPH which is way higher than my 5th gear can be engaged at and is also way higher than the TCC is typically engaged in 5th

This first map is the most extreme. I think the others are easier to reconcile. Here's the next 4 maps. The 2 after these have a slightly different format that I need to work out. (note, I left the column numbers in, but those values are not part of the map. Also, as previously, throttle position is from low to high, top to bottom for the columns.

[Mik325tds]

Originally Posted by DWR

This first map is the most extreme. I think the others are easier to reconcile. Here's the next 4 maps. The 2 after these have a slightly different format that I need to work out. (note, I left the column numbers in, but those values are not part of the map. Also, as previously, throttle position is from low to high, top to bottom for the columns.

Still can't get my head around the TCC maps. Didn't you say they where dependent on Throttle and OSS while the parameters were PWM duty cycle of the lock up clutch valve? Wouldn't we expect a 3d table then with the duty cycle as infield values? Or are there 2 sets of maps working together?

[Hoooper]

Originally Posted by Mik325tds

Still can't get my head around the TCC maps. Didn't you say they where dependent on Throttle and OSS while the parameters were PWM duty cycle of the lock up clutch valve? Wouldn't we expect a 3d table then with the duty cycle as infield values? Or are there 2 sets of maps working together?

I would expect another table with tcc pwm min/max as infield values dependent on transmission fluid temp and desired torque or maybe gear. Based on the way the lockup feels in this car, I wouldn't be surprised if there isn't much difference between the min and the max.

[DWR]

Originally Posted by Mik325tds

Still can't get my head around the TCC maps. Didn't you say they where dependent on Throttle and OSS while the parameters were PWM duty cycle of the lock up clutch valve? Wouldn't we expect a 3d table then with the duty cycle as infield values? Or are there 2 sets of maps working together?

As I understand it, we are looking at OSS infield vs throttle. 3 states for converter lock up. ON/OFF/PWM. PWM and its duration is handled somewhere else.

[DWR]

And if you thought the previous set was difficult ... here's the best I could do with the next section of the cal file. Data is in groupings of 5 and it has the same look of the previous maps, but the columns are even harder to interpret.

Starting to wonder if these are OSS numbers? The only consolation is it is just as funky in a stock Ford cal, see below. That map is supposedly 3 columns per gear. Again, just as the Ford guys in Aussieland figured this stuff out, we'll have to change whole sections to off/on and see when the converter locks and unlocks. I'd like to start by changing everything to '8160' and verify that the converter never locks up.

[Wibbles]

Originally Posted by Mik325tds

Yes, you can use any of the numbers I posted to go with your transmission. They should flash with no problem. There is one newer file for you transmission:
7606261. They seem to be ordered in consecutive numbers.
Look at the attached file: Your HW number is 7566894. All lines that include that HW number are flashable with WinKFP in comfort mode.

Thanks for helping me. I flashed my EGS with that firmware yesterday and it's made a difference to the way the car drives. Shits are smoother & the TC locks up earlier. It's no DCT but an improvement none the less.

Cheers

[Mik325tds]

Took some logs today to start sorting out the D vs DS shift maps. This is coasting down from 6th to 2nd gear in D-XE and 4th to 3 in DS. I'm starting to think that Stat_SA is a Status in a State machine that handles up and downshifts. SA starts changing values before a change in gears happens and ends at the end of the gear change. During that time a series of things happen: The lock up clutch unlocks, rpms rise due to the actual gear being engaged, then lock up clutch engages again.
Now the question is: Which of the following OSS is referred to in our shift maps? Start or End of SA stat or the OSS where the new gear is actually engaged?

[Mik325tds]

Should we have gotten lucky? There's only one map that DWR identified which closely matches the start of Stat_SA change: