[Mik325tds]

Originally Posted by DWR

Oh man, if we could get our hands on a calibration template to know what is what. Even in German would be great - right Mik? OK Hoooper, you're in charge of donations for the consultant, ha! I'm laughing now, but that could change...

Yes, that would be something. But who knows, maybe he is as crazy as us and doesn't mind sharing his knowledge?

[Mik325tds]

I just flashed the A7610591.0da file for the N57 against all odds .
First test drive really felt a difference. It seemed to shift earlier and quicker somehow.
Unfortunately, my butt-dyno tricked me - the data doesn't really support that:

At least I'm getting better on the steady foot. And I took some screenshots of the flashing process, so hang in there for a tutorial.

[TDIwyse]

Originally Posted by Mik325tds

...
First test drive really felt a difference. It seemed to shift earlier and quicker somehow.
Unfortunately, my butt-dyno tricked me - the data doesn't really support that:
...

Really appreciate your use of data to test and evaluate these efforts. Thank you Sir.

[DWR]

Hey Mik,

Great work. Can you put some unique symbol on the curves for each cal? The graphs are starting to get busy. Thanks.

Wondering if we should be trying cals for a different engine so we can really measure some differences ... and see those differences in the cals.

[DWR]

Originally Posted by Mik325tds

Yes, that would be something. But who knows, maybe he is as crazy as us and doesn't mind sharing his knowledge?

OK, well, you are the foreign language expert on this team. I'll pay for the call ...

[DWR]

Originally Posted by Mik325tds

I just flashed the A7610591.0da file for the N57 against all odds .
First test drive really felt a difference. It seemed to shift earlier and quicker somehow.
Unfortunately, my butt-dyno tricked me - the data doesn't really support that:

At least I'm getting better on the steady foot. And I took some screenshots of the flashing process, so hang in there for a tutorial.

The 30% graph is the best for comparison. Very steady foot, now it just needs an offset calibration.

There's no hope for the butt-dyno. The original shifts are quicker, based on SA duration. Mik, the timescales are in sync, right? I see the speeds (trans out) seem to match pretty close.

We do not know for sure if the "load axis" on the cal maps are throttle, MAF, MAP or some other proxy for load. We are taking a guess at throttle, but on a turbocharged engine that's a bigger risk. Unless we get some additional info, I'm thinking we need to add those to the logs. To keep from extending the sampling period, we would need to take something out. It is pretty easy to see the gear changes from the turbine speeds, so the "gear" parameter could go. Similarly, throttle% and torque are highly correlated. At some point, torque is affected by gear changes, but not the other way around.

[Hoooper]

Lots of new transmissions are controlled highly based on calculated torque. That could be the load category. Shift time desired, pressures, etc based on torque and gear number. Shift points and torque converter lockup based on throttle, speed, and current gear. I can post some fairly simple tuning charts for 4l60e transmissions that might give insight if you think it would help.

[davidjaen]

I think that there is one BMW 335d alpina... Maybe somebody know something about it.

[TDIwyse]

Question.

Are these logs being done in D or M/S mode? Reason I'm asking is because from some older documentation it appears this, as well as how one was driving previously to the test, could have a large impact on the shift behavior... it also calls out at least 4 basic baseline "maps"...

From BMW Steptronic pdf found on Post # 9 on this thread: http://www.bimmerfest.com/forums/sho...d.php?t=262516

The adaptive transmission control (component of electronic transmission control) adapts the shift
characteristics of the comfort program and sport program to the driver's wish and the driving situation. In both
programs, the adaptive transmission control changes from a basic map to a performance-orientated map, as
required:
· Comfort program in selector lever position "D"
The comfort program is based on 2 characteristic maps: The XE map (extreme economy) and the
E map (economy). The EGS control unit normally selects the economical XE map (= basic map). In
case of particular demands (e.g. load requirements), the system will change to the performanceorientated
E map. If the input signals change in favour of a more "gentle" driving style, the system will
change back to the more economical XE map.
· Sport program in selector lever position "M/S"
The sport program effects dynamic, sporty shift characteristics. In the sport program, there is the basic S
map (sport) and the performance-orientated XS map (extreme sport). If an extremely dynamic driving
style is called for, the system will change from the S map to the performance-orientated XS map.

[DWR]

Originally Posted by Hoooper

Lots of new transmissions are controlled highly based on calculated torque. That could be the load category. Shift time desired, pressures, etc based on torque and gear number. Shift points and torque converter lockup based on throttle, speed, and current gear. I can post some fairly simple tuning charts for 4l60e transmissions that might give insight if you think it would help.

Hoooper, that would be very helpful. Thank you.

[DWR]

Originally Posted by TDIwyse

Question.

Are these logs being done in D or M/S mode? Reason I'm asking is because from some older documentation it appears this, as well as how one was driving previously to the test, could have a large impact on the shift behavior... it also calls out at least 4 basic baseline "maps"...

From BMW Steptronic pdf found on Post # 9 on this thread: http://www.bimmerfest.com/forums/sho...d.php?t=262516

The adaptive transmission control (component of electronic transmission control) adapts the shift
characteristics of the comfort program and sport program to the driver's wish and the driving situation. In both
programs, the adaptive transmission control changes from a basic map to a performance-orientated map, as
required:
· Comfort program in selector lever position "D"
The comfort program is based on 2 characteristic maps: The XE map (extreme economy) and the
E map (economy). The EGS control unit normally selects the economical XE map (= basic map). In
case of particular demands (e.g. load requirements), the system will change to the performanceorientated
E map. If the input signals change in favour of a more "gentle" driving style, the system will
change back to the more economical XE map.
· Sport program in selector lever position "M/S"
The sport program effects dynamic, sporty shift characteristics. In the sport program, there is the basic S
map (sport) and the performance-orientated XS map (extreme sport). If an extremely dynamic driving
style is called for, the system will change from the S map to the performance-orientated XS map.

Thanks TDIwyse, great infomation. I believe the reason Mik is trying to fix the load on the transmission during the logs is to avoid the switching between map 'types'. In the calibration files, there are certainly enough maps to cover many senarios - it does not make it easier, as you can imagine. There are 2 forms of adaption, the one you describe and one that adjusts timing between the application and release of clutches. Luckily, the later isn't something we need to worry about ... yet anyway.

[DWR]

Originally Posted by Hoooper

Lots of new transmissions are controlled highly based on calculated torque. That could be the load category. Shift time desired, pressures, etc based on torque and gear number. Shift points and torque converter lockup based on throttle, speed, and current gear. I can post some fairly simple tuning charts for 4l60e transmissions that might give insight if you think it would help.

I believe shift time desired is going to be very hard to find, using this method, as I'm guessing it is not a 2 dimensional map. And yet, that has to be one of the most important parameters for many of us. We have our work cut out. Really hoping for a lucky break(s).

[DWR]

Originally Posted by TDIwyse

Question. Are these logs being done in D or M/S mode?

Sorry, I did not answer the initial question. Mik is most interested in gaining mpg, so the tests are in D (Mik correct me if I'm in error)

Originally Posted by TDIwyse

it also calls out at least 4 basic baseline "maps".

I think that is the part that is surprising me the most. Most of the differences between the cal files are outside of the maps. The majority of the maps do not change. There is a group of maps that seem to be the exception - now if we could just figure out what they control!

[Works]

If you guys happen to figure out how to get the DS (S/XS) shift and tc lock map to work with D's (E/XE) rev map the rest of the world would be eternally grateful. This combo would then give birth a new D mode with economical shift points with crisp shifts and early tc lock.

I would be so happy..

[Hoooper]

Originally Posted by DWR

I believe shift time desired is going to be very hard to find, using this method, as I'm guessing it is not a 2 dimensional map. And yet, that has to be one of the most important parameters for many of us. We have our work cut out. Really hoping for a lucky break(s).

I guess it depends on the format. I don't know much about code, especially hex code, but I would expect the presentation for desired shift time which is probably between 0.2 and 0.5s to be very different from all other values which the lowest maybe being something like 3 for the 2-1 shift mph minimum and the rest of that table being much higher.

[DWR]

Originally Posted by Hoooper

I guess it depends on the format. I don't know much about code, especially hex code, but I would expect the presentation for desired shift time which is probably between 0.2 and 0.5s to be very different from all other values which the lowest maybe being something like 3 for the 2-1 shift mph minimum and the rest of that table being much higher.

Yeah, BMW literature says 300ms. So a value around 300 is likely. Unfortunately, with 50,000 addresses, there are many places where that value exists and in 6 consecutive addresses or more. Yikes!

[Hoooper]

Originally Posted by DWR

Hoooper, that would be very helpful. Thank you.

Will do ASAP, they are on my tablet at home

[Mik325tds]

Originally Posted by Hoooper

Will do ASAP, they are on my tablet at home

Yes, please! That will be very helpful.

[Mik325tds]

Originally Posted by TDIwyse

Question.

Are these logs being done in D or M/S mode?

DWR is correct, I did the logs in D and am pretty sure the car is in XE mode as I drive mostly for economy. Also, I reset the adaptions after the flash and drove on a level surface.

[cptcrunch]

Interesting data sets and logging. I'll be following your progress closely

[mob17]

Are you able to code the car/tranny to switch DTC on (one press) automatically when switching to DS mode?

[Mik325tds]

Originally Posted by mob17

Are you able to code the car/tranny to switch DTC on (one press) automatically when switching to DS mode?

I'm not sure if that is a coding option. I'll look out for it when I go in next with NCSexpert. It turns out, flashing a new SW turns off the coding for the flappy pedals behind the wheel. Will need to correct that later. Don't need them for now.