New A6 Transmission parameters walkthrough

[boomer]

Well it looks like it works somewhat like i thought! attached picture is what i change the tables to (upshift T/M is enabled).

then watch my Log file from 2:55+ (had rev limiter issues before that). i made some WOT runs shifting at high and mid rpm ranges along with some partial throttle shifts. there is a big difference. timing pull always stayed in the positives at shift points which resulted in a harder shift.

but that's all i have time for right now. gotta go to school.

[johnh]

Great news! Thanks... Eagerly awaiting the release....

[thesubfloor]

Well it looks like it works somewhat like i thought! attached picture is what i change the tables to (upshift T/M is enabled).

then watch my Log file from 2:55+ (had rev limiter issues before that). i made some WOT runs shifting at high and mid rpm ranges along with some partial throttle shifts. there is a big difference. timing pull always stayed in the positives at shift points which resulted in a harder shift.

but that's all i have time for right now. gotta go to school.

Although it could be because we have different cars (I have an 07 Corvette), I don't like the way your scan looks with respect to how the timing gets pulled.

In the attached scan that I took back in June with full TM enabled, the timing doesn't start to get pulled until I hit peak RPM. In your scan however it starts to pull timing much sooner and then takes roughly twice as long to give it back.

Hopefully the difference will be because our respective cars, but if it ends up behaving the same on my car then it almost looks like leaving TM fully enabled would end up slowing you down less than trying to dial it down by a certain percentage.

Christopher

[boomer]

Although it could be because we have different cars (I have an 07 Corvette), I don't like the way your scan looks with respect to how the timing gets pulled.

In the attached scan that I took back in June with full TM enabled, the timing doesn't start to get pulled until I hit peak RPM. In your scan however it starts to pull timing much sooner and then takes roughly twice as long to give it back.

Hopefully the difference will be because our respective cars, but if it ends up behaving the same on my car then it almost looks like leaving TM fully enabled would end up slowing you down less than trying to dial it down by a certain percentage.

Christopher

i understand what you are saying and it makes sense. the G8s pull timing very aggressively. the second the car would hit a commanded shift, timing would go in the negatives then would slowly ramp back up. with everything set to .5, it was much different as you can see.

ya, i would guess the corvettes T/M is less aggressive.

[boomer]

compare this file to my Beta file and you will see that there is a big difference in timing pull at the commanded shift points. both have upshift T/M enabled.

[johnh]

Yeah no more negative timing at the shifts!! Boomer, is the feel/SOTP difference between the two dramatic?

I'd love to test this out as well.

[thesubfloor]

compare this file to my Beta file and you will see that there is a big difference in timing pull at the commanded shift points. both have upshift T/M enabled.

Your HPT Shift log looks just like mine - it doesn't start pulling timing until it hits peak RPM and then gives it back twice as fast as when you tried to reduce it by only a percentage.

I'm not going to try it out in my car until Chris chimes in on this thread as I won't really feel comfortable until I know how the values are supposed to work.

Christopher

[thesubfloor]

As a side note to Chris and Bill, I discovered that there are some redraw issues with the chart numbers in the scanner. I attached a screenshot so you can see what I'm talking about.

Christopher

[Bill@HPTuners]

When using ANY beta version contact support with any issues you come across please.

Thanks,
Bill

[boomer]

In the attached scan that I took back in June with full TM enabled, the timing doesn't start to get pulled until I hit peak RPM. In your scan however it starts to pull timing much sooner and then takes roughly twice as long to give it back.

sorry, i was in a rush when i posted my first beta log earlier. after reviewing it i saw what you were talking about how it was pulling timing before the shift. its because i was using TUTD to shift so it wouldn't hit the rev limiter. TUTD must pull timing differently.

got that all fixed and did another log this is with all the same T/M settings as my other Beta log.

[Chris@HPTuners]

a good place to start would be with the Cold Shift Factor tables. Usually they are 1.0 at the highest temp. This factor is the last multiplier and should see an differnce for all upshifts). Use this just to test things out, once you are happy with what to change and how much to change it, probably best to set the Cold factor back to stock.

Making the numbers closer to 0 should result in less torque reduction.

Some more detail:

The Final Torque Factor is calculated like this:

FTF = {Shift Torque Factor + (Shift Torque Factor Adder * Shift Torque Factor Adder Mod)} * Torque Factor Cold

Shift Torque Factor is per shift & usually in most cals set to 1.0

Shift Torque Factor Adder is per shift, it is multiplied by the Shift Torque Adder Mod before being added to the Shift Torque Factor.

Shift Torque Adder Mod is a table with RPM & Driver Demand Torque. It is used to scale the Shift Torque Adder depending on RPM/Demand Torque To give the different shift feel. In some cases this table is 0's in others it changes the shift at lower RPM/Torque. Remember it is the combination of all these factors that ends with the Final Torque Factor.


Torque Factor Cold is used to account for cold trans temp. You can see in the forumla above it is the final multiplier of everything. So it will be a good place to start. ie. leve everything stock and set the cold factor to say 0.75 and then 0.5, 0.25 or 0.0 or something and see what happens. Then set it back to 1.0 and try to refine it further using the individual factor, adder and adder mod.

Also, the desired output torque tables control the profile of the torque during the shift and you'll need to experiment with those as well as these new params to get what you want.

Hope that helps,

Chris...

[thesubfloor]

a good place to start would be with the Cold Shift Factor tables. Usually they are 1.0 at the highest temp. This factor is the last multiplier and should see an differnce for all upshifts). Use this just to test things out, once you are happy with what to change and how much to change it, probably best to set the Cold factor back to stock.

Making the numbers closer to 0 should result in less torque reduction.

Some more detail:

The Final Torque Factor is calculated like this:

FTF = {Shift Torque Factor + (Shift Torque Factor Adder * Shift Torque Factor Adder Mod)} * Torque Factor Cold

Shift Torque Factor is per shift & usually in most cals set to 1.0

Shift Torque Factor Adder is per shift, it is multiplied by the Shift Torque Adder Mod before being added to the Shift Torque Factor.

Shift Torque Adder Mod is a table with RPM & Driver Demand Torque. It is used to scale the Shift Torque Adder depending on RPM/Demand Torque To give the different shift feel. In some cases this table is 0's in others it changes the shift at lower RPM/Torque. Remember it is the combination of all these factors that ends with the Final Torque Factor.


Torque Factor Cold is used to account for cold trans temp. You can see in the forumla above it is the final multiplier of everything. So it will be a good place to start. ie. leve everything stock and set the cold factor to say 0.75 and then 0.5, 0.25 or 0.0 or something and see what happens. Then set it back to 1.0 and try to refine it further using the individual factor, adder and adder mod.

Also, the desired output torque tables control the profile of the torque during the shift and you'll need to experiment with those as well as these new params to get what you want.

Hope that helps,

Chris...

Thanks Chris,
Christopher

[johnh]

Chris, when do you think this will be out of beta and released?

[Bill@HPTuners]

Chris, when do you think this will be out of beta and released?

I would venture to guess sometime after the PRI show.

[Pro Street]

When using ANY beta version contact support with any issues you come across please.

Thanks,
Bill

Bill how do you get this beta? I would like to give it a try on my 09 G8.

[boomer]

Bill how do you get this beta? I would like to give it a try on my 09 G8.

email support with your name and ID#

[Pro Street]

email support with your name and ID#

Thanks Boomer!

Email sent to support.

[carneb]

I'm thinking that reducing the TM to give the shift feel we are after is doing things the wrong way. What we should be doing is changing the parameters that adjust shift feel and then let the TCM calculate the correct TM to give the resultant feel we are after.

If you have a look at this document http://www.google.com/patents?id=6KQ...BAJ&dq=6364811 it seems to describe how the 6L80 works. It talks about changing output torque trajectory and turbine acceleration trajectory to calibrate shift feel, then the TCM calculates the TM and oncoming clutch pressure required to deliver that.

HPtuners already allows us to adjust desired output torque, initial and final acceleration, and total inertia phase shift time. I'm wondering what else is hiding in the program that would allow us to adjust the turbine acceleration to give the shift feel we are after?

Having said all that, I am happily playing with the new TM settings for the moment.

[johnh]

I'm thinking that reducing the TM to give the shift feel we are after is doing things the wrong way. What we should be doing is changing the parameters that adjust shift feel and then let the TCM calculate the correct TM to give the resultant feel we are after.

That makes sense. We want just enough torque for a good shift but not all the torque to break hard parts--best of both worlds so to speak.

So shift time (know where that is) and shift feel (don't know which params that is ) are probably the main inputs (I am theorizing here!). I am thinking that GM set it up so that they can change feel just by a few tables, the rest are just used to "describe" the physical system (eg. pressure, max torque clutches, gears, splines can handle etc). That's the way I'd design it anyway...

A couple of oddities I've noticed (not beta code):
Shift Properties, Discrete shift torque Enable why is this set enabled for 1-2, 2-3 only? Anyone test this out with all 0s (all disabled)

Shift Properties, Max Adapt Vol, why is 2-3 different than all the rest? 40 vs. all rest at 200?

[Pro Street]

That makes sense. We want just enough torque for a good shift but not all the torque to break hard parts--best of both worlds so to speak.

So shift time (know where that is) and shift feel (don't know which params that is ) are probably the main inputs (I am theorizing here!). I am thinking that GM set it up so that they can change feel just by a few tables, the rest are just used to "describe" the physical system (eg. pressure, max torque clutches, gears, splines can handle etc). That's the way I'd design it anyway...

A couple of oddities I've noticed (not beta code):
Shift Properties, Discrete shift torque Enable why is this set enabled for 1-2, 2-3 only? Anyone test this out with all 0s (all disabled)

Shift Properties, Max Adapt Vol, why is 2-3 different than all the rest? 40 vs. all rest at 200?

johnh I looked at these tables way to long last night. My head still hurts!