6L80 Reference Material

[joecar]

Finally...

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16484&d=1394164179

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16426&d=1393033854

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16427&d=1393033964

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16428&d=1393034073

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16470&d=1393991704

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16430&d=1393034451

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16431&d=1393034600

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16432&d=1393034839

[joecar]

http://forum.efilive.com/attachment.php?attachmentid=16433&d=1393034982

[T2000]

Excellent info - thankyou Joecar !!! :coool:

[joecar]

PT shift table relationships:

basic relationships:
1->2 curve should be above 2->1 curve;
2->3 curve should be above 3->2 curve;
3->4 curve should be above 4->3 curve;
4->5 curve should be above 5->4 curve;
5->6 curve should be above 6->5 curve;

additional relationships:
3->2 curve should be above 1->2 curve;
4->3 curve should be above 2->3 curve;
5->4 curve should be above 3->4 curve;
6->5 curve should be above 4->5 curve;

this can be summarized as: the curves must not cross anywhere;

if they cross anywhere, then you will not be able to predict what gear the TCM selects when the operating point hits the cross

( prioritization of the shift curves appears to be different across various TCM OS's )

[Telco]

Joe - have a question about your previous post, and I'm hoping there is new info on this. I've also been searching the forum.

How big a deal is it if the lines do cross? I ask because my factory 2015 program has the lines crossing in several spots, and I'm wondering if it's related to the 6L80E being able to directly switch between gears (like a 6>3 shift going 6-3) rather than having to cycle through (like a 6>3 shift going 6-5-4-3). Mine has been attached.

For anyone else, has anyone got a good 6L80E shift program they would care to share? I'm looking for something for a street vehicle that has excellent shift characteristics, that will still get up and go when I get into it. If nobody has one I can start with, I'll post mine up when I finally get one worked out. The 4.11s did much for drivability, but the stupid trans shifting at 5200RPM makes it feel like it's tripping at the finish line, so to speak.

[joecar]

Hi Telco,

Even with the ability to skip gears, the curves should still not cross (i.e. not be allowed to cross).

When they cross, the crossovers provide the TCM with multiple choices of which gear to select next...
the TCM/OS will make a decision based on current conditions (which may not be consistent each time);
different TCM/OS will select different choices (i.e. not consistent).

[Telco]

OK thanks. Makes sense, just didn't understand why GM was doing it.

[joecar]

I don't understand why they did it, seems like they rushed it.

[raceghost]

Your over thinking the second relationship. I'll explain.


additional relationships:
3->2 curve should be above 1->2 curve;
4->3 curve should be above 2->3 curve;
5->4 curve should be above 3->4 curve;
6->5 curve should be above 4->5 curve;

If the Transmission is in 5th gear, then the TCM shift points would be looking at its options from 5th gear, either it goes to sixth or down to 4th. So as to say "6->5 curve should be above 4->5 curve;" This is irrelevant, logic states that if the TCM looks at the 4>5 curve in relationship, it is already in 5th gear, it is not going to say, oh, shift back to 4th, and then shift back to 5th. The logic controller does not work that way. It is programmed with a set of if statements, "if this is (x), then do Y. SO there is no way you could confuse the TCM in the manner that you are suggesting with your second set of "additional relationships."

Now with that said, prove me wrong, please, educate me on how this would be programmabley possible with how language controllers are/or written in?

I do respect you, but with a degree in language engineering and software programming, I believe you are misleading. Your first statement of relationships holds true, but your second, not so much.

[raceghost]

Your over thinking the second relationship. I'll explain.



If the Transmission is in 5th gear, then the TCM shift points would be looking at its options from 5th gear, either it goes to sixth or down to 4th. So as to say "6->5 curve should be above 4->5 curve;" This is irrelevant, logic states that if the TCM looks at the 4>5 curve in relationship, it is already in 5th gear, it is not going to say, oh, shift back to 4th, and then shift back to 5th. The logic controller does not work that way. It is programmed with a set of if statements, "if this is (x), then do Y. SO there is no way you could confuse the TCM in the manner that you are suggesting with your second set of "additional relationships."

Now with that said, prove me wrong, please, educate me on how this would be programmabley possible with how language controllers are/or written in?

I do respect you, but with a degree in language engineering and software programming, I believe you are misleading. Your first statement of relationships holds true, but your second, not so much.

Joecar, I am blown away, it is as you suggested. I have tested and tested, and you are correct. I have not been able to look at the code directly from a disassembly point of view, but testing shows it, and I am able to recreate by doing some crazy crossovers within a few test roms. I can create it, and then take it away too.

[joecar]

Hi raceghost, thanks :)

The TCM can select a gear that is more than one gear's distance away, in these situations I've seen crossed VSS/TP shift tables causing unpredictable PT upshifts and indecisive PT dowshifts.

I've seen Throttle Downshift tables not at 100% TP cause conflicts with the VSS/TP shift tables.

[raceghost]

Hi raceghost, thanks :)

The TCM can select a gear that is more than one gear's distance away, in these situations I've seen crossed VSS/TP shift tables causing unpredictable PT upshifts and indecisive PT dowshifts.

I've seen Throttle Downshift tables not at 100% TP cause conflicts with the VSS/TP shift tables.

Which tables are you referring too? Also, have you ever made changes to the Inertia Tables within the Shift Speed sub category? Those look weird too.....

[joecar]

Throttle Downshift tables:
I have to look up their id numbers.
These look like the vss/tps upshift tables with the axes swapped.

Inertia tables depend very closely on physics of rotating components... my opinion is to leave these alone.