T43 Dissasembly

[antus]

Yeah the whole plugin thing was new to all of us at some stage. I know Mark added it for the LS1 plugin, as he didn't want to own too many customizations and wanted the developers with the domain knowledge to be able to extend the app. I caught it and set it unique for the ose plugin but obviously missed it on the checksum plugin back... lets see 9 years ago here viewtopic.php?p=47699&hilit=ls1+xdf#p47699 . At the time there were no other plugins, so I guess it didn't cause a problem until years later. I just checked your plugin because in more recent times there has been a lot of confusion around UUIDs and the whole process is a little clunky and not obvious to the developer or the user.

It would be nice if Tunerpro listed loaded plugins and their UUID. At the moment it only shows data acquisition and logging plugins loaded, it does not show their UUID and does not list loaded checksum plugins at all. It also does not show UUIDs in the XDF or ADX editors, only the name of the plugin. This is fine when everything is working and XDF/ADX are shipped with a matching DLL but it makes troubleshooting hard when multiple xdfs/adx/dlls have the same UUID in error, or the XDF/ADX is shared without including the plugin DLL.

[hjtrbo]

Reckon I got a new table for us to play with. Line Pressure Temperature Compensation is a 1D table with presumably a temperature axis (have not located yet), that is a gain value within the line pressure calculation. There are some missing bits to the puzzle but there is enough there to claim a strong assumption of what's going on. Massive thanks to (you know who you are) for making these additional discoveries possible. I'll hold off on the xdf update for a short while as I'd like to find some more stuff. Address info attached if you want to look at it. Hopefully this will remove the need to jack the torque tables in the engine cal when doing a FI adder.

Ukh6j4X.png (13.62 KiB) Viewed 1171 times

[Tre-Cool]

probably safe to assume the temperatures from an existing known calibration. like -40 to 150 c

has to go high enough temp wise for the hot mode to be enabled & im pretty sure thats around 130c

[hjtrbo]

TCC logic is coming along. I've found the 2 variables I was looking for. 1 is a pressure offset due to an upshift condition and the other is the running TCC pressure command before being modified further here. Before it arrives here it has been modified by what seem to be the 2 modes as described in the attached patent. There are also gear vs rpm clamps and other modifiers that I'm still working through. But I think I have enough information now to interpret and tune the desired pressure table with a better degree of confidence than I was previously. Alas, tracing the table in the scanner will still be difficult as we only have line pressure available (PCS1 x 2.045 + 300 or PCS1 x 2.335 + 300).

Extra tid bit, this is the PCS1 to line pressure corelation. If you're interested in setting up a math channel in the scanner choose 1 of the formulas above. 2.045 is from the HP forum and 2.335 is from the 6L80 forum where someone listed test results with pressure sensors hooked into the valve body.

zNAwCWQ.png (42.18 KiB) Viewed 1014 times

Extra extra tid bit, looking at complex functions can be tricky, but using the Graph Control Flow tool makes things a lot easier to visualise what's going on. Here is a good example of the shift inertia and shift time torque adder select function. Viewing this in code was nasty!

[gmtech825]

Really great work! I'm very interested in the TCC control strategy for these 6l transmissions. I'm curious about how the tcm determines the slip rate when increasing torque demand on 2014+ trucks. steady state they hold a consistant slip rate but as the pedal is pressed the TCC slip increases but doesn't fully release...probably to reduce the load on the engine.

[hjtrbo]

Yeah, there is conditions for transitions to enter / exit the different TCC modes. As torque increases it should go into lock on mode (full lock) then as you go back to steady state cruising it will transition back to on mode (constant slip). Lock on mode may not be turned on in some cals.

The information in the patent I attached earlier is very similar to what I'm seeing in the disassembly.

I've since made another discovery since my last post, I've found a pertinent switch, scalar and table for the tcc coast mode. For as long as I've owned my car, when zipping around the hills that coast mode has been giving me the shits. It commands a low pca pressure, then as you get back on the throttle, it goes through the enable condition checks before it will transition to on mode or lock mode. With a loud exhaust it sounds shit as you get the engine rpm rise then fall as you get increased slip while waiting for the PCA pressure to wind up to on mode pressure. Anyway, I'm disabling all that coast crap out so hopefully that's the end of that. It should just stay in on or lock mode.

These are the factory settings:

kILxsNo.png (30.66 KiB) Viewed 1005 times

[Tre-Cool]

Interesting. Can't say that i've ever noticed what you describe on mine. but they have aftermarket multiplates, so i set the base pressure to around 30psi. It will coast down & perform rev matches and maintain lock up on decel. The trick there appears to make sure you have enough min idle air flow, without the rev match also doesnt seem to work.

Might need to add them into the OSes i mostly use to see if they are on. Generally Range 4 or 5 is used in the commodores depending on OS. For Example in 24256025 D3 is normal auto mode then i pull back to D4 for tapshift to engage for my paddles. but in 24256125 in the same detent as the previous OS it shows as D5.

Cats has a PRNDL config & they are different between the 2 oses. However I have no idea what the numbers correspond to. I might need to test changing the numbers & see what happens.

[hjtrbo]

so i set the base pressure to around 30psi. It will coast down & perform rev matches and maintain lock up on decel

Do you guys have hills over there

Jokes aside, are your OS's using the reg settings or desired pressure?
This is what my OS does using desired pressure:
Coasting in tutd in 6th at around 1800rpm, running down a big ass hill, PCS TCC pressure backs right off to like 150kPa, reach the bottom and get back into the throttle to climb up the other side. That's where I see it. PD blower life.

[Tre-Cool]

my drive from home to workshop really only has 2 small hills going up, then down where it will come off throttle for decel. so I'm mostly talking about a flat ground deceleration events.

i just checked my ve ute tune (24256025). that actually has 400kpa (58psi) set for Reg offset, 1.5 for gain & does have "use regulator offset" bit set. (fdfi2650 engine)
vy ute with same os i have at 300kpa (this tune was basically taken from the ve ute years ago). (turbo 6lt) both these cars have circle d multi-plate 3400-3600 converters.

the ve ute tune is the one posted years ago to the hpt forum for all the guys having issues with TCC lockup not holding properly, I found that there was an issue in the holden calibration where it would not hold lockup in drive & the rpm would constantly bounce when on cruise etc, but in tapshift it would lock. So i flashed in a camaro os, then put my calibration back in and it stopped.

Which now when i think about it. is wierd since we now know that it's purely a bit switch that tells the os/cal to use the regulator offset. but i probably tried a bunch of shit to stop the rpm "slip surge"', so the reg gain/offset was probably changed.

Actually just checked, tis2web says it should have os 24253416 in it, but the original tune i pulled from it had os 24254909 / cal 24254908 but it had the reg gain set to 8, offset - 4096. im guessing the switch for use reg offset is off because i haven't bothered adding it into for that os. those values are highly likely to have been fcked with by previous "tuner'' when i purchased the car.

I'll log the vy when i head to my workshop later today and see what it's doing. I've not spent as much time on that trans tune besides adjusting shift speeds because it felt fine. However i've used the same settings/tricks in the 24256125 os in customer cars etc. I like the converters to keep a solid lock for decel so u keep engine braking, then drop out once you get to 3rd or below 50kph.

[hjtrbo]

Done some testing. It's staying in hard lock mode when off the pedal (good), but as soon as I touch the brake pedal it gets kicked into presumably 'on mode'.

More work to do...