AA80E 8-speed

[Jim Blackwood]

I ran some more datalogs (below) with the same results as before.

It's still running in 3rd gear.
Paddle upshift still doesn't work.

Jim

[Bernard Fife]

Jim, The msq you posted doesn't have the mshift variables in it.

[Jim Blackwood]

Thanks Lance, I just loaded the S19 file to the controller, accessed it with TS, checked the upshift indicator, shut down everything, and then did the test runs.

Did the new upload drop out the variables when it was loaded? I'm not the most proficient with this and forget how it works. Does this mean I have to go back and re-enter the Mshift variables? (I'm assuming, yes, that was what you told me to do. I'm just unclear on exactly what these Mshift variables are that I'm supposed to re-enter, and where they are located.)

I'm guessing I should find these variables by opening TS with an earlier msq and search for anything that is different, then copy them to the new tune?

BTW, I thought of a way to test the function of the shift trigger solenoid. I should be able to cut the wire that comes out of the controller and wire in a hand held push button switch, then press the button when I shift into drive. If that starts the car in 1st gear instead of 3rd I should then be able to successively use button presses to upshift through the gear range and back down again, since presumably the controller will be commanding shifts in automatic fashion. What do you think? As soon as I get these Mshift variables sorted I should be able to try that.

Thanks for the assistance,

Jim

[Bernard Fife]

Jim, Plug straight into the GPIO serial port to grab the relevant parameters into an msq.

[Jim Blackwood]

Thanks, I've been tied up with windows on the house but hope to sort it out in the next day or two. When I open TS I can go to the CAN1 device and it looks like all the MegaShift tuning stuff is there as well, so would that work the same?

Looks like I will have to go through everything item by item, line by line and make sure it matches the msq from before the s19 file was revised if I understand correctly.

So what do you think of my workaround with the pushbutton switch? Sound feasible to you? It's the only alternative I can think of for getting the transmission valve body out of "Limp" mode aside from switching the AD4 and PT6 ports and setting up a trigger pulse on PT6.

Jim

[Bernard Fife]

Jim, You know much more about this trans and your circumstances than I do, so you are in a far better position to judge the feasibility of your workaround.

[Jim Blackwood]

Lance, I've got to admit I've been wondering for awhile now why you won't just switch the ports. (AD4 and PT6)

We talked about this way back before I built the controller, and I thought you said you could do that. Is it that difficult?

Jim

[Bernard Fife]

Jim, We can't just switch the ports without screwing it up for everyone else who wants it to stay the same. So all the places in the code where either of the ports are used need to be sorted by adding a switching option; and there are many, many places they are used in the code because of all the options already allowed for their use.

[Jim Blackwood]

OK, I can see how that makes sense. But this much I am sure of, the danged thing is never coming out of Limp mode without a Shift Trigger. Which makes it a very large, heavy, expensive, time consuming and complicated 1 speed transmission.

I realize this means there has to be a code variant just for the AA80E specifically. But aren't there specific variants for other transmissions too? I also realize that 5 years ago you probably were a lot more interested in pursuing this that you are now. I may be the only pioneer right now, but my work is freely available to anyone else who follows along, and second time around the build just isn't that much different from the norm. I've been careful to blaze a good trail and leave a record that should be easy to follow.

If it wasn't in essence closed source I might have other alternatives, my son does programming and could look at it. But with an entry barrier of $1000 or more just to get the software to look at the code that's a mite stiff for a single application. Even as a student he can't afford the software. He says the proprietary compiler is the problem, otherwise he could help. Wasn't all this Mega stuff supposed to be open source?

Jim

[Bernard Fife]

Jim,

No, there are no code variants, it's all one code with options.

If you want to write your own code, your best bet is to start with this code: http://www.msgpio.com/manuals/iocode/Te ... t_1101.zip

This code is small enough that you can use the free special edition of codewarrior from this page: http://cache.freescale.com/lgfiles/devs ... =Downloads (you have to register)

You could also start with the 1.100 MShift code here: http://www.msgpio.com/manuals/mshift/GP ... oj1100.zip as this is small enough to compile 'as is' on the special edition of CW but already has numerous transmission functions coded.

You will also want the MC9S12C Family Reference Manual, it contains much of the information you will need when programming the GPIO's processor (things like register definitions, etc.). You can get the reference manual here: http://www.msgpio.com/manuals/iocode/MC9S12C128V1.zip

If you want to work with the latest code, here is 5.201:

MShift_V5201_Proj.zip

5.201 code project for codewarrior

(1.7 MiB) Downloaded 1023 times

However, you will have to strip a fair bit of the code out (there ought to be lots of code in there that you don't need for your transmission control) to be able to compile this with the special edition of CW and its 32K limit, and making deletions from the code is not a trivial task.