I thought i had read your instructions correctly and I did and I changed the gear setting back to 4 and before i finished I changed it to 8 out of curiosity to see how long it took for the trans to engage, 5 sec. sorry for the confusion.
Iv'e added a crude schematic of the lever position switch with resistors.
- shift sw schem.jpg (1.17 MiB) Viewed 3388 times
What are the values of the resistors in your diagram? Have you measured to voltage signals, or are you relying on the MShift signal for that?
The datalog screen shots you posted were useful to see what was going on, but I really need the full datalog (and the exact MSQ used at the time the datalog was made) to be able to analyze your issue properly. A datalog has all sorts of clues in it that won't usually show on a screen shot.
I've added more info hopefully its what you need.
The log is from shifting from park to 4th and back.
The volt values are at the resistor connection points with a fluke.
Any idea why I didn't have this problem with the 4.146 code?
Rick,Any idea why I didn't have this problem with the 4.146 code?
*Most* aspects of the 5.1xx code are changed from the 4.1xx codes. Interrupts had to be rearranged to suit the new serial comms, and most sections of code had to be re-written to at least some degree. So it's not surprising it doesn't behave exactly the same, but it should work at least, and we'll get it there once I fully understand the problem.
I found what may be an issue with an argument passed to a function by comparing the 4.1x code with the 5.1xx code. So here is 5.101viii test code that *might* work better:
This code is totally untested (my test bench is in the process of being moved at the moment) so use great caution if you decide to try it.
Good news, the code worked flawlessly with my lever setup. (5012viii)
Thanks is a lot... short for what you've done, hopefully I wont be the only that benefits from this.
And another thanks to you and mill3833 for the msq tip every time I loaded via CAN ( msq or mshift current tune)it would get corrupted!
Again thank you Rick.
- IMG_0620.JPG (3.34 MiB) Viewed 3353 times
- (113.65 KiB) Downloaded 165 times
Thank *you* for being patient with this process. It's often silly little things in the code that I mess up when making so many changes, and it's impossible to check everything on my bench. So we all benefit when someone spots an issue and has the patience to stick with it until it's solved, as you have done. Thanks again,
Very nice rig, btw!