a few issues with shifting

[ashford]

these are some logs to look at i talked about in the 4r70w thread
in the log labeled 1st i am having some issues with moving the lever while the car is moving, when i do the gpio commands 1st gear. the only way to get it to not command 1st is to get on the throttle or come to a stop.
seconds 292, 297,311 and 353 show this 311 shows hitting the throttle to get speed up.

368.5-370 shows some wierdness.


the other log labeled 4to3, was what i mentioned on the 4r70w thread of the torque converter not unlocking on 4to3 down shift while decelerating. according to the log it is unlocking but it feel like someone dumping the clutch on a manual trans.
a few unwanted downshifts at 1047 1035 and 1028 the logs show no reason for this. and lastly at 905 i had some shifting stutter.

a few side notes, the 4r70w has no engine braking in 1st unless the gear selector is in first, the same for 2nd unless a 3-2 downshift has occured.
vss doesn't report less than about 14mph with settings taken during the logs, i am still working on dialing in vss settings


thanks, John

EDIT: horrible wording

[Bernard Fife]

i am having some issues with moving the lever while the car is moving

ashford,

Are you saying that the trans goes to 1st whenever you move the lever? That may be happening if there are dead spots between the gear position, so you might want to set the 'Voltage Dead Band' to zero (see: http://www.msgpio.com/manuals/mshift/V41tune.html#i ). That might not fix it, though, and I will have to think about how to handle this (it's not as simple as it might appear, unfortunately - but I am sure it can be sorted).

368.5-370 shows some wierdness.

What sort of weirdness?

In general, MShift relies on the inputs (and settings) to be correct in order to operate properly (it's the old 'garbage in/garbage out' scenario). So you need to make sure your inputs (especially the VSS and gear lever inputs) before doing anything else. There are tips on the VSS here: http://www.msgpio.com/manuals/mshift/vss.html For the settings, I can see what is in the MSQ you posted, but I can't know if these are correct for your transmission unless I know more about your application.

Lance.

[Bernard Fife]

ashford,

Here is some test code (v4.120). It blocks any shifts if the lever is in an error state. No other changes were made.

http://www.msgpio.com/manuals/mshift/fi ... 20.abs.s19
http://www.msgpio.com/manuals/mshift/fi ... t_4120.ini

In this code, if the gear is momentarily between gears, it *shouldn't* jump to first (or any other gear), it should just wait in the current gear until a valid lever position is found. However, I haven't thought through all of the implications of doing this (I will have to think about that some more).

Lance.

[ashford]

ok i will give a try after work, does this have the tps/map table like 4119?

[Bernard Fife]

Whoops, I forgot to update the revision number on the code from 4.119 to 4.120 (the only other change made to the code!).

I have put the fixed code at the same link.

Lance.

[Bernard Fife]

does this have the tps/map table like 4119?

ashford,

Yes, it is the same as 4.119 otherwise.

Lance.

[ashford]

4120 helped the lever shifting out. 1-2 worked 2-d i cant tell i had comms droupout and the gpio locked up in first when i shifted. i had to do a power cycle to bring it back. this was getting on to a highway so i didn't have a chance try again. and it never locked up on me again. after a while on the highway i forgot to test again when i slowed down, so just consider it a single bad datapoint. i think i have my vss figured out it seems to work pretty good with a mask of 30.

Again thanks for helping.

[Bernard Fife]

the gpio locked up in first when i shifted

ashford,

What do you mean when you say it "locked up". Did you lose comms, but the gear shifting, LED display, etc. kept working? If so, was your PC connected directly to the GPIO board, or was it connected to MShift via CAN through an engine controller?

Or did the comms continue but the controller stopped functioning completely (no shifting, the speedo no longer changed, etc.) If so, exactly what were the symptoms?

Or perhaps it lost all power (including the LEDs, etc)?

Lance.

[ashford]

the gpio locked up in first when i shifted

ashford,

What do you mean when you say it "locked up". Did you lose comms, but the gear shifting, LED display, etc. kept working? If so, was your PC connected directly to the GPIO board, or was it connected to MShift via CAN through an engine controller?

Or did the comms continue but the controller stopped functioning completely (no shifting, the speedo no longer changed, etc.) If so, exactly what were the symptoms?

Or perhaps it lost all power (including the LEDs, etc)?

Lance.

ts displayed no connection, unplugged usb restarted ts wont connect. trans wouldn't shift out of first(requires sol action for 1st). i cycled the key and all was well it didn't do it again. i will try to duplicate it after work again.

[ashford]

ended up i corrupted the firmware, the problem got worse where it would do it without looosing connection, it looked like it lost communication of the can- no inputs from ms3, the strang thing is i could feel it retard the timing about where it normally would but not shift. i reburned the firmware and all was good.

strang enough the vss worked perfectly when this happened. the pwm was turned off(100 on the gauge) would this indicate i have internal noise issues with pwm? i have a 2.5ohm resistor near the gpio to limit current to 3 amp at 15v and a flyback diode at the tranny.

on a side note i got the timing retard to work throug can, but i think the timing of it is wrong it retards the timeing before the shift and returns it right at the shift point, to my reasoning this should overlap the shift. i reduced shift completion delay to 150 and it made no differece, would adjusting shift pressure adjustment higher than the completion fix this?

i have a long datalog where it lost connection a few seconds in and a few time throught the log.