Terrible VSS Noise 4L80e

[shaodome]

I have tried the current release code and the 5.006 variant as well. It seems after a reflash, the first pull it will be OK sort of, but then after that the speed sensor goes nuts.

Below is a list of what I have tried:

1. Bypass my original wiring and ran a new shield (original was shielded too) right from the sensor through the window
2. 1k shunt between VR+ and VR- (results in no signal)
3. 10k shunt between VR+ and VR- (results in no signal)
4. 10k inline on VR+ no change
5. switch polarity from falling to rising in the software
6. switch VR+ and VR- on the sensor wiring itself
7. try various polarities again with the wiring reversed

At lower speed after a fresh relfash its fine, but once I crank the RPMs it goes wacko and never recovers. If I reflash and keep the rpms down it is still jumpy, but remains drivable at low speed. Once it's gone through a high rpm pull the car isn't drivable anymore.

[tobmag]

Shaodome.

We are many had has suffered from this issue and it seem worse for us using an 4L80E.

There are a number of threads and some different solutions to this, but before starting to modufy your GPIO do check all ground points and if you have an oscilloscope do check how clean the + feeding the GPIO and it functions is. I have a very noicy car so I installed a filter on the feedline as well.
That helped a bit but didnt solve the VSS issue completly.
At th same time do check how the signal looks from the VSS/ISS sensors with the scope, you can get a lot of noice here as well, shielded cables is a must.

The current VSS signal conditioner on the GPIO board is extremly simple to cope with the huge varaity of different sensors out there.
Unfortunaly it seems it not working good enough for the 4L80E tranny, we are many here on the forum with the same issue....
Solution so far has been to replace the excisting VSS cisrcuit with something more reliable.
There are different signal conditioner ready built using the MAX9926 circuit. It didnt work out in my car due to a very noicy invoronment since that circuit is unfortunatly very sensitive to external noice....

You have a 2 different solutions in this thread;
viewtopic.php?f=4&t=678&start=30

Or you can shearch for Jean's Dual VR signal conditioner on the internet, he uses the MAX9926 circuit.
It might work or not.....for me it didnt

BR//Tobmag

[Bernard Fife]

after a fresh relfash its fine

Shaodome,

Does cycling the power to the GPIO board work? Or do you actually have to reflash to code? Do you have anything connected over the CANbus to MShift (MS-II or MS3)? This is important, because if you have to reflash it means the code is being corrupted, which is a different thing than a spikey VSS (though they might have the same root cause).

Have you tried adjusting the VSS filter settings? The optimal settings will be different if you change the hardware (by adding resistors, for example) so you need to readjust these each time.

Lance.

[shaodome]

after a fresh relfash its fine

Shaodome,

Does cycling the power to the GPIO board work? Or do you actually have to reflash to code? Do you have anything connected over the CANbus to MShift (MS-II or MS3)? This is important, because if you have to reflash it means the code is being corrupted, which is a different thing than a spikey VSS (though they might have the same root cause).

Have you tried adjusting the VSS filter settings? The optimal settings will be different if you change the hardware (by adding resistors, for example) so you need to readjust these each time.

Lance.

Power cycling it doesn't work. You have to reflash it. I have MSPro connected over CAN BUS. I have tried the filtering a length, but when I couldn't even get a signal as shown above the VSS filtering was disabled.

After changing the actual speed sensor in the trans didn't work, I just build an LM1815 circuit and the signal is great now. Just bypassed VR1 on the GPIO and took the output of the LM1815 circuit with a 1k inline and brought it to the jumper.

[Bernard Fife]

Power cycling it doesn't work. You have to reflash it. I have MSPro connected over CAN BUS.

Shaodome,

Then MSPro is corrupting the MShift parameters via CAN (that's the only way to mess up the memory enough to require a reflash). You need to take this up with the MSPro developers to get them to sort out the issue(s). I suspect they will need TS debug comm logs to start investigating.

Lance.

[shaodome]

Power cycling it doesn't work. You have to reflash it. I have MSPro connected over CAN BUS.

Shaodome,

Then MSPro is corrupting the MShift parameters via CAN (that's the only way to mess up the memory enough to require a reflash). You need to take this up with the MSPro developers to get them to sort out the issue(s). I suspect they will need TS debug comm logs to start investigating.

Lance.

Maybe so, but since switching to the LM1815 circuit I haven't had the issue come back. Instead the car ate some bearings today