Vida CEM swapping

[vtl]

Could CEM_PN_AUTODETECT being turned off on a P1 have that effect?

Sh4rp read the top of this page for more on this.

Nope.

I recently pulled off my other toy (https://github.com/vtl/volvo-ddd) because it was killing the CAN network. Didn't have time/will to figure out why, but I saw it many times that a slightest bug on the CAN bus, like crossed wires, incorrect speed or simply flimsy soldering may cause clicking solenoids, disappearing blocks and even deadlocks in firmware that can only be cured with the total power cycle (battery off, wait, battery on).

[RickHaleParker]

I recently pulled off my other toy (https://github.com/vtl/volvo-ddd) because it was killing the CAN network. Didn't have time/will to figure out why, but I saw it many times that a slightest bug on the CAN bus, like crossed wires, incorrect speed or simply flimsy soldering may cause clicking solenoids, disappearing blocks and even deadlocks in firmware that can only be cured with the total power cycle (battery off, wait, battery on).

31254903 is one of the P1 CEMs that has been confirmed so we know it will crack. If he is using the latest code, its got to be something on the hardware side.

[RickHaleParker]

Now all sort of lights were blinking, the radio would stay on and so on and you could hear a very slight clicking in the doors that was going on for the whole time of running the software. I even had to reset the ECU`s once by disconnecting the battery.

What is the battery voltage. All kinds of funky things go on when the voltage is low.
Hook up VIDA/DiCE and see if there are any CAN communication issues.

[Sh4rp]

Well today it’s raining and since I don’t have a dry place to work I might have to leave it until tomorrow. My laptop only works with a charger….

I cleaned up the wiring again and exchanged the cable in between the OBD and Transceivers with a CAT 5e that I had laying around. I read early on in the thread that vtl used something like that? Still going to look for a proper OBD cable though.

Greetings and thanks for the advices!

[vtl]

Well today it’s raining and since I don’t have a dry place to work I might have to leave it until tomorrow. My laptop only works with a charger….

I cleaned up the wiring again and exchanged the cable in between the OBD and Transceivers with a CAT 5e that I had laying around. I read early on in the thread that vtl used something like that? Still going to look for a proper OBD cable though.

Greetings and thanks for the advices!

Don't you have a common ground between car and the cracker? OBD-II pin 4 or 5 to GND of Teensy? Any two electrical circuits connected together must have a common ground. That missing ground point could alone explain all your glitches.

I think I made a mistake in my assumption about common knowledge and need to add the explicit ground connection to the schematics.

[RickHaleParker]

Any two electrical circuits connected together must have a common ground. That missing ground point could alone explain all your glitches.

K-line needs a common ground but CAN does not.

My Bosch CF160 cracker does not have a common ground with the CEM. It cracks without any problems.

CAN bits are determined by the difference between CANH and CANL. Not a common reference potential.

[vtl]

Common GND is still needed, otherwise the levels will float and may exceed maximum ratings of the transceivers. Plus there's always a risk of frying something when you plug in a device w/o ground reference. In OBD the first pins connecting are GND pins.

[RickHaleParker]

Common GND is still needed, otherwise the levels will float and may exceed maximum ratings of the transceivers. Plus there's always a risk of frying something when you plug in a device w/o ground reference. In OBD the first pins connecting are GND pins.

CAN transceivers will tolerate a “ground offset" up to a point. However, the offset can affect achievable bit rates.

Considering that a common ground cannot hurt, might help, cost next to nothing and we are not all using the same CAN transceivers .. adding a common ground not a bad idea. Gets any concerns out of the way.

Signal Ground ( Pin #5 ) not Chassis ground ( Pin #4 ) of the OBDII connector would be the better choice. Both are the same DC wise but, Signal Ground has noise isolation.

[vtl]

Signal Ground ( Pin #5 ) not Chassis ground ( Pin #4 ) of the OBDII connector would be the better choice. Both are the same DC wise but, Signal Ground has noise isolation.

Usually they both are the same ground. In my cheap OBD-II scanner that I've cannibalized for the cracker they are soldered together right at the connector.

[RickHaleParker]

Usually they both are the same ground. In my cheap OBD-II scanner that I've cannibalized for the cracker they are soldered together right at the connector.

Direct Current wise, they are both Chassis ground but the Signal ground has some AC isolation that keeps noise from other systems off the line. In other words it is quieter. Like the issue of common ground or no common ground, it is one of them things you might get away with or might not. I think using the Signal ground only for CAN is the better bet. Chassis ground would be for something like a DC to DC converter.