There are many topics where people state that VE.CAN is (galvanic) isolated and the two BMS ports not. Fine. But why exactly is that?
The ground for all ports is the same. I’ve also read someone from Victron saying to install ‘a small resistor’ on the ground wire.The thing is, no value was given. And how dangerous is it to use the BMS port for a JK-BMS?
Please note that I am not interested in personal opinions. Just facts. My BMS runs fine on the BMS port. And I have to use it, since the Victron meter only works with 250k/bits Yeah yeah. Ethernet. No thanks.
First of all the fact that CAN is a BUS system there is no galvanic isolation on the BUS side itself. GND is not used for communication but to provide a connection for shielding your communication cable. CAN interface can be isolated on the device side (for example optical isolation to a UART conversion.) Other forms of galvanic isolation is magnetic like Ethernet where communication is coupled by small transformers on both point to point sides. In most cases you would profit from grounding the shield on just the master side. This is valid for all BUS systems like RS-485/profibus/MOdbus/UniversalSerialBus/etc… A resistor on one side would reduce ground loop currents by a factor depending on impedances. Instead of a resistor try to only connect the ground on the GX device’s side.
Guys. Thanks for the reply. Please note that someone keeps telling other people, that switching off the BMS, causes ‘the current to flow over the ground wire and blow up either the Cervo and/or BMS’. No idea which current that’ll be - with the BMS turned off - and how that is supposed to work, but hey. What do I know
This was so mind boggling that I had to ask for a clear explanation. Never replied. Duh..
I only use three wires on the UTP (RJ45. connector) being 3 (GND) 7 (CAN-H) and 8 (CAN-L) with a. Cerbo powered by the battery. On a very short wire. All should be fine I guess.
