Electronic – CANBUS multi-node communication fails when electing bus master

Section 6.1 of the CAN spec:

BIT ERROR: A unit that is sending a bit on the bus also monitors the bus. A BIT ERROR has to be detected at that bit time, when the bit value that is monitored is different from the bit value that is sent. An exception is the sending of a ’recessive’ bit during the stuffed bit stream of the ARBITRATION FIELD or during the ACK SLOT.

So, the node which first transmits a '1' when the other is transmitting a '0' will note a Bit Error and then signal an error as normal - by transmitting an error-flag (see Section 3.1.3) , as described formally in Section 6.2.

Informally, if that node is error-active (which should be the usual case) it will transmit an error flag of 6 dominant bits, which all other nodes will also detect (as a stuff error). This has the effect of destroying that message completely:

• no-one will receive it
• none of the transmitters will think they have successfully transmitted anything.

Each transmitter will then attempt to retransmit - depending on the precise timing of the retransmissions, one may start sufficiently before the other the gain control of the bus. Otherwise, the same sequence may happen again. (Or another higher-priority message may put them both off for a while!)

Extended answer inspired by @clabbacchio's answer below.

You mention "nasty nodes", and clabbacchio makes the valid point that if two nodes transmit at different times, each receiver needs to decide what to do with its multiple receptions.

This was demonstrated by a hack last year. The paper discusses, in the section "PSCM specifics", how an attacker can synchronise to the regular messages on the bus and play their evil message just before the one that the "good" ECU is about to send. The receiving ECU accepts the earlier message, updates its message counter and then discards the "good" messages as erroneous, because its message counter has not incremented.

The issue with the AMIS42700 chip is that when trying to broadcast a message to all nodes, it seems to keep sending the same message repeatedly

This implies that the problem could as well be on the CAN controller level. There must be at least one CAN controller present who will acknowledge the reception of the message, or otherwise the controller will keep trying 128 times. Some troubleshooting to check:

• Check for error or bus off flags from the transmitting CAN controller. Ensure that no CAN controllers are set to filter out certain identifiers. Ensure that they aren't in "listen-only" mode where they won't ack received frames. Ensure that no controller is in loopback mode (only talking with itself).

• Do you get error frames on the bus or actual data? If you look for error frames with a plain oscilloscope, they look like small, single pulses. While data frames look like a train of binary ones and zeroes.

(For an error frame, it should start with 6 bits, which are either dominant or recessive depending on what error state the sending node is in: active error state=dominant bits, passive error state=recessive bits. A node is in active error state until after 128 attempts, when it reverts to passive and can no longer mess up bus arbitration. No matter if active or passive, this part of the frame violates the "bit stuffing" rule on purpose. CAN usually only tolerates up to 5 bits of equal level before using bit stuffing. So the only time when you should encounter 6 bits with same level in a row is during the error frame.)

• What does the enable signals EN1 and EN2 look like on the master? Are they stable?

• Same problem when you disconnect the second "slave" AMIS from the bus?

• I will always suspect the most classic serial bus problem: what is tx and what is rx? (As a rule of thumb/Murphy, you always get these wrong no matter what you do :) ) It would seem that the AMIS circuit will act as a data communication "modem" in this case, or if you will just as any other plain CAN tranceiver. In that case you'll have to ensure that CAN controller Tx goes to AMIS Tx. Your schematic looks correct, but I would always double-check this.

• I doubt the terminator resistors have much to do with the problem, since termination-related problems either have no notable effect on the communication (the bus may work fine without proper termination, especially on lower baudrates) or they cause the whole bus to go down, ie they work or they don't. In case they don't, no slave would be able to send anything either.

  However, in the datasheet application example of AMIS-42700 (p4), they seem to prefer 60 Ohm, which implies that each of the individual buses should have the normal 120 Ohm termination in each end. As in: add two more 120 terminators on each bus and treat them as two buses not one.

• Probably needless to say, but signal grounds need to be the same for both buses. So if they are located on physically different locations, it is not enough to just connect the Text and Rint signals, you must also connect signal grounds.

Strangely, when disconnecting one of the CAN lines (CANH or CANL) at a time, the communication could be established between nodes. In that case though I did not check CANH and CANL at the oscilloscope.

CAN is very rugged, so even when you do such evil things to it, it may still be able to "limp home" if you are lucky.