TJA1043T Bus failure protection

[manhleduc]

Hi team,

I am trying to verify CAN Bus protection on my design. I saw in the datasheet of TJA1043T that when the CANL pin is shorted to VBAT 12V, it can detect bus failure and pull down the ERR_N pin.

I tried it, but ERR_N is not stable at 0, and it seems to still be able to communicate. Can you give me some explanation ?

Here is my schematic and my result (The green rectangle indicates the time I started shorting CANL to VBAT).

Thank you. 

 

[TomasVaverka]

Hi,

The TJA1043 can indeed detect several types of CAN bus failures, including CANL shorted to VBAT.

The bus failure flag is set if the transceiver detects a bus line short-circuit condition to VBAT, VCC or GND during four consecutive dominant-recessive cycles on pin TXD, while trying to drive the bus lines dominant.

This flag is cleared at power-on or when the transceiver re-enters Normal mode.

BRs, Tomas

[manhleduc]

Thanks for your reply, but as you see in my result, the ERR_N pin is not stable at 0; it goes up and down repeatedly. This does not occur when I short Vcc 5V to CANL. You can see below.

Can you explain or analyze the error for me ?

[TomasVaverka]

Hi,

When VBAT drops, the transceiver likely detects a power-on event, causing it to clear the bus failure flag. If CANL is still at 12V, the fault gets re-detected, pulling ERR_N low again. This cycle repeats, leading to the flickering ERR_N you observed.

So check closely if VBAT is dropping below the undervoltage threshold of the transceiver. If it does, the transceiver may be triggering a power-on reset cycle. 

Try using a low-impedance power supply to see if preventing VBAT drops stabilizes ERR_N.

BRs, Tomas

[manhleduc]

Hi, 

I have try to use another power supply and ERR_N is still ringing. But when i try to short CANL with 12 V, ERR_N is stable at 0 and it occurs with CANH too.