This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

SN55HVD251: Bus-fault protection vs. short-circuit/thermal shutdown

Part Number: SN55HVD251
Other Parts Discussed in Thread: TCAN1042H

Hello,

I want to use SN55HVD251 in an application where the CAN signals will be bundled in a connector with +25V and GND.  This is a high-cycle connector in a very wet/dirty environment, so the design must assume that steady-state connections between +25V and either CAN signal, or GND and either CAN signal, or the CAN signals to each-other, are likely.  The hardware must not suffer damage in these situations (of course the system need not continue to operate while the short is present...)

The SN55HVD251 datasheets specifies +/- 36V bus fault protection, which I assumed meant that it tolerated a steady-state short between CANH/CANL and +/-36V without damage.  However, section 8.3.5 states:

"The SNx5HVD251 has a thermal shutdown feature that turns off the driver outputs when the junction temperature nears 165°C. This shutdown prevents catastrophic failure from bus shorts, but does not protect the circuit from possible damage. The user should strive to maintain recommended operating conditions and not exceed absolute-maximum ratings at all times. If an SNx5HVD251 is subjected to many, or long-duration faults that can put the device into thermal shutdown, it should be replaced."

What am I misunderstanding?  The thermal shutdown section seems to indicate that the CANH/CANL pins can not withstand a +/-36V short because the driver will enter thermal shutdown and incur damage over time.  I suspect I am misunderstanding the definition of one of these terms.

Thank you!

  • Hi Samuel,

    This is a good question. You are correct that the bus pins for this device can withstand a prolonged short to 25V with no damage as it's below the bus fault rating. This is even true when the device is attempting to drive the state of these pins. This is possible because the output current on the CAN pins is internally limited in a short circuit case (see Short-circuit steady-state output current specification). This means that in a case where CANL is shorted to battery level (12V in datasheet), the current will be limited to 200mA. This prevents the transceiver from quickly burning up, but this is still a relatively high current and will eventually cause the device to overheat. This is where the thermal shutdown feature comes in and turns off the device completely before overheating causes permanent damage. 

    Keep in mind that this short circuit current only flows when the transceiver is actively trying to drive the state of the bus (in dominant state or logic 0). When the transceiver is in a recessive state (logic 1) there is only a weak bias applied to the bus and only a few mA will flow even in a short circuit case. This means that the heating only occurs during active communication on a shorted bus. Once the system recognizes that communication is no longer possible, it will stop attempting to drive the bus and leave the transceiver in a recessive state. This way the transceiver is unlikely to overheat as it only spends a short amount of time in the high-current fault condition before the system goes idle. 

    The result is that in typical systems, SN55HVD251 will be able to withstand prolonged bus faults up to ±36V without any damage. 

    If the slope control function on this device is not necessary, you may also consider a newer transceiver such as TCAN1042H with ±70V bus fault protection and high ESD ratings for increased reliability for the high-cycle connections. You may also consider TCAN1044A with an optimized cost structure while still providing ±58V bus fault protection. Both of these devices have lower current limits for their short-circuit conditions and similar thermal shutdown behaviors. 

    Let me know if you have any more questions or would like to know more about the recommended devices. 

    Regards,
    Eric Schott

  • Wonderful, thank you for your response.  I wish I could consider a newer transceiver -- TCAN1042 is my preference -- but supply chain issues are limiting me right now.  I'll be redesigning this system to use modern parts when the supply chain catches up.  Until then, it's a SN55HVD251 for me