• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 34 subscribers
  • Views 209 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

SN65HVD230: Proper TVS choice for protection of CAN-H and CAN-L

Knud Pedersen
Prodigy 10 points
Part Number: SN65HVD230
Other Parts Discussed in Thread: ESD2CAN24-Q1, SN65HVD234, ESDS452

Tool/software:

The maximum voltage rating of the device is plus/minus 16V, thus I find the "usual" ESD TVS choices of eg. NUP2105 or ESD2CAN24-Q1 not suitable due to their high clamping voltage above 16V.

I've chosen a TVS: ESDS452DBZRQ1, which clamps betwee 7V and 14V which I find much more sufficient.

However, in the field we're experiencing many faults, and it's always the TVS which has been damaged and typical shows 0-2ohms on one or the twpo TVS' in the device.

Was it a bad choice? Is the NUP2105 or ESD2CAN24-Q1 really suitable for protecting the transceiver? And if so; why? (When the clamping specifications is above the trasceivers maximum)

Now I'm considering changing the trasceiver to the   SN65HVD234, which features a plus/minus 36V specification, thus the NUP2105 or ESD2CAN24-Q1 would be sufficient.

  • 0
    TI__Expert 8400 points

    Hi Knud, 

    Are you able to curve trace the TVS to see if it is failing short (straight up and down)? If not, do you know what transients the diode is being exposed to? 

    ESDS452 is not a bad choice. It offers both surge and ESD protection and has good clamping performance. It'll depend on what the TVS is being exposed to and how/when these failures are occurring to determine if something needs to change. 

    Best,

    McKenzie

  • 0 in reply to McKenzie Eaker
    Prodigy 10 points

    I don't know exactly what you mean by "curve trace the TVS" - the failing TVS' are all almost shorted. We don't know what the TVS' are subjected to, and we've experienced malfunction even in an office environment. Maybe due to some boards being supplied from one PSU and others by anonther PSU and the only common connection being the CAN-H and CAN-L? 

    Another issue we're investigating is whether the connection to the boards by a PC and an USB connection (FTDI UART) may have a negativ effect.

    Regards Knud

  • 0 in reply to Knud Pedersen
    TI__Expert 8400 points

    Hi Knud, 

    We have a piece of equipment that you are able to see the DC IV curve of the diode. This is what we usually use to verify damage on the device. Another way is checking to see if the diode is leaking an excess amount of current prior to the listed breakdown. 

    Will you send over a schematic and layout if you can? Also, during the office environment, are there specific tests you are running that is causing these failures?

    Best,

    McKenzie

  • 0 in reply to McKenzie Eaker
    Prodigy 10 points

    Hi McKenzie

    We have absolutely no clue of what's destroying the TVS, but it appears to be somwwhat connected with events where the firmware of the PCB is being updated. Doing so, a PC is connected to the unpowered board, (All other boards sharing the CAN bus is still powered, and the powersupply ground is also shared). After connecting the programming tool, the power is connected and the programming is done. Hereafter the power is removed once more, and the programming tool removed as well. Then the PC is connected to the board, which feature an FTDI (USB to 3V3), and some parameters stored onto the board.

    After this process, some boards are malfuntioning, and analysis shows TVS shorted.

    Regards, Knud 

  • 0 in reply to Knud Pedersen
    TI__Expert 8400 points

    Hi Knud,

    Thank you for the background on what is happening. Are you able to see if anything else on the board is malfunctioning or failing? There is some unplugging and plugging occurring where a transient event can happen, but I wouldn't expect it to be significant enough to damage the device unless it's for a long period of time. 

    If you remove the diode from your system, does anything happen? Also, would it be possible for you to check solely the diode and see if it is failing on one channel or both and if it is in one direction (positive or negative)?

    Best,

    McKenzie

  • +1 in reply to McKenzie Eaker
    Prodigy 10 points

    We've changed the trnasceiver to SN65HVD234, which features a plus/minus 36V specification, and the TVS to ESD2CAN24-Q1.

    The usual "killer" actions have been extensively performed on several test items, and none have been destroyed - so that's what we'll go  with.

    Thanks for the elaboration on the matter.

    Regards Knud

  • 0 in reply to Knud Pedersen
    TI__Expert 8400 points

    Hi Knud, 

    Glad the ESD2CAN24-Q1 is working with the setup! Do you happen to still have the ESDS452 units that were failing? I would be interested to may be do a failure analysis on the part to see if there was any damage to the part. 

    Best,

    McKenzie

  • 0 in reply to McKenzie Eaker
    Prodigy 10 points

    Hi McKenzie

    Unfortunaetely, we don't have any malfunctioning TVS' - they've all been thrown out during the repair process.

    Regards, Knud