• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 79 subscribers
  • Views 718 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.

TSS721A: Protecting the device from EMC surges (61000-4-5)

Bart
Expert 6470 points
Part Number: TSS721A


Team, 

we would need an example schematic on how to protect TSS721A from EMC surges (61000-4-5). 

After first trial of injecting 500 V differential surge (lowest possible with the test), the device has stopped working. 1kV is the minimum tolerance for surge in our system.

Can TVSxx00 type of devices be used here? If so, can you please share example schematic.

Thanks for your help.

  • 0
    TI__Genius 17960 points

    Hi,

    The above picture visualizes the principle of protection with discrete components. The drawing is pretty old, thus probably more modern parts for the protection diodes can be selected. Of course also protection diodes versus GND/Supply could be applied. Beside this of course the layout is one decisive component.

    Furthermore I would double check, whether it is the TSS721, or the MSP430, which stopped working after the surge.

    Best regards

    Peter

  • 0 in reply to Peter Spevak
    TI__Expert 6470 points

    Hi Peter,

    I can confirm the MSP is working fine after the surge, it is the TSS721A that got damaged.

    The schematic you showed doesn't include TVS diodes protection, do we have any examples to show that? To meet  61000-4-5 type of tests, the resistors only won't be enough?

    Thank you.

  • 0 in reply to Bart
    TI__Genius 17960 points

    Hi,

    there is a suppression diode for the differential voltage between BUSL1 and BUSL2. As stated suppression diodes could also be applied towards supply and GND in similar way. The resistors are also meant to take away a significant portion of potential incoming bursts. Whether they are sufficient in a certain case, and what the optimum value would be, depends on many design specific parameters, resulting from the HW design, which is mainly the layout topology. Thus it is not possible to predict, what would be necessary.

    The one approach could be going with a layout with the maximum components foreseen, and then based on the actual test results try to remove certain components, and after passing the test successfully with the minimum components either go with "do not place" approach or generate a dedicated final HW.

    Best regards

    Peter

**Attention** This is a public forum