• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 13 subscribers
  • Views 491 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.

ISO1228: Additional Surge and ESD Protection using TVS - Request for Feedback

MAURICIO VIDAL
Intellectual 260 points
Part Number: ISO1228
Other Parts Discussed in Thread: ISO1211, TVS3300, ISO1212, TVS3301, TVS2700

Tool/software:

Hi everyone,

I'm implementing the ISO1228 in Sink Mode and Type 1 configuration using the default resistor values indicated on its data-sheet. The sensors connected to it operate at 24 VDC.

While I’m not required to meet any regulated standards, I would like to implement as much ESD and surge protection as is practically possible. I’d appreciate your help in reviewing my reasoning and confirming whether my conclusions are sound.

For reference, I’m looking at IEC 61000-4-5, and basing my observations on this Application Brief from TI, which discusses the ISO1211 rather than the ISO1228. (The ISO1211 tolerates up to 60 V on its inputs, whereas the ISO1228 handles up to 38.5 V, according to their datasheets.)

I'm using 1 kΩ MELF resistors. Based on the findings in the brief and presented on its Table 1, this configuration can tolerate surge levels up to 2.5 kV. Is it reasonable to extrapolate this result to the ISO1228?

Now please consider this schematic for my use case:

TVS3300 depicted in the schematic

As shown in the schematic above, I considered adding a unidirectional TVS3300 on each input line instead of using the diodes listed in Table 4 of the brief. However, the TVS3300 has a clamping voltage of 40 V—just slightly higher than what the ISO1228 can tolerate. More importantly, it seems the surge protection offered by the TVS3300 would not exceed the protection of 2.5kV surge already provided by the 1 kΩ resistor.

Given that, does it make sense to include the TVS3300 (unless there’s another benefit I’m missing)? And if not, would it be better to choose one of the diodes listed in Table 4 instead? Am I understanding this correctly?

(Also, it seems there might be a couple typos in the brief I have indicated in red in the image above)

One last question: assuming the TVS3300 is suitable for the input lines, could this type of TVS also be used on the AVCC line of the ISO1228? Since the same power source is used by the connected sensors, I expect any ESD or surge from them could propagate to AVCC.

I appreciate any help on this.

Thank you!

  • 0
    TI__Mastermind 31865 points

    Hello Mauricio, 

    Good questions. Please allow me another day to review each question. 

    At first glance. Realistically once a surge protector is used (other than resistors like RSURGE) most of the surge protection will come from the TVS diode or other protection you are using. A MELF resistor helps to limit the current into the input pin while the diode is clamping. Here is a really great resource showing which protection to use for various levels IEC 61000 ESD, EFT, and Surge Bus Protection for CAN Reference Design (Rev. B). 

  • 0 in reply to Andrew J
    Intellectual 260 points

    Great resource. Thank you Andrew!

  • 0 in reply to MAURICIO VIDAL
    TI__Mastermind 31865 points

    Hello Maurico, 

    Thank you for waiting. 

    MAURICIO VIDAL said:
    I'm using 1 kΩ MELF resistors. Based on the findings in the brief and presented on its Table 1, this configuration can tolerate surge levels up to 2.5 kV. Is it reasonable to extrapolate this result to the ISO1228?

    The results surge results for ISO1212 using a MELF resistor won't directly extrapolate or match to an ISO1228 since it uses a different input structure internally. The ISO1228 datasheet recommends the following values.  

    MAURICIO VIDAL said:

    As shown in the schematic above, I considered adding a unidirectional TVS3300 on each input line instead of using the diodes listed in Table 4 of the brief. However, the TVS3300 has a clamping voltage of 40 V—just slightly higher than what the ISO1228 can tolerate. More importantly, it seems the surge protection offered by the TVS3300 would not exceed the protection of 2.5kV surge already provided by the 1 kΩ resistor.

    Given that, does it make sense to include the TVS3300 (unless there’s another benefit I’m missing)? And if not, would it be better to choose one of the diodes listed in Table 4 instead? Am I understanding this correctly?

    Typically, a bidirectional diode is recommended since this will make sure the TVS diode will survive reversed wiring. A surge-proof resistor will help to protect the device while the TVS is clamping. A TVS diode will provide more surge protection than the surge resistor since the diode becomes a short during a surge event and safely dissipate the surge to ground. Whereas a surge resistor will only attenuate the impulse as a series resistance. (TVS3301 would be the bidirectional version.)

    As mentioned earlier, the surge-proof resistor should be rated to protect the device between 36V (device abs max) and 40V (TVS clamping voltage). So the power rating of the surge resistor will need to create a ~4V drop across the resistor and have a current handling that matches the surge current. 

    MAURICIO VIDAL said:
    One last question: assuming the TVS3300 is suitable for the input lines, could this type of TVS also be used on the AVCC line of the ISO1228? Since the same power source is used by the connected sensors, I expect any ESD or surge from them could propagate to AVCC.

    TVS3300 would be suitable for input lines and power lines (assuming there is no risk of reverse wiring). 

    Best,
    Andrew

  • 0 in reply to Andrew J
    Intellectual 260 points

    Hi Andrew,

    Thank you! It’s clear to me now that the TVS diodes provide an additional layer of protection by diverting surges to ground before they reach the rest of the circuit.

    Regarding your point:

    Andrew J said:
    So the power rating of the surge resistor will need to create a ~4V drop across the resistor and have a current handling that matches the surge current. 

    Could you please confirm if I understood this correctly? Given that the MELF resistor I selected is 1 kΩ at 0.25 W, it should be able to handle up to approximately 15.8 V (based on P = V²/R) before reaching its power dissipation limit. So, we can use that as a check to see whether it can support the ~4 V drop you mentioned. Does that sound right?

    Also, I must say I’m not sure how I overlooked the TVS2700/1 on my initial research. It has a lower clamping voltage, which seems ideal for this case. That said, the TVS3300/1 appears to be more cost-effective. Assuming the previous point holds, would you say that going with the TVS3300/1 is still a safe choice?

    Lastly, while reverse wiring protection isn’t a requirement, I’m sure my friends in the field would appreciate it. If I understand correctly, this would mean adding a Schottky diode to each input line. If that’s the case, would you happen to have a reference or example of how this might be done for the ISO1228?

    Thanks again for your help!

    Best regards,

    Mauricio

  • 0 in reply to MAURICIO VIDAL
    TI__Mastermind 18555 points

    Hello,

    Thanks for reaching out. 

    Our team is on TI Holiday. Please allow our team until Monday next week (4/21/2025) to reply.

    Regards,
    Aaditya V

  • +1 in reply to Aaditya V.
    TI__Mastermind 31865 points

    Hello Mauricio, 

    Thank you for waiting. 

    • Does that sound right?
      • Yes, that seems correct. 
    • "Assuming the previous point holds, would you say that going with the TVS3300/1 is still a safe choice?"
      • TVS3301 is used on our ISO1228EVM which is why we recommend them. You can create a new thread with TVS3301 in the title and a protection devices expert can help you navigate the full portfolio 
    • "Lastly, while reverse wiring protection isn’t a requirement... If I understand correctly, this will mean adding a Schottky diode to each input line. If that’s the case, would you happen to have a reference or example of how this might be done for the ISO1228?"

    Best,
    Andrew

  • 0 in reply to Andrew J
    Intellectual 260 points

    Thank you Andrew!