• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 16 subscribers
  • Views 1446 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

ISO7241C: basic electrical insulation?

ISO7241C: ESD protection for isolator

lesheng cao
Prodigy 170 points
Part Number: ISO7241C
Other Parts Discussed in Thread: STRIKE, ISO7741, ISO7841

Hi,

My product need pass the IEC61000-4-2 ESD certification, the testing voltage level high to +-15kv for air/contact discharge, I found the isolator was always damaged by ESD strike during testing, in my design, I put two 50meg ohm resistors in series between two isolated ground( one is the isolated ground that connected to a connector that is ESD testing points, one is the digital ground that connected to mental chassis through screws), to protect isolator and also to make sure the product can pass the ESD testing, any additional components like TVS or gas discharge tube or some circuits else needed?

1, if add a external TVS, how to put this TVS, put it between connector pins and digital ground, or put it between connector pins and isolated ground?

2,the same question for gas discharge tube?

3, Can I add a gas discharge tube or an Y capacitor  between the isolated ground and digital ground and parallel with two series resistors? if add a Y capacitor, what value is best?

Thanks !

  • 0
    TI__Expert 6195 points
    Hello,
    Thanks for bringing your question to E2E.

    First off, you may be able to use the ISO7741 or ISO7841 which I believe will have better performance in this test. 15kV air is package dependent, so extra wide package of the ISO7841DWW may pass this test.

    Let me check with our IEC61000-4-2 experts to see if they have any external circuitry recommendations to help you. Please give us until next week to follow up on this.

    Best regards,
    Dan
  • 0 in reply to Dan Kisling
    Prodigy 170 points

    Dear Dan,

    Thanks for your quick response and thanks for your suggestion.

    duo to other reasons like schedule/cost so far we don't want to change to other isolator, we prefer to improve ESD performance by other way like adding some external components at the signal path.

    I am looking forward the suggestion from your IEC61000-4-2 expert.

    Regards

    Kelvin

  • 0 in reply to lesheng cao
    TI__Expert 6195 points

    Hi Kelvin, 

    I've checked in with our experts and I have a few recommendations for you:

    1. Please consider ISO7741. The same pinout and package is available. The cost difference should not be much- but the performance improvement is significant. 
    2. A gas discharge tube (GDT) will improve ESD performance. It should be connected across the GNDs of the device and be placed as close as possible to the device GND pins. 
    3. A Y-cap is expected to somewhat improve ESD performance. 
    4. If it is allowed, using a series resistor between the strike point and its GND would help limit the current into device thereby protecting it from damage. The value of the resistor needs to be found out by experiments but we have seen 10kΩ helping in meeting 15kV.
    5. TVS diode and 50MΩ resistor across the barrier are not expected to improve ESD performance. 

    Please let me know if you have any additional questions. 

    Best regards, 

    Dan

  • 0 in reply to Dan Kisling
    Prodigy 170 points
    Hello Dan,
    Thanks for your quick reply.
    I am talking about ISOLATED ground vs NON-ISOLATED ground (the non-
    isolated ground is connected to a metal chassis which may connect
    to mains/safety earth). To which one should the protection device
    connect and why?

    You suggested place a GTD between GNDs(between GNDI and GNDO), I understand that can set a current path from isolated ground to non-isolated ground for ESD current during testing; in my current design, a RC circuits that combined with 2 series resistors and 2 series Y capacitors also have the similar function, do you think your suggestion of GDT is a better solution?

    Is there any other good way to do ESD protection for input signals of isolator?
    like put a TVS to clamp the input signals to power supply and isolated ground? PLUS put a GTD between input signal and non-isolated ground?
    Or put a TVS or GTD between the input signal and non-isolated ground directly, that also set a ESD current path to non-isolated ground.

    Thanks and best regards
  • 0 in reply to lesheng cao
    TI__Genius 17100 points
    Kelvin,

    Hi! Protection devices may be connected from device pins to the local ground (Isolated or non-isolated) and across the gap from isolated to non-isolated grounds. This depends on the protective device and scheme being used.

    Gas Discharge Tubes may be connected between GNDI and GNDO (like a y-capacitor) to protect the entire device from very high ESD strikes. The GDT can react sooner to transients/surges and stand up to higher air ESD voltages than y-caps and RC circuits.

    TVS diodes protect individual pins when clamped to Vcc or GND. This protection method CAN be used together with GDTs yielding a more redundant system, but I/Os will likely not be stressed in an across-the-barrier test. Additionally, TVS diodes are not kV parts and should not be used across ground barriers. Are you willing to share a schematic demonstrating where the strike point is in the system? Have you considered Dan's suggestion to use the better-performing ISO77xx?

    If you would like to shift this conversation to emails or direct messages on E2E, please let us know! Dan is on business travel and will give delayed responses. The isolation team has his back.


    Thank you for choosing TI isolators.


    Best,
    Manuel Chavez
  • 0 in reply to Manuel Chavez
    Prodigy 170 points

    Hi,

    Thanks for your reply.

    I will follow your suggestion and then to try if it can pass ESD testing, I may update the testing results later.

  • 0 in reply to lesheng cao
    TI__Genius 17100 points
    Kelvin,

    You're welcome! Please feel free to follow up with us by replying to this post or asking a new question.


    Thank you for choosing TI,

    Manuel Chavez