• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 14 subscribers
  • Views 410 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.

ISO1640: Using ISO1640 from a single I2C bus on the controler to isolate i2c salve doing EMC test

Ni Jiale
Prodigy 10 points
Part Number: ISO1640
Other Parts Discussed in Thread: TCA9517

Tool/software:

Hi everyone，

I'm working with an MCU(3.3V) as an I2C mater, and I need to isolate a segment of the bus to interact with:

1. one sensor(powered at 1.8V) for I2C signals

The background of the circuit design is like this, the sensor is undergoing EMS/EMC test, I need to read the data through I2C communication to monitor if the sensor is working properly.

so I need an isolation to isolate the RF interference.

now the problem is that the ISO1640 can't support the logic level of 1.8V; do you guys have any good ideas?

Thank you in advance for your help.

  • +1
    TI__Expert 7725 points

    Hi Jiale,

    Thanks for reaching out.

    Yes 1.8V is not supported for ISO1640, hence I have following suggestions:

    • You can use the ISO1640 for Isolating the the Sensor and Master.
    • In order to step down to 1.8V, you need to use a bidirectional I2C buffer(like TCA980x, TCA9517) between ISO1640 Side2 and the Sensor.

    This will enable you to operate the sensor at 1.8V and also Isolate the two buses.

    Regards
    Varun

  • 0 in reply to Varun Kumar
    Prodigy 10 points

    Hi Varun,

    Thanks for quickly reply.

    I've considered this approach, but I'm concerned about RF interference with the sensor I2C signal affecting the I2C buffer.

    So you're suggesting something like the diagram below?

  • 0 in reply to Ni Jiale
    TI__Expert 7725 points

    Hi Jiale,

    Ni Jiale said:
    I've considered this approach, but I'm concerned about RF interference with the sensor I2C signal affecting the I2C buffer.

    Good to know you already had this in mind. The concern is valid but I2C buffers are generally robust to noise interferences. What is the origin of this RF noise exactly here or is it some RF immunity test that you're targeting?

    The Above Diagram will not work since Side1 of ISO1640 (Right Most in the Diagram) is connected to 1.8V which is not supported by ISO1640.

    Ground Connections are also wrong at multiple cases considering you are using Isolation to separate 2 grounds after MCU.

    Regards
    Varun

  • 0 in reply to Varun Kumar
    Prodigy 10 points

    Hi Varun,

    The RF noise is caused by the EM immunity test I am doing, and I need I2C communication during this test.
    Also I would like to explain: the ISO1640 is powered by 3.3V on both sides, and then the side2 side is connected to the slave via an i2c buffer; I don't connect the ground of the slave to the master because I am considering RF interference on the slave side.

  • 0 in reply to Ni Jiale
    TI__Expert 7725 points

    Hi Jiale,

    Thanks for the additional information. 

    For the RF Immunity details it would be best if you can create a new post with the I2C buffer part number so that the right product experts can answer your questions for that. My Suggestion would as long as the supply decaps and GND connections are proper for the buffer, it should not be an issue but I recommend to confirm from the product experts once.

    Ni Jiale said:
    Also I would like to explain: the ISO1640 is powered by 3.3V on both sides, and then the side2 side is connected to the slave via an i2c buffer; I don't connect the ground of the slave to the master because I am considering RF interference on the slave side.

    Both Sides of ISO1640 being 3.3V is absolutely fine but in order to isolate slave and master grounds, there should be 2 different 3.3V supply on each side for VCC1 and VCC2 of ISO1640 which I saw missing in the above diagram.

    Regards
    Varun

  • 0 in reply to Varun Kumar
    Prodigy 10 points

    Hi Varun,

    Thanks for your feedback and suggestions.

    I’ll go ahead and create a new post for the I2C buffer part number to get more targeted advice from the product experts, as you recommended.

    Regarding the ISO1640, thanks for pointing that out — you’re absolutely right. In our current setup, VCC1 and VCC2 are both supplied from the same 3.3V source, which means we’re not achieving full galvanic isolation between the master and slave sides. We will modify the design to use two isolated 3.3V supplies to properly separate the grounds while maintaining the functionality of the ISO1640.

    Appreciate your help!