• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 2 answers
  • Subscribers 16 subscribers
  • Views 150 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.

ISOW7742: Recommendation ferrite beads EVM and Datasheet differ, feedback requested

Frank Rothoff
Expert 6130 points
Part Number: ISOW7742

Tool/software:

Hi ISO team,

My customer has below question, could you comment on this.

I have a (very specific) question regarding the ISOW7742. The datasheet recommends usage of 330 ohm Ferrite Beads in the supply rails (and GND) on the non-isolated side. In the EVM and its documentation these are replaced with 1500 ohm Ferrite Beads. Both documents use 330 ohm Ferrite Beads on the isolated side. The difference on the non-isolated side makes me curious. Is the higher value the recommended one? Should we also go for this value? This could dampen EMC emissions. Is there a danger of ringing?

Another difference between the datasheet and the documentation of the EVM is that VIO and VISOIN use 0.1 µF instead of 0.01µF capacitors. Is that for stability or for EMC reasons?

  • 0
    Guru 270970 points

    What ferrite bead to choose depends on how much noise at what frequencies you expect to see. Isolators are likely to be used in environments with more noise than usual (and the internal DC/DC converter also generates some noise), but 330 Ω is just a general recommendation. (And note that differents beads have different impedances at frequencies other than 100 MHz.)

    Ringing is unlikely, but this depends very much on the characteristics of your board. Finding a correct bead might require testing and experiments. (And there is likely to be a range of values that are OK.)

    The exact value of a decoupling capacitor is not very important; more important is a low-impedance connection to the chip, i.e., short traces and a small (low-inductance) package.

  • 0
    TI__Guru 53095 points

    Hi Frank,

    Thanks for reaching out.

    Frank Rothoff said:
    Is the higher value the recommended one? Should we also go for this value? This could dampen EMC emissions. Is there a danger of ringing?

    As Clemens stated, the selection of ferrite beads depends on the amount of noise observed and the frequencies at which noise is observed. 330Ω of impedance is the minimum impedance FB we recommend and a higher value of impedance is usually better although the distribution of impedance across frequency also makes a difference. If customer is observing emissions at 400MHz then high impedance at this frequency is important. The value 330Ω specified in FB datasheet refers to the impedance at 100MHz. The impedance vs frequency plots provided in datasheet will help you understand the impedance at frequencies of your interest.

    We don't expect any ringing issues due to FB.

    Frank Rothoff said:
    Another difference between the datasheet and the documentation of the EVM is that VIO and VISOIN use 0.1 µF instead of 0.01µF capacitors. Is that for stability or for EMC reasons?

    Since VIO and VISOIN power the digital isolator part of the device, 0.1µF should be sufficient to provide the device with the needed transient currents. Alternatively, having 1µF + 0.01µF should also be sufficient. Both these options minimize the ripple at the input thereby avoid EMC issues. Thanks.


    Regards,
    Koteshwar Rao