3
\$\begingroup\$

I have a device which contains a USB chip which uses 19.2MHz external oscillator.

When I measured the EMI generated from the device, I got series of peaks ranging from 100MHz to 1GHz+.

The peaks have interval of around 3.2MHz. Why do I have harmonics of interval of 3.2MHz instead of 19.2MHz?

\$\endgroup\$
  • 1
    \$\begingroup\$ How are you taking these measurements? If you're not doing it in a faraday cage, you should take a baseline reading with the device off. You may have spurious emissions from other things in your environment getting into your measurements. \$\endgroup\$ – Connor Wolf Oct 19 '13 at 5:17
2
\$\begingroup\$

19.2Mhz is not used directly inside the chip - it's instead feed into internal PLL which then generate all the frequencies needed.

So ether 3.2Mhz is used internally to drive large capacitive loads (unlikely, too low for anything useful in USB3.0 chip), or it's just some resonance on the PCB or decoupling capacitors.

For further investigation I would suggest to monitor if these peaks move when you heat capacitors and crystal separately to 100C.

\$\endgroup\$
1
\$\begingroup\$

3.2MHz is exactly one sixth of 19.2MHZ and this maybe some bus activity on your circuit but it looks small fry compared to the 100M to 1GHz stuff so I wouldn't regard it as an EMI issue.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.