When building a product with an RF transmitter inside, one of the important things is to ensure that it complies with the CE requirements. So being able to measure the radiated power, both of the carrier and the harmonics, without paying the EMC-test lab every time, would be a great tool.

So I was experimenting with measurements on the soccer field next to our office, but I'm unsure if it is even possible to do within 2-3 dB error margin.

I can see that a "real" OATS needs a metallic ground plane beneath the transmitter and the receiver.

My idea was to place the DUT and the receiver 3 meters apart and 1 meter above ground. Then follow the steps below:

  1. Measure received power from the DUT.
  2. Substitute the DUT with a signal generator + antenna with know gain.
  3. Measure received power from substitution.
  4. Calculate DUT power as difference between 1) and 3).

Does anybody have any experience with this?

  • \$\begingroup\$ And your actual question is...? \$\endgroup\$
    – PlasmaHH
    Commented Oct 11, 2017 at 9:03
  • 1
    \$\begingroup\$ OATS is usually in some remote location far from other interference sources like radio stations and cell towers, maybe not a big problem for 3m tests, but if you have to measure at 30m it makes a difference. \$\endgroup\$ Commented Oct 11, 2017 at 9:14
  • \$\begingroup\$ you can do A/B tests of between last version and new version in a farraday room, if the numbers go down it's probably worth getting them confirmed by a certified lab. if they go up try a different approach. \$\endgroup\$ Commented Oct 11, 2017 at 9:17

1 Answer 1


Sounds good, but

Remember a random location above real or metallic ground does not give clean measurements because of reflections from the ground.

So for CE/FCC/CISPR measurements, you need to raise and lower the receiving antenna to find the strongest signal. Also rotate the DUT and its wires all around. Of course you need to do this at each frequency, but in practice you just hunt down the strongest position for the few strongest spurs. Without rotation and searching over height, your results will be dominated by random changes in the gain of the transmit antenna system (DUT and its feed wires) and reflections causing cancellation at different frequencies.

Finally, if you have a calibrated spectrum analyser and reasonably well known antenna, you don't need to calibrate with a known source. You can simply calculate the transmitted power from the measured field strength. It would be a good check of your method though.

  • \$\begingroup\$ How would I calculate the transmitted power from the measured field strength, do I use Friis equation? \$\endgroup\$
    – JakobJ
    Commented Oct 11, 2017 at 17:30
  • 1
    \$\begingroup\$ Yes, or a version of it. In free space, Power density = \$E^2/377=P_T/{4\pi r^2}\$. With a perfect reflection, E is doubled. EMC values always seem to be +- several dB \$\endgroup\$
    – tomnexus
    Commented Oct 12, 2017 at 1:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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