I'd like to clarify how the FCC measures power in a frequency band (both for intentional and unintentional radiators). There are devices which either frequency hop (output is a sine wave, but center frequency may jump quickly) or which have modulation that broadens the frequency band (eg FM); for an ultra-wideband example, the output could be a single Gaussian pulse or half-sine wave. A lot of the FCC regs describe field limits in terms of the field in uV/m at a frequency, or at all frequencies within a range; but obviously devices like this don't have a single operating frequency, and the rms field value depends on the time interval used for averaging as well as the bandwidth of the instrument.

An example of this type of regulation from 45 CFR 15 is here:

enter image description here

How exactly is the field strength measured for FCC purposes? Is there some specification like "the meter will measure RMS power using 100ms averages of each 1MHz wide band in 1MHz steps over the whole range of interest" (or something like that)?

If the device radiates quite a lot of total power, but spread evenly across frequencies (near the limit in uV/m at each single frequency), would it pass? What if the power comes in pulses which are much above the limit, but the rms values are very low?

I'm interested in both Part 15 and Part 18 type devices.

  • 1
    \$\begingroup\$ The FCC has discussed this question (link below). I don't have time to read it all, but they specify a peak power of 1 watt for spread spectrum transmissions, which is higher than the normal Part 15 limit for non-spread spectrum. Dig in: transition.fcc.gov/Bureaus/Engineering_Technology/Orders/1997/…. \$\endgroup\$ Commented Sep 26, 2020 at 17:56
  • \$\begingroup\$ Guess you find the answers in the international and national standards, as they not only define the limits but also how to measure em. \$\endgroup\$
    – schnedan
    Commented Sep 27, 2020 at 9:43

1 Answer 1


Testing must occur at an FCC certified lab

Although devices following the SDoC procedure are required to be tested, it is not necessary to use an FCC-recognized accredited testing laboratory (unlike the certification procedure).

47 CFR Part 15.31(a) specifies the measurement procedures that are to be used for compliance testing of Part 15 unintentional radiators. Here are some of the tests that might be required:

a. FCC/OET MP–2 – Measurement of UHF Noise Figures of TV Receivers.

b. ANSI C63.4-2014 – American National Standard for Methods of Measurement of Radio-Noise Emissions from Low-Voltage Electrical and Electronic Equipment in the Range of 9 kHz to 40 GHz.

c. ANSI C63.17-2013 – American National Standard Methods of Measurement of the Electromagnetic and Operational Compatibility of Unlicensed Personal Communications Services (UPCS) Devices.

If the device radiates quite a lot of total power, but spread evenly across frequencies (near the limit in uV/m at each single frequency), would it pass?
Source: https://www.compliancegate.com/fcc-unintentional-radiators/

As far as the power goes for precompliance (compliance should be similar), the equipment is required to to have a resolution of around 10kHz at the lower frequencies and 100kHz at higher frequencies. The equipment detects a peak value in that resolution bin, so if any value goes over the values that you listed in the table above it.

Source: Electromagnetic Compatibility Engineering

Yes you can 'spread' the power around. For example some DC/DC switching circuits use a spread frequency (so instead of the loop operating at say 100kHz, it dithers in a wider band say from 90kHz to 110kHz. Here is another example:

enter image description here
Source: https://www.analog.com/en/technical-articles/spread-spectrum-frequency-modulation-reduces-emi.html


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.