[Harc] INTERESTING RMS POWER DISCUSSION

Charles Randall Couch randy9944 at yahoo.com
Tue Jan 21 21:45:32 CST 2020


 For what it's worth, I'll disagree with one of the concluding remarks in this paper.  I would re-word the second sentence as shown below:
"The RMS power is different than the average power, and therefore isn’t the equivalent heating power. In fact, the RMS value of the power doesn’t represent anything useful."  (my emphasis)
"The ..........  In fact, the average value of power in many RF applications may not be useful."
After reading the paper I started writing about average versus RMS voltage/power and when I left the computer to pour a glass of wine, I'd written nine paragraphs - no kidding !!!  And I still hadn't made my point coherently, ha-ha-ha.
Well I've been retired for 19 years and my thought processes are questionable.  Maybe my point can be illustrated in a measurement context instead of an analytical one:
If a mixer sideband level in a signal path needs to be determined, which produces the most accurate result, measuring the average power output or the RMS value ?  Assume that two instruments can be available.
The obvious one is a conventional RF power meter that uses a bolometer to measure accurately the temperature rise of a known load when driven by an unknown RF signal. This technique isn't frequency selective, it responds to all signal power within it's bandwidth, which can be as much as 10 GHz, for example.  Random waveforms are averaged as effectively as sine waves.
So if a power meter is used (with an appropriate probe) to measure the output of a mixer it will measure both mixer sidebands and all intermodulation products.  The average measured power error will typically be at least 3 dB - double the actual power in the desired sideband.
A spectrum analyzer might also be available. The spectrum analyzer can be adjusted to select only the desired sideband. The analyzer preselector filter and resolution bandwidth filter attenuate undesired signals so that only the true power level of the selected signal is displayed, the RMS power level, not the average power level.
The same situation exists when measuring comb generators, certain types of frequency multipliers, frequency dividers, some oscillators, mixed digital/analog signals and any circuit with significant spurious and harmonic levels.
Of course one must quantify how important this is to system performance.  If a signal has only five or six measurable harmonics no greater than -30 dBC then the error between average power measurement and RMS power measurement is very small and performance degradation could be negligible.
K6CP


    On Tuesday, January 21, 2020, 12:02:32 PM CST, Mark Nelson via Harc <harc at humboldt-arc.org> wrote:  
 
 http://eznec.com/Amateur/RMS_Power.pdf
-- 
Harc mailing list
Harc at humboldt-arc.org
http://mail.humboldt-arc.org/mailman/listinfo/harc_humboldt-arc.org
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.humboldt-arc.org/pipermail/harc_reflector.humboldt-arc.org/attachments/20200122/d5ee6541/attachment.html>


More information about the Harc mailing list