D. Chester k4kyv at charter.net
Sat Feb 16 10:40:15 EST 2008

> From: "Gary Schafer" <garyschafer at comcast.net>

> The average VOLTAGE of a sine wave is meaningless. You don't use it to
> calculate average power. We are looking for the RMS value.
> When we are talking about PEP and the "average power over one or more
> cycles" we are talking about the RF cycles of the transmitter. Each part 
> of
> the audio envelope, even the narrowest peaks, are going to be composed of
> (contain) many RF cycles in each of those peaks. Being that all of those 
> RF
> cycles are pure sine waves the rules of .707 times the peak voltage to 
> find
> RMS voltage will always apply. The amplitude of those RF cycles will vary
> with the modulation envelope level. To measure PEP we pick a point that is
> the highest part of the modulation envelope (peak) and there will be many
> many pure RF sine waves contained there. We want to find the average power
> in those sine waves at that particular time.

The average voltage of a sinewave is zero, since there is precisely as much 
of the waveform above the baseline as below.  That is why the DC plate meter 
on a properly operating, 100% modulated modulated AM transmitter doesn't 
move with modulation, and that's also why a transformer will not pass DC.

> The RF signal is always composed of sine waves no matter if there is
> modulation or no modulation. It also doesn't matter what the wave shape of
> the modulating signal is either, tones, voice or whatever, the RF coming 
> out
> of the transmitter is still going to be sine waves. If it were not there
> would be serious harmonics generated.

The waveform of the rf signal is not the same thing as the waveform of the 
envelope pattern of a modulated waveform.  When we talk about the average 
power output, or mean power output of a transmitter, we are talking about 
the average that is integrated over several cycles of the waveform of the 
audio that modulates the carrier.  With PEP, you are selecting the one 
highest peak, which may occur infrequently, or possibly only once during the 
entire transmission, and basing your power output  reading on the average rf 
power during that peak.  This has very little to do with the apparent 
loudness of the signal, or the amount of interference it may generate.  PEP 
is most useful for measuring the output capability, AKA "headroom", of an 
amplifier, to determine the maximum output level before the amplifier goes 
into saturation and flat-tops.

The average-reading RF output meter works much like an audio VU meter.  The 
mechanical mass of the analogue meter movement integrates or "averages out" 
the rf power to the equivalent amount of DC power that would heat a 
resistive load to the same temperature during the measurement interval.

There is no such thing a an "rf power" meter, at least when we are talking 
about an instrument that most amateurs could afford.  The so-called 
"wattmeters" are really rf voltmeters calibrated to the read the level of 
power delivered when the  measured voltage is imposed across a known 
resistance.  Most rf wattmeters are calibrated to work into a 50 or 72-ohm 
load.  At any other load impedance, the  reading is erroneous.

An alternative to the rf voltmeter is an rf ammeter.  Those have been in use 
since the 20's.  They are not usually calibrated in watts, but in rf 
amperes, so you have to use Ohm's law to calculate output power.

In either case, the impedance of the load must be precisely known in order 
to get an accurate reading, and the load  must be purely resistive, not 
reactive.  Very few amateur radio antennas present a perfectly resistive 
50-ohm load, as reflected back through the feedline,  to the transmitter.

Average power output measurements work with SSB, if the ballistic 
characteristics of the meter movement are known, in the same manner as the 
VU meter.  With the VU meter, a standard ballistic characteristic was 
defined, and the manufacturers built analogue meters to those standards.  In 
the same manner, the DC input measurement works with SSB, as long as the 
ballistic characteristics of the plate current meter are known. The vast 
majority of D'arsonval meter movements are perfectly satisfactory for this 

With the l.e.d. meters that are ever more replacing the electro-mechanical 
analogue meter, it is easier to measure p.e.p. than it is average  power. 
The latter measurement would require some kind of integrating circuitry to 
meaningfully measure average, or mean voltage or current and thus power.

Don k4kyv 

More information about the AMRadio mailing list

This page last updated 24 Feb 2018.