Gary Schafer wrote in message . ..
On Wed, 20 Oct 2004 15:54:12 -0700, Bill Turner
wrote:

I have a question about AM sideband power. I have always believed what
Roy says above, that the sidebands add to the total radiated power. You
can see this when watching an S-meter.

This is true.

Consider a classic plate modulated Class C AM transmitter running 200
watts input. It requires 100 watts of audio for modulation. Assume it
has plate efficiency of 75%.

With no input to the modulator, the transmitter produces 150 watts of
RF from the 200 watts input. With sine-wave audio modulation, the
input rises to 300 watts (200 watts DC plus 100 watts audio) and the
total RF output rises to 225 watts. That's not a lot - only 75 watts
more (75% of 100).

However.

Last year I took a tour of the KFI transmitter site in Southern
California and was fascinated by the 50kW transmitter. On the
transmitter's front panel was a meter calibrated in output power. It
read *steady* at 50kW, except when the operator dropped the power
momentarily to 5kW, just to show he could.

Ever since then, I've kicked myself for not asking why the power didn't
rise with modulation. The transmitter was a Harris model DX50 (IIRC)
which uses dozens of low power solid state modules which are switched on
and off digitally to produce the RF output. Could it be that as they
are switched on and off, they also are driven in such a way as to
maintain constant power? In other words, when modulation is added the
carrier power is reduced? It's the only thing that comes to mind, but
there may be another reason.

Ideas?

I'll take a stab at this. It is probably because of how the power is
being measured.

True - but not for the reasons you state.

An actual sample of output power is rarely used for
monitoring. They may have just a simple rf voltmeter measuring the
antenna line voltage, calibrated in watts. Many watt meters work in
this fashion.

When the carrier is modulated, the composite signal, carrier and side
band voltage swings up to twice the voltage and down to zero volts
with 100% positive and negative modulation. So the average voltage is
the voltage that the carrier produces itself.

No, it isn't.

The meter can't follow
the swings fast enough so it stays at the average which is equal to
the carrier. The plate current meter is in the same situation. It
stands still also with modulation for the same reason even though the
plate voltage may swing between 2 times and zero.

True, but that's not the problem.

If the modulation is not symmetrical, positive peaks greater than
negative peaks, then you will see a slight upward kick with modulation
as the average is no longer equal to the carrier voltage.

An antenna current meter on the other hand will show a definite upward
kick with modulation. The current meter is usually the thermal couple
type of meter that actually requires power for it to operate. Heating
of a resistor in it is what makes it work. Heating that thermal couple
requires power. Since power output only increases with modulation, it
never goes below carrier level, the antenna ammeter sees an average
power increase with modulation.

Not the issue.

Both an RF voltmeter and a thermocouple ammeter require some power to
work. If the antenna is a linear load, the voltage and current must
retain a linear relationship, so that increasing one increases the
other.

IOW, voltmeter vs. ammeter makes no difference. Sorry.

The reason the power meter didn't move was probably due to averaging.
The measeurement system probably averages the power over several
seconds, rather than trying to follow the audio. If they had
transmitted just an unmodulated carrier for a long-enough time, you'd
see the power output drop. But BC stations don't do that intentionally
- "dead air" is a real no-no.

73 de Jim, N2EY