Gary wrote:
> By adjusting the tuner for zero reflected power on the first watt meter
that
> leaves the drake watt meter and a 1000 ohm resistive load that should have
> no reflected power.
Huh, 1000 ohms is a bad mismatch against a 50 ohm characteristic
impedance, whether its purely reactive, purely resistive, or some
combination of the two. I would expect the second meter to indicate a
large amount of reflected power under this condition. In fact a 1000
ohm load produces a VSWR of 20:1 in a 50 ohm system. This
corresponds to a voltage reflection coefficient of around 0.9. This
corresponds to a power reflection coefficient of 0.9^2 = 0.81,
which is consistent with your measurement of 100 watts forward
and 80 watts reflected. Remember the 80 watts isn't real after the
first couple of bounces of the RF signal. The reflection from the
1000 ohm load combines with the incident wave at the tuner output
such that the net steady state impedance seen by the tuner is 1000
ohms (I am assuming very short cable lengths). The tuner then
transforms the 1000 ohms into 50 ohms so that the first meter
indicates a 1:1 VSWR.
>
> With 20 watts indicated on the first watt meter the drake watt meter
showed
> about 100 watts forward and about 80 watts reflected. The difference in
> forward and reflected readings is the power delivered to the load. But
there
> should be no reactive component in the load in this setup. There should
not
> be any reflected power from the load. There should be no "apparent power".
>
> The reading of high forward and high reflected power on the meter has to
be
> meter error when not operating at the design impedance.
>
> This brings us back to the question of how do we determine how much
> reflected power we have, if any, when looking at a watt meter?
>
When it comes to amateur systems where the rise time of the RF envelope
is long compared to the settling time of the transmission line's transient
response
due to reflection/re-reflection of the RF signal, reflected power is a red
herring. It only has useful meaning with respect to the transient response
of the RF waveform. As it turns out, there is a direct mathematical
relationship
between this reflected power, VSWR, and load mismatch. Wattmeters really
indicate power delivered to the load and VSWR/reflection coefficient as
expressed in terms of forward and reflected power. The concept of reflected
power flowing back into the source is misleading in this context and should
be avoided.
> In this case with the 1000 ohm resistor there should be no reflected power
> in the circuit. This indicates a gross error in the meter reading.
>
The meter readings you observed are consistent with the amount of power
that would be initially reflected if a 100 watt signal traveling along a
long 50
ohm transmission line encountered a 1000 ohm mismatch. These readings
are also consistent with the 20:1 VSWR that is created when a 50 ohm
line is terminated with a 1000 ohm load.
73 de Mike, W4EF.............
> Gary K4FMX
|