RadioBanter - View Single Post

Cecil Moore[_2_] · October 16th 07, 12:56 PM posted to rec.radio.amateur.antenna

wrote:
Why calibration is required for VSWR meter...

Assume 100 watts in a 50 ohm transmission line
driving a 50 ohm load. The current in the line
is 1.414 amps and the voltage across the line
is 70.7 volts. That's also the current through
the load and the voltage across the load.

A typical SWR meter samples the current using
a toroidal transformer function resulting in a
sample voltage proportional to the *line current*
and equal in phase. Call this phasor voltage V(i).

The line voltage is sampled through a voltage
divider. The result is a sample voltage proportional
to the *line voltage* and equal in phase. Call this
phasor voltage V(v).

To obtain a voltage proportional to the reflected
power, the two sample voltages are subtracted, i.e.
added out of phase, so the total voltage,
V(t) = V(v) - V(i) = 0 for zero reflected power.

The only characteristic impedance and load for which
V(v) = V(i) is Z0=50 ohms. If one increases Z0,
V(v) will increase and V(i) will decrease and not
be equal in magnitude. If one decreases Z0, V(v)
will decrease and V(i) will increase and not be equal
in magnitude. Thus the SWR meter is calibrated to
50 ohms and only 50 ohms. If we switch to 75 ohm
coax and a 75 ohm load, the 50 ohm SWR meter will
indicate reflected power where none exists on the
Z0=75 ohm line.

(One can argue that the reflected power within the
50 ohm meter actually does exist inside the Z0=50
ohm meter environment but that is another discussion.)

An SWR meter is essentially a bridge with one leg
of the bridge typically set to 50 ohms. I have
designed SWR meters to read forward and reflected
power calibrated for Z0=300 ohms and 450 ohms. Those
meters are not useful when used with coax.

Continuing: To obtain a voltage proportional to the
forward power, the two sample voltages are added in
phase so the total voltage
V(t) = V(v) + V(i) = 2*V(v) = 2*V(i)
for the forward power indication in the example above.
A 100 watt calibration mark is written on the meter face
for that condition above.

If one goes through the exercise of the current being
out of phase with the voltage, one will observe that the
forward power and reflected power readings are accurate
as long as Vfor/Ifor = Vref/Iref = Z0 = 50 ohms.

V*I*cos(theta) = forward power - reflected power
--
73, Cecil http://www.w5dxp.com