On Tue, 11 Oct 2005 10:43:00 +0100, Ian White G/GM3SEK
wrote:

Reg Edwards wrote:

"Ian White wrote
Yup. It measures the reflection coefficient of whatever impedance is
connected to the port on the opposite side from the transmitter.

=====================================
No, it doesn't.

It measures the MAGNITUDE of the reflection coefficient. It discards
the information which is contained in the phase angle of the
reflection coefficient.

Sorry, I left that important word out.


To be picky, in most implementations, its response is a function of
the forward or reflected power provided that Zo is real, and the
magnitude of the complex reflection coefficient can be calculated from
those measurements.

Owen
--

"Owen Duffy" wrote
To be picky, in most implementations, its response is a function of
the forward or reflected power provided that Zo is real, and the
magnitude of the complex reflection coefficient can be calculated
from
those measurements.

================================
Owen,

Forward and especially reflected power are even more imaginary than
the SWR on a non-existent 50-ohm line.

The only use for forward and reflected power is to calculate the
magnitude of the reflection coefficient. And the only use for the
reflection coefficient is to calculate the imaginary SWR. And the only
use . . . . .

I understand meter manufacturers provide graphs, which, if you don't
know how to use a pocket calculator, will do the calculations for you.
But you will still go round in circles.
----
Reg.

Reg Edwards wrote:
Forward and especially reflected power are even more imaginary than
the SWR on a non-existent 50-ohm line.

My old Heathkit HM-15 SWR meter samples the peak forward and
reflected currents. A pot is used to set the meter to full
scale using a voltage proportional to the peak forward current.

When the voltage proportional to the peak reflected current is
then switched into the meter circuit, it reads SWR from the
precalibrated scale which is linear with |rho|, i.e. at half-scale,
SWR=3 and |rho|=(3-1)/(3+1)=0.5
--
73, Cecil http://www.qsl.net/w5dxp