Thanks for the clarification.

The HP4805a isn't an impedance meter. It's a vector voltmeter, which
measures voltage amplitude and phase. As you know, it requires external
directional couplers to measure impedance. (Or you could use current
measurements using current probes in addition to the voltage measurements.)

When you use directional couplers with the vector voltmeter -- or with a
network analyzer for that matter -- you're not measuring forward and
reflected power. You're measuring forward and reflected voltage. As I
mentioned in my last posting, power measurement doesn't give you enough
information to determine impedance.

The answer to your question is that you can't measure impedance at all
using only the voltage probes of the 4805a, and you can't measure it at
all by measuring forward and reverse power with directional couplers. If
you want to compare using voltage *and current* probes with measuring
forward and reverse *voltages* with directional couplers, the former is
probably better when impedances are considerably different than 50 ohms
resistive, the latter better in a near 50 ohm system. Making accurate
measurements with either method is much more difficult than most people
realize, and it helps to have an understanding of what you're actually
measuring. Both methods have numerous sources of possible error which
have to be understood and controlled.

Roy Lewallen, W7EL

Network

dansawyeror wrote:
Roy,

Thank you. Yes, I understood both magnitude and phase are required for
both.

The question arose after looking into the block diagram of an HP 4805a
Impedance Meter. It measures voltage and current as the basis for
determining impedance.

My work with a vector voltmeter has been through using directional
couplers and measuring forward and reflected power.

My question is: Is there an inherent accuracy benefit to one measurement
basis over the other?

Or: Are current and voltage probes inherently easier to build and more
accurate then directional couplers?

- Dan


Roy Lewallen wrote:
dansawyeror wrote:

All,

What is the best way to base impedance measurements, by voltage and
current phase or by forward and reflected power?


Measurement or calculation of forward and reverse power don't give you
enough information to determine impedance. There are an infinite
number of impedances which result in any given set of forward and
reverse powers.

Likewise, voltage and current phase don't provide enough information.
Again, there are an infinite number of impedances possible for any set
of voltage and current phases.

First I assume both methods provide the data to measure impedance.


No, neither one does. Among adequate sets of measurements which will
allow you to determine impedance are both magnitude and phase of both
voltage and current; or the ratio of voltage and current and the
difference between their phase angles; or the magnitude and angle of
the reflection coefficient.

It would seem that both suffer from the real effects of the taking
the measurement. Measuring voltage suffers from the effects of the
reflected signal, I would assume current does as well. Couplers have
a built in cross talk parameter.


It's much more difficult than most people realize to make accurate RF
measurements. Any measurements have numerous potential sources of
error. If you need accurate results, you should always measure a
number of known impedances approximately equal to the measurement
being questioned to verify that the system is sound. Measuring
antennas brings the additional problems of coupling among the antenna,
test equipment, and you; disturbances from other signals being
received by the antenna; common mode currents; and properties of
connecting feedlines.

Roy Lewallen, W7EL