On Fri, 10 Dec 2010 17:09:25 -0800, Jim Lux
wrote:

Didn't the 432 used the idea of DC power substituting for RF power to
bring the sum to a fixed temperature? The difference between the 432
and the 434 is that the thing measuring the temperature is also the RF
load in the 432. The 434 just uses the RF as a heater, and relies on
the DC powered heater and RF powered heater being matched.

Hi Jim,

I worked with a world of different methods of measuring RF power.
Thermistors, Thermocouples (the 8481A, 8482 you identify - refer to HP
Application Note 64-1, "Fundamentals of RF and Microwave Power
Measurements"), and the older technologies of Thermopile, Wollaston
wire, Bolometers (which encompass the same things as those already
mentioned), and Barreters (a variation upon, or exactly the same as
the Wollaston wire - I've seen 10mA fuses used for the same purpose),
diodes certainly (generally for peak power). I have had the occasion
to burn out more than a couple of these.

Those in a bridge configurations (many in fact as the 434 is an
example) are temperature tracking. An excellent description can be
found on page 14 of:
http://www.hpmemory.org/an/pdf/an_64-1a.pdf
for a thermistor bridge that compensates for ambient heat.

"The fundamental premise in using a thermistor for power measurements
is that the RF power absorbed by the thermistor has the same heating
effect on the thermistor as the DC power." This from the HP432A which
has a dual thermistor, dual bridge design.

Pages 18 through 20 describe how heats are separated in a thermocouple
bridge to compensate for ambient. From page 26 is discussion of
diodes.

As for accuracies: "All thermocouple and diode power sensors require a
power reference to absolute power, traceable to the manufacturer or
national standards."

A good remainder of the application note goes into the issues of
accurate determination. Pages 51 and 60 each has a table of all the
various sources of error (rarely considered outside of the Metrology
Lab, but ever present nonetheless). This was standard consideration
and the examples the author offers yields roughly 5% accurate
measurements from the best of instrumentation - an accuracy figure
that I frequently read here as commonly available from the Bird RF
meter (and I have the experience to the matter having calibrated these
meters to know that is a fantasy of the first order).

73's
Richard Clark, KB7QHC