Re: Magnetic quenching.

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: Ed Phillips <evp-at-pacbell-dot-net> 

Tesla list wrote:
 >
 > Original poster: "Jim Lux" <jimlux-at-earthlink-dot-net>


 >   I thought
 >  > the "three voltmeter method" was long forgotten, and never heard of its
 >  > use for RF.
 >
 > Featured quite prominently in several of the ARRL Antenna Compendium series
 > of books, where the oldest articles describe the graphical approach, and the
 > newer ones are more software oriented.  Once I ran across it there, I
 > started looking in the literature and found it has been used for quite a
 > while.  There are some papers from authors at NBS and elsewhere that talk
 > about error analysis in detail, and so forth.  I also ran across some papers
 > that describe it in the context of measuring the admittance matrix of an
 > arbitrary (linear) N-port with only voltage probes (a process of great
 > interest to folks trying to build field deployed HF phased arrays)
 >
 >   Actually, all it takes is a resistor and the load to be
 >  > measured - the reactance isn't needed. I worked out the arithmetic and
 >  > posted it here a while back.  The arithmetic is simple, and a graphical
 >  > construction works fine too.
 >
 > How do you do it with no reactance and a scalar measurement? You can only
 > make two independent measurements with a single resistor in series, and you
 > need to determine three things (I, V, phase, or some combination of those).
 >  >
 >  > All this applies to LINEAR impedances.  I suspect that the results with
 >  > a TC load might be quite misleading, whereas a true electrodynamic
 >  > wattmeter is quite accurate.
 >
 > Now that's an interesting aspect... The three meter technique does depend on
 > knowing the reactance of the standard component in series, and if you don't
 > know the power spectrum of the signal being measured, you can't really know

	The method I've used requires three readings which obviously takes only
a time-shared voltmeter in practice.  Measure voltage across the line,
across the series resistor, and across the unknown load.  From this you
can determine the reactance and resistance of the load, but you can't
tell whether the reactance is positive or negative. I guess that with a
known series reactance to insert in the circuit it would be possible to
resolve the ambiguity. Before pocket calculators a graphical solution
was the easiest, but not now.  All I have to do is derive the thing
again, which I'm too lazy to do.  I first ran into it in an old EE Lab
course at the University of Missouri, circa 1942/3.  A few years later I
had a teaching assistantship at Caltech and got the assignment of baby
sitting an EE lab for seniors.  We had a couple of lab exams and the
leader of the charge (Professor Maxstadt) suggested giving a problem
which "will make them think on their own".  I gave this for the problem
and no one in the lab had a clue, even after being shown the setup and
the circle diagram.  Days of long ago!

	I have never seen the RF references to which you refer. I have the
Compendiums (Compendia?) Volumes 2 and 3 and if it's buried in there
it's well enough disguised I can't find it by casual reading.  Remember
if it's in either of these and if so what is the title of the article?
Putting together books without a comprehensive index is easy for the
publisher but doesn't always help the reader!  DSon't ever remember
reading it in QST.  I'm interested in the NBC reference too.

Ed