That is his "obvious" explanation. He should remove that from his webpage as
it is rather embarassing. Given that the magnetic field moves at the speed
of light, there will be no equipment in any hamshack that will measure the
delta between the field affecting coils spaced 1mm apart vs coils spaced
10mm apart or 1000mm apart.



I should think that many hams have things that can measure 3 ns (1000mm
light time), particularly in a repetitive system. That's one cycle at
300 MHz, or 36 degrees at 30 MHz.


Systems that rely on nulling or matching, with a variable line
stretcher, for instance, can do this fairly well.

For example of a measurement technique, say one put a LED in series with
the turn at one end, and another at the other end, along with enough DC
bias current to make sure they both stay lit, with the RF current
essentially modulating the brightness (Hmm. the LED has parasitic terms,
and you'd need a fast one.. but that's the general idea).

You could then observe the two LEDs with some system that compares the
modulated signal from the two in a nulling arrangement (for instance,
put an optical chopper wheel in front of one light path), then adjust
relative lengths of the optical paths (with a moving mirror).

Or, what about using a H field probe (i.e. something like a Rogowski
coil), fed back to a measurement system using resistive leads (377
ohms/square) that don't perturb the field.

If you have a LOT of RF power available for the test, you could use the
Faraday or Kerr effect to measure the magnetic field too.. Flint glass
has a Verdet constant of 0.11 radians/(Tesla*mm).
Rotation(radians) = V*B*l

Say your probe is 1mm long, and you can reliably measure a rotation of
0.11 radian (5-6 degrees), you'd need a field of 1 T, which is fairly high.
Biot-Savart is B=4piE-7*I/(2pi R) = 2E-7 *I/R
Say your probe is 1mm (1E-3m), to get 1T you'd need 1/2E-4 amps (5kA)..



Anyway... a sufficiently clever amateur probably does have equipment in
their shack that could be cobbled together to make this sort of
measurement, without needing exotic measurement gear.

(Mind you, having a fast sampling scope would make it easy).

Jim Lux wrote:
I should think that many hams have things that can measure 3 ns (1000mm
light time), particularly in a repetitive system. That's one cycle at
300 MHz, or 36 degrees at 30 MHz.

The referenced W8JI 3 nS "measurement" was the delay
in a 2' dia, 100 T, 10" long loading coil on 4 MHz,
i.e. 4.5 degrees.
--
73, Cecil http://www.w5dxp.com

On Thu, 29 Nov 2007 14:18:22 -0600, Cecil Moore
wrote:

I should think that many hams have things that can measure 3 ns (1000mm
light time), particularly in a repetitive system. That's one cycle at
300 MHz, or 36 degrees at 30 MHz.

The referenced W8JI 3 nS "measurement" was the delay
in a 2' dia, 100 T, 10" long loading coil on 4 MHz,
i.e. 4.5 degrees.

Jim's point is that it can be done!

Your point is that you can't do it?

Asking for a handout, and escaping work is called mooching.

Richard Clark wrote:
Cecil Moore wrote:
The referenced W8JI 3 nS "measurement" was the delay
in a 2' dia, 100 T, 10" long loading coil on 4 MHz,
i.e. 4.5 degrees.

Jim's point is that it can be done!

In that particular coil at 4 MHz - no, it cannot be done.
--
73, Cecil http://www.w5dxp.com

Cecil Moore wrote:
Richard Clark wrote:

Cecil Moore wrote:

The referenced W8JI 3 nS "measurement" was the delay
in a 2' dia, 100 T, 10" long loading coil on 4 MHz,
i.e. 4.5 degrees.


Jim's point is that it can be done!


In that particular coil at 4 MHz - no, it cannot be done.

measuring the phase shift between two sinusoidal currents at 4MHz to a
precision of hundredths of a degree is easy. HP sold a box (the 8405
vector voltmeter) that did this decades ago. Actually, they've sold two
different boxes (the 8508A ), both of which I've used. My point was
that you don't even need to go that far, and that most experimentally
oriented hams probably have stuff that can be used to make an improvised
measurement of that accuracy.

I note that the TAPR or N2PK VNAs could easily do the measurement.

The practical challenge is figuring out how to get a current probe that
doesn't perturb the measurement. Optical pickups are one approach. high
impedance probes with resistive leads are another. Both are commonly
used in antenna measurements where you want to measure the fields directly.

One could, of course, also do a near field range type measurement, but
the inversion from measurements at one set of locations to values at
another presumes that you believe Maxwell's equations, which I seem to
think might be at issue among the contenders here.


now, if you said you wanted to measure tenths of a degree at 50 GHz, I'd
say you have a real challenge in front of you

Jim Lux wrote:
Cecil Moore wrote:
In that particular coil at 4 MHz - no, it cannot be done.

measuring the phase shift between two sinusoidal currents at 4MHz to a
precision of hundredths of a degree is easy.

Jim, you misunderstood what I was trying to say and that is:
It is impossible to measure a 3 ns delay through a 2"dia,
100T, 10" long coil at 4 MHz because the delay is much longer
than 3 ns. It is closer to 30 ns.

I DID NOT say it is impossible to measure a 3 ns delay at 4 MHz!
I said it is impossible for that coil to exhibit a 3 ns delay
at 4 MHz, therefore 3 ns is not a possible measurement value.
--
73, Cecil http://www.w5dxp.com

Cecil Moore wrote:
Jim Lux wrote:

I should think that many hams have things that can measure 3 ns
(1000mm light time), particularly in a repetitive system. That's one
cycle at 300 MHz, or 36 degrees at 30 MHz.


The referenced W8JI 3 nS "measurement" was the delay
in a 2' dia, 100 T, 10" long loading coil on 4 MHz,
i.e. 4.5 degrees.


4.5 degrees is easy to measure at 4 MHz with a variety of systems.

Basic measurement theory says that the phase measurement uncertainty is

uncertainty in radians = 1/sqrt(T * Psig/No)
where T is the integration time, Psig is the signal power, and No is the
noise spectral density (W/Hz)

Let's throw in some numbers..

Psig = 1 mW (1E-3W)
No = -160 dBm/Hz (kTB noise + 14 dB)
T = 10 millisecond

uncertainty = 1/sqrt(1E-2 * 1E16) = 1 / 1E7 = 1E-7 radian

1 degree is about 0.017 radian, so I think you wouldn't have much
problem measuring the phase shift, from a physics standpoint.. all a
matter of experimental technique..

Anyway, there are LOTS of ways to do the measurement, most of which
would require only things that hams have sitting around, with a few
hours of cobbling together.

Jim Lux wrote:
4.5 degrees is easy to measure at 4 MHz with a variety of systems.

If at 4 MHz, you measured 4.5 degrees change in
the phase of *standing-wave current* on each side
of a loading coil in a standing-wave antenna system,
would you report that value as the delay through
the loading coil? One glance at the standing-wave
current equation should convince one that is an
invalid measurement technique.

For instance, the change in the phase of the standing-
wave current is ~5 degrees from feedpoint to tip in
a 90 degree long 1/4WL monopole. How can that standing-
wave current possibly be used to measure the delay
through a loading coil in the middle of that antenna?
--
73, Cecil http://www.w5dxp.com