On Sat, 17 Dec 2005 23:38:13 -0500, "J. Mc Laughlin"
wrote:

I have dealt with approximations of the subject device. In each case,
an extremely high input impedance amplifying device is placed at the base of
the antenna that has a known voltage amplification and a 50 ohm output
impedance. Knowing that a close approximation of the open circuit voltage
is amplified by a known amount, a calibrated, tuned voltmeter (at 50 ohms)
is able to measure the size of vertically polarized E (with the usual
uncertainties). (and a bit of arithmetic)

Hi Mac,

I too, will jump in with alternatives to this short, thin rod feeding
an infinite Z. It makes for a simple specification, but when the
frequency begins to climb such is not very practical. Input Z's tend
to be dominated with strays and that "short" rod begins to become
enormous. Such artifacts of the MF era are quickly discarded.

The NIST methods (NIST technical note numbers 1309 and 1098) employ
resonant sized dipoles feeding a DC Hi R (and hence AC Hi Z load) at
the gap of the elements. By DC Hi R, the detector filter employs
50KOhm components in a balanced cascading filter that in turn feeds a
Hi R voltmeter through 250KOhm leads (carbon impregnated plastic
conductors to decouple both loading and induction).

Uncertainty, worst case, is 1dB.

Schelkunoff's algorithm is used to find the length of the dipole (no
real surprise here for halfwave length). The effective length is not
half, but rather closer to 62 - 63%.

73's
Richard Clark, KB7QHC

"Roy Lewallen" wrote in message
...
Frank wrote:

Repeating what was previously posted. The following model treats a 1
meter (perfect conductor) monopole, of 0.814 mm diameter, connected to a
perfectly conducting ground. I have applied a vertically polarized
incident E-field of 1 V/m (peak). The base of the antenna is loaded with
the antennas complex conjugate of 1.747 + j823.796. NEC2 computes the
current through the load as 0.2863 Amps (peak), which is 0.5001 V peak.
This appears to agree with Reg's program.
. . .

Possibly someone can point out if there are any errors in the following
code:

CM 1 Meter Vertical
CE
GW 1 50 0 0 1 0 0 0 0.000814
GS 0 0 1
GE 1
GN 1
EX 1 1 1 0 90 0 0 1 1 1
LD 4 1 50 50 1.747 823.796
FR 0 3 0 0 19.9 0.1
RP 0 181 1 1000 -90 90 1.00000 1.00000
EN

You've specified a plane wave of 1 V/m peak arriving in a horizontal
direction over a ground plane. This results in a field strength of 2 V/m
peak at the antenna. For more information about this, look at my postings
over the last couple of weeks on the thread "Antenna reception theory".

Roy Lewallen, W7EL

Not sure I really understand what is going on, but have been aware of your
previous postings, also on the NEC-list. What I should have said is that
the above program agrees with Reg's previous assumption -- but not with his
new program "grndwav4.exe". In any case, just to satisfy my curiosity, I
ran the following code, which is, in essence, almost identical to your
NEC-list post with 5.555.... kW input producing 1V/m peak at 1000m. The
following agrees exactly with Reg's new program.

CM Short Monopoles
CE
GW 1 50 0 0 1 0 0 0 0.000814
GW 2 50 1000 0 1 1000 0 0 0.000814
GS 0 0 1
GE 1
GN 1
EX 0 1 50 00 65698.12106 0.00000
LD 4 2 50 50 1.747 823.796
FR 0 3 0 0 19.9 0.1
RP 1 1 360 0000 0 0 1.00000 1.00000 1000
RP 0 181 1 1000 -90 45 1 1
EN

Noting the comments by others, obviously familiar with ATR measurement
techniques, this exercise with NEC is purely academic. There is no way you
could experimentally prove these results. Since I have never made
measurements on an "Open-air" test site it will be interesting to verify
Mac's assumptions, which I am sure are correct.

The confusions I have are now related to the fact that NEC results depend on
how the incident E-field is generated. I will check all previous posting by
Roy to see if I can figure out this anomaly. For some reason I have not
received any update concerning the NEC list postings.

Frank

Dear Richard:
I am obliged to you for your useful comments.
In the large TEM cells that I use to about 200 MHz (with care above
about 150 MHz) a very small probe is inserted through the roof of the cell.
That probe is used, with the sort of processing you mentioned, to provide
one verification of the E inside of the cell. (The other verification comes
from measuring the power going in and coming out of the cell, and the
internal dimensions.) The probe-scheme's success is helped by the fact that
E can go up to 200 v/m (or more) so a lot of signal is available.

Interesting things happen to some electronic equipment well before 200
v/m. A TEM cell with a spectrum analyzer can inform about what a device is
radiating.

Still, outdoors, a one meter rod over a large ground screen placed well
clear of other structures provides the means to measure (with reasonable
uncertainties) the low TOA vertical E field up to 20 or so MHz.

A scheme that I saw at NBS circa 1975 (now NIST) to measure an estimate
of total field uses orthogonal, very small doublets with diodes at their
center. They received a patent on the idea. To bring the DC to a measuring
device, lossy transmission lines were used so as to make the transmission
lines "invisible" to the field. In other words, the step up from
rod-and-screen is a big step.

In my youth, I measured VHF signals propagated by FM broadcast stations
in the "hills" of West Va. to enhance the VHF propagation models that then
existed. The purpose was to be able to predict interference to a radio
astronomy site. We used tuned dipoles elevated some standard distance that
I do not remember.

As many here have said: It is not easy to measure fields.

73 Mac N8TT

--
J. Mc Laughlin; Michigan U.S.A.
Home:
"Richard Clark" wrote in message
...
On Sat, 17 Dec 2005 23:38:13 -0500, "J. Mc Laughlin"
wrote:

I have dealt with approximations of the subject device. In each
case,
an extremely high input impedance amplifying device is placed at the base
of
the antenna that has a known voltage amplification and a 50 ohm output
impedance. Knowing that a close approximation of the open circuit
voltage
is amplified by a known amount, a calibrated, tuned voltmeter (at 50
ohms)
is able to measure the size of vertically polarized E (with the usual
uncertainties). (and a bit of arithmetic)

Hi Mac,

I too, will jump in with alternatives to this short, thin rod feeding
an infinite Z. It makes for a simple specification, but when the
frequency begins to climb such is not very practical. Input Z's tend
to be dominated with strays and that "short" rod begins to become
enormous. Such artifacts of the MF era are quickly discarded.

The NIST methods (NIST technical note numbers 1309 and 1098) employ
resonant sized dipoles feeding a DC Hi R (and hence AC Hi Z load) at
the gap of the elements. By DC Hi R, the detector filter employs
50KOhm components in a balanced cascading filter that in turn feeds a
Hi R voltmeter through 250KOhm leads (carbon impregnated plastic
conductors to decouple both loading and induction).

Uncertainty, worst case, is 1dB.

Schelkunoff's algorithm is used to find the length of the dipole (no
real surprise here for halfwave length). The effective length is not
half, but rather closer to 62 - 63%.

73's
Richard Clark, KB7QHC

Frank wrote:

Not sure I really understand what is going on, but have been aware of your
previous postings, also on the NEC-list. What I should have said is that
the above program agrees with Reg's previous assumption -- but not with his
new program "grndwav4.exe". In any case, just to satisfy my curiosity, I
ran the following code, which is, in essence, almost identical to your
NEC-list post with 5.555.... kW input producing 1V/m peak at 1000m. The
following agrees exactly with Reg's new program.

CM Short Monopoles
CE
GW 1 50 0 0 1 0 0 0 0.000814
GW 2 50 1000 0 1 1000 0 0 0.000814
GS 0 0 1
GE 1
GN 1
EX 0 1 50 00 65698.12106 0.00000
LD 4 2 50 50 1.747 823.796
FR 0 3 0 0 19.9 0.1
RP 1 1 360 0000 0 0 1.00000 1.00000 1000
RP 0 181 1 1000 -90 45 1 1
EN

Noting the comments by others, obviously familiar with ATR measurement
techniques, this exercise with NEC is purely academic. There is no way you
could experimentally prove these results. Since I have never made
measurements on an "Open-air" test site it will be interesting to verify
Mac's assumptions, which I am sure are correct.

The confusions I have are now related to the fact that NEC results depend on
how the incident E-field is generated. I will check all previous posting by
Roy to see if I can figure out this anomaly. For some reason I have not
received any update concerning the NEC list postings.


I've just now finally gotten around to posting a response to the
NEC-list. It might help clarify things for you.

The essential point is that when you specify a plane wave source, it
acts like a plane wave of the specified amplitude coming from the
specified direction. That wave interacts with the ground plane just as
any other field would. When a ground plane is specified, the result is a
field strength -- and polarization -- which isn't generally the same as
that of the original wave. You can illustrate this by specifying a plane
wave which originates at an angle of 45 degrees above the horizon, and
looking at the current induced in a short circuited vertical wire or the
base voltage of an open circuited wire (the latter simulated by putting
a high impedance load at the base). Begin with the wire vertical, then
tilt the wire so the direction of the plane wave source is broadside to
the wire, and again so the direction of the source is in line with the
wire. You'll get the same result from the last two tests, and the
induced current or voltage in those two is less (by about 1/sqrt(2))
than when the wire is vertical. This shows that the field is purely
vertically polarized (normal to the ground plane) at the location of the
wire. (I think there's actually a small horizontal component except
exactly at the ground plane surface.) It does show conclusively that the
orientation of the field isn't the same as it was when it left the
source -- otherwise the induced current or voltage would be greatest
when the wire was tilted broadside to the plane wave source and zero
when tilted in the source direction.

So the interaction of the plane wave source's field with the ground
plane alters both the amplitude and the polarization of the field. When
the source is in the horizontal direction and the ground plane is
perfect, the field strength just above the ground plane is exactly twice
the amplitude of the plane wave source. So a 1 V/m plane wave source at
zero elevation angle (90 degree zenith angle) produces 2 V/m just above
the ground plane, which induces 1 V at the base of an open circuited
electrically short 1 m vertical wire.

Roy Lewallen, W7EL