Re: NEC-LIST: Plane wave question

Thanks to everyone who commented on my earlier message. I apologize for
taking so long to respond.

Email from Jerry Burke explained the cause of the apparent discrepancy.
I'll paraphrase what he said. When a plane wave source is specified, the
behavior is that of a plane wave coming from the specified direction. It
interacts with the ground (just as a field created by the wire would),
resulting in a field strength which can be (and usually is) different
from the plane wave source amplitude. In the example I presented, the
direction of the plane wave source was horizontal and the ground was
perfect. The resulting field strength just above the ground plane was
exactly twice the plane wave source amplitude.

Since then, I've developed another example which further illustrates
what happens. In this example (file reproduced below), I begin with a 1
meter high vertical wire, which is 0.1 wavelength at the 10 MHz
frequency, connected to a perfect ground. A high impedance (1.E8 + j0
ohm) load is placed at the base so that the open-circuit voltage can be
monitored. This voltage will be 1/2 of the field strength in V/m of the
component of the field which is parallel to the wire. (The effective
height of the wire is approximately 0.5 -- approximately because it
isn't infinitesimally short.)

The first calculation applies a 1 V/m plane wave from the horizontal
direction. From the output file, the segment 1 current is 1.00E-8 amps,
so the voltage across the load -- the vertical's open circuit base
voltage -- is 1.00 volts. This shows that the component of the field
parallel to the wire is 2.00 volts, exactly twice the plane wave source
voltage, as expected.

Then the plane wave direction is changed so it originates 45 degrees
above the horizon, but still in the +X azimuth direction. The
calculation results show that the vertical component of the field, that
is the component parallel to the wire, is now 0.7067 volts, very nearly
1/sqrt(2) the previous value.

For the third calculation, the source zenith angle remains at 45 degrees
and the wire is tilted 45 degrees toward the -X axis so it's parallel to
the field of the plane wave source -- that is, the plane wave source is
directly broadside to it. The open circuit base voltage is now found to
be 0.512 volts, just a little more than 1/sqrt(2) times the value when
the wire was vertical. This tells us that the direction of the field at
the wire isn't the same as the direction of the plane wave source field;
otherwise, the base voltage would have increased when we tilted the wire.

Finally, the wire is tilted in the other direction, so its end points
directly toward the plane wave source. The base voltage is the same as
for the previous calculation.

The last three tests show that the orientation of the plane wave at the
wire is purely vertical, that is, normal to the ground plane rather than
the same orientation as the plane wave source. And this is what should
be expected, knowing how the plane wave source is applied. It seems to
me that trying to create an arbitrary field value at some point above a
ground plane would be an impossible task. Among other problems, it might
require an impossibility like a horizontally polarized component at the
ground plane surface. So the way NEC defines and applies plane wave
sources makes sense. But it can be a trap for those of us who didn't
think carefully about it.

Roy Lewallen

    -------- Input file (complete) -------

CM Short vertical over perfect ground with
CM plane wave excitation at zenith angle of
CM 90 degrees (horizontal)
CE
GW 1,30,0.,0.,0.,0.,0.,1.,.0005
GE 1
FR 0,1,0,0,10.
GN 1
LD 4,1,1,1,1.E8,0.
EX 1,1,1,0,90.,0.,0.,0.,0.,0.
XQ
NX
CM Plane wave direction is changed to zenith angle
CM of 45 degrees
CE
GW 1,30,0.,0.,0.,0.,0.,1.,.0005
GE 1
FR 0,1,0,0,10.
GN 1
LD 4,1,1,1,1.E8,0.
EX 1,1,1,0,45.,0.,0.,0.,0.,0.
XQ
NX
CM Vertical antenna is tilted to be parallel to
CM applied plane wave
CE
GW 1,30,0.,0.,0.,-0.7071,0.,0.7071,.0005
GE 1
FR 0,1,0,0,10.
GN 1
LD 4,1,1,1,1.E8,0.
EX 1,1,1,0,45.,0.,0.,0.,0.,0.
XQ
NX
CM Vertical antenna is tilted to be normal to
CM applied plane wave
CE
GW 1,30,0.,0.,0.,0.7071,0.,0.7071,.0005
GE 1
FR 0,1,0,0,10.
GN 1
LD 4,1,1,1,1.E8,0.
EX 1,1,1,0,45.,0.,0.,0.,0.,0.
XQ
EN

-- 
The NEC-List mailing list
NEC-List_at_robomod.net
http://www.robomod.net/mailman/listinfo/nec-list