(no subject)

>From: DCHATTERJEE_at_KUHUB.CC.UKANS.EDU :
> Probably this issue has been addressed before on this group. I have
> documentation and code for NEC (January 1981).
>
> With this, can I model biconical antennas ? Is there any frequency
> limit ??

You certainly can model biconical antennas. The frequency limit depends
on the number of segments used. The easiest way is to enter one wire
extending from the origin at the cone half angle from the z axis. Then
rotate this wire perhaps 20 times using GR0,20 to get a wire model of
the cone. You need enough wires to make it look like a solid cone, but
~20 seems to be enough for cones that are not too long. Circular wires
can be added to make it look more "solid". There is a tradeoff between
longitudinal and circular wires that someone studied way back in
designing EMP simulators, but I do not have the reference (AFWL notes?).
Anyway, it is easier to just use longitudinal wires if no interactions
upset the symmetry. When the wires all end on the ground plane for a
monocone, or at the apex of a bicone, you need to put a voltage source
at the base of each wire (and remember to use GE1, GN1 for the monocone).
Also you need to sum the currents to get admittance or divide the impedance
of one source by the number of wires. An advantage of this method is
that it takes advantage of the symmetry; but make sure that the
dimension limit for symmetry is set large enough.

If you need a single source you can raise the cone up with a command
like GM0,0,0.,0.,0.,0.,0.,.1, and insert a wire from the ground to
the cone apex at a height of 0.1. For a bicone: raise it, reflect it
and add the source wire. However, it may be necessary to taper the
lengths of the segments on the cone wires so that the lengths at the
junction are about equal to the height of the cone. With this method
you have a single source, but lose the advantage of symmetry in filling
and factoring the matrix.

Since the cone must have finite length, rounding the top will reduce the
reflections. Data for an approximately 79 ohm cone with round top
follows. The tapered wire uses 4 segments, starting with a segment
length of 0.01 and radius of 0.0005 and the last (4th) segment has
radius of 0.001. This a fairly rapid taper (to save on segments) but it
seems to work OK. This form of the GC command is for NEC-4, and will be
different for NEC-2 or 3. You could also taper it yourself, with
individual GW commands.

---------------------------------------------------------
CEMonocone, half-angle = 30 deg., 79 ohms
GW1,4,0.,0.,0.,.120625,0.,.208929,
GC1,0,0.,.0005,.001,.01,0.,
GW1,10,.120625,0.,.208929,.98,0.,1.7,.001,
GW1,1,.98,0.,1.7,1.,0.,1.8,.001,
GW1,1,1.,0.,1.8,1.02,0.,1.9,.001,
GW1,1,1.02,0.,1.9,1.,0.,2.1,.001,
GW1,1,1.,0.,2.1,.96,0.,2.3,.001,
GW1,1,.96,0.,2.3,.86,0.,2.5,.001,
GW1,2,.86,0.,2.5,.7,0.,2.7,.001,
GW1,2,.7,0.,2.7,.42,0.,2.9,.001,
GW1,2,.42,0.,2.9,0.,0.,3.,.001,
GR1,20,
GM0,0,0.,0.,0.,0.,0.,.02,
GW61,2,0.,0.,0.,0.,0.,.02,.001,
GS0,0,.0133333,
GE1,
GN1,
EX0,61,1,0,1.,
FR0,10,0,0,500.,500.
XQ
EN
-------------------------------------------

Jerry Burke
LLNL
Received on Thu Oct 12 1995 - 06:36:00 EDT