K6STI wrote:
>>> I tried optimizing a loaded dipole for 3.525 and 7.025 MHz. I used an RLC
>>> load to properly represent a self-resonant coil. After optimization, the
>>> load wound up out 40.5' on a 53.9' wire (one side of dipole).
I wanted to see if I could duplicate Brian's results using a NEC
engine. In a private email Brian told me, "The coil is 37.3 turns of
#12 copper, 3" diameter, 7.2" long (rounded dimensions). Use RLC
values from COIL for the load." Using those dimensions here's what
his COIL 4.62 program shows for RLC values. Note that I included 1.5"
leads to go from coil edge back to coil center.
https://i.postimg.cc/MK1G8n3h/image.png
For the antenna, I created a 5-wire model that looks like this.
https://i.postimg.cc/jSsR0Z77/image.png
Wires 2 and 4 are "sliding" 3-segment wires, with the RLC load in the
center segment of each, that can be precisely positioned rather than
just being placed at the "closest available" segment. The EZNEC loads
look like this.
https://i.postimg.cc/NFYz9PY8/image.png
The "R" value is fixed at 1.05 ohms per the COIL program but because
the "R Freq" is set to 3.525 MHz, at 7.025 MHz the actual R value sent
to the NEC engine will be 1.05 * Sqrt(7.025/3.525) = 1.48 ohms which
is what COIL shows at the higher frequency.
I used the AutoEZ optimizer to adjust a) the length of each leg and b)
the exact coil position to find minimum SWR(50) at the 80m and 40m
design frequencies. Values were very close to the 53.9' and 40.5'
that Brian found. With a 1 amp source, here's the current *magnitude*
at each design frequency.
https://i.postimg.cc/W3VnXFKT/image.png
And here are the free space azimuth patterns. Almost identical.
https://i.postimg.cc/YCgYZ2Q9/image.png
On 40m the broadside gain is down a whopping 0.21 dB!
If anyone would like to play with the AutoEZ and/or EZNEC models just
send me an offline direct email.
Dan, AC6LA
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|
