At 02:01 PM 6/1/2006, Keith Dutson wrote:
>Jim Brown wrote:
> >As a general rule of thumb, I use 75 ohm coax for high dipoles and 50 ohm
>coax for lower ones. The generally accepted definition of "high" is more
>than a wavelength; "low" is less than a half wavelength.
>
>Let's say you have an 80M dipole made out of 12 AWG insulated copper wire
>supported at 80 feet on a piece of PVC pipe sticking out two feet from a
>Rohn 45G tower. The ends have insulators with non-metallic line (e.g.
>Dacron) attached to reach to ground supports. Bring the ends out on
>opposite sides and tie off to supports (e.g. trees or stakes) 120 feet from
>the tower base. This would provide an inverted V with approx 56 degrees for
>each end with respect to the tower (112 degrees total apex). How would one
>go about calculating the feed point impedance?
One would model it in something like NEC, 4nec2, EZNEC, MMANA, etc. Just
start with modeling the insulated wire and modeling the tower as a single
fat wire. You can probably ignore the PVC support or end insulators. Run
the model with and without the tower "wire" and see how much difference it
makes.
I built such a model. With 120 degrees between the arms, and a 100 ft
tower next to the antenna, 1 meter away, with the classic 13/0.005 ground:
75.2 ohms impedance at resonance. The legs wound up being 19.46 meters
long. Removing the tower didn't change much.
Remove the ground, and the impedance winds up at 59.2-j12.1
Reoptimizing for resonance in free space, the impedance winds up 60.4 ohms,
and the arms are now 19.585 meters long.
Droop it a bit more, and you can get the feedpoint impedance to 50 ohms. I
seem to recall something like 100 or 110 degrees included angle gets you 50
ohms.
>Regardless of the feed point impedance, what difference would be noticed at
>the rig if 75 ohm coax is chosen as the transmission line versus 50 ohm
>coax?
>
>73, Keith NM5G
Jim
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|