My attraction to the sleeved dipole was that the 1/2" hardline I have will take a 1/2" copper pipe coupling and fit snug, so I was going to use a 1/2" to 3/4" adapter and use 3/4" copper pipe for the bottom part.
If you do that, make sure you use some sort of anti-corrosion paste on the joint.
Copper / aluminum joints are particularly prone to corrosion due to dissimilar metals.
That's why using aluminum for the antenna might be a good idea.
Will this not provide enough space? Do I really need something like a 2" pipe at the bottom with that much space between the hardline and the bottom section?
You really don't
need that much space, but the common mode current will
increase as you make the pipe smaller, but there isn't a set point where it stops "working".
(Well, except where it arcs over, since the end of the element is a high voltage
point and there there is a high potential gradient across the opening.)
I measured a commercial version that I have out in the barn - it uses a 3" diameter
lower element to pass the support pipe (with the coax inside), and the clearance
between the element and the pipe is 5/8". So if the inside diameter of the sleeve
is 1" larger than the outside of the coax, that's probably a reasonable design goal.
One other factor is the thickness of the outer insulation on the coax: this is not
designed for good RF performance, and the larger the percentage of the gap between
the shield and the sleeve that is filled with insulation, the higher the losses will
be in the insulation. This is why it is good to use a spacer to keep the coax
centered in the sleeve rather than laying against one side. (Pipe insulation or a
spiral of foam weatherstripping down the coax might be easy ways to do this.)
How much of a difference does this make? I'd guess no more than 1dB when the
shield is right against the coax. But it is another reason to provide some gap
between the sleeve and the coax if you can.
I was wondering about the common mode currents. Will I need a choke balun? That would not be easy/possibly with the hardline so I may not go with the sleeved dipole approach.
A choke balun formed by coiling the coax doesn't work well at VHF/UHF due to stray
capacitance. Instead I'd suggest using snap-on ferrite beads around the coax, preferably
two sets spaced 1/4 apart for maximum decoupling.
While a choke will help, generally it is better to minimize common mode currents by design
where possible rather than relying on the choke to do all the work.
With a sleeved dipole, are top and bottom elements equal in length or what is the rule for that?
Generally the fatter element would be somewhat shorter for resonance. But in this
case it doesn't matter that much.
I ran a model of a dipole using different diameter conductors and adjusted the length
for minimum SWR at 146 MHz on 75 ohm coax:
1/4" 38 1/8"
1/2" 37 5/8"
1" 37 1/4"
2" 37"
You could use these to calculate the length of each half of the antenna. Or not -
a dipole will work when fed off-center, and it may be easier to make the sleeve
a convenient length then make all adjustments to the center conductor. And the
exact tuning isn't too critical, as even the 1/4" dipole has a 1.5 : 1 SWR bandwidth
of 10 MHz, and the fatter ones will be even wider.
Is there a better antenna recommendation than the sleeved dipole?
Most end-fed vertical antennas are susceptible to common mode current. This
includes the J-pole (some designs are especially bad in this regard) and the
ground plane. But all of these still get used, even in commercial service (except
the J-pole), so it isn't that they won't work.
I happen to like the ground plane because they are easy to build. I use a set of
dimensions published by Woody Smith W6BCX (inventor of the half square and
bobtail curtain) that are supposed to reduce common mode currents: 18" for
the radiator and 24" for the radials. Whether or not this makes much difference
I can't say for sure, but the shorter radiator length allows me to use half a piece
of brazing rod. I ran a model on the computer that included the feedline and
there was about a 1dB variation in gain as the coax length was varied, so it still
has some common mode current. Adding some ferrite chokes would help further,
or a second set of radials 1/4 wave below the first (which is a common commercial
practice.)
But you can probably make some small variations to your design that will reduce
the common mode currents to a similar extent. For example, if you use the 3/4"
pipe as you were planning for the lower element, but make it no more than 9" to
12" long and then splay it out somehow to shape it like radials the rest of the
length so it is spaced further from the coax at the open end. This might be done
either by splitting the pipe and bending the pieces outwards, or by soldering wire
radials to the end of the pipe. Adding a second set of radials below it would
also help decouple the feedline - because the outer conductor is aluminum, I'd
recommend using aluminum strips or wires held in place with a hose clamp. With
the increased spacing at the bottom of the sleeve, just adding some ferrites
may be sufficient.
While it is good to know the design considerations, that shouldn't prevent you
from experimenting with simpler versions. If you are hoisting the antenna up on
a rope halyard, put up an initial version and find a weak repeater to use as a
signal strength reference. Then take it down, add some ferrite beads, and see
if that makes a difference. Common mode current will depend on the length
of the feedline and how it is placed relative to the ground and the antenna, so
it tends to make performance less predictable. If you have to climb the tree
and mount the antenna permanently, I'd worry a lot more about getting it
right the first time.
