Possible - Small transmitting loop made from Aluminium Foil
Richard Burden
Hi all,
I've been thinking of making a small transmitting loop for some time for
use on 80m. The biggest obstacle appears to be the high voltage capacitor
required. Does anyone have any comments on the following possibility?
What would happen if instead of copper tubing, a wooden frame (or PVC pipe)
was made on which aluminous foil was glued / stapled etc. Instead of a
vacum variable capacitor or trombone, the capacitor is made by fixing the
ends of the foil to two swinging plates. To vary the capacitance the plates
are moved. The following diagram may assist in communicating the idea
(square loop shown for ease)
+---------------------+ +------------------------+
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦
¦ ^ ^ ¦
¦ ! ! ¦
¦ These are the capacitor plates ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
+--------------------------------------------------+
The whole antennae is made from a single piece of aluminous foil. Being a
wide conductor perhaps the skin losses are negligible. To couple the
feedline to the antennae a coupling loop would avoid the need for
electrical connections to the foil.
Has anyone tried such an arrangement or have any comments to make?
-----------------------
Regards,
Richard VK6FKB
"Nothing better than home-brew, except perhaps an absence of Spam"
Steve Harper
I would also be very interested in seeing some discussion of this idea.
I happen to have a couple hundred feet of 7 inch wide mylar-backed
copper foil and would like to try this, too. The idea I had for the
capacitor was very similar, except to keep it in the circumference of
the loop, rather than bring it down in the middle of the loop, thusly:
-------------------
| ------------------
| |
| |
| |
| |
| |
| |
| |
---------------------
Mostly, I have questions rather than answers:
1. Would a trombone capacitor, or other such structure in the middle of
the loop disturb the pattern or cause other variations in performance?
2. Does anyone remember enough EM theory to compute the RF resistance of
thin band of conductor as opposed to a round pipe? Clearly there will
be some current concentration near the edges, but if width/thickness
ratio is faily large, as in the aluminum foil or copper foil case, it
seems this should be fairly minor, and might be a cheap way to greatly
reduce resistive loss. Skin depth is an issue here, too. My copper is
only about .001 inch thick and I think skin depth at 3.5 MHz is about
.0017 inch. Are this correct? The aluminum foil is probably thicker,
but skin depth is probably more, too, since it's not as conductive as
copper. I've considered using two layers of my copper foil to reduce
the RF resistance.
3. One other effect of using a thin band of conductor rather than a
round conductor is a substantial reduction in the overall inductance of
the loop. The total inductance is apparently a complicated function of
the loop geometry and the conductor cross-section. Again, I don't
remember enough EM to compute the theoretical inductance change. I did
a sensitivity analysis using an equation from Ted Hart on circular
conductors and found a 25% reduction in inductance just going from 1
inch copper pipe to 3 inch copper pipe. The flat conductor will be even
more, I suspect, and depend on the width/thickness ratio. This is bad
since it proportionally increases the size of the capacitor for a given
frequency.
4. The physical dimensions of the capacitor quickly become daunting. I
did a rough calculation, again using Hart's equations. For a 10 foot
diameter octagon loop with my copper foil (a rough quess at the
inductance), to resonate at 3.7 MHz required more than 2 square feet of
capacitor at a spacing of 0.25 inch with air dielectric. If anyone has
ideas how to construct and keep such a capacitor mechanically stable, I
would very much like to hear about it. I've thought about using common
window glass as a dielectric (K about 6 or 7), though I don't know if
this would be lossy or have other problems. Again, if anyone has
knowledge in this area, I would like to hear it.
5. Clearly, aluminum foil or copper foil aren't self-supporting and
there must be some sort of mechanical structure to keep the thing
tight. Stapling to a wood structure sounds awfully heavy for a
reasonable size loop. I have thought of PVC pipe with the copper foil
stretched around "wheel-spokes". That might not be suitable for
aluminum foil unless it is much thicker and stronger than the standard
kitchen variety. Again, I'm looking for ideas.
Hope this elicits some knowledgable answers.
-Steve Harper-
KF7WY
Paul Gili
Steve Harper <steve_...@om.cv.hp.com> writes: > I would also be very interested in seeing some discussion of this idea.
> copper foil
With probably the same intentions as you, I once bought a ~50' roll of
aluminum tape, I think 2" wide and a couple mils thick. It is normally
used by heating and ventilating contractors to patch ducts. It was not
backed by mylar. Relatively inexpensive stuff with a nasty glue backing.
> Mostly, I have questions rather than answers:
>
> 1. Would a trombone capacitor, or other such structure in the middle of
> the loop disturb the pattern or cause other variations in performance?
>
If you could resonate the thing, the mylar or air in vicinity of the gap
will break down due to high voltage--mylar's dielectric strength is
4kV/mil. Air is mostly 75kV/inch = 75V/mil, depending...
> 2. Does anyone remember enough EM theory to compute the RF resistance of
> thin band of conductor as opposed to a round pipe? Clearly there will
> be some current concentration near the edges, but if width/thickness
> ratio is faily large, as in the aluminum foil or copper foil case, it
> seems this should be fairly minor, and might be a cheap way to greatly
> reduce resistive loss. Skin depth is an issue here, too. My copper is
> only about .001 inch thick and I think skin depth at 3.5 MHz is about
> .0017 inch. Are this correct? The aluminum foil is probably thicker,
> but skin depth is probably more, too, since it's not as conductive as
> copper. I've considered using two layers of my copper foil to reduce
> the RF resistance.
This was my motivation to use the aluminum tape--to reduce the conductor
losses. I just calculated 1.39mils skin depth for copper at 3.5MHz, so
you should go for thicker material. I've used pi x D/2 as an equivalent
strip width for a D diameter cylinder when comparing rf resistances
without getting any raised eyebrows. The good thing is that with a
strip, the rf conducts on both sides of it, giving a total perimeter of
2 x W compared to pi x D, if you get my drift. This assumes the thick-
ness is greater than the skin depth, but NOT "much" greater, since it's
an exponentially decreasing function.
>
> 3. One other effect of using a thin band of conductor rather than a
> round conductor is a substantial reduction in the overall inductance of
> the loop. The total inductance is apparently a complicated function of
> the loop geometry and the conductor cross-section. Again, I don't
> remember enough EM to compute the theoretical inductance change. I did
> a sensitivity analysis using an equation from Ted Hart on circular
> conductors and found a 25% reduction in inductance just going from 1
> inch copper pipe to 3 inch copper pipe. The flat conductor will be even
> more, I suspect, and depend on the width/thickness ratio. This is bad
> since it proportionally increases the size of the capacitor for a given
> frequency.
>
Arrgh. I remember about a year ago when we were trying to convert from
rectangular to circular cross-sections on the NEC reflector I think.
No-one had a clear answer, and I'd like answers from the group as
to what an equivalent strip-width is for a given diameter antenna
conductor.
> 4. The physical dimensions of the capacitor quickly become daunting. I
> did a rough calculation, again using Hart's equations. For a 10 foot
> diameter octagon loop with my copper foil (a rough quess at the
> inductance), to resonate at 3.7 MHz required more than 2 square feet of
> capacitor at a spacing of 0.25 inch with air dielectric. If anyone has
> ideas how to construct and keep such a capacitor mechanically stable, I
> would very much like to hear about it. I've thought about using common
> window glass as a dielectric (K about 6 or 7), though I don't know if
> this would be lossy or have other problems. Again, if anyone has
> knowledge in this area, I would like to hear it.
>
> 5. Clearly, aluminum foil or copper foil aren't self-supporting and
> there must be some sort of mechanical structure to keep the thing
> tight. Stapling to a wood structure sounds awfully heavy for a
> reasonable size loop. I have thought of PVC pipe with the copper foil
> stretched around "wheel-spokes". That might not be suitable for
> aluminum foil unless it is much thicker and stronger than the standard
> kitchen variety. Again, I'm looking for ideas.
>
My conclusion was to use #14 wire. A small loop isn't going to work
anyway unless you put it so far away from other conductive and lossy
objects that you might as well put up a larger structure in the first
place. By the way, who's Hart?
> Hope this elicits some knowledgable answers.
Well, answers....
>
> -Steve Harper-
> KF7WY
73, Paul, AA1LL
Greenville, NH
Steve Harper
Ted Hart, W5QJR, is the author of the section in the 15th and later
editions of the ARRL Antenna Book on small transmitting loop antennas.
He has also written a slightly expanded version in "The Loop Book", see
www.antennex.com to purchase an electronic copy. This is good stuff for
anybody wanting to try out this sort of antenna. His construction uses
copper pipe and he has some excellent equations to help in the design,
but unfortunately those equations are specific to copper pipe in octagon
format only, I believe. It's fairly technical, but not sufficiently so
to answer the questions I'm interested in regarding a flat conductor.
-Steve Harper, KF7WY-
Roy Lewallen
At least a couple of references show that a flat strip of width W is
equivalent to a cylindrical conductor of diameter 0.5W. But this is for an
equivalent impedance (hence, same resonant frequency for a given length of
conductor); the two "equivalents" don't necessarily have the same RF
resistance. For RF resistance, the best approximation I know of would be to
compare the surface areas (assuming that the strip is many skin depths
thick, which apparently isn't true for the case under discussion), then
apply some fudge factor for the increased resistance due to current
bunching at the edges of the strip. But what fudge factor to apply? Don't
know the answer to that one. It might be practical to infer the answer from
Q measurements of one-turn coils made from the two materials.
Roy Lewallen, W7EL
Ho4bart
at the last hamfest i looked at capacitors, looking for something
that would let tune the loop but without driving current thru
a moving-contact. i didn't find one, natch, but i did think,
why don't i mount some kind of flexible cable / braid right to
the stator section, like to one plate in the rear of the section?
there's enuff room to either drill a hole in the plate to fix this,
being careful to minimize the effect that bolt protrusions or
screwhead lessens the arc path.
i found one Hammarlund HV cap that looked good, it actually
had a stop on the stator shaft, and the plates were brass, so
the stop could be soldered to, and some kind of spring, maybe
something like a coiled clock (style ) spring could be attached
there. unfortunately the insulation on the cap was not
porcelain, it was of some black - plastic like material i couldn't
identify, so i put off trying that one. but probably that insulation
would suffice for QRPP levels.
hue