Bill Turner wrote:
If your mind is set on coax traps, ok, but you might consider
something different. W9INN uses what he calls a "resonactor" in his
trap dipoles and I think they are far better. Basically, they are just
a single layer coil which is carefully trimmed to resonate at the
desired frequency with its own self-capacitance. There is no other
capacitance needed. The advantages are two: Very low loss and very
high power handling capability.

Consider that is a very good description of a resonant
coiled coax choke.
--
73, Cecil http://www.w5dxp.com

Owen Duffy wrote:
As you know, trapped antennas are sometimes explained by considering
the traps to be equivalent to frequency selective switches that are
o/c or s/c at certain frequencies of operation, and conveniently
disregarding the conductors on the outboard side of the o/c switches.

Traps are high-impedance devices at their resonant frequency.
For them to work properly, they need to see a low impedance on
the far side of the trap. If they see a high impedance on the
far side, they cease to be "traps" and become the phase shifting
coils described by Kraus useful in a phased array.
--
73, Cecil http://www.w5dxp.com

Owen Duffy wrote:

Sure. I should have mentioned that after trying NEC models to get a
low VSWR solution for just two bands (one twice the frequency of the
other) I gave up figuring that it was likely that an ATU would be
needed in any case, and the traps were just an unnecessary
complication.

As you know, trapped antennas are sometimes explained by considering
the traps to be equivalent to frequency selective switches that are
o/c or s/c at certain frequencies of operation, and conveniently
disregarding the conductors on the outboard side of the o/c switches.
It is a simplistic view, it does not lead to a design that works, nor
does it explain designs that do work, but it remains the most popular
explanation to be heard on air in my experience.

Yes, I'm afraid that's true.

I try to avoid simplified explanations just because people forget that's
what they are, and begin treating them like fact. A good example is the
recent thread about there being two "theories" of ground plane
operation, one being the simplified explanation of operation as an
"image". (I still remember an April magazine article in which the author
began with the popular and common "image" idea to explain the operation
of a grounded vertical. Then he proposed drilling a hole in the ground
and putting the vertical there, fed at ground level. This, he said,
resulted in the "image" being above the ground. The perfect stealth
antenna! Anyone who knows where this appeared -- must have been around
the '60s -- would do me a favor by letting me know. I'd love to read it
again. I've searched the entire collections of QST and HR to no avail.)

But back to traps. Trap operation is indeed a lot more complicated than
most people realize. I was recently contacted by a puzzled EZNEC
customer who saw a situation where a trap had very little effect. He had
put the traps where the current was normally very low, so of course the
traps did little. Plenty of current was being induced in the wire beyond
the trap by mutual coupling from the field produced by current on the
wire on inside of the trap, whether the trap was there or not.

The simplified idea of a trap as a short circuit on other bands also
leads to bad decisions. On any frequency band below its resonant
frequency, it acts like a loading inductor; on bands above resonance,
like a capacitor. It's common to see more loss on the lower frequency
bands where the trap is a loss inductor than at the trap's resonant band.

Modeling makes it easy to find out what works well and what doesn't for
a given application. But of course, you have to know the trap's L, C, and R.

Roy Lewallen, W7EL

On Sat, 11 Nov 2006 15:11:14 -0800, sailtamarack wrote:

You can design the antenna and the coils using Hamcalc.

Thanks for pointing me in that direction. That program is a goldmine!

Also thanks to everyone else for their advice.

I looked at rfparts.com for suitable capacitors for these traps
(assuming I don't go the coax trap route), and I gotta say, they sure are
proud of their capacitors ... $16+ for each 5 KV doorknob capacitor.

For a dual-band 80m/40m dipole using resonant traps, how can I figure out
the capacitor voltage rating I need for each power level?

This antenna is never going to be used over 200 watts and rarely over 100
watts, in fact it's probably going to spend most of its life around 10
watts or less (FT-817 / IC-703).

You guys sure are way beyond me in antenna and trap theory. :-)

I probably neglected to mention earlier, but this antenna is for portable
operation in the field on 75 and 40 (none of the other bands are needed),
so it needs to be easy to put up and take down and easily transportable,
also fairly rugged so things like airwound coils waving in the breexe with
ceramic insulators inside the coil probably won't work; instead the coils
will probably have to be tightly wound around small lengths of PVC pipe
and lacquered in to place.

I need to build several of these antennas (at least 10, I think, perhaps
more), and so $16+ each for doorknob capacitors from places like
rfparts.com isn't likely to be practical. That reason, along with ease of
construction (and now ease of design with programs like Hamcalc around) is
why I have been leaning towards coax traps.

The bandwidth (of the traps or of the antenna itself) is not important and
so high Q and narrow bandwidth is OK.

Power levels will be low, never more than 200 watts and probably never
more than 100 watts, and often around 10 watts.

If you guys say coax traps are a sub-optimal approach I'll take your word
for it.

But, the things you all are saying about traps coming with a lot of
baggage that no one ever thought of is making me nervous. :-)

It's not exactly a new concept, trap antennas have been used with great
success for what, 80+ years now?

If I forget the coax trap idea and go with a resonant coil-and-capacitor
approach on, say, a 1.5" coil form (which Hamcalc seems reasonably happy
with after warning me that 2" was too big and 1" was too small), what
should I look out for?

On Sun, 12 Nov 2006 22:05:38 -0500, "C. J. Clegg"
wrote:


You guys sure are way beyond me in antenna and trap theory. :-)

I probably neglected to mention earlier, but this antenna is for portable
operation in the field on 75 and 40 (none of the other bands are needed),
so it needs to be easy to put up and take down and easily transportable,
also fairly rugged so things like airwound coils waving in the breexe with
ceramic insulators inside the coil probably won't work; instead the coils
will probably have to be tightly wound around small lengths of PVC pipe
and lacquered in to place.

I need to build several of these antennas (at least 10, I think, perhaps
more), and so $16+ each for doorknob capacitors from places like
rfparts.com isn't likely to be practical. That reason, along with ease of
construction (and now ease of design with programs like Hamcalc around) is
why I have been leaning towards coax traps.

The bandwidth (of the traps or of the antenna itself) is not important and
so high Q and narrow bandwidth is OK.

Power levels will be low, never more than 200 watts and probably never
more than 100 watts, and often around 10 watts.

If you guys say coax traps are a sub-optimal approach I'll take your word
for it.

But, the things you all are saying about traps coming with a lot of
baggage that no one ever thought of is making me nervous. :-)

It's not exactly a new concept, trap antennas have been used with great
success for what, 80+ years now?

If I forget the coax trap idea and go with a resonant coil-and-capacitor
approach on, say, a 1.5" coil form (which Hamcalc seems reasonably happy
with after warning me that 2" was too big and 1" was too small), what
should I look out for?


When you make your mind up about the MAXIMUM operating power level,
you will be in a better position to finalise the design. Capacitors
are likely to fail if you exceed their voltage at all, not for 90% of
the time, or 10% of the time, but at all, and in microseconds.

If you design a 75/40m trapped antenna, and place the trap resonance
away from 40m band (in fact any other operating frequencies), you will
reduce the voltage impressed on the trap. At low power levels, you may
well be able to use a capacitor or stack rated for around 1000V which
will come a lot cheaper and a lot lighter than a doorknob.

But, as Roy has told you, they don't fall into place. The question is
whether "hamcalc" will allow you to design with sufficient freedom and
rate the operating voltage on the traps. The technique of modelling
the proposed design will reveal the voltage expected across the traps,
so long as your estimates of the trap parameters are sufficiently
accurate.

It might be safer to copy a published design rather than looking for
design tools that might not be rock solid.

I cannot offer you a proven design, the thing that I use is pretty low
tech, it is an 80m dipole with insulators in the middle of each leg
and wander leads to switch between a 40m half wave and 80m half wave.

Owen
--

C. J. Clegg wrote:
. . .
If you guys say coax traps are a sub-optimal approach I'll take your word
for it.

But, the things you all are saying about traps coming with a lot of
baggage that no one ever thought of is making me nervous. :-)

It's not exactly a new concept, trap antennas have been used with great
success for what, 80+ years now?

The coax trap is a relatively new idea. I don't think it was around
until the '70s. They definitely have their place, being cheap, simple,
and rugged. And the loss is minimal in a lot of cases. Note, however,
that there are two ways of connecting the coax. In one, the center of
one end of the coax is left open. In the other, the shield of one end is
connected to the center conductor of the other, and the two remaining
ends are the external connections to the trap. The latter method gives
an L/C ratio that's generally more favorable.

As for the "great success", almost no amateur is able to tell when he's
getting a couple of dB loss, which is about what you typically get with
an unfortunate trap/band combination. I've mentioned that this is what
happens and how I've avoided it, but most people happily accept it, if
it happens, without worry. Each to his own.

If I forget the coax trap idea and go with a resonant coil-and-capacitor
approach on, say, a 1.5" coil form (which Hamcalc seems reasonably happy
with after warning me that 2" was too big and 1" was too small), what
should I look out for?

For your application, I'd probably use coax traps. Using a separate coil
and capacitor aren't a guarantee against loss. Even a pretty good trap
can introduce a fair amount of loss if it ends up looking like a big
inductor in the wrong place in an antenna on a band below the trap's
resonant band. Or if you put it in a point where there's an extreme high
voltage between the ends on the band where it's resonant. Without the
ability to measure trap characteristics and do some modeling, what you
end up with will be determined entirely by luck. But as I said, you
probably won't notice the difference anyway.

Roy Lewallen, W7EL

Roy Lewallen wrote:
The simplified idea of a trap as a short circuit on other bands also
leads to bad decisions. On any frequency band below its resonant
frequency, it acts like a loading inductor; on bands above resonance,
like a capacitor.

Until it gets to the 1/2WL self-resonant frequency where
it yields the 180 degree phase shift described by Kraus.

Quoting Kraus: "A coil can also act as a 180 degree
phase shifter ... The coil may also be thought of as
a coiled-up 1/2WL element." This coil is placed at a
current minimum point in a phased array.

This does NOT mean that the length of the wire used in
the coil is 1/2 wavelength. It just means the coil is
series resonant with a 180 degree phase shift.
--
73, Cecil http://www.w5dxp.com

C. J. Clegg wrote:
For a dual-band 80m/40m dipole using resonant traps, how can I figure out
the capacitor voltage rating I need for each power level?

For a rough approximation, you can model the trap as
a stub with the same VF and Z0 as the coil. Equations
(32) and (51) in the following IEEE paper are useful
for that purpose.

http://www.ttr.com/TELSIKS2001-MASTER-1.pdf
--
73, Cecil http://www.w5dxp.com

I cannot offer you a proven design, the thing that I use is pretty low
tech, it is an 80m dipole with insulators in the middle of each leg
and wander leads to switch between a 40m half wave and 80m half wave.

Owen
Please describe "wander leads".

John Ferrell W8CCW