Sunday, December 4, 2011

Considering a small HF antenna? You have to read this ...

I came across this explanation of the limitations of stealth/small antennas by Dan Zimmerman, N3OX. He has provided the best explanation of the physics behind 'short' antennas that I have seen so far.

This should be required reading for anyone considering one of the stealth antenna designs such as the Tak-tenna, Isotron or Crossed Field Antennas.

I hope Dan won't mind me copying his explanation from eHam.net. It is worth your while to head over to his web site and see some of his home brew antenna projects, there are excellent explanations provided as well as the thinking that went into each design.

Once you've had a read below go on over to: http://www.n3ox.net/ and take a look at Dan's website.

"In theory, a short antenna can be made efficient enough to compare very well to a full size version, but the tradeoffs (lower impedance, narrower bandwidth) are inescapable and you need to address them carefully to make it work well in practice"

And the *most important thing* for hams who need small antennas like KB3HJK does is to never forget that those tradeoffs are fundamental.

In order to radiate a certain amount of RF power into the universe with a short dipole, you *have to increase the current* flowing along the straight bit many times over what has to flow in a half wave dipole.

If you make a very short/small antenna and want to *radiate* the same amount of power, that *requires* a much higher current flowing in the radiating part of the antenna, period.

In order to pump more current through the antenna without causing significant losses, you have to reduce the loss resistance significantly compared to a big antenna.

And an antenna that needs lots of current to radiate a given amount of power is said to have a "low radiation resistance"

The power lost to heat in an antenna is basically I^2*Rloss (the antenna current squared times the loss resistance) The power radiated is I^2*Rrad (current squared times radiation resistance).

So what does this have to do with bandwidth? Well, a couple of things. One is that when you make an antenna smaller and drive high currents in it, you make a LOT more electrons slosh back and forth in a small physical space. To focus on a short cap hatted dipole, there's a LOT of magnetic field caused by the very strong current (lots of electrons per cycle) flowing through the horizontal dipole part, and there's a LOT of electric field caused by lots of electrons piling up on the capacitance hats, first one and then the other, as the RF cycle progresses.

A capacitance hat charges up on one part of the RF cycle, a quarter cycle later, all those electrons are rushing at maximum speed toward the other hat, another quarter cycle later, they're piled up on the 2nd hat, and another quarter cycle later they're rushing back toward the first hat. Energy is exchanged between the electric field, largely between the hats and the magnetic field as the electrons are rushing through the horizontal conductor.

Since everything is so compact, the electric and magnetic fields are very strong, and store a lot of energy near the antenna.

But we also know something else... we know that the radiation resistance is very small, and to make the antenna efficient, we must reduce the loss resistance. So the *total resistance* is very low. The resistance is associated with the energy lost per cycle of RF. Some goes to heat in the loss resistance, some goes to radiation, "dissipated" in the radiation resistance.

But if you compare the energy *stored in the fields around the antenna* vs. the energy *lost per cycle*, you find that there's a lot of energy stored vs. how much is dissipated in the radiation and loss resistances. The strong fields make the stored energy high, the low resistances make the dissipation small.

So the antenna is very "high Q." If you cut power to a very high Q antenna, it will ring down for a relatively long time as the stored energy is damped by the dissipation into loss and radiation. But we know from other circuits that high Q resonant circuits are very sharply tuned, and a small antenna is no exception. It has a very narrow bandwidth over which you can slosh current back and forth effectively in a resonant way.

Since the radiation resistance and the stored energy in the fields is fixed by the size and shape of the antenna, the only way to broaden the response of a certain size antenna with fixed tuning is to add losses!!!

This is fundamental, and will steer you away from very small, broad bandwidth antennas if you keep it in mind. You absolutely, positively must give up bandwidth to keep efficiency at small size.

This is why the very best tiny antennas will all be motor driven. Magnetic loops and mobile screwdriver antennas with capacitance hats are two great examples of how to get around the narrow bandwidth problem. Sure, the antenna is 10kHz between the 2:1 SWR points, but if you can tune that 10kHz anywhere you want between 5 and 21 MHz, who cares?

But there's even a point where motor drive doesn't save you. There's even a point where superconducting antennas don't save you.

There's a lower limit that few talk about (except a few crazy magloop guys who come close to running up against it)

If you make a very very tiny, extremely low radiation resistance antenna and you stamp out almost all the losses by welding together huge conductors, your antenna's bandwidth could become so narrow as to not pass even a SSB signal. ;-)

You'd actually roll off your audio if you had a 1kHz wide magnetic loop and could make the tuning stable!

But this is the basic fact you need to remember when antenna shopping. Quite small antennas should be easily retunable in small steps across a ham band, otherwise they are required to be *quite* lossy to give good SWR bandwidth. No matter what any manufacturer says, a tiny antenna needs to be VERY small in bandwidth for it not to be lossy.

And KB3HJK, as far as that particular HF-p antenna goes? It's nearly impossible to know exactly but I expect that since it covers 200kHz of 40m with no retuning and is only 10 feet long, it's probably going to be about 1% efficent.

For comparison, I built a 40 foot long 40m dipole with a loading/matching coil at the feedpoint that should have been about 80% efficient (-1dB) and was about 70kHz between the 2:1 SWR points. End loading could improve that, but the HFp isn't end loaded.

If you really need to get on 40m better than you have been in the sort of 3-10 foot antenna class, your next antenna should have a motor.

Or if you're worried about feedline radiators because you can put the antenna 20 feet up on a pole, just go ahead and make the pole the antenna instead. N0LX has some interesting voltage fed "loaded end fed half waves" on his website, and they actually model reasonably well.

And even a Tak-Tenna type antenna is maybe OK, the problem with them is that there's NO REASON to use a 30 inch antenna on 40m. It's too short. Do the same thing but make it 15 feet long and you'll be much better off.

73
Dan 

No comments: