In a message dated 9/8/2002 20:36:28 PM GMT Daylight Time,
w8ji@contesting.com writes:
> I've never seen any measured data (where the correct parameters were
> measured) that indicated wire size is important to radial loss. See QST Dec
> 2000 pages 38 and 39 where I specifically discuss ground loss and radial
> size.The
> primary reason size is not important is easy to see, if we look at
> the problem in proper perspective.
>
> First, consider loss in the antenna itself. It is very well known
> that the power dissipated in even relatively small wire in an antenna
> is negligible, as long as current levels are reasonably low. Not true. dBs
> are dBs and have nothing to do with absolute power level. It's a simple
> Ohm's Law problem. If you don't want significant losses, don't use lossy
> conductors.
>
> Next, think about what a radial does and how it interacts with the
> medial around it. The radial system can NOT have more current than
> the antenna "pushing" against it, so I^2 losses in the conductor
> itself are even less of a problem than in the antenna. Not true. If you
> have a lossy antenna because of small gauge wire, then ground losses are
> less significant. Obviously, if the antenna is only 5% efficient and the
> ground system is 80% efficient, it makes no sense to spend all your money
> improving the ground system. On the other hand, If the reverse is true and
> antenna losses are very low, the next thing to reduce losses in is the
> ground system.
>
> The bulk of losses in any radial or radial system are coupled losses
> to the earth itself, and only **distance** reduces that coupling.
> When we add multiple wires (spaced some reasonable distance apart, so
> the wires couple more to earth than to each other), we decrease the
> current and voltage density in any given cross-sectional area of
> earth below the radial system, and lower losses. Coupled losses eventually
> have to make it back to the braid of the coax. With radials the intention
> is to intercept coupling to earth in the vicinity of the antenna. An
> efficient way to do that is to create a less lossy earth with conductive
> radials. Anything that we can do to imrove the conductivity improves
> antenna efficiency.
>
> If you want to observe this effect, model two parallel Beverages a
> small distance apart. You will see the "gain" doubles with relatively
> close spacing, while the pattern, S/N ratio, and directivity remain
> essentially the same. The gain doubles because loss halves, because
> the pair of antennas divide power and reduce field density and
> losses. Of course, but you haven't said anything about ground losses.
> Paralleling equal value resistors will halve the resistance, but what if
> the ground losses account for 99% of the loss and a single beverage is 1%
> efficient? Just paralleling two of them makes the system 2% efficient. Big
> deal. Get rid of the ground loss and the efficiency could improve
> significantly.
>
> Now make the individual antennas thicker, and watch the gain. You'll
> find it has almost no effect at all on loss, changing primarily only
> the impedance of the system. Wrong. Decreasing conductor loss by using
> larger conductor size (all else being equal) always decreases loss giving
> an apparent improvement in "gain." If the impedance changes, then matching
> differently is required, but matching is only a way of maximizing power
> transfer to the load (receiver). A match can always be found and should be
> found before comparisons are made.
>
> We can be easily misled with all the hyperbole about "isolating
> displacement currents" and looking at impedance changes in a complex
> system. Sometimes we even install a good ground system, then never
> actually make a connection to that system in the hope that the lack
> of a connection will reduce loss!! Sound contradictory? It does
> because it is contradictory. If you install a good
> ground, use it with a direct connection. You have everything to lose
> and nothing to gain by intentionally having a poor connection, or
> intentionally concentrating fields with elevated radials, when you
> have what would be a GOOD ground system below the elevated
> radials. I don't get your point. Why have a ground system if you don't
> connect to it? If you don't, it's a broken circuit. You can't expect the
> lightbulb to light up if you don't connect both terminals to the battery.
>
> All we really want to do spread the current and voltage out over a
> very large cross section or area of earth, and we can do that only by
> using a radial system that keeps field density low by spreading it
> over a large area of earth. Making the conductor thicker, insulating
> it, or shortening it and loading won't help decrease the field
> intensity below the radial. This is only half of the argument. Eventually
> the fields cause current flow in real conductors, earth, radials, above
> ground wire or otherwise. Minimizing the conductive losses improves antenna
> efficiency.
>
> Increasing the height a large amount compared to the radial length
> and operating wavelength will help, but really isn't much different
> than simply using more wires in a regular easy-to-maintain radial
> system. Agreed, but what you are doing is paralleling resistors, i.e. low
> loss wires with high loss earth. This improves antenna efficiency.
> 73, Tom W8JI
> W8JI@contesting.com
73 de Tim, W9QQ
--- StripMime Report -- processed MIME parts ---
multipart/alternative
text/plain (text body -- kept)
text/html
---
|