Short version: Believe what you measured. Long version:
We did measurements of this for understanding and designing BOGs some years back. In the upper east central part of North Carolina (contains Raleigh/Durham) we made extensive resonance and resistance measurements of a 151 foot dipole of insulated wire laid on the ground (Dipole On Ground or DOG). These varied almost wildly. The "wildness" was confirmed. From these we computed velocity factors from 45 to 80 percent. Feed resistance at resonance measured from 60 to 200 ohms. Whether these variations are even more extreme in other places I cannot say. But they were more than sufficient to convince that some applications required knowing those constants for effective design. Resistance and resonance did not correlate in any pattern we ever saw, broadly scattering on a plot. This is not terribly surprising since dielectric constant and conductivity are separate specifications for soil. Measurements over the same back yard occasionally varied significantly with the DOG laid in different orientations. No knowing why. Lots of guesses, but no proof. Buried wires, pipes, septic fields, constuction debris used as leveling underneath the top layer of lawn grade dirt, vein of sand with water? The list is endless. But one thing is plain. The coupled resistance of the ground significantly broadens the resonance curve but does not swamp it. The resistance from traversing 75 feet out and back laying on the ground does not sufficiently attenuate the resonance effect from the end of the wire and the resonance is plainly still there. A bare wire buried dipole did not show this resonance. Guessing, it might if the ground was extremely dry or sandy, like the ground seen in the Sand Hills of North Carolina. The insinuation for insulated radials laid on the ground is that some degree of attention to velocity factor may be in order, as in Pete's question. It's interesting to explore this as if it were a controlling issue to see what it would imply. Depending on one's faith in NEC ground effect calculations on wires very close (within several wire diameters) to ground, this is also proven out in the models. Here's the procedure: Create a 151 foot (46m) dipole (+/- 75.5 feet) out of your insulated radial wire. Lay out the DOG centered upon a diameter of your proposed radial circle. Push it through any grass and get it next to the dirt, because that is where it will get to in time. There is a significant difference between ON the dirt and a half inch or cm ABOVE the dirt. If you measure and cut for on top of the grass, they will gradually sink and become electrically longer (and lossier) than you need. Only measure on top of the grass if that is how it is going to be when you are operating, as in laying them out for a contest and rolling them up afterwards. Measure resonance and feed resistance at resonance. The resonance should be in or around the 160 meter band. Re-lay the dipole on the diameter at right angles and remeasure. If you are lucky the two will be close and you can work with the average. The resonant frequency will tell you how long you need a radial to be so that the radial feed is resonant and essentially resistive. The formula for the radial is the DOG resonance frequency in MHz for 151 foot length, times 41.25 feet to get you length for resonance at 1.83 MHz laying on your personal dirt. Trim or extend the dipole to 82.5 times the measured resonance in MHz to adjust to 1.83 MHZ. Redo the procedure, being sure to push the wire down to the dirt. Re-measure. You should be very close to resonance. If this step did not center the resonance, you probably are not being serious about pushing the wire down to the dirt. (Anyone who does this, we would be happy to know your results, resonance and resistance, length and location to add to our data.) Half the resistance will tell you how many insulated radials you need, of that length laid to the dirt, one radial for each ohm, for the radial field to have a radial system effective series impedance of one ohm resistive. There are some places around Raleigh where the measurements tell you that would need a hundred radials 95 feet long. Another place would be 30 radials 60 feet long. This would lead one to expect large variation in results with not so dense radial installations. If one did not want to bother with the math, to be sure of commercial results over heavily root bound soil where you can't possibly bury, I would go with the fuzzy average 65-ish velocity factor, 2/3 for easy memory, and put down 120 insulated 90 foot (28m) radials. That's 10,800 feet of wire, so I'd think you would want to measure and do the math. 30 times 60 feet, if you're lucky, is only 1800 feet. Note that the wire that is pushed down to the dirt always gets the slower velocity factors, thus shorter electrical 1/4 wave radials. Not pushing it down to the dirt costs you. Note that one can reverse engineer this formula, once you have measured down to the dirt and found your personal resonant length, you can predict the feed resistance of a lower radial count related to a budget or materials on hand. Buried bare wire radials are directly connected to the ground medium and have no insulated end to form a reflected wave unless it's buried in dry sand or some very dry or highly resistive medium. For Pete's question I would use the method above, simply because the reactance over some dirt would skew the results from using some one size fits all book formula. I would also seal the far end of the insulated wire (dipping the end in a can of liquid tape?). If you don't, the end will gradually short to ground and the characteristics of the radial system will gradually change with it. If one was not concerned with that, I would think this whole discussion is too fine a point. Another thing that should be taken from this discussion is that sparse radials, and particularly short, miscellaneous, sparse radials made from insulated wire laying on the ground are going to present exceptional loss. We're talking about stuff that has been measured, not speculation. 73, Guy. On Wed, Dec 5, 2012 at 3:48 PM, Pete Smith N4ZR <[email protected]> wrote: > I had always understood that because of coupling, radials laid on the > ground were functionally the same on top-band as those buried a few inches. > Recently, some experimentation with Beverage on ground antennas has me > doubting that. > > It's been suggested that elevated radials need to be resonated, like very > low dipoles. Is that true with radials-on-ground, but with a > much-different velocity factor? > > Trying to relate efforts to results for radials on a tower in thick woods, > where only radial ON ground are possible, what would be my best return on > effort deploying, say, 1000 feet of wire? 4 longer elevated radials > (resonated?) or 10 shorter radials on ground (unresonated). > > -- > > 73, Pete N4ZR > Check out the Reverse Beacon Network at > http://reversebeacon.net, > blog at reversebeacon.blogspot.com. > For spots, please go to your favorite > ARC V6 or VE7CC DX cluster node. > > ______________________________**_________________ > Topband reflector - [email protected] > _______________________________________________ Topband reflector - [email protected]
