LAYING RADIALS FER A VERTICAL

[ve6pg]

...hi from tim...i'm planning on putting more radials out from my  1/4 wave 40 metre vertical. the radiating element is a t.v. tower, insulated above ground...these will be the typical 1/4 length, but my question is one of placement...
...considering that the radials are all tied together, and in parallel, does it matter (much), that they physically are "bunched" together?....
...i know it would be best, to run them out from the antenna, equal spacing from each other, etc., in a circle. but this is the real world....i will do my best, to space them, and "fan" them out, but consider this....what if a guy used a multi-conductor cable, say 8-10 wires in a plastic jacket, 1/4 wave length long, and only terminated at the ground/radial feed point?...would this be of any use, or would the closeness of the wires only see one wire, instead of 10?...there will be some areas, where i will have to bend the radials, but they will be cut to about 32feet...
...any input would be great...

..tim...

..sk..

[k4kyv]

Think of the radial system as a shield, to isolate the radiating antenna from the lossy earth.  It also serves to provide the missing part of the antenna, if the antenna is less than a half wavelength long.

So you want to cover all the ground in the vicinity of the base of the vertical.  If the radials are bunched together, with large gaps elsewhere, the missing antenna length may be taken care of, so that the antenna can be tuned to resonance, but the "hole" in the shield will still be there, to use up precious rf energy to keep the earthworms warm.

[W3RSW]

The 'holes' will also cause some directivity variations in what would otherwise be a 'perfect' circle.   Your car's FM radio demonstrates this effect well, depending on where on the vehicle the antenna is mounted, frequencies in use relative to the car's metal (ground field) being less than, equal to or more than 1/4 wavelength, the direction your traveling relative to the transmitters, etc.

You may desire directivity and thus lay many of your radials in a preferred direction, but as Don says, you'll heat the ground in the other directions and/ or direct a a lot of the signal straight up as well as affect the inpedance of the antenna. 

[Ed-VA3ES]

The purpose of a radial system is to isolate ground losses from the antenna system, and to improve radiation efficiency.   A lopsided  ground system will be lossy (to ground) in the areas not covered by radials.   The more radials the better (up to a point).   Radials should be spread out evenly beneath the feed point.    A "bunched-up cable" would function as a single, but fat radial. 

An absolute minimum of four radials is required, spaced at 90 degrees, but more is better.  Broadcast towers  typically run 120 radials or more.   For a 14AVQ I ran years ago, I had 8 radials per band, except for 10M where I had 12.  The longer  radials for the lower frequencies aid the efficiency for the higher bands. This was effectively 34 radials on 10,  22 on 15, 14 on 20 and 8 on 40.   The antenna was roof mounted on a flat roof, up 35 feet.  It put out like a bomb.

[k4kyv]

Limiting radials to certain directions is a poor way to achieve directivity.  It only amounts to a dB or two at best, and even in the preferred directions the signal drops when the radial field is not complete, because some of the transmitter power is lost heating up the ground.  Less I^2 R(rad) in the vertical section means less radiation in any direction.

I put in 120 quarter-wave radials under my 160m vertical because I had the time, space and wire, but adding more radials is always a case of diminishing returns. 30 radials is a good compromise between cost, labour and efficiency.  There is very little difference in performance between 60 and 120 radials, so my 120 radials, and broadcast ground systems are overkill.  I suppose they keep the ground system efficient for several more years after the copper begins to corrode away in certain soils.

Elevated radials are affected by resonant length, but radials lying on the ground or buried become non-resonant, so the longer the better on any band, but again you reach diminishing returns.  The only disadvantage to using radials longer than 1/4 to 1/2λ is wasted copper.

It doesn't matter whether the radials are bare wire or insulated.  Insulation will protect the metal from ground corrosion, but has no effect electrically.  Ideally, the radials should lie on top of the ground, since burying them introduces a layer of lossy earth between the radiating antenna and the ground plane - the very thing the ground plane is supposed to eliminate.  But burying them  a few inches below the surface causes negligible losses and protects the wire from surface traffic.  I recall reading in an ARRL antenna book from the late 30's, recommending burying the  radials a couple of feet in the ground.  That would have required a ridiculous amount of work and would have resulted in ground losses.  I don't know if that was a misprint or if somebody really thought radials were supposed to be buried that deep for amateur frequencies. 

[KD6VXI]

Limiting radials to certain directions is a poor way to achieve directivity.  It only amounts to a dB or two at best, and even in the preferred directions the signal drops when the radial field is not complete, because some of the power is lost heating up the ground.  Less I^2 R(rad) in the vertical section means less radiation in any direction.

That completely depends on the install, Don.

I had a 1/4 wave vertical, one of the "dropping radial" no tuning method quarter waves from the 70s, called a starduster, mounted on top of an 8 element horizontal 10 meter beam.

I could point the beam at Los Angeles from San Diego, and had NO problems on the ground plane, making contacts in LA.

Point the beam at a rt angle to LA, NOTHING.

While I agree that buried radials, it makes little difference that we can measure... If you are doing it on a raised installation, putting a couple radials in a specified direction (or more to the point, away from a desired direction) can make a diff.

I have a friend using a ground plane, and a V Beam.  We ground both sides of the V beam and work the vertical against it on 10 and 15.  On 10, he is over an S Unit stronger in the "V Beam direction"....  15 meters is less, as can be expected.

Buried radials, I'd be willing to agree 100 percent with Don's statement.

--Shane

[flintstone mop]

I found that using #22 0r #24 insulated stranded wire was the best for laying it down at the end of the mowing season and by the time late Spring arrives they will almost disappear into the grass. Use "SOD STAPLES" to tie it down every few feet to keep it from being a tripping factor.

I have had the worst luck with any type of bare solid wire of any gauge or larger stranded insulated wire. Almost always get caught up in the lawn tractor or DR Trimmer. I waste more time un-tangling the mess. Every Fall got to lay more new radials.........but using the above mentioned wire only now.

Don's response is a good one to follow. He is heard almost everywhere.

It looks like your installation is on an established lawn.

Fred

[W3RSW]

Your #22 or #24, was that bell wire, or better quality hookup wire?  Is it Available in large rolls, say 500 ft. at Loews/Depot?
Sounds good. Knew a ham that used sod staples on his RG8 too.  Right in the grass and it grew over just like yours.

[k4kyv]

IThat completely depends on the install, Don.

I had a 1/4 wave vertical, one of the "dropping radial" no tuning method quarter waves from the 70s, called a starduster, mounted on top of an 8 element horizontal 10 meter beam.

I could point the beam at Los Angeles from San Diego, and had NO problems on the ground plane, making contacts in LA.

Point the beam at a rt angle to LA, NOTHING.

While I agree that buried radials, it makes little difference that we can measure... If you are doing it on a raised installation, putting a couple radials in a specified direction (or more to the point, away from a desired direction) can make a diff.

I would venture to say that is true if the whole thing is  raised at least a wavelength or two above ground, as for example a 2m or even a CB ground plane.  As the ground plane approaches earth it becomes more and more like a buried radial system, and the  less effect the physical radial pattern has on the radiation pattern.  The low-band "elevated" radial systems with a  counterpoise raised a few feet above the ground will act pretty much like a buried system, although not quite as many radials will be required.

Think of two extremes. (1) a 2m ground plane vertical, 50 ft. in the air.  Only 3 or 4 radials are needed to make a completely effective ground plane. (2) the standard AM broadcast tower, with 120 1/4λ radials.  At several wavelengths above ground, not much is needed to shield the vertical radiator from earth.  The ground plane serves primarily to complete the resonant circuit of the quarter wave vertical radiator.  The closer you move the ground plane to the ground, the more radials will be needed to completely shield the vertical from the earth, until the radial system is sitting right on top of the ground in which case the maximum amount of shielding is needed.

I wonder how effective the "ground plane" was on that AM broadcast station in Los Angeles that had the old fashioned multi-wire inverted L mounted on top of the Odd Fellows building, and depended on the steel skeleton of the building for a ground system.  They recently moved the station because the new owners of the building raised the rent to ridiculous heights, the last station in the US known to use that kind of antenna.  I suspect it fell way short of FCC specifications, but they were allowed to keep on using it because of a "grandfather clause".

[W2DU]

I've been reading the posts on this thread with great interest. IMHO, the most illuminating and solid is Don's, with one exception--quarter-wave radials. That length doesn't quite cut it from the standpoint of max efficiency.

I'm assuming many of you are not intimately acquainted with the Brown,Lewis & Epstein experiment performed in 1936, which became the reference for the standard AM BC radial installation. Their experiment at 3.0 MHz showed that beyond a radial distance of 0.4 wavelength from the vertical radiator, the displacement currents reaching ground become insignificantly small, but out to 0.4 wavelength the currents are significant to the total efficiency of the ground system. Thus, any length less than 0.4 wl increases the ground loss, while any length greater than 0.4 wl has passed the dimimishing returns point.

It must be remembered, as Don said, radials either in ground or lying on the ground are non-resonant, so the quarter-wave length is irrelevant, and that length does not contribute to the maximum efficiency available.

Now to the number of radials versus efficiency. To establish a reference point let's use that which is standard in the AM BC community, which is that the theoretical field strength at one mile with 1000w radiated by a quarter-wavelength vertical is 194.5 millivolts/meter. This, of course, assumes perfect ground. I'm going to include BL&E's graph showing the field strengths obtained with various number of radials, each 0.417 wavelength. But below I'll show the values of the field strengths appearing on the graph when radiated from a quarter-wavelength vertical.

No. of Radials       Field Strength      Relative Voltage       Loss in dB
     113                   192 mv/m              0.9846                  0.13
       60                   185                      0.9487                  0.46
       30                   174                      0.8923                  0.99
       15                   158                      0.8103                  1.83
         2                   126                      0.6462                  3.79

The relative voltages are referenced to 1.0 as the theoretical reference (no ground loss), and the loss in dB is referenced to 0.0 dB with respect to the theoretical.

The radials used in the BL&E experiment were No.8 copper, buried at six inches. You are probably wondering the reason for the unusual number 133. Well, after Lewis and Epstein had plowed in 100 radials they still had some wire left over. So the asked Brown what they should do with the remaining wire. He said "Plow it in." So the remaining wire amounted to thirteen more radials.

Don's statement that the radials provide a shield between the vertical and the ground. He is correct, but a slightly different way of considering the radial system is that of a mirror, such that the nearly perfect radial system creates a mirror image of the actual vertical radiator, thus supplying the other half of the above-ground radiator. Consequently, when the ray reflected from the ground joins the ray coming directly from the radiator we obtain a gain of 3 dB compared to a half-wave vertical in space.

On the ground plane antenna, invented by George Brown (of BL&E), his experiments showed that two radials were all that's needed. But the RCA marketing people said it would sell better if it looked more uniform, so the result became four quarter-wave radials. In the elevated ground radial system the radials are resonant at a quarter wavelength.

I may have said earlier that I performed all the engineering pertaining to the FCC requirements for a construction permit, as well as the final building of AM BC station WCEN, 500w on 1150 KHz. My Dad (W8YNG) and I plowed in 90 radials, four inches deep. (90 is all that's required by FCC) My Dad was a mechanical engineer and an expert at creative welding. He designed and built the plow we used to insert the radials. I drove the tractor while he managed the plow. Field strength measurements showed approximately 190 mv/m with 1000 w radiated. Our tower was a four-legged 300' self-supporting Blaw-Knox, leg spacing at ground level 16'. The station was sold a few years ago and the AM portion is gone, only the FM is left.

In the past I've had difficulty in keeping the columns in a table in alignment. In case my columns don't line up correctly when the table above is posted, perhaps someone can straighten them out.

Walt

[W2DU]

On the QRZ site it's easy to edit, even after being posted. However, there is an error in my post immediately above that I've found no way to edit.

The error is in the text where I state 133 as the number of radials used in the BL&E experiment. The correct number is 113, as appears in the table.

Sorry about that,

Walt

[k4kyv]

Walt,
Look at the upper right hand corner of your posted message.  There should be the words  "quote" and "modify".  Click on "modify" and you can edit your post, just as on QRZ.com.  When you have made the change, click on the "save" button at the bottom.

I didn't intend to imply that quarter wave buried radials gives the maximum efficiency, but the difference in signal between 120 quarter wave radials and 120 0.4λ  radials would hardly be discernible on a practical receiver - again the phenomenon of diminishing returns.  With an elevated counterpoise, the radials take on a resonant characteristic, and when combined with a quarter wave vertical radiator, the whole thing becomes resonant, just like a dipole.

Another factor is the number of radials in the ground plane.  If you had only one small roll of wire, it would be better to use it to make a dozen or so shorter radials than to make only four 0.4λ radials.  But if you have enough wire and decide to lay out 60 or more radials, you will  get some advantage by extending each one out to 0.4λ. 

To bury mine, I made a homebrew plough that attached to a rear-tine Troy-Built garden tiller.  I removed the tilling tines and installed the plough as an attachment to the back of the machine, and made a reel to hold enough wire to make one radial, supported by a bracket between the handle bars of the tiller.  The blade would cut a slit in the soil, and feed the wire into the slit, all in one operation.  I made the blade bell shaped, with a pointed end on the sharpened cutting edge, and on the trailing edge I brazed a piece of steel brake line.  The wire fed down through the brake line, and I fixed it in place at the base of the tower, and as I went outwards with the tiller, it cut the slit and laid the wire all at the same time.  I got the idea by watching a phone lineman use a power tool they had for  laying buried telephone drop cable from the side of the road to the house.  I cut the pieces by hand to make the plough, using a hacksaw and some steel plate and angle iron, then took the pieces to a welding shop and told them how I wanted it put together.  It didn't work quite  right the first time, so I partially disassembled it with the hacksaw, made a modification, and took it back for them to weld together again.  I still have it round here somewhere.  It took me and an assistant about 3 days to lay out 16,000 ft. of #12 wire.

But if I had it to do over again, I would have a local farmer come and disc up the soil to a fine powder, then I would lay the radials on top of the ground, held in place with sod staples.  Within a few weeks, the wire would start to bury itself in the ground, and by the end of the following growing season, grass would have grown over the radials and the sod and thatch would completely cover the whole thing; it would be difficult to pull them up if I tried.

[Ralph W3GL]

Walt,

One more edit in your response...

6th para, 2nd line, 9 words in from left "133" !

Regards,

Ralph, W3GL.

@ 1:05pm, 6/10...
Oops... I see by the time line, you hadn't gotten to the edit  yet...
Sorry, I misspoke.

[W2DU]

Don, thanks for the assist on how to edit.  I looked and looked last night for the place to find an edit button, but I was so sleepy then that I totally overlooked the 'modify' button.

And Ralph, the error you spotted is the one I referred to earlier. Is there another one that I missed?

Now for everyone: I'll be pleased to email a copy of the entire BL&E IRE paper to anyone who'd like it, but please request it by email to me at walt@w2du.com. The paper is in the Proceedings of the IRE, Volume 25, No. 6, June 1937.

I've attached a coupla pics, one of my Dad assisting me, the other showing the ruts made where the radials went into the ground. This occurred in 1948.

[KD6VXI]

IThat completely depends on the install, Don.

I had a 1/4 wave vertical, one of the "dropping radial" no tuning method quarter waves from the 70s, called a starduster, mounted on top of an 8 element horizontal 10 meter beam.

I could point the beam at Los Angeles from San Diego, and had NO problems on the ground plane, making contacts in LA.

Point the beam at a rt angle to LA, NOTHING.

While I agree that buried radials, it makes little difference that we can measure... If you are doing it on a raised installation, putting a couple radials in a specified direction (or more to the point, away from a desired direction) can make a diff.

I would venture to say that is true if the whole thing is  raised at least a wavelength or two above ground, as for example a 2m or even a CB ground plane.  As the ground plane approaches earth it becomes more and more like a buried radial system, and the  less effect the physical radial pattern has on the radiation pattern.  The low-band "elevated" radial systems with a  counterpoise raised a few feet above the ground will act pretty much like a buried system, although not quite as many radials will be required.

Right again!  70 foot crank up tower, beam mounted about 1.5 feet above, vertical on top of a 12 foot section.

Interesting reading on the BC antenna.  I have to wonder what the level of radiation was for the people inside..


--Shane

[K3ZS]

I have used an electric lawn edger to make ground slits for radials.   With mine you just raise the guard that is used for edging.

[k4kyv]

At first I tried using a power lawn edger.  But the slit would inevitably get clogged with clumps of loose dirt that made it impossible to push the wire down inside.  I gave up trying to use it with the first radial.  That's when I borrowed the heavy duty garden tiller and built the plough.

Since the building in L.A. has a metal skeleton, there might not have been much RF inside.  If there was, it was probably no worse than what we are exposed to when operating our stations.  I think for most of the years that station was on the air, they ran only 100 watts, and maybe upgraded to 1000 watts relatively recently.  http://gallery.bostonradio.org/2004-12/la/100-02598-lrg.html

I think the RF hazard from MW and HF has been blown way out of proportion.  I would tend to be careful around microwaves and UHF/VHF, but I think the worse thing you can get from lower frequencies is an rf burn when you come in direct contact with "hot" circuitry.

Frankly, given to-day's paranoid society that cards 65-year-olds when they buy beer and suspends kids from school for packing a plastic picnic knife in their lunch bag, it surprises me that by now someone hasn't successfully lobbied to outlaw amateur radio altogether, claiming that a transmitting station in a residential neighborhood poses a "radiation hazard" that will give everyone or everyone's kids (remember, do it for the "children") cancer.

[W2DU]

Well, Don, I hadn't heard about 65-year olds getting carded, but going further on paranoia, I read in the news just last week about a kid farting in the school bus, and they suspended him for three days. Unbelievable!!

And about MF and HF being blown out of proportion wrt being a health hazard, what those proponents don't understand is at those frequencies people are VERY SHORT antennas. A piece of wire approaching a half-wavelength would of course have enough coupled voltage to cause some current flow on it, but a wire the length of people's heights would have insufficient voltage for any practical amount of current flow. Consequently, at those frequencies the voltage induced across the length of a human body would be so small that the current flow in the body would be infinitesimal.

Walt

[k4kyv]

Well, Don, I hadn't heard about 65-year olds getting carded

It is true.  I have been carded more times since I retired than I ever was during all those years when I looked like I could possibly be too young to buy beer.

I always go out of my way to make a point to ask them if any damn fool couldn't tell that I'm old enough to buy beer, make as much a stink about it as possible, try to make the poor sap at the check-out feel as stupid as possible, and to try the patience of the people waiting in line behind me as much as I can.

Maybe if more people would make it an issue to refuse to cooperate with this sort of nonsense, society would regain some common sense.

As for the rf, you have a point.  If you stood on top of a loading coil so that your body was part of a resonant circuit, you might be at high rf potential, but the rf current flowing through your body would still be negligible.  And what current there was, would be confined to your skin area.

[Steve - WB3HUZ]

OET 65.

[k4kyv]

OET 65.

http://www.fcc.gov/Bureaus/Engineering_Technology/Documents/bulletins/oet65/oet65.pdf

[K6JEK]

Excellent articles on radials in the last four issues of QEX.  Rudy Stevens, N6LF, built a test range and meticulously measured performance of various numbers, lengths and heights of radials.  By and large his measurements confirm NEC modeling and the wisdom expressed by Don and Walt above.  The articles are full of graphs.  He tries to let the numbers speak for themselves although he does draw a some conclusions.

A few things that struck me: 

Very few elevated radials match scores of on ground radials. They can even start low and slant up and still work gangbusters.  This, of course, can be very hard to do.

What I would call diminishing returns occurs way before 120 on ground radials. There is less than 1 dB improvement going from 30 to 60, for example, 1/4 wave vertical, 1/4 wave radials. 

The worse the antenna, the more the radials matter.  The curves for shortened antennas are much steeper than the curves for full sized antennas.

I don't have this series of four articles in electronic format and I don't know if the ARRL does either.  They were in the last four issues of QEX, 251,252,253,254.

[Steve - WB3HUZ]

What I would call diminishing returns occurs way before 120 on ground radials. There is less than 1 dB improvement going from 30 to 60, for example, 1/4 wave vertical, 1/4 wave radials. 

Indeed. The point of diminishing returns will usually be far different for the amateur station than the broadcast station.

The worse the antenna, the more the radials matter.  The curves for shortened antennas are much steeper than the curves for full sized antennas.

Shorter antennas have lower radiation resistance. Since efficiency is determined mainly by the ratio of radiation resistance and equivalent ground loss resistance, this makes sense.

[ke7trp]

While visiting with my uncle, We talked about this same subject about ground radials.  He worked as a broadcast engineer in the east all his life.  He said on many ocations they Trimmed off radials to stop signals from reaching NYC. He explained they did extensive testing and found that you could make a significant impact on the direction of the pattern with the length of the radials.

I have above ground radials. Mainly do to the lack of wire and space around my tower and vertical for 160.  I only use 4 radials about 2 feet above ground. My radials are short.. Only 30 to 40 ft long. Even with this in place, The tuning on the system changed and the performance was greatly improved. When I find a good deal on wire, I plan to add alot more and some full length radials. out. But for now, The 4 actualy work great!  When the Dog tears or knocks a couple down, I can see the tuning has changed. We have the worst soil being in Arizona so everything helps.

Clark

[Steve - WB3HUZ]

Consider tying your four radials to a common coil and resonating them.