Buried Radial Length Same as Monopole Height?

[rfry]

It has been stated on R-I and elsewhere that a monopole antenna does not need buried radials any longer than the height of the monopole they are used with. However that conclusion is not supported well by the 1937 I.R.E. paper on ground systems by Brown, Lewis & Epstein of RCA.

Below is a link to a clip from that paper, with their data and statement showing that the conductivity of the path through the earth to the base of the antenna is important out to a radius of 0.3 wavelengths "in order to operate a short antenna efficiently." The clip has been marked up to show the point below which on the curves applies to a Part 15 AM 3-m monopole with its base at earth level.

Using 3-m buried radials with such a 3-m monopole would "miss" collecting a significant amount of earth current from beyond the ends of the radials out to 0.3 wavelengths. That current would be converted to heat by the relatively high resistance of the earth compared to the radial wires, and so could not produce radiation from the monopole.

Also ... there is more to a good r-f ground than a connection point having a low value of resistance to earth potential. The physical location of that r-f ground with respect to monopole also is very important.

If the buried r-f ground conductors are not located where the induced/conducted earth currents from the monopole are the strongest (within a 0.3-wavelength radius of the monopole), then those earth currents will be absorbed by soil losses before they can reach those buried conductors. The radiation efficiency of the antenna system then can be greatly reduced, compared to centering the monopole over the buried conductors (radials).

http://i62.photobucket.com/albums/h85/rfry-100/BLE_Earth_Currents.gif

RF

 

[Goat Rodeo Cowboy]

It would be fascinating to find a site where an AM station has ceased to exist, or has moved to a new location, and after tower removal, the ground system was still intact. mount the classic 3 meter whip used today in Part 15 stations on the old concrete foundation of tower and be able to connect to the legacy ground system.

I suppose the old ground system will have been captured for salvage by then, but if not, what a great location for experimentation.

The other great experiment would be DX reception at such a site. Would attaching the ground terminal of a DX receiver to an old AM station ground system provide superior reception?

 

[druidhillsradio]

Lots of good info here. I would to see more comments.

 

[rfry]

The other great experiment would be DX reception at such a site. Would attaching the ground terminal of a DX receiver to an old AM station ground system provide superior reception?

It would if the antenna input connector was attached to a resonated monopole at the center of the radials, and proper impedance matching was used.

//

 

[Goat Rodeo Cowboy]

It would if the antenna input connector was attached to a resonated monopole at the center of the radials, and proper impedance matching was used.

I've had this fantasy for years. Someday I would find a piece of property that had an AM tower on it (no longer in use) and the property would be scenic and suitable for building a comfortable place to live.

Years and years ago I was involved in the construction of a new AM station. We had everything ready to go and were awaiting the telegram from the FCC authorizing us to proceed with "program testing". So that no one would get any bright ideas about getting ahead of the game, whoever was in charge of RD issues disconnected the transmission line from the terminal atop the transmitter. Being the curios sort, I took the antenna lead from the CONELRAD receiver (I said it was years and years ago! ;D ) and twisted it around the center lead of the transmission line to the tower.

The result was something of an electronic equivalent of taking some kind of narcotic party drug! The stations that I could tune in was amazing. But that was with a little over 100' of tower, not a 3 meter whip.

 

[Nostalgia]

GRC, you speak of every low band ham radio operators wet dream! Picture an AM tower at the high end of the band, just adjacent to the 160 meter ham band. Yow wee!!!

I heard many stories back in the day at daytimers, or stations that shut down at midnight, engineers/hams would swoop in and hook up to the stick until sun up!

 

[Ermi Roos]

While it is true that Trainotti's more recent work seems to allow considerably lower ground screen radii for electrically short vertical monopoles than the classic paper by Brown et al., it should be remembered that Trainotti has access to considerably more sophisticated tools for mathematical analysis than were available 73 years ago. To paraphrase Sir Isaac Newton: if Trainotti has learned more than his predecessors, it is because he has built upon their work.

[Newton's actual statement is a lot less prosaic than my paraphrase: "If I have seen further than others, it is because I have been standing on the shoulders of giants."]

 

[rfry]

While it is true that Trainotti's more recent work seems to allow considerably lower ground screen radii for electrically short vertical monopoles than the classic paper by Brown et al., it should be remembered that Trainotti has access to considerably more sophisticated tools for mathematical analysis than were available 73 years ago.

That is true, but then the BL&E study is based on carefully-done, real-world measurements.

Pursuing the theme of the original post in this thread, the measured data in Figure 34 of the BL&E paper (link below) allows us to compare the performance of a monopole using 113 buried radials each as long as the height of the monopole, with that same monopole using 113 radials 3 X longer than that height.

Radial lengths of 45 feet are equal to the height of a 49-degree monopole on the 3 MHz test frequency used by BL&E. Figure 34 shows that the equivalent field at 1 mile from a 49-degree monopole using 113 radials each 45 feet long is about 140 mV/m.

Using 113 radials each 135 feet long with the same 49-degree monopole produces a field of 186 mV/m at 1 mile.

The maximum theoretical inverse distance groundwave field at mile for 1 kW of radiated power from a perfect 49-degree monopole with a zero-ohm r-f ground connection is about 191 mV/m, as shown in Figure 34. So the 49-degree monopole with 45-foot radials is only 53.7% efficient, while the same monopole with 135-foot radials is 94.8% efficient.

The BL&E paper provides rather conclusive evidence that (1) even when monopoles are not extremely short in terms of wavelengths, their radiation efficiency is greatly dependent on the length and number of buried radials they use, and (2) highest radiation efficiency from a monopole of every height from zero to 90 degrees using buried radials requires at least 113 radials each at least 0.274 free-space wavelengths long -- preferably 0.41 wavelengths.

http://i62.photobucket.com/albums/h85/rfry-100/BLE_Fig_34.gif

RF

 

[Ermi Roos]

I only disagree with the term "conclusive," since since field strength measurements tend to be inaccurate, even when carefully done.

 

[rfry]

...field strength measurements tend to be inaccurate, even when carefully done.

Actually, medium-wave field intensity measurements made with professional, calibrated meters can be quite accurate, when carefully done.

Below is a clip from the specifications of the most commonly-used FI meter in the US AM broadcast industry.

Field Intensity Range: 10 uV per meter to 10 V per meter
Accuracy of Calibration: 1%, referenced to NIST Standard field
(Calibrated at 220 mV per meter)
Accuracy of Range Attenuator: 2% over entire FI range and tuning band

A measurement accuracy of 3% of field (the sum of the above accuracy specs) is an error of less than 0.3 decibels in the field intensity measurement.

Given Dr Brown's credentials and experience in lab and field work, and his position as a lead engineer at RCA Laboratories in Princeton, NJ, it would be surprising if his measurements were not as least as accurate as shown above. But even if BL&E were off by twice that, it would add about 0.6 dB of uncertainty to their measured fields.

A 0.6 dB error applied in opposite directions to the measured values for 135-foot and 45-foot radials shown in Figure 34 of their 1937 I.R.E. paper still would not change the conclusion that for good antenna system radiation efficiency, the lengths of 113 buried radials should be at least 0.274 wavelengths (preferably 0.412 wavelengths) regardless of the monopole height, and should not be limited simply to the length (height) of the monopole they are used with.

Here is a link to more specs of that field intensity meter:

http://www.pi-usa.com/fim2241/fim2241g.htm

RF

 

[frdanbecker]

Doesn't the bandwidth get really narrow (for part 15) when the ground resistance is very low?

I am currently doing this experiment with a Rangemaster about 4 feet off the ground, by measuring the increase in range (and possible reduction in bandwidth) as I add more and more chicken wire to the ground.

I suspect there will be audio distortion before I get too far beyond a 15 foot radius.

Thanks to all for your very helpful posts.

 

[rfry]

Doesn't the bandwidth get really narrow (for part 15) when the ground resistance is very low?

That is the trend, but it can be reduced by using "large" diameter conductors for the radiating structure -- although that will change the conditions under which some Part 15 AM transmitters were FCC-certified.

RF

 

[Ermi Roos]

Characterizing an open-area test site is pretty tricky business. The current standard for accuracy is +/- 4 dB, but the actual error, when changing experimental conditions for tests, can be even greater. Trainotti was the head of an antenna laboratory for 40 years, and he never saw, from actual measurements, the 6 dB difference for antennas over a ground plane when transmitting and receiving. The proof of the 6 db difference did not come from measurements, but from theoretical calculations. This is why theoretical calculations are important.

[To explain my reference to the 6 dB receiving and transmitting antenna gain difference, this was a very long debate between Mr. Fry and myself on part15.us. Fry maintained that the principle of reciprocity requires the same antenna gain when transmitting and receiving. I maintained that it has been known since 1959 that a vertical antenna over ground has 6 dB more gain transmitting than receiving. I brought Trainotti into the debate, and he found that the 6 dB gain difference does, indeed, exist, and applies to all antennas over a ground plane, whether vertical or horizontal. An IEEE article about this subject is now in peer review, and a second article is being written.]

 

[Ermi Roos]

A very narrow bandwidth for the short antennas for Part 15 AM has been predicted, but is rarely seen. For example, with no loss resistance, the Q of a 3 m radiator near the upper end of the AM BCB is 30000, which corresponds to a bandwidth in the vicinity of 50 Hz. It is well-known that there are losses due to ground and loading coil resistances, but what is frequently neglected from consideration is the source resistance of the transmitter. More likely antenna system Q is about 50, giving an RF bandwidth of somewhat more than 30 kHz, or an audio bandwidth of 15 kHz.

Not only wouldn't audio pass a very narrow bandwidth, but it would be very difficult to keep the antenna tuned to the operating frequency.

 

[rfry]

Characterizing an open-area test site is pretty tricky business. The current standard for accuracy is +/- 4 dB, but the actual error, when changing experimental conditions for tests, can be even greater.

An explanation is in order for the reason why a field intensity meter specified and calibrated to be accurate within 0.3 dB of an N.I.S.T. standard field on medium wave could result in an error of as much +/- 4 dB in the value of field intensity that it measures.

If such a +/- 4 dB error was, in fact, even possible for the measurements needed to prove the groundwave field intensity values required by the FCC in various directions from directional AM broadcast station arrays, that would be more error than could be tolerated by the FCC and/or licensed broadcasters and their consultants -- even though those measurements were not taken on a qualified, open-area test site.

Reciprocity (whether proven or unproven) has nothing to do with this matter.

//

 

[Ermi Roos]

Mr. Fry and I clearly understand the subject of making radiated emissions measurements differently.

I mentioned the reciprocity debate on part15.us primarily because Trainotti initially had the same opinion about reciprocity as Fry. Trainotti had been the head of the Antennas and Propagation Division of CITEFA for 40 years, and he had made numerous measurements of both the receiving and transmitting gains of numerous antenna designs. He studied both full-sized antennas and microwave analogs. He told me that he never noticed any difference between the transmitting and receiving gain of any antenna. He now agrees that, over a ground plane, the transmitting and receiving gains differ by 6 dB. The significance of this fact to this thread is that a 6 dB gain difference can be too small to be noticed by field strength measurements. This indicates an inherent difficulty in making accurate field strength measurements. We might also look at the Liberty 1640 NOUO that was recently discussed on this board. Everybody insists that the transmitter involved is a Rangemaster elevated 20 feet; but no Rangemaster, elevated by any amount, can produce the field strength recorded in the NOUO. The two alternative possibilities are that somebody placed a higher-power transmitter in a Rangemaster box, or that the reported field strength measurement was in error. Which of these two scenarios is more plausible?

Trainotti was so convinced that reciprocity was a settled matter that he did not, at first, consider it to be a subject open for serious discussion. He was willing to receive the information I had about reciprocity failure only because he knew Kenneth Norton personally, and he wanted to know what his old friend was talking about. Similarly, I asked Fry to e-mail me his home address so that I could send him the information (my file is too big for an e-mail attachment). Fry, unfortunately, did not respond.

 

[Tom Wells]

It would be fascinating to find a site where an AM station has ceased to exist, or has moved to a new location, and after tower removal, the ground system was still intact. mount the classic 3 meter whip used today in Part 15 stations on the old concrete foundation of tower and be able to connect to the legacy ground system.

I suppose the old ground system will have been captured for salvage by then, but if not, what a great location for experimentation.

The other great experiment would be DX reception at such a site. Would attaching the ground terminal of a DX receiver to an old AM station ground system provide superior reception?

The old WNWI 1080 (1965) tower at Valparaiso, In still stands and as the station has gone to Oak Lawn IL, and the tech school is gone, they've
put up apartments and condos nearby on the old property. Maybe you could get a zoning on a single family residence within 300 ft of the stick.
As far as I know, there have been no secondary users on it for years. I heard stories in 1971 about amazing dx off this tower when the
WWV multiband receiver on the wall behind the console got hooked up to the feedline. If I recall, it used 120 radials that went to the maximum practical
distance as defiined by property lines or structures.

 

[rfry]

He (Trainotti) now agrees that, over a ground plane, the transmitting and receiving gains differ by 6 dB. The significance of this fact to this thread is that a 6 dB gain difference can be too small to be noticed by field strength measurements. This indicates an inherent difficulty in making accurate field strength measurements

The clip linked below shows field intensity measurements made by the well-respected broadcast engineering consulting firm of duTreil, Lundin and Rackley. The radiating source was a 1/4-wave monopole on 1680 kHz using 120 each 1/4-wavelength ground radials.

The chart at the top of the clip compares the field intensity measured by a Potomac Instruments FIM-41 to the theoretical maximum inverse distance field for the conditions. The inverse distance field is shown by the straight line on the chart, and the measured values are shown as small circles.

Now, if it were true that there is a 6 dB measurement error due to the failure in reciprocity of the receive antenna, then the measured fields at every distance shown on that chart could never be greater than 50% of their inverse distance value. Yet such a result is not shown in these measurements, and/or in the thousands of field intensity measurements made of AM broadcast systems going back many decades..

The differences that do appear are related to groundwave propagation losses over less than a perfect ground plane.

The paper from which these clips were copied is interesting and educational. It may be viewed on line at http://www.star-h.com/publications/nabpaper.pdf .

LINK: http://i62.photobucket.com/albums/h85/rfry-100/FI_Measure.gif

RF

 

[rfry]

We might also look at the Liberty 1640 NOUO that was recently discussed on this board. Everybody insists that the transmitter involved is a Rangemaster elevated 20 feet; but no Rangemaster, elevated by any amount, can produce the field strength recorded in the NOUO. The two alternative possibilities are that somebody placed a higher-power transmitter in a Rangemaster box, or that the reported field strength measurement was in error. Which of these two scenarios is more plausible?

The link below leads to a NEC analysis of a situation which may or may not apply to "Liberty 1640." It consists of a transmitter with 35 milliwatts of unmodulated output power elevated 20 feet above the earth, with a 3-m vertical conductor connected to its antenna output connector through a loading coil with 10 ohms of loss, and a 20-foot conductor connected to an earth ground with 25 ohms of loss resistance.

The groundwave field intensity at 30 meters calculated by NEC for these conditions is about 14,300 microvolts/meter, as shown.

The NOUO for Liberty 1640 reported a field of 73,800 microvolts/meter at that distance.

The difference in those two fields is 14.25 dB, which is much more than can be attributed to the lack of reciprocity of the receive antenna.

http://i62.photobucket.com/albums/h85/rfry-100/1640_kHz_Analysis.gif

RF

 

[rfry]

The difference in those two fields is 14.25 dB, which is much more than can be attributed to the lack of reciprocity of the receive antenna.

I was hoping someone else would post this comment about the difference I stated above, but that difference is in the wrong direction if it was caused by the receiving antenna having 6 dB less gain as a result of failing to meet reciprocity.

Instead of the readings being 6 dB too low for the field calculated from an elevated transmitter setup with 35 mW of output power per the linked NEC results, it is over 14 dB higher than that field.

//