Free (At Least, Extremely Cheap) Noise! Part 2
In the first part of this series we developed some basic understanding of the sources of unwanted radiofrequency noise, and how much of a significant hindrance to ordinary radio communications it constitutes.
Some “home experiments” done by the Curmudgeon during an unanticipated regional power failure led to the conclusion that the external urban radio noise level at 14 MHz near his house typically measured about 12 microvolts at a receiver input, while during the blackout the ambient noise level was reduced to about 0.2 microvolts under identical equipment conditions. While these were not “laboratory grade” measurements, they clearly indicated that high ambient noise level problems do exist, especially in metropolitan environments. See https://lbagroup.com/blog/free-at-least-extremely-cheap-noise-part-1.
To be thorough about these results, we must examine several other possible explanations for the high noise level. Some radiofrequency noise does occur from natural processes, and perhaps one or more of these processes might be contributing to the large urban high noise levels. Cosmic noise sources can contribute to the overall background, but they become important above 15 MHz, and this measurement occurred below that frequency. Natural terrestrial noise sources are very important at VLF but become far less significant above 10 MHz.
Suffice it to say, the observed S-meter value during the power blackout, when both cosmic and world-wide terrestrial natural noise sources were in operation, was very close to the thermal noise background of the receiver itself; at most these natural sources contributed one or two additional dB of noise. This is far below the typical daily urban noise level, which is some 35 dB higher than the thermal background. Thus it is unlikely that the majority of today’s long-term HF noise problems are due to naturally-occurring sources.
In response to Part 1 in this series, reader James Medlock commented that much of today’s noise problem lies in the VHF and UHF portion of the spectrum, and not just at HF. He is correct, of course, but there is a reason why the VHF/UHF problems are not highlighted to the same degree as the HF problem: the VHF/UHF noise problem is far less easily observed.
In the HF spectrum, voice communications are overwhelmingly transmitted using envelope modulation modes (i.e., double-sideband, inserted carrier and single-sideband, suppressed carrier amplitude modulation). Of course, there are also a number of character-based transmission modes in use at HF, but the transmission conditions using these modes are not typically directly observed by the operators. Most observations of transmission problems are made aurally.
And envelope modulated modes are notably susceptible to random radiofrequency noise; when present it is always audible in the received signal.
In contrast, most VHF/UHF voice communications are transmitted using either frequency modulation or multi-level digital modulation. [A notable exception is the Aeronautical Mobile Service, which has always used envelope modulation transmission modes.] Both of these VHF/UHF modes have some degree of noise immunity inherent in their operation. Thus ambient noise may well affect the reception of RF signals, but the listener may well not be aware of the interference unless the received FM signal is truly marginal or the digital signal “falls off a cliff.” Additional comments and/or measurements of VHF/UHF ambient noise levels from readers who have access to VHF/UHF receivers with AM detectors (or who are using spectrum analyzers) to gauge the noise levels at these frequencies, would be welcomed. [Of course, much engineering/scientific work on these problems has also been performed and the results have been published.]
Thus it appears that the majority, perhaps even the vast majority, of the ambient metropolitan RF noise is of man-made origin. What sorts of man-made systems produce the noise? The Curmudgeon will list several major classes, with the expectation that readers will add to the list. The first one, and perhaps the most obvious, is the modern electrical power distribution network. By definition this wire conductor-based network carries large 60 Hz currents, in the case of utility electrical transmission networks on the scale of hundreds to thousands of Amperes, equivalent to as much as Megawatts of electrical power. Moving charges produce EM fields, but the primary 60 Hz field is not the real problem here (in reality, the harmonics of 60 Hz also ride the lines, but above the fourth harmonic the effects become small.).
But it’s the voltages involved that are more the source of radiofrequency noise. These voltages range from a minimum of 4 kV on some old local distribution networks, up to 500 kV and even higher for major transmission lines. If the power transmission/distribution networks had been machined out of a single slab of copper or aluminum, the world would be subject to far fewer noise problems. Instead they are constructed from various lengths of wire conductors that are spliced or clamped or bonded together, sometimes using jumpers where wire segments meet.
If a junction isn’t perfect, a small spark gap can easily result. Given some substantial voltage on the conductor, an arc can be struck across that gap. And then that arcing gap recreates radio history: “King Spark” again takes to the air, aided by the long runs of the electrical conductors which form natural, multi-wavelength antennas.
Dirty, cracked, broken insulators, especially old glass and ceramic units, to which the conductors are tied or are suspended, also can harbor spark gaps. And corona discharge, the leakage of electrons directly into the atmosphere, can also produce both audible and electrical noise.
The modern utility practice is to place electrical distribution networks in underground conduits, and the thickness of earth above them does help considerably to suppress radiation of RF noise. But the original, old, weathered aerial distribution networks, found in the older parts of cities, are the major offenders, and while there are programs underway to “underground” these circuits, in reality it may be another half-century before the work is finished.
Radiation of EM noise from power networks, which can be detected as high as 500 MHz, is technically a violation of the FCC Part 15 Rules, and the FCC could in principle visit fines on the utilities if it wished. This is very unlikely, except perhaps in the most egregious of cases. Nevertheless, to varying degrees the individual electrical utilities do show some willingness to “clean up” noise problems as they are identified by customers. Some utilities are very cooperative and have on staff dedicated troubleshooters with professional training and equipment, who will attack reported problems. At the opposite end of the scale, other utilities will resolutely deny that problems even exist.
Next on the list are the large, power-consuming pieces of electrical machinery. These are classified by the FCC as “unintentional radiators:” they produce and transmit some radio-frequency energy, although they were not designed to do this and the transmitted energy doesn’t aid in the work of the machine. This class includes large motors, electrified transport systems, welding stations, RF-fired ovens, electric fences, etc. They too are covered by Part 15 of the FCC Rules, although enforcement is very rare.
Next add in the radio-frequency industry itself: broadcasters, public and private radio common carriers, radars, consumers’ local area data networks, etc. The space around large metropolitan areas is rich with these intentional radiators, which contribute to a very large EM energy density in the area. Given these densities, mixing occurs at many spots and additional new mix product frequencies are generated and transmitted. The result is a “dirtying up” of the spectrum leading to rising noise floors. We’ll have more to say about these problems later.
Automobile ignition systems play their unique parts in raising the overall noise level, although the situation in this area has improved considerably over the last fifty years.
The final contributor to the noise pollution is a new entrant into the classification, one which probably wouldn’t even have been listed as late as twenty years ago. These are the ubiquitous digital devices which today characterizes our culture. As is well understood, almost all digital devices operate by generating rectangular pulses and/or square wave signals to represent the digital data being processed. In order to operate at the high speeds characteristic of modern computing devices, these pulses must necessarily have very fast rise and decay times. Thus as Monsieur Fourier has elegantly shown us, very rapid processes in the time domain produce signals in the frequency domain that are very rich in harmonics!
That situation, by itself, could be handled. The application of well known circuit design techniques involving shielding, bypassing, and control of conducted energy can tame the racket produced by most digital devices. But these techniques raise the cost of a product while, for the average consumer, their salutary effects are not even noticed. Thus the dilemma: sales margins in the marketplace are already thin, and for any product the addition of more components whose only functions are to reduce (unrecognized) noise is only poorly justified at best.
Therefore the production costs of preventative treatment are shaved by omitting the filtering, under the observation that enforcement of existing standards is very unlikely. And the world become ever more populated with “shrieking little digital devices” producing a seemingly endless supply of low amplitude, discrete, often harmonically-related frequencies that scatter across the RF landscape and create an ever increasing noise floor. Today a digital component has been added to the historical analog noise background, and the digital devices have become the scourge of our era!
In the next part we will move onto looking at the effects, costs, and possible remedies for ALL THIS NOISE!
What do you think?
“Let’s save the universe for RF!”
The Old RF Curmudgeon
Since 1963, LBA has been providing RF equipment and engineering consulting services for
radio and television broadcast and wireless communications.