|Back to Ordering||
ELECTROSTATIC RECEIVER NOISE
By Mike Norton KE4NS
This document is available as a complete document as a PDF file. It is approximately 2.5 megs in size and includes all of the table and images that are not included in this HTML presentation. Click to download ELECTROSTATIC RECEIVER NOISE as a PDF file.
Causes and Prevention:
Have you ever wondered why your receiver noise floor becomes elevated above ambient, or more important what you can do to minimize or eliminate it entirely? With the increased sensitivity of receivers today and the extended range expected by all, electrostatic discharge (ESD) can and does significantly degrade system performance. This happens due to the ESD producing broadband noise raising the noise level above ambient. This article will describe several phenomena causing this to happen and what can be done to minimize or prevent it from occurring. Everyone is aware of noise and how it affects performance but it may be more of a factor than you realize.
Three years ago I began working for Pegasus Message Corporation, a meteor burst communications company. Only after daily monitoring of signal levels just above galactic noise did I realize the true importance of noise. Increases in the noise level of only a few dB can entirely mask out weak signals. As the level becomes greater, a good receiver signal to noise ratio can easily be swamped. I felt compelled to write this article and share the information that I collected.
Propagated noise crashes from distant thunderstorms and man made electromagnetic interference we may have to live with but they are generally impulse type noise and although bothersome are not normally associated with sustained noise levels. The type of noise of major concern for this article is electrostatic discharge (ESD), its causes and prevention. First we will need to discuss how and why current flows between the earth and sky even during fair weather conditions.
Air, a mixture of gases, is largely composed of nitrogen and oxygen. It is generally considered as insulator, and would be an excellent one if all the oxygen and nitrogen molecules were in the neutral state. However, the air is actually composed of varying quantities of neutral molecules and positive and negative ions. As the number of ions in the air is increased, the air becomes a progressively better conductor. In general, gradually more ions are found the higher we ascend until, at about a height of 40-50 miles, a region called the ionosphere is reached. Here, there are sufficient numbers of ions to reflect radio waves. The ionosphere, although conductive, can be considered as a whole as being uncharged. This is due to the number of positive ions being equal to the number of negative ions plus electrons that are distributed in layers varying in height and in degree of ionization. In contrast the earth has a surplus of electrons and is actually about 300,000 to 400,000 volts negative with respect to the ionosphere (note 1). This potential difference together with the total conductive qualities of the atmosphere are sufficient to cause the earth to continually lose electrons to the ionosphere. The entire earth's surface and the ionosphere may be considered as the oppositely-charged plates of a vast capacitor with the air between them acting as a rather inferior insulator, for it leaks continuously. In addition to the presence of ions, which make the atmosphere slightly conductive, various meteorological processes called precipitation or hydrologic cycle, contribute to the leakage rate of the "capacitor": falling rain, for example, tends to bring down the less-mobile large ions toward the earth while electrons are carried upwards on rising moisture-laden air. This steady loss of electrons from the earth is called ionic current, and, infinitesimal as it is, it has been measured and amounts to about 9 microamps for every square mile of earth's surface (note 1). This current flows from the earth via the most convenient conductive path or those offering the least resistance: so most of the electrons are discharged at natural and man made points (our antennas and towers) that project into the atmosphere.
This occurs when electrically charged particles (raindrops, snow, dust, etc.) strike the antenna tower, antenna boom or elements (fig.1), inducing a current impulse in the element and thereby producing broadband noise. This noise called precipitation static is generally defined to include all external atmospheric electrical effects which produce electromagnetic interference.
Corona Discharge Noise
The other type of noise called corona is due to the flow of electrons through the antenna (tower, boom and antenna elements) into the atmosphere when a charged cloud is near the antenna (fig.2). Corona or brush discharge occurs when a charge is built up and electrostatic lines of force are developed. More lines of force per square inch appear at the sharpest points and the more likely it is that a strong field will pull free electrons from the point (note 3). Electrons pulled out from a sharp point form a corona or brush discharge. This noise can also be generated by a mobile in motion and is the reason that mobile antennas have a ball of some sort at the tip. The ball eliminates the sharp point and tries to minimize the effect. If enough electrons leave in such quantities that the air is heated and becomes ionized, a spark of electronically heated air will be visible, also known as Saint Elmos Fire. Corona is a major noise source but not the only one that should be in this category. There are also streamer currents, arcing and others. When a charge is built up on the antenna and tower or other supporting structures, it is possible for arcing to occur. The arcing will occur between any insulated parts or poorly connected ones if the charge potential raises to a sufficient level which it does frequently.
Another source of noise that should not be overlooked is antenna and tower hardware. Without the proper grounding and bonding you may constantly be looking for noise sources. Pay particular attention to any point where two conductors are in poor contact with each other. The two conductors rubbing against each other can cause different types of noise including rectification noise. Also if a feedline is not secured at short enough intervals, the wind can eventually cause enough movement to wear through the jacket exposing the shield to the tower and become another noise source. Another troublesome area is dissimilar metal joints (copper grounds to steel towers which will corrode and oxidize quickly due to galvanic action. The best way to bond these joints is to Cadweld them (note 2). Cadwelding provides a mechanically rigid, electrically superior low maintenance connection that will not corrode or loosen.
Both corona and precipitation static (or p-static) are well known and easily observable. I have seen many instances of this so called precipitation static (or p-static which is a term loosely used to describe either or both phenomena) raise the noise level 40 to 50 dB above ambient for sustained periods of time. This phenomena has been well understood for the HF spectrum and was not considered to be a problem above 50 MHz. Recent well documented reports indicate that sustained VHF and UHF p-static noise can and does occur (note 4). On HF the elevated noise floor is very apparent due to the nature of SSB signals and the increase in background noise masking out weaker signals. The noise can easily swamp a good receiver signal-to-noise ratio and eliminate otherwise usable reception. On VHF and UHF, FM being the most popular mode, the effect is not as apparent even though just as real. Receiver desensitization of up to 20 and 30 dB can be occurring without detection due too not unsquelching the receiver. This results in decreased range, noisy signals from areas normally covered, missed calls and other things generally attributed to poor conditions. In many cases the noise floor will raise sufficiently to unsquelch a receiver. I'm sure many people have heard their local repeater transmit white noise for no apparent reason.
Looking at the frequency spectrum of corona and p-static noise you might think that a given receiver range is immune to such noise. Frequency discrimination applies strictly to linear circuitry and over driving a receiver by large transients into a non-linear region is another mechanism of desensitization.
The typical characteristics of the individual electron avalanches which make up a corona discharge include an amplitude on the order of 10 milliamperes with a rise time of 10 nanoseconds and a decay or fall time of 100 nanoseconds (note 5). The noise frequency is determined by this rise and fall time. For a typical 50 ohm antenna, this results in an impulse voltage with a peak amplitude of one-half volt. This represents a huge signal for normal receiver sensitivities which are generally better than one-half microvolt. This electron avalanche causes the broadband noise and to make it worse, it is modulated with the repetition rate of the avalanches which vary from a low audio signal (a few pops per second) through the audio range and extending to at least one megahertz. You will hear a high pitched screaming noise in the receiver, which comes in cycles of two to twenty minutes duration. The pitch is constantly changing and at times sounds just like ignition or alternator noise. The noise will slowly go away as it discharges through the front end of the receiver and will repeat as soon as the antenna system charges up again. This occurs to mobile as well as fixed stations.
Noise Prevention (Wick-em-up)
Now for the good part, how to reduce or eliminate these effects. The data presented here indicates you can significantly reduce or eliminate p-static and corona noise by using static wick dischargers (fig.3). These static discharge devices bleed off the excess electrons thereby reducing the undesired electrical noise that results in receiver desensitization. Or stated another way, these discharge wicks lower the discharge threshold of the structure to which they are attached allowing the charge that builds up to be drained off before reaching the level needed to cause the severe broadband noise. This is the same device that has been used by the aircraft industry for years to minimize electrostatic interference to airborne and ground station equipment. The only difference is this model has been optimized for antennas instead of airframes to insure no detuning of the antenna even if element mounted. There are few manufacturers of aircraft today who do not offer, at least as an option, a high quality static discharger installation. These static discharge devices are just as effective in eliminating the same problems from antennas and towers. They will provide results on long wires, dipoles, yagi's for HF through UHF, verticals, TV antennas and even satellite dishes during corona/p-static charging conditions. In many cases, the discharge wicks will lower the noise level several dB even under normal weather conditions.
Testing the Wicks
In the aircraft industry the standard procedure for discharger wick ground testing is to insulate the wick from the airframe and monitor the discharge current. The aircraft is also insulated from the ground and then exposed to an ion flooding fixture (note 6). The current discharge levels are then used to determine the areas and number of discharger wicks needed. There are many more variables affecting the aircraft than antennas and towers. To test the dischargers, I assembled two identical 7 element 50 MHz yagi's. They were mounted with the elements vertical up 30 feet or 1.5 wavelength. Yagi separation was 1 wavelength and both were fed with identical electrical length feedlines (fig.4). Both yagi's were swept with a spectrum analyzer/tracking generator combination to insure resonance at the same frequency. Both antennas performed identical. Next the antennas were lowered and a static discharger was attached to each end of the boom on one yagi (fig.5) and then raised back to the 30 feet level. The yagi without the discharger was used as a control antenna. The results are shown by the graph in figure 6A. Using the static wicks can improve the noise performance as shown. As can be seen the reduction in corona discharge and p-static when using the dischargers is not trivial. These measurements were made when a front was approaching that resulted in a few sprinkles of rain at the peak of the noise level. Note that it does not have to be raining for corona or p-static discharge to take place. Just a front moving in with charged clouds in your antenna vicinity will cause the build up of electrons to take place. When the threshold is reached the noise starts as the discharging occurs. The whole theory behind the wicks is to lower this threshold which they do very well. They are effective in keeping the electrons drained off before they reach the necessary level to cause the desensitization or broadband noise. You may still hear some noise if the charging intensity is high enough but you can rest assured it would have been at least 20-30 dB worse without the wicks as shown by the graph in figure 6A. An attempt was also made to find the optimum location for mounting the wicks. The first noise graph (fig.6A) was made with the dischargeers mounted vertically one foot in from the ends of the forty foot boom. The noise reduction was very significant as shown. The next location used was on the ends of the boom in the plane of the boom (fig.5B). The increase in performance as shown by figure 6B was most likely due to the discharger being in a more prominent DC field area of the antenna. I am also in the process of collecting additional data with the dischargers mounted on the driven element of a seven element yagi and comparing the effectiveness of boom verses driven element and then the parasitic elements. I didn't have enough data collected to include in this article but I will follow up with additional data for the driven element and parasitic mounted dischargers. Various methods were used to record the antenna and discharger performance. The noise graphs were prepared from data collected by an eight channel analog to digital converter interfaced to a computer. Also a three range ammeter was built (0-100 picoamp, 0-100 nanoamp and 0-100 microamp) to measure the discharge current of individual dischargers during varying weather conditions. After a normal baseline is established for fair weather discharger current, you can often see a front approaching 20 to 30 miles away. To do this will require an ammeter resolution of at least 1 picoamp. It was also interesting to note that the onset of audible noise did not happen at the same time the discharger current increased above the normal baseline. Usually the baseline current had to increase 10 times before there was detectable audio noise. Even though not detectable in the audio, the receiver front end can still be desensitized 10 to 20 dB.
Mounting and Availability
The dischargers are made and patented by Robert L. Truax, president of TCO Manufacturing Corporation and the Truax Company. They are relatively inexpensive and only 3/16" by 6-3/4" long and require only one small tapped 4-40 hole for mounting. Just 1/4" of the base of the discharger is metal so there will be no significant detuning of the antenna even if element mounted at frequencies of 470 MHz or less. If element mounted, the antenna should be at DC ground which most are to provide equipment protection. If element mounted and the point is not at DC ground, a 100,000 ohm to 1 megohm resistor can be added to provide the DC path. Figure 7 shows a discharger end mounted on the boom of a 430 MHz circular polarized antenna used for satellite work. You can demonstrate how effective the discharger is by insulating it from the boom and running a single wire to a zero center 150 microamp meter to ground (fig.8), the meter should be protected by back to back diodes. You will see up to 100 micro amps of discharger current when the noise level rises 40 dB above ambient. The readings may vary for different installations and heights but you will see the correlation. The meter needs to be zero centered because the discharge can be positive-point corona or neqative-point corona. These dischargers are not lightning deterrent devices. Although they are installed in the most prominent DC field area they may in many cases be a sacrificial device and thereby reduce damage to more expensive equipment.
There is no real maintenance to perform on the dischargers once they are installed and they rarely need to be replaced. The only precaution is to make sure they show continuity to whatever they are attached. There must be an electrical path from the structure to the discharger. Other than checking for physical damage and a good mechanical connection while doing other antenna or feedline inspection they are basically maintenance free. Due to the greater than 20 megohm resistance of the discharger you may need a driving potential of at least 100 volt to measure the resistance.
If you are interested in purchasing a "Static Buster" Click Here
Back to the top of the page
Back to Ordering
This web page created and is © copyright March 29, 2000 by Kevin Custer W3KKC
This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.