"Grounds" for Confusion, January 1960 Electronics World

Robert Gary waxes philosophical on the subject of ground in his Electronics World article, "'Grounds' for Confusion." He is justified from the viewpoint of someone attempting to make sense of how something as seemingly fundamental as Earth ground is not a constant. The layman probably doesn't care. Practitioners in the electrical and electronics realms who deal only with low frequencies and short distances might occasionally be affected by differences in ground potentials, although they might not realize it is the cause of their problems. Those with more than a casual involvement (designers, installers, and maintainers as opposed to only users) in high frequencies and/or long distance signal interconnections are likely to be intimately familiar with the effects of ground potential differences. RF Cafe visitors are undoubtedly members of the latter group. Bad grounds can render a system utterly nonfunctional, marginally working, or intermittently functional (e.g., dependent on soil conductivity due to weather). So, too, are electrical power distribution engineers and technicians who must assure a proper ground connection to substations, localized service lines, and individual houses and buildings. Depending on soil condition, some installations require periodic application of chemicals to the ground in order to bring conductivity up to standards. To illustrate how only ground has been an issue with radio installations, see the ARRL's 1951 QST magazine article entitled, "Ground Resistance and Its Measurement." The March 2019 issue of QST (p50 - not listed in the TOC for some reason) had an interesting article by Ward Silver (N0AX) on multipoint ground bonding that described a method for chemically fusing (welding) interconnecting ground wires to the ground rods driven into the soil, using the Harger Uni-Shot exothermic welding kit.

TV set, train, and antenna - each has a reference "ground," but this ground may be a different concept in each case.

One word, many meanings: The type of equipment or type of voltage involved may decide what ground is.

A well-known quote by a well-known writer informs us that "a rose is a rose." Is not a ground also a ground? And, if that is the beginning and the end of the matter, why should any confusion exist about so simple a concept? However, if you choose even the clear-cut rose as a theme and let your mind play with it idly for a minute, you can come up with a surprising range of variety in associated ideas. "Sweet Rosie O'Grady," for example, has little to do with the legendary rose fields of Bulgaria that provide essential oils for the perfume industry. Start with the word "ground" and you will come up with an even greater variety of notions, even if you limit yourself to what this word means in the language of electronics.

You may think of the concept as it appears in manuals on telegraphy. This is the earth ground or ground return, in which a single-wire line is used, with earth ground being used as the second line. This also ties in with the notion of the lightning arrester developed by Ben Franklin, the device being basically a good conductor to ground. Yet the concept seems a far cry from that of a TV receiver chassis, which had better be a good ground for the r.f. and other signals circulating in the receiver - but which may be "hot" or "cold" with respect to the ground on which the technician stands.

Fig. 1 - Notion of ground (A) as unipotential return path. However, no two ground points (B) are actually the same.

Even if we confine ourselves to the ground within a single item of electronic equipment, such as a radio, audio amplifier, or TV chassis, we may run into confusing effects. For example, two points may both appear to be at ground potential on a schematic diagram; yet, if these two are too close physically - or perhaps too far apart-oscillation may develop. If they are both at ground, why should it matter how near or far apart they are? Why does touching a "grounded" chassis sometimes detune a circuit using this ground? If an a.c.-d.c. receiver has a resistor and capacitor network going from the chassis to the grounded side of the power line, can the chassis be considered as a ground? Does a power transformer ground a set just because its center tap is connected to the chassis? Why do certain broadcast receivers operate better if a wire is connected to a water-pipe ground? These questions may certainly puzzle the layman and sometimes even confuse the student of electronics or the technician.

The problem arises because ground may be one or many things to different people in the same or different electrical circuits. A good starting point is the fact that electric current must have a complete circuit in order to flow, and we usually regard this circuit as being divided into a double path. In d.c. work, for example, we say that current goes from the generator or other source to the load over one path and then returns through the other. An oversimplified understanding of this action leads many of us to think that the return path is "cold" because the current has done its work and no longer has any energy left. We then begin to assume a "safe" return or ground path containing no power, which can be overlooked.

A quick look at the popular schematic representation of this action (Fig. 1A) may mislead some people. However, if we redraw the circuit as in Fig. 1B, we take into account the fact that, between any point in the circuit and any other point - even if both points are on the same "ground" line - there are inevitably some electrical factors introduced, such as resistance, capacitance, and inductance, no matter how small they may be. Thus some form of circuit action may take place between any two points, however slight it may be, and the energy within the circuit may be active anywhere.

A good example of this can be observed by a study of the tracks of any electrified railroad. The "hot" power is delivered through a suspended overhead cable or an insulated third rail, while the return path goes through the tracks. The tracks are, of course, "grounded." However, where rails are joined, heavy bronze cables carry the "cold" return current across the junction. Should there ever be a gap in the metal return path, the arcing across it will be every bit as severe as if the top wire were broken.

Fig. 2 - A u.h.f. oscillator circuit may look like any of the three representations below, depending on the viewpoint of the person who is dealing with it.

A ground, in this sense, is simply one side of a two-wire system. For our purposes, we usually choose the side to which the electrical resistance from the ground (earth, or its equivalent) on which we stand is the least. A good example of this is the automobile. Its battery, radio, and other electrical equipment are all grounded to the car body; but, because of the nonconductive rubber tires, the car itself is not connected to the earth at all. This sometimes causes shocks or small arcs when a conductor - such as a person - supplies the connection between the "grounded" car and the actual ground. For this reason, some autos use ground straps which drag along under the vehicle and electrically connect the car body to the pavement.

To further illustrate the problem of clearly defining a ground point, consider the case of a u.h.f oscillator circuit, as shown in Fig. 2. The circuit is shown in Fig. 2A as it would appear on a schematic diagram. Fig. 2B is the true electrical equivalent of all portions of the circuit outside the tube envelope. For convenience's sake we have lumped the stray and wiring inductances, capacitances, and resistances that would actually be significant at these frequencies together. The obvious components are still identified for us (C₁ L₁ R₁ etc.) but they are heavily outnumbered by the inevitable resistances, capacitances, and inductances that are nevertheless present - and we have not necessarily shown all of the latter.

Already, in comparing Fig. 2A with Fig. 2B, we can see how two different engineers, each concerned with various aspects of the circuit, could take divergent viewpoints concerning the nature of ground. Let us see how the same circuit would be viewed by still another person - a safety engineer whose sole business is the question of shock hazard. To him, everything but the plate and supply voltages (Fig. 2C) could conveniently be assumed as being at "ground" potential. We can simplify even further if we think in terms of grounding the receiver to protect against lightning hazards. Here the chief circuit element to be considered would be the insulation resistance between the primary and secondary of a transformer.

Fig. 3 - How power transformers serve to isolate earth ground from chassis ground.

Much has been written in the past on the subject of the shock hazard of electrical appliances. Like the weather, this topic has received a lot more discussion than earnest attempts at .a cure. But before we can think of eliminating shock hazards we should understand exactly what they are. Basically there are three ways one can get an electrical shock and, although this may seem facetious, they are by touching one side of the line, by touching the other side of the line, or, through special talent, by touching both sides. If we could assume that one side of the power line is properly connected to the earth on which our body stands, then it would be perfectly safe to touch that grounded side of the line, but the other side would still be "hot." We see that grounding one side of the power line immediately reduces the shock hazard by 50 per-cent. All appliances that operate directly from the power line, such as toasters, broilers, and the like, could have their case and chassis connected to the grounded side of the line, and this would limit the hazard to contact with the "hot" side. Should a "hot" wire accidently touch the grounded chassis, this would cause a short circuit and the nearest fuse would blow, but there would be no danger to any human. Similarly, if radio and TV receivers could be grounded directly to the power line, there would be no shock hazard in touching the chassis, metal cabinet, or uninsulated controls.

It has long been proposed that all house wiring make use of polarized outlets and plugs so that the correctly grounded side of the line is always connected to the chassis or cabinet. Some progress has been made in this direction in that most newly designed sockets and plugs use one wide and one narrow contact blade. But, since so much unpolarized wiring exists and since the correct grounding of one side of the power line is not always guaranteed, the introduction of polarized wiring still seems to lie in a nebulous future. The military services have standardized all of their specifications so that polarized plugs and grounded power lines are required in all new equipment.

Recently many TV manufacturers have returned to the use of power transformers to effectively isolate the receiver power supply from the a.c. power line. Fig. 3A shows the a.c. voltages on a typical full-wave rectifier circuit and indicates why there is no shock hazard here between the chassis and the earth, regardless of whether one side of the power line is grounded or not. The capacitors C₁ and C₂ represent the leakage of the transformer. Since this is usually quite low, a very high impedance isolates the secondary from the a.c. power line. A few TV receivers use simple, isolation transformers. These perform the same function as the isolation transformer that every safety-conscious service technician keeps around. The voltages on such a transformer are shown in Fig. 3B. It is true, in any electrical device, that the power-line wires themselves must be properly insulated, but, as we showed before, a short circuit would merely blow the fuse.

Fig. 4 - An RC network (A) provides only partial earth-to-chassis isolation, as equivalent circuit at 60 cps (B) demonstrates.

Having observed the voltage relationships with respect to ground when a power transformer is used, consider now the arrangement of Fig. 4, which shows a technique used with many so-called "hot" chassis in a.c.-d.c. sets. This is a rather elegant approach, because some attempt has been made here to isolate the chassis from the power line. The network of the resistor and the capacitor (220,000 ohms and 0.01 μf. in Fig. 4A) helps keep 60-cps voltages off the chassis while providing a low-impedance path for r.f. and i.f. signals to one side of the power line, which is therefore essentially at ground for these higher frequencies.

The problem with this arrangement can be seen by inspecting Fig. 4B, in which the RC network has been reduced to an equivalent impedance of 170,000 ohms (calculated for 60 cps) for simplification. Some potential does indeed continue to exist between one side of the power line and the chassis. However, even if the receiver involved happens to be plugged in with the polarity shown, the chassis will not be quite as hot as would be the case without the network. The impedance of 170,000 ohms will be in series with any person who accidentally contacts the chassis and a point at earth ground potential at the same time. This impedance will serve to drop a good deal of the voltage, so that a full 117 volts would not be impressed across the unfortunate individual, and will also serve to limit the maximum amount of current that would flow through him. Also, if the capacitor should break down, a direct connection to one side of the power line becomes possible.

In many other receivers, the chassis is actually connected directly to one side of the line, and then all metal portions of the chassis, controls, mounting hardware, and the like must be covered with insulation. This usually protects the customer from shock, but the service technician must use an isolation transformer as soon as he needs to handle the chassis.

It is true that rubber-soled shoes insulate the body from ground and thereby eliminate some of the hazard, but it is always possible to touch some grounded object inadvertently while touching the "hot" chassis. While most such accidents are not lethal, technician time can be spent in better ways than in recovering from electric shock.

In looking over the signal-input circuit of most TV receivers, we notice that neither of the two conductors of the 300-ohm antenna lead-in is grounded. The elements of the antenna themselves are also ungrounded, except possibly for some parasitic portions such as reflectors and directors through the mast. At the tuner, however, the center tap of the antenna-input circuit is connected to the chassis ground. This makes both sides of the transmission line "hot" for r.f., but grounds them as concerns low frequencies. If lightning should strike, the path would be through both conductors, the TV chassis, and the power line to earth ground.

Fig. 5 - Arrester provides lightning with a breakdown path to earth-ground.

To avoid this hazard, most outdoor antenna installations include - or should include - a lightning arrester, which short-circuits the path from the antenna to a ground, but presents sufficient impedance for r.f. so that the signal path is still direct to the receiver. A typical equivalent circuit of such a protected installation is shown in Fig. 5. Simply, an alternate path to earth is provided for lightning through two resistances. These may be actual resistors or high-resistance spark gaps, as indicated by the broken lines. In either case, the resistance prevents bypassing of r.f. to the earth but breaks down and shorts out under the stress of lightning.

Some TV installations still use a coaxial cable from the antenna to the receiver and then the outer conductor is usually grounded at the receiver chassis. If the TV set uses a power transformer it would be perfectly safe to connect the shield of the coaxial cable directly to a water pipe, drain, or other good external ground.

The phenomenon of grounding one side of an a.c.-d.c. radio's loop antenna will work safely only if the loop is connected to the set through capacitors, or is otherwise isolated. Then the grounding merely serves to establish one of the loop terminals at external ground, making the other side "hot." It will usually be noticed that, depending on the side connected to ground, the tuning controls and the radio chassis itself will become more or less sensitive to body capacity.

The aforementioned practice of grounding one side of an antenna to improve signal pick-up raises still another point for consideration. Why does this work and how can we get along without this type of ground in the case of aircraft radios, battery portables, automobile receivers, and other units that do not normally come into actual contact with the earth? The magic word that begins to provide the answers is "counterpoise." For an antenna to work best, it must have the greatest possible signal voltage developed across it. This means that, in effect, the "top" of an antenna system (that portion which is normally farthest from the earth) must be at the greatest possible potential difference from the "bottom" of the system. To establish this relationship, we seek a large body as a counterpoise that is not likely to be at or near the same potential with respect to transmitted signal that will appear at the "top" of the antenna. Where it may be used, the earth itself serves very well as such a counterpoise. Thus, the required counterpoise has come to be known as ground, because earth ground is the frequently used medium.

However, where it is not convenient to use earth ground, some other body of reasonable size is chosen as the counterpoise. This may be the body or shell of an orbiting satellite, of an airplane, or of a car - or it may simply be the chassis of a portable radio serving as its own "ground" - or should we say counterpoise? Actually, the confusion arises from the fact that, since earth ground was once the almost universally used counterpoise, we have come to call every such counterpoise a ground - whether it is or not.

When these "grounds" are brought near the earth's ground, the tuning as well as the set's sensitivity changes. Another effect observed with such devices is the accumulation of a static electric charge. Airplanes have short wire strips attached to their trailing wing edges to discharge such static electricity into the surrounding air during flight. During landings, the aircraft is carefully grounded to the runway to avoid, on a more violent scale, the shocks we often experience during dry winter days when getting out of a car.

Although technicians, experimenters, and hams should be well aware of the shock hazards around the home, and especially in their work, trivial as well as serious accidents do happen. In presenting the following suggestions, we hope to help in reducing the number of electrical accidents, and also make the work of the technician easier.

Some of our readers may think that rubber mats on the floor or the wearing of rubber soles insulate a man from ground so that, if he touches something "hot," he will feel no shock. This is true to some extent, and the wearer of rubber soles is definitely somewhat safer than the barefoot technician. But there are usually a host of other grounded items with which the "safe" person may make inadvertent contact while believing he can touch the "hot" points with immunity. We feel that a properly grounded and carefully fused a.c. power line, a grounding strap within easy reach, and an isolation transformer are the best insurance against shock. The antenna lines, all test equipment, and metal shelves should be grounded. Although not essential, we favor a grounded piece of sheet metal as a base on which to put sick chassis.

If there are portions of the set which should be insulated, we add the necessary insulation in the form of cardboard and Bakelite strips. This has the advantage that it pinpoints any "hot" danger spots and reminds us of them during work. Metal parts that lie loosely on an insulator can get a static charge from nearby high-voltage fields and then, when the unsuspecting hand reaches for them, nasty "bites" result. The author has a half-inch scar to illustrate this particular effect and therefore is now well supplied with ground clips, sheet metal work bases, and other protective devices.

One last word to those who can touch a 117-volt line with apparent immunity. Body-moisture content varies with age, climate, living habits, and a host of other factors. If you were getting no more than a mild tingle from 117 volts three years ago, don't believe that you have life-long immunity to electric shock. Even a lion tamer may wind up his career with a "bite."

"Grounds" for ConfusionJanuary 1960 Electronics World

"Grounds" for Confusion
January 1960 Electronics World