October 1960 Popular ElectronicsTable of Contents
People old and young enjoy waxing nostalgic about and learning some of the history of early electronics. Popular Electronics was published from October 1954 through April 1985. All copyrights are hereby acknowledged. See all articles from Popular Electronics.
If you are looking for a very thorough treatise on transformers, from AC line frequency up through audio and RF frequencies, but without a lot of formulas to distract you, then this article by Ken Gilmore is it. He begins with the fundamentals of a current setting up a magnetic field, talks about mutual coupling and induction, step-up- and step-down transformers, autotransformers, iron and laminated core transformers, multi-tap voltage and impedance matching transformers, and even touches on the mage-size transformers in electrical distribution system substations. Transformers fundamentally work the same way today as they did when first invented in the 1830s. Equations for simulations have been refined to the nth degree, but for the vast majority of mankind this article will suffice to provide a better-then-layman's knowledge of the principles.
See all articles from Popular Electronics.
By Ken Gilmore
A fundamental coupling device, the transformer is one of electronics' most capable magicians - here's what's behind its electrical sleight-of-hand and how it performs its multitude of valuable tricks.
The electrical power that makes your light bulbs glow, runs your refrigerator, and operates your hi-fi set comes into your home at a potential of about 115 volts. Yet if you were to climb the utility pole outside and measure the voltage there, it could turn out to be as high as 6000 volts. If you kept on climbing poles at other places around town, you might find voltages as high as 120,000 volts!
Even in your home, some appliances - air conditioners, clothes dryers, electric ranges, and other heavy-duty equipment - may operate at 230 volts instead of the usual 115 volts. And if you probe into your television set, you'll find an even wider voltage range. For although your TV draws its power from the wall plug - and power there is at 115 volts - your set has the ability to change this voltage into a number of different values, so that each tube and circuit can operate under the exact conditions it likes best. Consequently, in some places, you will find values as low as one or two volts; in others, values as high as 15 or 20 thousand volts.
Electric power, one of our most useful servants, becomes tremendously more useful when we can change it at will to dozens, or even hundreds, of different voltages. Fortunately, we can make these changes easily and economically with a device known as the transformer.
Transformers are all around us. One - the gadget about the size of a large garbage can hanging near the top of utility poles - changes the 6000-volt transmission-line power into the 115 and 230 volts you need. Another - this one about as big as a flashlight - takes 6 or 12 volts from your car's battery, and changes it into the 10,000 or more volts needed to fire your spark plugs. Still another - a square can a little bigger than your fist - channels high-fidelity electrical signals into the speakers of your hi-fi set.
We'll talk more about these special applications - and others like them - a little later. Right now, let's get down to the business of seeing just how a transformer goes about performing this valuable electrical sleight-of-hand-changing one voltage into another.
When an electric current flows in a wire, a weak magnetic field is set up around it. If we twist the wire into a coil, the weak field around each turn of the wire is reinforced by the fields around the other turns; the result is a much stronger field.
If an a.c, current flows in the coil, the magnetic field builds as the current flows in one direction; dies down, or decays, as the current returns to zero; then builds in the opposite polarity as the current flows in the other direction. You can think of the building and decaying magnetic field as a pulsing, invisible force, expanding and contracting as the current reverses its direction of flow. As the field builds and decays, the magnetic flux lines (the circular lines in the diagram) cut back and forth through the coil.
Now suppose we put another coil of wire next to and in line with the first, although not actually touching it. As the magnetic field expands and contracts, the flux lines will cut back and forth through the second coil as well as through the first one, and a voltage will be induced in the second coil. This is called "mutual induction," and is the basis of all transformer action. Because of this property, a simple transformer can be made - and many are - simply by placing two coils of wire close together and applying an alternating current to one of them.
The main value of a transformer lies in the fact that the ratio of the voltages in the two coils can be controlled by the number of turns of wire in each. To put it another way, if the secondary (the coil into which voltage is induced) has ten times as many turns of wire as the primary (the coil across which the original voltage is applied), then the secondary voltage will be ten times the primary voltage. In such a case we have a step-up transformer.
On the other hand, if the secondary has only one-tenth as many turns as the primary, the secondary voltage will be one-tenth the primary voltage, and we have a step-down transformer.
In the above calculations, we have assumed that all magnetic lines of flux, as they expand and contract, cut all turns of the transformer. The magnetic coupling in such a case would be 100%. Of course, in practical transformers a few lines of force manage to stray outside the useful area. But by careful design, engineers are able to produce transformers with efficiencies of 80%, 90%, and even more. In fact, for the purposes of most calculations, transformer efficiency can be considered to be virtually 100%.
Even though we can get a higher voltage from a transformer than we put into it, the transformer is not capable of creating power. What we gain in voltage, we lose in current. On the other hand, if we step down the voltage, we get more current.
If the current flowing in the primary of the step-up transformer in the diagram above is 5 amperes and the voltage 110 volts, the power consumed in the primary is 550 watts. Since the output voltage is 1100 volts, or ten times as much, we would have available only one-tenth the current, or 0.5 ampere. Thus, even though we can juggle voltages and currents at will, the output power is 550 watts - the same as the primary input. (Actually, the output power would be slightly less than 550 watts, due to the small losses in efficiency mentioned earlier.)
So far, we have described a transformer as two coils of wire, placed close together along a common axis. Although some transformers are actually built this way, most use other types of construction. Instead of being placed side by side, the two coils are usually arranged with one coil inside the other; this gives much closer and more efficient magnetic coupling.
For use at low frequencies, designers wind the two coils around a common iron core. Since iron is a much more efficient conductor than air, the magnetic field built up is much stronger. That is, almost all the magnetic lines of force developed by the primary winding are gathered up by the iron core and shaped so that almost all cut through the secondary winding. Therefore the efficiency of the transformer is greatly increased.
The diagrams at left show the three principal types of iron-core transformers. First is the open-core transformer which, while possible, is never used because of its relative inefficiency - a large part of the magnetic field would still have to be in air, rather than in iron. The closed-core transformer is considerably more efficient; and the shell core transformer is most efficient of all. The shell-core type has another advantage: since the flux path is almost entirely contained in the iron core, it is less subject to disturbances by external magnetic fields than other types, and it doesn't disturb other nearby circuits as much.
The first transformer ever made was simply an iron ring with two 2-layer coils of wire wrapped around it. Its inventor was Michael Faraday, the great English electrical pioneer. He discovered electromagnetic or mutual induction - the principle upon which the transformer works - in 1831. When he connected his primitive iron-ring transformer as shown, the galvanometer needle jumped as the switch was closed.
Although Faraday's device was a true transformer, its losses were high. Today's modern, refined transformers have assumed a wide variety of sizes, shapes, and characteristics as engineers have attempted to minimize the losses that are a part of every transformer's operation.
Transformer losses come from many different sources. First, not every magnetic flux line cuts the secondary - some simply travel out into space, consuming energy from the primary, but doing no useful work. This loss is called flux leakage. Designers minimize it by careful physical arrangement of the coils and core. Sometimes the primary is wound on the core first, then the secondary applied on top. At other times the secondary is split into two layers with the primary in between.
The so-called copper losses are caused by the electrical resistance of the transformer windings. Although copper is a good conductor, it has a measurable resistance, as does any conductor. When current flows through this resistance, heating takes place and power is wasted. As a result, almost any transformer will feel warm to the touch when operating normally, and some are actually hot.
Since the iron core itself, as well as the coils, is cut by the expanding and contracting magnetic field, a current is induced here, too. As this eddy current flows in the core, it steals energy from the primary circuit and dissipates it as useless heat. The eddy current flows at right angles to the magnetic flux. It can be reduced by substituting several thin layers of iron for the solid core. These thin layers - laminations - are separated by layers of glue which electrically insulate the laminations from each other. In practice, a small eddy current is set up separately in each lamination, but the total loss is much less than for a solid-core transformer.
Still another core loss is caused by the alternating current itself. Since this current reverses its direction 120 times a second, the iron core - in effect, an electromagnet - must continually reverse its polarity. And since the minute magnetic elements in the core tend to resist this change, power must be expended to realign them. This is called hysteresis loss. Engineers reduce it by building transformer cores of steels which change magnetic polarity with comparative ease, so that less power is consumed in making the switch.
Since the turns of wire in a transformer are close together, there is some distributed capacitance between the turns, between different layers of windings and between separate windings. This capacitance, though small, is cumulative. Like a small capacitor connected across the transformer, it shorts out some of the voltage developed across the windings. At low frequencies (the usual 60 cps of house current, for example) this loss is unimportant, but at higher frequencies engineers must go to great lengths to minimize it.
Another small loss is caused by the imperfection of transformer insulation. A small leakage current will flow through almost any insulator, and thus absorb some of the transformer's power. This is known as dielectric loss.
Then, too, particularly at high frequencies, a transformer can begin to act as a small but efficient radio transmitter, and actually radiate power like a broadcast antenna. This is called transmission loss.
Most of these losses, under normal conditions, are minor, but at times they become serious. For example, eddy current losses are small at power-line frequencies, but at the high end of the audio spectrum - say around 20,000 cps - they become significant. This means that a poorly designed transformer in the output stage of a hi-fi amplifier will operate much less efficiently at 20,000 cps than at 1000 cps; the result is poor frequency response.
To minimize eddy currents designers specify thinner laminations. Where laminations 20 to 25 thousandths of an inch thick are used in power transformers designed to work at 60 cps, audio transformers rarely have laminations thicker than 10 or 15 thousandths of an inch. For really good hi-fi reproduction, lamination thicknesses may range from ten thousandths of an inch all the way down to only one thousandth of an inch.
As frequencies go still higher, even one thousandth of an inch is too much, and eddy current losses become excessive. Consequently, r.f. transformers frequently have cores made of minute grains of iron suspended in an insulating material and compressed under high pressure into a solid mass. Since the grains are insulated from each other, they break up the eddy current path and help reduce eddy current losses.
As might be expected, the size of the iron granules becomes important as the frequency increases, since at high frequencies eddy currents are even set up within the individual granules. Granules several thousandths of an inch thick are satisfactory below 100,000 cps, but as the frequency goes higher the particles cannot be larger than several millionths of an inch thick.
A new type of magnetic core made of iron ferrite has recently allowed designers to build iron-core transformers to operate at frequencies higher than ever before. These ferrites - varieties of iron oxide, or rust - are valuable because they have magnetic properties, and yet are insulators and do not conduct current. Because of the unusual construction of these transformers, no eddy currents form.
If you have bought an ultra-portable radio recently, you are benefiting from ferrite-improved transformers. Miniature radios of even a few years ago had loop antennas at least 8 to 10 inches long and almost as high to collect enough signal to operate. Now ferrite-core antennas, far more efficient because of their magnetic core but not susceptible to eddy current ills, can be built as small as a short pencil. As a result, portable radios can now be produced smaller than they have ever been produced before.
In many applications, particularly for very high frequencies, air-core transformers are used. The coils are wound on a non-magnetic form such as Bakelite or polystyrene. The coils may be concentric, or end to end. Frequently one is movable, so that the degree of coupling between them is adjustable.
One of the biggest problems in high-frequency transformer design, particularly where multiple layers of winding are involved, is stray capacitance. If a regular winding were used, with adjacent layers lying parallel to each other, this capacitance could become intolerable. Consequently, layers are frequently spiraled back and forth as in the transformer shown in the drawing at right. This makes adjacent layers cross each other almost at right angles instead of being parallel, and stray capacitance is materially lowered as a result.
The transformer invented in the 1830's wasn't put to work outside the laboratory until 1885 when William Stanley, an engineer who worked for George Westinghouse, designed and tested a transformer power-distribution system. He used a 500-volt generator and fed the power directly into a 4000-foot transmission line. A transformer to step down the voltage to 100 volts was used at the other end of the line.
Westinghouse wasted no time in putting Stanley's superior transmission system into operation. That same year he built the first plant especially designed for transformer power distribution in Buffalo, N. Y. It went into use on November 30, 1886. His generator produced a 1000-volt, 133-cps output which was fed directly into the transmission line, and stepped down at the customer's home.
In spite of its obvious superiority, however, high-voltage transmission with transformers did not gain immediate acceptance. Thomas Edison, for one, was violently opposed to a.c. power, and he used his tremendous prestige to gain support for his own d.c. system. Consequently, it was not until many years later - after the turn of the century - that high-voltage a.c. power distribution became common. Even today there are a few places - some areas of New York City, for example - still receiving Edison's legacy of d.c. power.
But giant power transformers and their complex distribution stations now dot the landscape all over the country. The one shown on the next page, one of the largest ever built, can handle enough electric power for a city of 500,000 inhabitants.
Why use transformers for power distribution? The efficiency of transmission is tremendously increased by stepping up the voltage to as much as several hundred thousand volts. Also, a given size of wire can carry far more power at high voltage than low, saving money in transmission costs. Let's see why.
As an example, let's take a transmission line of No.1 wire 10 miles long - that's a conductor about the size of your little finger. The resistance of one such wire 10 miles long is about 7 ohms. (Actually, the resistance of each wire in the transmission pair is 7 ohms but for the sake of illustration let's consider just one.) Now let's say that we transmit a current of 120 amperes at 120,000 volts (a common transmission-line voltage) over the 10 miles. The total power fed into the line at the generating station is 14,400,000 volt-amperes.
With 120 amperes flowing in the 7-ohm line, the voltage drop over the ten miles will be 840 volts. Thus, the output voltage will be 119,160 volts; 120 amperes at 119,160 volts gives a 14,299,200 volt-ampere output. Along the line we have lost 100,800 volt-amperes, dissipated by the resistance of the transmission line. This seems like a lot of power, but if we figure it in terms of percentage, the loss amounts to a negligible 0.7% of the total fed into the line.
Now let's see what happens if the supply voltage is reduced to only 12,000 volts. The power input is now 1,440,000 volt-amperes. We will assume that the transmission line is still carrying 120 amps - its maximum load under any conditions. Since the current and resistance are the same, the voltage drop over the 10 miles will also be the same - 840 volts. The loss in the transmission line will still be 100,800 volt-amperes, but now this represents a whopping 7% of the total fed into the transmission line.
Obviously, the high-voltage transmission is far more efficient. As also demonstrated in this example, the transmission line can carry far more power under high-voltage conditions. For these reasons, all transmission lines operate at higher voltages than those delivered to your electric meter by the power companies.
At Niagara Falls, N. Y., for example, hydroelectric generators produce power at 6000 volts. It is immediately stepped up by transformers to 120,000 volts and fed to long-distance transmission lines. At various points it is stepped back down to 6000 volts for distribution over local areas, then stepped down once again to 230 and 115 volts for home use.
Although a power-distribution transformer is more spectacular, you're much more likely to be familiar with the ordinary power transformer used in radios, amplifiers, and TV sets. Such devices have a primary winding and usually several secondary windings to meet the various voltage and current requirements of a receiver or amplifier; a drawing of a typical power transformer is shown at right, above. The primary is usually designed for 115 volts; the high-voltage secondary may produce anywhere from 250 to as high as 600 or 700 volts (higher for some purposes). The other secondaries, usually rated at 5.0 and 6.3 volts, are for tube filaments.
Power transformers are available with a wide variety of windings and current capabilities. They may have four, five, six, or even more windings, each rated at a different voltage for some specific purpose. The high-voltage winding of a light-duty power transformer may be capable of producing perhaps only 30 or 40 ma., while a heavy-duty unit may turn out 300, 400, or even 500 ma. Transformers for high-power transmitters produce voltages and currents far in excess of these values, but for such applications separate transformers are generally used for high-voltage and filament supplies.
So far, all the transformers we have talked about in detail are designed for use in power circuits which operate at 60 cps. But transformers can operate on a wide variety of frequencies - every audio amplifier uses at least one transformer of this sort, and many include several such transformers.
Although the same basic principles of step-up and step-down are used in audio transformers, this is usually of secondary importance to the transformer's ability to serve as an impedance-matching device. Take, for example, an input transformer. Here it may be necessary to match a phonograph pickup, a microphone, or other input source of as little as 200 or 300 ohms (even less, in some cases) to a grid circuit of as much as several hundred thousand ohms. If the pickup or microphone were connected directly to the grid, a serious mismatch would occur, which would not only reduce the efficiency of the circuit but upset frequency response as well. The input transformer matches the components so that each operates properly.
The interstage transformer is another variety of the audio transformer and performs much the same kind of job: matching the output tube-several thousand ohms-to a grid circuit of a much higher impedance.
A third variety is the output transformer, whose main task is again impedance-matching. The plate circuit of the output tubes may have an impedance of many thousands of ohms, while most loudspeakers are 4, 8, or 16 ohms. To accommodate various tube-speaker combinations, most output transformers have a series of "taps" on the secondary winding, and perhaps on the primary as well, so that windings of the proper impedance can be selected. Only the part of the transformer windings actually used (in 'the diagram, the portion between the first and second terminals) affects the circuit's impedance values. One form of output transformer - known as a "universal" type - is so designed that it is capable of matching virtually any possible tube and speaker combination.
Great progress has been made recently in audio-transformer design. Just a few years ago it was difficult to get a transformer with any appreciable output above, say, 10,000 to 15,000 cps. Today, transformers with flat outputs up to 20,000 cps are common, while units flat to 50,000 or even 100,000 cps are available.
Tremendous problems had to be overcome to produce today's outstanding transformers. In addition to the losses mentioned earlier, a transformer has inductive reactance which varies according to frequency (remember that a transformer is also a coil). At frequencies of 100 and 1000 cps, the inductive reactance of the primary will be 10 and 100 times, respectively, its value at 10 cps. The inductive reactance appears to the output tube's plate as a load resistance, and thus various amounts of amplification take place at various frequencies. As a result, the gain of the amplifier is about two-and-a-half times higher at 200 cps than at 10 cps. At 3000 cps it would be three times higher. At still higher frequencies, distributed capacitance becomes an important factor, and gains fall off rapidly.
Engineers go to great lengths to compensate for these effects; by means of special core materials, unique coil designs, special wrapping patterns, interlaced layers, and other techniques, they have produced a variety of audio transformers with unbelievably even response over an extremely wide range of frequencies.
As mentioned earlier, transformers are also widely used in r.f. circuits. Even the simplest five-tube a.c.-d.c. radio will usually have as many as four transformers, in addition to its audio output transformer. A typical radio, for example, might have an antenna coil (actually a small transformer which couples the antenna's output into the grid of the first amplifying tube), an oscillator coil (a transformer which supplies feedback for the oscillator), and two i.f. transformers which couple the various stages.
These transformers are likely to be air, powdered-iron, or ferrite-core transformers, since a regular iron core would cause intolerable eddy-current losses. The windings will also probably be of a special spiral design calculated to minimize capacitance effect,
Although the transformers we have been discussing make up the bulk of those used, there are many other types, all of which perform their useful, specialized jobs.
The autotransformer, for example, uses only one winding instead of two, but accomplishes an effect similar to that of a regular transformer. If the whole coil is used as the primary and only a portion as the secondary, then it is a step-down unit. Hooked in reverse, it is a step-up device. This transformer, of course, cannot be used in circuits which must be electrically isolated from each other. But it serves very well in your automobile where it draws current from the 6- or 12-volt battery or generator and puts out the 10,000 or more volts needed to fire your spark plugs.
While we're on the subject of your automobile, let's take a look at the car radio which uses another kind of specialized device, the vibrator transformer. This device effectively "transforms" d.c. As the vibrator element moves back and forth touching each contact in turn, current flows through each half of the primary alternately, with each pulse going in a different direction. With the proper turns ratio, the output from the original 6- or 12- volt d.c. source can be as much as several hundred volts a.c.
Photoflash transformers, used to operate photographer's electronic flash or "strobe" units, are also vibrator-operated. They can take the vibrator-interrupted output from a 1 1/2-volt battery and turn it into several thousand volts a.c.
Pulse transformers are used primarily in radar. They range from tiny units (several of which can fit in a thimble) that put out a few millionths of a watt to huge, multi-ton giants that transmit powerful million-watt pulses. These transformers are designed to step up odd-shaped waveforms without changing the waveshape.
One of the newest types - transistor transformers - are similar to those used in regular r.f. and a.f. circuits except that their impedances and voltage ratings are calculated to match the operating requirements of transistors. Some of these units, by the way, can fit in a cube three-eighths of an inch square, and they weigh only a fraction of an ounce.
Thus, through the ingenuity of the design engineer, the transformer - though always operating on the same simple principle discovered by Faraday - can be adapted to perform hundreds of useful and important services.
Posted May 14, 2014