by David McNamee
By the time you finish reading this, you will have a very good concept of how a Tesla coil works, even if you have no previous knowledge of things electrical. This is a primer, an overview, on how Tesla coils work. No mathematics are presented, nor are difficult electrical terms or measurement units used, and the subject matter is simplified to clarify the concepts presented. Diagrams are currently 'under construction', but the text has been purposely written so as to be able to stand on its own, assisted by your imagination. Analogies are used to compare electrical effects with more common everyday experience.
The Tesla coil is a simple device, and is a wonderful introduction to the fundamentals of electricity. I've included some definitions here at the beginning which you may skip if you find them redundant.
Electron - Thought of as the fundamental 'particle' of electricity. It is a 'negatively charged' particle.  Electrons are one of the building blocks of  atoms, and are normally arranged in 'orbits' around the 'positively charged' central nucleus of atoms. The negative and positive charges balance each other, and matter is thus normally electrically 'neutral'. But electrons can be displaced, and a  flow of electrons constitutes an electric current.
Polarity - The 'sign' of an electrical charge. An overabundance of electrons in matter creates an overall 'negative' charge; a shortage of electrons creates an overall 'positive' charge by the exposure of some of the positive central charge of the atoms of the matter.

Circuit - A path made of elements through which an electric current can flow.

Direct current - An electric current that is made to flow in one direction in a circuit, such as the current that can be obtained from a battery. It has a polarity, 'minus' or 'plus', referring to the negative or positive terminal of the battery respectively. Traditionally, electrons flow from the negative terminal to the positive terminal.

Frequency - How often something happens. For our purposes, usually how many times that something happens in one second.

Cycle - The period between departure from, and returning to, a starting point.

Alternating current - An electric current that is made to flow back and forth in a circuit at a periodic rate or frequency, usually expressed in 'cycles per second'. Its polarity constantly changes back and forth at that rate. House current usually is 'alternating' at 50 or 60 cycles per second, depending on where you live.

Resistance - Opposition to electric current flow. A high resistance allows little flow; a low resistance allows a lot of flow.

Conductor - In a general sense, anything through which an electrical current can flow.

Insulator - In a general sense, anything through which an electrical current cannot flow.

Losses - 'Lost' energy caused by transformation to some other form (heat, sound, light, etc.).

I've always found it convenient to think of electricity in plumbing terms, and the following concepts might help in visualizing some of the phenomena we will examine:

Voltage (in volts) is electrical pressure (similar to water pressure in a pipeline).

Current (in amperes or amps) is electrical volume (similar to how fast the water flows through a pipeline).

Power (in watts) - the product of the volume and pressure (volts times amps) (similar to the total amount of the water flowing in a pipeline, represented by the rate of flow at the specified pressure). The pouring of water from a bucket represents a high current (volume) at low voltage (pressure).  The jet from a squirt gun represents a low current (volume) at high voltage (pressure), while the stream from a fire hose represents high current (volume) at high voltage (pressure).

Now, on with the show! We have arrived at a local event, a 'Teslathon', which features, among other wonderful things, several Tesla coils made by both amateurs and professionals.

Before us is a Tesla coil set up by a dedicated hobbyist for the benefit of curious onlookers. Warned by its maker of impending operation, we retreat to a safe distance. When the coil is turned on, noisy lightninglike streamers reach several feet out into the air around the top terminal of the machine, writhing in ever-changing fractal paths, occasionally brightening when they happen to strike conducting objects set up nearby as targets. Somewhere near the base of the machine is a bright flickering light; this is not a short circuit, but a normal part of the operation of the coil, as we shall see. So much is happening here. Where does one start?

We are going to investigate the Tesla coil 'backwards' from the most interesting part first, the arcs and streamers emanating from the donut shaped top terminal. Obviously there must be a very high voltage there to ionize the air and form streamers and arcs like that. What causes the voltage to be so high? We will do a little  investigation of the streamers first, and I will give you an imaginary measuring probe which will provide you with important information. Pointing the probe at a streamer, for example, you find that the electrical energy in the streamer is actually occurring in pulses too fast to see, several hundred pulses per second in fact, and the energy in each pulse is alternating (its polarity continously changing back and forth from positive to negative) at a very high rate, about 200,000 cycles per second! Also you find that although the voltage is very high, on the order of several hundred thousand volts, the current is very low, only a fraction of an amp. Useful toy, that probe is, hmm?

Continuing our sojourn, what could  the top terminal be connected to that could provide it with such a high voltage at that high frequency? We see that the terminal is mounted on top of an upright plastic cylinder which on earlier examination was seen to be wound its full length with a single strand of insulated wire, five hundred turns or so, in a single layer. This coil of wire is known as the secondary coil (we will find out why it is secondary later). The end of the wire at the top of the coil is connected to the terminal. The other end of the wire, at the lower end of the coil, is connected to the end of a long copper rod almost completely buried in the damp earth. This end of the wire, then, is literally 'grounded' and thus must be at the same potential, voltage-wise, as the earth itself. Something seems amiss. Pointing the probe at the ground, you can see that it says zero voltage.

"Wait a minute", you say, "how can there be no voltage at one end of a wire while the other end is spraying sparks all over the place?".

Pointing the probe at the wire just above the ground rod reveals that there is actually a minute amount of voltage there, but - what is this? A tremendous current is flowing, several tens of amps in fact! And it is also pulsing and alternating at that same high frequency. Following the wire with the probe from the ground up to the base of the coil and then along the coil to the top terminal reveals a fascinating phenomenon: along the wire,  current decreases continuously as  voltage increases proportionally.

While the mind is digesting this strange state of affairs, we'll take a little side trip to shed a little light on resonance, the reason behind the phenomena we are observing.

Imagine, in your mind's eye, a long straight springy rod. Now picture one end of it solidly buried in concrete, with the other end free. Now push the top of the rod to one side, and hold it there. Notice the small amount of force required to do this. Release the rod, and see that it does not immediately return and stop at its original position. It swings back and forth past the center point in ever decreasing motion over a period of time until it is once again still.

Of great significance is that swinging motion. It took energy to bend the rod over, and when it was released, it gave up that energy, not all at once by snapping quickly back and stopping, but slowly and in an orderly fashion, oscillating at its own natural resonant frequency. This is the fundamental mechanical resonant frequency of this rod. At the resonant frequency, the energy stored in the rod continually changes back and forth from kinetic (moving) to potential (static or still). Left to itself, the rod eventually comes to rest, the energy having been dissipated in frictional losses in the rod and the surrounding air.

Move your hand down to the center of the rod and again push it to one side, so the free end is displaced exactly as far as it was the first time, and hold it there. Notice that significantly more force is required. Release it, and you see that it does exactly the same thing as before, despite the extra force. Now, move your hand down close to the base of the rod, and again push it to one side so the free end is displaced the same distance as in the past two trials, again holding it there. Notice that it requires a great deal of  force to do this. Yet, when released, its behavior is the same.

Let us learn some more from this. Gripping the rod again near the base, begin shaking it, very slowly at first, and then slowly increasing the shaking rate. This will at first cause the free end to move slightly, requiring a large input of force to do so. As the frequency of the shaking increases, you notice that eventually there occurs a point at which the motion of the free end increases dramatically, with the free end whipping violently back and forth. You also notice it takes much less force to maintain this motion at this one frequency once it has started. The rod, when acted upon or 'driven' by an outside source (you) of mechanical energy at the resonant frequency, is acting as a mechanical resonator, storing mechanical energy. The only additional input required to keep the rod swinging is that required to offset the losses, which are a minimum at resonance.

The free end of the rod swings through a wide arc at a fairly high velocity, but with little force. The base of the rod swings through a much smaller arc at a fairly low velocity, but with much more force. Try stopping the rod while it is resonating, at both points, and see.

It is important to note that the amount of energy stored in the rod at resonance can be much greater than the amount of each small push that it was given.

This analogy will help us to understand what is happening in the secondary of the Tesla coil. The secondary coil is behaving the same way electrically as the rod behaves mechanically. They are both resonators being driven at their natural frequency. See the similarity between velocity and voltage, and force and current:
High velocity and low force at the top of the rod,  high voltage and  low current at the top of the coil.
High force and low velocity at the base of the rod, high current and low voltage at the base of the coil.
And a smooth transition from one to the other between the two extremes. Minimum losses and thus great efficiency at resonance allow a large buildup of mechanical energy in the rod which is manifested by extreme motion. The same thing occurs electrically in the coil and manifests as extremely high voltages.

The builder of  this coil whose display of sparks we are appreciating thus took care to make sure the secondary coil is operating at the resonant frequency. If so, there must be energy at that frequency being put into it somewhere, somehow, to make it do that. But there is nothing connected to the coil except the ground at one end and the terminal at the other!

However, there is something near the secondary coil. In fact, there are three components here. We will investigate them one at a a time.

The first is a flat horizontal spiral of copper tubing located just below the base of the secondary coil. This is the primary coil. We see that the primary coil is aligned concentrically with the secondary coil. Also we notice that this flat coil has only a few turns, and though it is a coil, it looks very different from the secondary coil.

Since we seem to be running into coils a lot, it makes sense to get some information on what coils do, and why.

A coil has the ability to store energy in, and release energy from, a magnetic field.
Taking a piece of insulated wire and coiling it up, and connecting it to a source of direct current, such as a battery, we find a steady magnetic field forms in the space in and around the coil. Actually, a magnetic field is formed in and around any wire that is carrying current, but this is usually a weak effect and coiling the wire intensifies the field enough to make it more noticeable. Iron or steel objects can be affected by this field if they are close enough, and a nail held in the hand can be used to 'feel' the field in and around the coil. The field extends for some distance, its intensity falling off with increasing distance from the coil.

The magnetic field represents stored energy. It took energy to form that field in the first moment the coil was connected, but to maintain it requires a constant flow of current from the battery. When the coil is disconnected from the battery, the field does not remain, it 'collapses' and during this collapse momentarily generates, or induces a voltage in the coil as it does so, which shows up as a spark during the disconnect. In effect, the energy of the collapsing field is converted back to current flow and returned to the battery. The faster the disconnect happens, the faster the field collapses and the higher the voltage pulse produced.

Now, if instead the coil is connected to an alternating current, the resulting magnetic field will be a continuously expanding and collapsing field. Remember that the field extends well beyond the coil. If another coil similar to the first is placed near it in proper alignment, this changing field will generate a continuously changing current in the other coil which is a virtual copy of the alternating current in the first coil. The strength of this current depends on how much of the field from the first coil interacts with the second coil. This is the basic principle of the transformer, a device which is used to change the voltage of an alternating current to a higher or lower value.

Now we see how energy can be transferred from the primary coil to the secondary coil without a physical connection. An alternating current, of any frequency, can induce a similar current flow in a nearby coil. Coils are sometimes called inductors, and inductance is the measure of the magnetic energy storing ability of a coil. Adding turns to a coil increases its inductance.

Now that we know what coils are and how they behave, we can go on to the second act in our three-part sideshow.

The second component is a capacitor. Its name implies an ability to hold something, and capacitance is the measure of how much. In its simplest form a capacitor is composed of two metal plates with an electrically insulating material between them; for example, air between the two plates can work as the insulating material.

That's too easy, it seems. What can two metal plates with insulation between them possibly do that would be useful?

A capacitor has the ability to store energy in, and release energy from, an electric field.
If each plate is connected to the terminals of a battery, each plate takes on the same voltage present on the terminals of the battery and an electric field is formed in the space between the plates. To indicate what has just happened, we say that the capacitor has been charged. Disconnecting the capacitor from the battery, we find that it remains charged; if a piece of wire is connected to one plate, and then the other end of the wire is brought into contact with the other plate, there is a spark as the voltages on the plates equalize and the capacitor discharges. If the wire has a high resistance the capacitor discharges slowly, as if the capacitor was behaving as a battery. If the wire has a low resistance essentially all the energy stored in the capacitor flows through the wire in a very short time, and a large pulse of current results.

When an alternating current is connected to the plates, the electric field is being continuously charged and discharged, and though there is no direct connection between the two plates, a current can still be measured in the circuit. If a capacitor is disconnected from an alternating current, it remains charged at the voltage that was present at the moment of disconnect.

How much charge a capacitor will hold is determined by the size of the plates, how far apart they are (the larger and closer the plates, the higher the capacitance), the properties of the insulating material between them, and the applied voltage. Air is a rather bad insulator, especially at high voltage; it ionizes easily and limits the voltage that can be applied to the plates before a spark jumps across and shorts out the capacitor. Plastics are much better electrical insulators, they can hold more charge than air, and if we make the insulating sheet somewhat bigger than the plates we can put much more voltage on it because any arc that forms has a longer distance to travel around the edge. By adding additional sets of plates and insulating sheets, we can increase the total capacitance.

When the capacitor is charged, the insulator between the plates is under a kind of electrical strain from the electric field in it. Some insulators are very 'elastic' in an electrical sense, and take, hold and release energy easily and efficiently. Others are poorer, and 'leak', losing their charge slowly all by themselves, or resist being charged and discharged, causing energy to be lost in the capacitor in the form of heat. The capacitor in the Tesla coil we are studying has aluminum foil plates with polypropylene plastic between them, which is a very low 'loss factor' material, in an electrical sense.

When we compare coils with capacitors it is discovered that they do the same thing: they take, store, and release energy. But they do this in opposite senses. Ponder for a moment the following interesting behavioral contrasts:
A capacitor can contain an electric field, a coil can contain a magnetic field.

The electric field in a capacitor exerts a force on the plates that tends to draw the plates closer together. The magnetic field of a coil exerts a force on the turns of the coil that tends to spread the coil apart.

The electric field in a capacitor is confined to the space between the plates. The magnetic field of a coil extends out into the space around the coil.

A charged capacitor needs no current flow to maintain its electric field. A coil must have a current flow to maintain its magnetic field.

For a capacitor to function properly, electric current must not be able to pass directly through it. For a coil to function properly, electric current must be able to pass through it.

The strength of the capacitor's electric field is indicated by the voltage on the plates. The strength of the coil's magnetic field is indicated by the current in the coil.

A current is produced from a charged capacitor's electric field when the plates are connected together. A voltage is produced from a coil's magnetic field when it is disconnected from a power source.

There is one thing, however, that is common to both coils and capacitors. It takes time to 'fill them up' (or empty them) depending on the amount of energy they will hold and the rate at which energy is provided to or taken from them. In one sense they are buckets for holding energy, and of course, bigger buckets take longer to fill or empty at the same rate. The sizes, or values, of coils and capacitors used in Tesla coils are quite 'small' electrically even though they may be large physically.

Now for the last and easiest part of our trio.

The third component is called a spark gap or 'break', and that is exactly what it is in this particular Tesla coil, a space between two pieces of metal that can be adjusted closer together or farther apart as needed. This is as simple as any electrical component gets! (A spark gap can range from two bolts on a board to digitally speed controlled motorized multiple breaks, according to intended use, taste, power level and bank account. To keep things easy, the spark gap in this Tesla coil is a simple 'straight' gap).

The spark gap functions as a switch. If the voltage between the two points reaches a high enough value, it will jump the gap. Looking at  the Tesla coil while it is operating, we see that the spark gap is ablaze with crisp blue-white arcs. This is normal and no cause for alarm. Just don't stare at it; spark gaps make a lot of nasty ultraviolet light.
Now we know the three components and their characteristics. Lets connect them together like we see them.
The primary coil, capacitor and gap are connected together in a ring, or 'in series'. One end of the primary coil connects to one side of the gap, the other side of the gap connects to one plate of the capacitor, and the other plate of the capacitor is connected to the other end of the primary coil. This elegant arrangement of components is called the primary circuit.

Pondering this layout, and armed with our new information, we recall that the energy in the secondary coil was occurring in separate pulses of high frequency. It was being induced in the secondary coil by the primary coil, and the same kind of energy is indeed present in the primary circuit. What causes the pulses?  What happens during each one of these pulses? How is such a high frequency generated?

The experimenter has shut down the coil for a moment to make an adjustment. At this time for convenience we will introduce our power source. This source is a transformer that steps up the common AC house current to a high voltage (15,000 volts or so). We see that the transformer has two high voltage terminals; one terminal is connected to one side of the spark gap, and the other terminal is connected to the other side of the gap.

And it has just this moment been switched back on. This is what happens, on a very short time scale:

With high voltage suddenly present, the capacitor charges and as it does so the voltage across the spark gap also rises. This occurs almost instantly, and it takes only a very small time to reach a high enough voltage to 'jump' the gap. The air in the gap ionizes, forming an electrically conductive path of hot gas. Now the fun begins.

The two plates of the capacitor are now connected together in a short circuit through the very low resistance of the primary coil by the arc in the gap. The capacitor discharges a heavy current through the primary coil, which dutifully stores the energy in a magnetic field. But in a very short time the discharging capacitor is unable to supply enough current to maintain the coil's magnetic field. The field collapses, generating voltage which then charges the capacitor back up, but with the polarity reversed. At the moment the capacitor again reaches maximum charge there is no current flowing across the gap, but the conductive path of ionized air in the gap is still there, not having had time to cool to the point of being an insulator again. So the whole process happens again in reverse, the capacitor discharging into the primary coil and the coil then generating a voltage to charge the capacitor.

The chain of events in the above paragraph is one cycle of the alternating current we found in the circuit, and in this particular circuit one cycle occurs in 1/200,000th of one second! This process repeats, energy flowing smoothly from the electric field in the capacitor to the magnetic field in the primary coil, back and forth.

The exchange of energy from capacitor to coil in the primary circuit occurs at a natural resonant frequency, which is determined by the values of the components used. The primary circuit, like the secondary coil, is a resonant circuit. And since the primary coil is close to the secondary coil, the magnetic field from the primary coil induces a strong current in the base of the secondary coil at this frequency. With each cycle, energy is transferred to the secondary coil, and some energy is also consumed by losses in the arc and primary circuit. When there is not enough energy to sustain the arc, it goes out, or 'quenches', the pulse of high frequency current is over, and the initial charging process begins again.

The resonant frequency of the primary circuit must be the same as the resonant frequency of the secondary coil. In practice it is almost impossible to design the two with exactly the same frequency - a movable tap is usually provided on the primary coil so that its inductance value may be adjusted and thus bring it into 'tune' with the secondary coil..

The transfer of energy from primary circuit to secondary coil is not as direct as it may seem, because current flowing in the secondary coil can also induce current in the primary circuit - magnetic induction works both ways. So the energy actually transfers several times to and fro between the primary circuit and the secondary coil before the arc in the spark gap quenches, stopping the process. Ideally the spark gap quenches at a point when most or all the energy has been transferred to the secondary coil. The open gap prevents the transfer of energy back into the primary circuit. The energy trapped in the secondary coil has no choice but to resonate until it is dissipated in resistance losses in the streamers and the coil.

We see similarities now between the primary circuit and the secondary coil. The base of the secondary coil is a region of high current and magnetic field activity, like the primary coil. The top of the coil is a region of high voltage and electric field activity. But there does not appear to be a capacitor associated with the secondary coil, unless... could it be that the donut shaped top terminal behaves as a capacitor? How? Digging out our handy probe and pointing it at the terminal... yes, the top terminal is behaving as one plate of a capacitor because it is a metal surface with a voltage on it. But what could be nearby that could act as the other plate?

The earth functions as the other 'plate', since it has no voltage. The difference of charge between the top terminal and the earth, with air between, form a capacitor. Since the distances are great between the terminal and the ground, the amount of capacitance is very small. And the probe reveals another amazing tidbit : you, and I, are part of this circuit. We are standing on the ground, and are thus part of the other 'plate'!

Now we can also understand why the streamers brighten when they strike grounded objects. The energy stored in this expansive capacitor is being 'shorted' to ground across what amounts to a very large spark gap, and the higher resulting current momentarily causes a more intense ionization.

The entire Tesla coil is a tuned, balanced system of opposites. The primary circuit has a large capacitance and a small inductance and is designed to make a high frequency current; the secondary circuit has a large inductance with a small capacitance and is designed to convert a high frequency current to a high frequency voltage. Both are resonant circuits that run at the same frequency despite their obvious physical differences.

Some additional info on that funny looking aluminum donut and why they are so often seen on a Tesla secondary coil:

The wire of which the secondary coil is wound is of fairly small diameter. The wire insulation can handle a few thousand volts or so. The voltage builds along the secondary coil turn by turn; since each turn 'sees' its next neighbor at only a few hundred or few thousand volts difference; they 'shield' each other. Near the top of the coil the now very high voltage on the remaining turns begins to 'see' the end of the winding coming and the lack of any potential just beyond it, and at the end of the winding the maximum voltage of the system is developed on the last turn which is, essentially, an edge. Edges are 'EXIT HERE' signs for high voltage. Air around the insulation ionizes, and streamers and arcs break out from the wire, damaging the insulation.

By connecting a large, rounded ring shaped metal 'donut' (toroid) above the secondary coil, the top of the winding can be shielded by the presence of a surface which is at the same potential as the last turn of the winding. This prevents or at least minimizes breakout from the top turn, because the large curvature of the toroid surface results in a smoother voltage gradient around the top of the coil. Generally it also means a higher voltage being achieved before breakout, and longer sparks. And unlike the wire, the top terminal can take the abuse the arcs dish out.

The top teminal is also called a 'top load' because, as we have seen, it behaves like a capacitor of small value, and this added capacitance 'loads down' the secondary coil and lowers its resonant frequency. It is similar to adding a weight to the top of the springy rod in the earlier analogy. Again, most experimenters provide adjustments and extra turns on the primary coil so the circuit can be adjusted and brought into resonance with the secondary and any anticipated top load.

The toroid or donut shape is the optimum shape as it most closely approximates the shape of the end turn of the winding. A shape such as a sphere will work, but generally must be larger than the corresponding toroid that would provide the same protection to the winding.

I hope this has helped you to understand a little more about how Tesla coils work. After the 'show' is over we stroll over and examine this Tesla coil closely, and, comparing it with others large and small at this gathering, find that the variability and complexity encountered in real Tesla coils is only in the manner and design of the parts that make it up; they all express the same elementary function.