Chapter 4
Budapest, February 1882 — the moment an idea became a field that would never stop spinning.
The date was February 1882. Nikola Tesla was twenty-five years old, recently collapsed from what he later described as a complete nervous breakdown — his body and mind refusing, after years of overwork at the Budapest Telephone Exchange, to continue. He had been convalescing in the city's Városliget, the great public park at the eastern edge of Pest, walking each day with his friend Antal Szigeti as his physician's prescribed exercise.
Tesla had been wrestling with a single problem for two years. Since 1880 he had been turning over in his mind the question of how to build a practical motor that ran on alternating current. The available solution — the commutator-and-brush arrangement that made DC motors work — was mechanically barbaric: a device that periodically reversed the coil connections to keep the rotor spinning in one direction. It sparked, it wore, it required constant maintenance. Tesla believed there had to be a better way. He had not yet found it.
That February afternoon, as the sun was setting over the park, Tesla and Szigeti were walking near the Vajdahunyad Castle structure. Tesla was reciting from memory passages of Goethe's Faust in German — he had committed much of the work to memory. He later recalled the specific passage about the sun setting and light retreating across the land. Tesla's own account, written for Electrical Experimenter magazine in 1919 — the most authoritative primary source for his inner life — describes what happened next:
"As I uttered these inspiring words the idea came like a flash of lightning and in an instant the truth was revealed. I drew with a stick on the sand the diagrams shown six years later in my address before the American Institute of Electrical Engineers, and my companion understood them perfectly. The images I saw were wonderfully sharp and clear and had the solidity of metal and stone, so much so that I told him, 'See my motor here; watch me reverse it.'"
— Tesla, N. My Inventions. Electrical Experimenter, February 1919, Part II.
The context matters enormously. Tesla had been working on this problem since his time as a student at the Graz Polytechnic, where in 1877 he had watched a Gramme machine — a DC dynamo — being demonstrated. The machine used a commutator. Tesla had remarked to his professor, Poeschl, that there ought to be a way to run such a machine without the commutator. Poeschl dismissed this as a perpetual-motion fantasy. Tesla spent the next five years proving him wrong, though the proof lived entirely in his mind until that afternoon in Városliget.
The insight was this: you do not need to mechanically switch anything. If you arrange two coils of wire at 90° to each other in space and feed them with two alternating currents that are 90° out of phase in time, the combined magnetic field of the two coils does not simply alternate back and forth. It rotates. Continuously. At the frequency of the alternating supply. Place a conducting object in that rotating field and currents are induced into it by Faraday's Law; those currents generate their own magnetic field, which is dragged along by the rotating stator field; the object rotates. No commutator. No brushes. No contact. Just the mathematics of two sinusoids 90° apart in phase and 90° apart in space.
Tesla spent the next six years building hardware to prove what he had already designed completely in his mind. The hardware matched the vision. It always did.
This is the section that most popular accounts of Tesla omit, and its omission is the reason most readers never truly understand what he accomplished. The rotating magnetic field is not a metaphor or a convenient description. It is a precise, calculable, verifiable physical phenomenon. Here is the proof.
Consider two coils, A and B. They are wound identically and oriented at 90° to each other in space — their axes perpendicular, both facing a common center point where a rotor will sit.
Coil A carries current:
I_A = I₀ sin(ωt)
Coil B carries a current identical in frequency and amplitude but advanced 90° in phase:
I_B = I₀ cos(ωt) = I₀ sin(ωt + 90°)
Each coil produces a magnetic field along its own axis, proportional to its current. Call coil A's axis the x-axis and coil B's axis the y-axis:
B_x(t) = B₀ sin(ωt) [field from coil A, along x]
B_y(t) = B₀ cos(ωt) [field from coil B, along y]
The total field vector at the center at time t is the vector sum:
B_total(t) = B₀ sin(ωt) x̂ + B₀ cos(ωt) ŷ
Compute the magnitude:
|B_total| = √(B_x² + B_y²) = √(B₀² sin²(ωt) + B₀² cos²(ωt))
= B₀ √(sin²(ωt) + cos²(ωt)) = B₀ × 1 = B₀
The magnitude is constant. It does not oscillate. It does not pulsate. At every instant the total field has exactly the same magnitude B₀.
Now the direction. The angle θ of the field vector from the x-axis:
θ(t) = arctan(B_y / B_x) = arctan(B₀ cos(ωt) / B₀ sin(ωt)) = arctan(cot(ωt)) = 90° − ωt
The direction changes linearly with time at rate ω. This is the definition of uniform circular rotation of a vector. The field rotates at ω radians per second — exactly equal to the angular frequency of the supply current. At 60 Hz:
ω = 2π × 60 = 377 rad/s → 60 complete rotations per second
Three-phase uses three coils at 120° spatial intervals fed with currents at 120° phase intervals:
I_A = I₀ sin(ωt)
I_B = I₀ sin(ωt + 120°)
I_C = I₀ sin(ωt + 240°)
Resolving each coil's field contribution into x and y components using its spatial angle, then summing and applying the identities sin(A+B) = sinA cosB + cosA sinB:
B_x(t) = (3/2) B₀ cos(ωt)
B_y(t) = −(3/2) B₀ sin(ωt)
The magnitude:
|B_total| = (3/2) B₀ = constant
Again constant; again rotating at ω. The magnitude is 50% larger than the two-phase case for the same peak coil field — a material advantage in motor design. The full trigonometric derivation is in Chapman, Electric Machinery Fundamentals, 5th ed., Chapter 4, and Fitzgerald, Kingsley & Umans, Electric Machinery, 6th ed., Chapter 4.
Tesla's original polyphase system used two phases. The world settled on three. The reasons are engineering, not physics.
Power delivery is constant. Instantaneous single-phase power oscillates at twice the supply frequency — it pulses between zero and its peak even at unity power factor. Three-phase power, summed across all three phases, is genuinely flat in time:
P_3φ = 3 × V_phase × I_phase × cos(φ) = constant
Constant power means constant torque. Constant torque means no vibration imposed on mechanical loads by the electrical supply — critical for large rotating machinery.
Conductor economy. Three separate single-phase circuits require six wires (send and return for each). A balanced three-phase circuit delivers the same power over three wires because the three return currents sum to exactly zero and no neutral is required. Three wires versus six: 50% saving in copper for transmission infrastructure.
Generator and transformer efficiency. Three-phase machines distribute their magnetic loading uniformly around the stator; they are smaller, lighter, and more efficient than single-phase equivalents at the same power rating.
The speed at which the rotating magnetic field revolves — the synchronous speed — depends on supply frequency and the number of magnetic poles built into the stator winding:
Ns = 120f / P (RPM)
where f is supply frequency (Hz) and P is pole count (always an even number: 2, 4, 6, 8…). The factor 120 = 2 × 60 seconds per minute.
At 60 Hz (North America):
| Poles (P) | Synchronous Speed (RPM) |
|---|---|
| 2 | 3,600 RPM |
| 4 | 1,800 RPM |
| 6 | 1,200 RPM |
| 8 | 900 RPM |
| 12 | 600 RPM |
At 50 Hz (Europe, most of Asia):
| Poles (P) | Synchronous Speed (RPM) |
|---|---|
| 2 | 3,000 RPM |
| 4 | 1,500 RPM |
| 6 | 1,000 RPM |
| 8 | 750 RPM |
The pole count is set entirely by the stator winding layout — how many coil groups are wound and at what angular spacing. A motor designer sets synchronous speed by choosing pole count. The supply frequency is the master clock; the pole count is the gear ratio. The motor's actual running speed will be slightly below synchronous speed due to slip — the subject of Chapter 5.
Tesla filed the foundational polyphase patents on October 12, 1887. George Westinghouse purchased them in July 1888 for $60,000 in cash plus $2.50 per horsepower royalty on every AC motor sold. At the time a generous valuation; within a decade, one of the most consequential technology purchases in industrial history.
| Patent | Title | Date Granted |
|---|---|---|
| US 381,968 | Electro-Magnetic Motor | May 1, 1888 |
| US 382,279 | Electrical Transmission of Power | May 1, 1888 |
| US 382,280 | Electrical Transmission of Power (alt. configs.) | May 1, 1888 |
| US 390,721 | Dynamo-Electric Machine (AC generator) | October 9, 1888 |
All four are publicly available at patents.google.com — search by patent number. The claims language is precise; the drawings are clear mechanical diagrams. Before reading any secondary account of Tesla's polyphase system, read the original patents.
On May 16, 1888 — before the patents were formally granted — Tesla delivered his lecture "A New System of Alternate Current Motors and Transformers" to the American Institute of Electrical Engineers in New York. This was the first complete public technical presentation of the polyphase system. It was reprinted in Thomas Commerford Martin's The Inventions, Researches and Writings of Nikola Tesla (1894), available in full at gutenberg.org. Every working engineer interested in this period should read it; it remains technically illuminating.
The Niagara Falls Power Station, opened in 1895, was the system's culminating early demonstration: 15,000 horsepower of polyphase AC transmitted 26 miles to Buffalo. The first large-scale commercial AC power transmission in North America.
The principle looks simple in retrospect. Two coils. Two phases. First-year mathematics. Why had no one done it before?
The obstacle was not mathematical difficulty. It was conceptual. Everyone working on AC motors in the 1870s and early 1880s was thinking through the lens of DC motors. A DC motor works by switching the direction of current in the rotor coils at the right moment — the commutator does this mechanically. The natural assumption was that an AC motor would do the same electrically: reverse the current, chase the reversal, produce rotation. This approach yields a motor with an oscillating field — not a rotating field — that vibrates, runs poorly, and produces no net torque at stall.
What Tesla saw was that you had to abandon the single-axis oscillating field entirely. Two separate fields, spatially offset, temporally offset, combining to produce a rotating resultant: this required spatial imagination — the ability to see a vector sum as a rotation rather than as two separate oscillations. It required thinking about the fields in space, not about the currents in the wire.
Galileo Ferraris (Italy) published a paper on two-phase induction in 1888, the same year as Tesla's AIEE lecture, and had independently built a working two-phase motor. But Ferraris concluded in print that the theoretical maximum efficiency of an induction motor was 50%, and on the basis of this conclusion dismissed the device as commercially useless. The analysis was wrong: Ferraris failed to correctly account for the conversion of rotor electrical energy to mechanical work. Tesla's motors, as built and measured, achieved over 90% efficiency. See Ferrari (1888), Accademia delle Scienze di Torino, and Carlson, Tesla: Inventor of the Electrical Age, Chapter 6.
Dolivo-Dobrovolsky (AEG) was working on three-phase systems from approximately 1889 onward. His work was original and important in the engineering of practical three-phase transformers and in the Lauffen–Frankfurt demonstration — but it followed and built upon the published Tesla patents. It was engineering refinement of a published concept, not independent invention of the rotating field.
Objective: Produce a visibly rotating magnetic field using two coils and two phase-shifted AC signals, physically confirming the vector-sum rotation derived in Section 2.
The compass needle rotates continuously, following the rotating magnetic field. At 1 Hz it completes one revolution per second — you can watch it spin in real time. Reverse the phase of Channel 2 to −90° (270°) and the needle reverses direction. This is exactly what Tesla saw mentally in 1882, now physically realized on your workbench.
The coils are stationary. The currents in them are alternating — no DC, no physical rotation. The rotation exists purely in the vector sum of two phase-shifted fields, exactly as derived in Section 2. The compass needle is tracing the path of the B_total vector.
Replace the compass with a Hall-effect sensor (Allegro A1302). Sample the output on an Arduino or oscilloscope and plot field strength as a function of angle and time. You will measure constant magnitude rotating at the supply frequency — precisely the result of Section 2. For three-phase: add a third coil at 120° spatial offset driven at 120° phase offset; observe smoother, lower-ripple rotation consistent with the (3/2)B₀ constant-magnitude result.