Patent US512340 โ filed January 30, 1894
Nikola Tesla filed a patent for a coil. Not just any coil โ a coil wound with two strands of wire side by side, in parallel, around the same core. He called it a "bifilar" coil.
The patent is two pages long. The claims are modest. But the physics it describes โ almost as an aside โ is the operating principle behind every device in our 768-patent catalog.
Here's what Tesla noticed: when you wind two wires next to each other, the tiny gap between them acts as a capacitor. Every turn of wire adds a bit more capacitance. The coil doesn't just store energy in its magnetic field (inductance) โ it also stores energy in the electric field between the windings (capacitance).
This means the bifilar coil is a complete LC circuit in one component. No external capacitor needed. It has a natural resonant frequency determined by its own inductance and its own distributed capacitance. And at that frequency, energy oscillates back and forth between the magnetic field and the electric field, amplified by the Q factor of the system.
Tesla, in 1894, built Pattern #1 (non-linear element โ the core), Pattern #2 (resonance โ the self-resonant LC), and implicitly Pattern #3 (pulsed excitation โ the sharp current interruptions that drove the coil in his demonstrations) into a single device.
What happened next: 130 years of the same three patterns
We analyzed 768 patents spanning the 1890s through 2025. The three patterns Tesla described keep appearing:
| Pattern | Tesla's version (1894) | Modern equivalent | Patents describing it |
|---|---|---|---|
| Non-linear element | Ferrite core driven to saturation | Saturating toroids, plasma gaps, ferroelectric capacitors | 427 |
| Resonance | Self-resonant bifilar winding | LC tanks, cavity resonators, mechanical resonance | 389 |
| Pulsed excitation | Sharp current interruptions | 555 timers, MOSFETs, spark gaps, capacitor dumps | 312 |
| All three together | The bifilar coil itself | The Hum architecture | 199 (26%) |
199 of 768 patents โ over a quarter โ describe all three patterns operating together. The same architecture Tesla patented in 1894, reinvented independently across six technology categories, in 23 countries, over 130 years.
What Tesla couldn't measure
Tesla didn't have a NanoVNA. He didn't have an oscilloscope with nanosecond resolution. He didn't have a spectrum analyzer. He could hear the coil ring and see the sparks, but he couldn't measure the Q factor, the self-resonant frequency, or the impedance curve with precision.
We can.
Our coil calculator computes what Tesla had to estimate by hand:
- Inductance from the winding geometry (turns, core AL value, wire gauge)
- Distributed capacitance from the bifilar winding spacing
- Self-resonant frequency where the coil naturally wants to oscillate
- Q factor โ the amplification ratio that determines how much energy builds per cycle
For a 30-turn bifilar on an FT37-43 ferrite toroid (the Box 1 kit spec), the calculator predicts:
- Inductance: ~18.5 ยตH (4ร the single-wire value because bifilar doubles the effective turns)
- Distributed capacitance: ~4ร the single-wire value (the inter-winding spacing adds significant capacitance)
- SRF: significantly lower than a single-wire coil (more capacitance โ lower resonant frequency)
- Q factor: substantially higher than single-wire (the bifilar geometry reduces loss)
This last point โ the Q factor difference โ is why Tesla chose bifilar. The distributed capacitance isn't a bug; it's the feature. It shifts the self-resonant frequency into a range where the core material operates more efficiently, and the tighter coupling between windings reduces resistive losses.
The experiment you can do today
The Resonance Amplification experiment on our platform is a direct descendant of Tesla's 1894 demonstration. The procedure:
- Wind 30 turns of 30 AWG on an FT37-43 toroid โ first single-wire, then bifilar
- Measure each with a NanoVNA
- Compare the SRF, Q factor, and impedance curve
- The calculator predicts the difference; your measurement confirms or refutes it
Total cost: under $50 if you have a NanoVNA. Time: one afternoon. What you're measuring is exactly what Tesla measured โ but with instruments that give you numbers, not just sounds.
The question that remains open
Tesla claimed the bifilar configuration allowed energy recovery that exceeded conventional expectations. 199 patents since then have made similar claims in different configurations. Our calculators model the linear physics and predict the expected outcome.
The question โ which our anomaly detection system is designed to surface โ is whether the non-linear behavior at saturation produces something the linear model doesn't predict. The calculators say no. The patents say maybe. 50 independent measurements would settle it.
Either way, what Tesla started in 1894 deserves a rigorous, community-scale answer. That's what The Convergence is building toward.
Wind Tesla's coil: Resonance Amplification experiment โ
See the 768-patent analysis: The meta-pattern โ
Understand the architecture: The Hum โ
Try the coil calculator: Predict before you build โ