Steve Mould Replicates Semiconductor Circuits via Mechanical Chains

Steve Mould

Mechanical chain systems reproduce complex electronic phenomena like inductive kickback and transistor switching because physical torque and chain velocity directly mirror voltage and current.

AI-assisted summary of Mechanical circuits: electronics without electricity — original source. Check the original for full context.

Source published: 2022-10-27

Key ideas

Steve Mould demonstrates Spintronics, a mechanical circuit system developed by Upper Story where chain velocity represents electric current, chain tension represents voltage, and friction provides resistance. (0:00)
Spintronics replicates advanced electronic components mechanically, using torsion springs for capacitors, weighted flywheels for inductors, epicyclic differentials for parallel junctions and DC voltage transformation, and friction clamps for transistors. (4:12)
Unlike electrical full-wave bridge rectifiers requiring four diodes, Spintronics accomplishes full-wave AC-to-DC rectification with only two diodes by mechanically crossing chain paths to reverse current direction. (14:58)
The kit reproduces real electronic phenomena such as inductive kickback breaking open switches and transient LC oscillations resolving the classic two-capacitor energy paradox. (7:10)

Detailed notes

## Background Published on YouTube by science communicator Steve Mould on October 27, 2022, this video explores Spintronics, a mechanical circuit kit developed by Upper Story. The page description highlights how mechanical resistors, inductors, transistors, diodes, batteries, and capacitors linked with drive chains provide an intuitive physical model for understanding how electronic circuits function.

## TL;DR Steve Mould demonstrates the mechanics of Spintronics, where physical chains, torsion springs, flywheels, and differentials model electric currents, voltages, capacitors, and parallel junctions. He connects these tangible mechanisms to deeper physical concepts, including Veritasium's Poynting vector debate on energy delivery and his own high school publication resolving the two-capacitor paradox. Throughout the video, Mould constructs functioning mechanical high-pass and low-pass filters, transistor-based peak voltage detectors, and digital logic components such as flip-flops and XOR gates. The demonstration highlights how physical, lumped-element mechanical models give immediate, intuitive access to complex electromagnetic principles.

## Atmosphere & style The presentation is energetic, tactile, and intellectually enthusiastic, driven by Mould's genuine delight in feeling invisible physics concepts turn into tangible mechanical resistance. His delivery strikes a natural balance between playful maker tinkering and rigorous conceptual inquiry, shifting effortlessly from clacking chain demonstrations to mathematical debates on field theory. Watching the video feels like an engaging workshop demonstration where abstract electrical equations become audible clicks, whirs, and physical forces under tension.

## Bulletin - [00:00](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=0s) Steve Mould demonstrates Spintronics, a mechanical circuit system developed by Upper Story where chain velocity represents electric current, chain tension represents voltage, and friction provides resistance. - [04:12](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=252s) Spintronics replicates advanced electronic components mechanically, using torsion springs for capacitors, weighted flywheels for inductors, epicyclic differentials for parallel junctions and DC voltage transformation, and friction clamps for transistors. - [14:58](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=898s) Unlike electrical full-wave bridge rectifiers requiring four diodes, Spintronics accomplishes full-wave AC-to-DC rectification with only two diodes by mechanically crossing chain paths to reverse current direction. - [07:10](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=430s) The kit reproduces real electronic phenomena such as inductive kickback breaking open switches and transient LC oscillations resolving the classic two-capacitor energy paradox.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=0s) Spintronics translates electronic principles into mechanical components, mapping chain velocity to amperage, chain tension to voltage, and mechanical drag to resistance. - [00:00](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=0s) Series versus parallel circuits in Spintronics depend on loop topology rather than physical component arrangement; a single continuous chain loop forms a series circuit even if parts sit side by side. - [01:41](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=101s) Mould addresses Veritasium's Poynting vector debate, arguing that while power travels through fields rather than drifting electron collisions, lumped-element mechanical models remain valid, intuitive abstractions for circuit design. - [04:12](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=252s) A torsion spring serves as a mechanical capacitor, storing energy against the drive chain and increasing counter-torque until series current drops to zero. - [04:12](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=252s) A three-gear epicyclic differential splits input torque and speed inversely across loads to model parallel junctions, and acts as a mechanical DC transformer providing a 2:1 advantage without AC. - [07:10](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=430s) An acoustic ammeter uses a serrated rotor and a flexible reed inside a gramophone horn to audibly map current flow via pitch. - [07:10](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=430s) A heavy flywheel models an inductor by storing angular momentum, producing inductive kickback that can physically break switches unless routed through a parallel shunt resistor. - [07:10](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=430s) The wind-up spring battery incorporates a centrifugal seatbelt-style brake to prevent runaway speeds and self-destruction during short circuits. - [09:42](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=582s) Unlike electrical components that remain open circuits when unplugged, disconnected Spintronics parts default to closed loops that discharge immediately unless constrained. - [09:42](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=582s) A mechanical transistor uses low input torque on an upper sprocket to lever open spring-loaded friction clamps around a rubber ring, releasing power to a lower output sprocket. - [11:41](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=701s) High-pass filters use parallel flywheels to shunt steady DC while passing high-frequency oscillations to the load; low-pass filters achieve the inverse using torsion springs. - [11:41](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=701s) Connecting a charged capacitor to an uncharged capacitor dissipates exactly 50% of the initial energy through transient oscillations driven by stray inductance and damped by internal resistance. - [14:58](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=898s) Full-wave AC-to-DC rectification requires only two mechanical diodes instead of the four semiconductor diodes required in electrical bridge rectifiers because drive chains can easily cross to reverse polarity. - [14:58](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=898s) Spintronics can implement digital computing logic, demonstrated through working Set/Reset flip-flops and XOR logic gates.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=0s) Introduction to Spintronics and Mechanical Circuits

Steve Mould introduces a mechanical circuit kit called Spintronics, developed by Upper Story. The system uses mechanical components—including resistors, transistors, capacitors, ammeters, batteries, and inductors—linked by drive chains to create an intuitive physical model of electronics. Mould clarifies that the manufacturer is not sponsoring the video, but explains that he backed the Kickstarter because of his lifelong passion for conceptual analogies. Describing the visceral appeal of the system, Mould remarks: “I can feel voltages and see currents.”

```mermaid flowchart LR accTitle: Spintronics Core Analogies accDescr: Mapping between electronic principles and mechanical Spintronics equivalents. subgraph Electrical["Electrical Domain"] V["Voltage (V)"] I["Current (Amps)"] Q["Charge (Coulombs)"] R["Resistance (Ohms)"] end subgraph Mechanical["Mechanical Domain"] F["Force / Tension (Spin Volts)"] S["Chain Velocity (Spin Amps)"] L["Chain Length (10 m = 1 C)"] Fric["Internal Friction (Spin Ohms)"] end V <--> F I <--> S Q <--> L R <--> Fric ```

The system operates via looped chains driven by a wind-up spring-powered battery, charged by pulling a pull-string:

- **Battery mechanism**: Pulling a pull-string winds an internal mainspring, which releases mechanical energy into the circuit chain when allowed to turn. - **Resistors**: Resistors use internal mechanical drag. A 100-ohm resistor provides a baseline drag; swapping in a 200-ohm resistor doubles the felt resistance and cuts the chain's speed in half. - **Chain velocity**: The linear speed of the interlocking chain is directly proportional to electrical current (amperage). - **Series connection layout**: Two resistors can be connected sequentially on one continuous chain loop or linked hub-to-hub. Even when physically configured side-by-side—which superficially resembles parallel components—they form a strict series circuit if the same single chain loop drives both.

[01:41](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=101s) Models in Physics: Veritasium and Poynting Vectors

Addressing the classic hydrodynamic ("water in pipes") model, Mould brings up the debate sparked by Veritasium’s two videos analyzing an open-wire transmission line one light-second across. In Veritasium's analysis, electrical power is delivered via Poynting flux—the orthogonal electric and magnetic fields traveling through the space around the wires—rather than by electrons physically nudging their neighbors end-to-end at drift speed. Mould points out that while field propagation disproves the crude notion of electrons acting like incompressible fluid pellets exiting a hose, electron motion remains foundational because those fields belong to the charges themselves.

Mould emphasizes that scientific understanding relies on choosing the appropriate abstraction level rather than insisting on a single absolute truth: “Even quantum mechanics is just a model, and it's wrong in the sense that we know it's not a true description of the universe because it doesn't play well with relativity, but we still use quantum mechanics because it's insanely accurate under most circumstances.” Different domains demand different tools: field theory explains electromagnetic wave arrival across transmission lines, but lump-element mechanics such as Spintronics or hydrodynamic models build functional intuition for circuit behavior.

[04:12](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=252s) Capacitors, Junctions, and DC Voltage Transformation

The Spintronics capacitor functions by winding a internal torsion spring against the driving chain. The more it turns, the higher the counter-torque becomes, mirroring how back-electromotive force rises as real capacitors accumulate charge. The faceplate features numbered graduations acting as a voltmeter calibrated in "spin volts" (where, by convention, 10 meters of chain passage equals 1 coulomb of electric charge). When placed on a single loop with a resistor, charging the capacitor causes the entire loop to grind to a halt once fully wound, illustrating how series DC current drops to zero.

To connect components in true parallel, Spintronics uses a mechanical junction built around a three-gear epicyclic differential:

```mermaid flowchart TD accTitle: Epicyclic Junction Branching accDescr: How the Spintronics three-gear differential splits torque and velocity. B["Bottom Input Sprocket (Source Voltage / Torque)"] J{"Differential Cage"} T1["Upper Sprocket 1 (Branch Circuit A)"] T2["Upper Sprocket 2 (Branch Circuit B)"]

B --> J J --> T1 J --> T2 ```

- **Differential kinematics**: The bottom sprocket couples through bevel gears to two concentric upper sprockets. Holding any single shaft stationary forces the remaining two shafts to drive one another, splitting input speed inversely with load. - **Parallel load division**: When two parallel resistors (for example, a 100-ohm and a 200-ohm unit) are driven through the differential, the lower-value resistor spins significantly faster, cleanly modeling how current takes the path of least resistance. - **Feeling the potential**: Resting a finger against an open differential sprocket provides tactile feedback of the circuit's force: “If I hold the top sprocket of the junction with my fingers, I can feel it pushing against me. That's the voltage.” - **Mechanical DC transformer**: By linking a single load across two sprockets of the junction while turning the third, the device yields a 2:1 mechanical advantage. The load turns at half speed while cutting perceived resistance in half, accomplishing direct-current voltage and current transformation without requiring alternating current or magnetic induction coils.

[07:10](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=430s) Auditory Ammeters, Inductors, and Inductive Kickback

To visualize and hear current draw, the kit includes an acoustic ammeter. It features a serrated perimeter track riding against a flexible mechanical reed connected to an upright gramophone-style horn. Fast chain flow yields a sharp, high-pitched acoustic buzz, while sluggish movement drops the tone to an audible low crawl. When paired with an RC-style charging capacitor, the acoustic horn starts at a high whine that continuously drops in pitch until silence indicates current has ceased entirely.

Mould introduces the mechanical inductor—his favorite component in the system. The unit uses a heavy perimeter-weighted flywheel to model electromagnetic reluctance. Mould explains the underlying symmetry: “An inductor resists the buildup of current, but once there is a current, it resists the reduction of current.” While electronic inductors rely on magnetic field storage, the Spintronics module relies on angular momentum.

This momentum produces identical real-world hazards, notably inductive kickback. When Mould opens an inline switch on a circuit powering an active inductor, the flywheel refuses to halt instantly. The sudden mechanical shock violently shears the latching switch teeth, physically breaking the switch. Mould demonstrates the standard electrical remedy: placing a shunt resistor in parallel with the inductor. When the switch cuts battery power, the flywheel discharges its accumulated kinetic momentum safely through the parallel resistor. Mould also highlights the built-in safety features: run without a load or shorted out across an open differential, the battery's pull-spring attempts to spin out of control, tripping an internal centrifugal seatbelt-style brake to prevent the mainspring from destroying itself.

[09:42](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=582s) Transistors, Amplification, and Peak Voltage Detection

A peculiar quirk of Spintronics is that its components default to being closed circuits rather than open circuits. While a traditional electronic capacitor holds its charge when physically disconnected from a breadboard, a charged Spintronics capacitor will instantly unwind and discharge the moment it is let go, because disconnected components are effectively looped back into themselves. Steve Mould remarks: “So unless a Spintronics component is connected to something else, it's probably connected to itself.”

```mermaid flowchart TD accTitle: Spintronics Transistor Operation accDescr: Diagram showing how a small input voltage releases mechanical clamps to permit a larger current flow in Spintronics.

subgraph Input["Control Side"] Vin["Small Input Voltage / Hand Force"] --> SprocketTop["Top Sprocket Turns"] end

subgraph Mechanism["Mechanical Action"] SprocketTop --> Clamps["Mechanical Clamps Open"] Clamps -.->|"Release Grip"| RubberRing["Rubber Ring on Lower Spindle"] end

subgraph Output["Load Side"] RubberRing --> SprocketBottom["Bottom Sprocket Spins Freely"] SprocketBottom --> Vout["Amplified Current / Voltage across Load"] end ```

Spintronics reproduces active semiconductor behavior using purely mechanical interactions. The transistor module consists of two levels: an upper control sprocket and a lower power sprocket encircled by a rubber ring held tightly by spring-loaded clamps. With no input signal, the clamps bite down on the rubber ring, preventing any chain or current from turning the bottom sprocket. When a delicate rotational voltage is applied to the upper sprocket, it levers the clamps open, allowing the bottom sprocket to spin freely. Steve demonstrates this by using one capacitor-based voltmeter to track the small finger-applied voltage at the control input, while a second voltmeter reveals the amplified output driving a heavier load. By routing a transistor back to itself through a junction, it can even mimic a diode—allowing rotation in only one direction.

Using these fundamentals, Steve constructs a peak voltage detector. Manual turning inputs an erratic voltage, visible on an input voltmeter, while a downstream capacitor captures and preserves the peak voltage achieved. Because the transistor-diode pairing blocks back-drive, the measuring capacitor cannot discharge backwards when he releases the input. Steve marvels at the physical tactility of the system: “What's brilliant about all this is I can feel it as it happens.” Spintronics also includes a standalone dedicated diode component, which achieves one-way current flow using an internal mechanical ratchet mechanism.

- Disconnected Spintronics components act as closed loops, discharging springs unless externally constrained. - The transistor amplifies current by using low-torque input rotation to part mechanical friction pads off a rubber ring. - Transistors looped back through junctions behave as diodes, complementing the dedicated mechanical ratchet diode.

[11:41](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=701s) Mechanical Filters and the 1997 Two-Capacitor Paradox

Electronic filtering translates seamlessly into mechanical chain drives. In a high-pass filter, an inductor (represented by a heavy flywheel) sits in parallel with the load. Because an inductor acts like a short circuit to low frequencies and direct current (DC), turning the input with a continuous, steady DC force simply spins up the heavy flywheel, shunting all the mechanical motion away from the ammeter. When Steve rapidly wiggles the input back and forth at high frequency, the flywheel's rotational inertia prevents it from responding quickly, behaving like a circuit break and forcing the rapid oscillations to pass directly through to the ammeter. Swapping the inductor for a capacitor reverses the behavior to create a low-pass filter, since capacitors behave like short circuits to rapid alternating currents and open breaks to DC.

This behavior reminds Steve of his first scientific publication from his school days: “When I was 18, I wrote a scientific paper and got it published in a journal.” Encouraged by his physics teacher, Mr. Parkinson, who offered to submit the student piece to the *Physics Education* journal, Steve had tackled the classic two-capacitor paradox: when a charged capacitor is wired in parallel to an identical uncharged capacitor, exactly half the charge transfers, reaching equilibrium. Calculating the stored energy ($E = \frac{1}{2} C V^2$) reveals that exactly half the original energy has vanished.

Steve had conceptualized this paradox using two equal water tanks linked by a bottom pipe. Opening the valve causes water to rush into the empty tank; rather than stopping neatly at the midpoint, the momentum of the moving water causes it to overshoot the level, sloshing back and forth and dissipating half the total potential energy as frictional fluid heat until equilibrium is reached. In electrical circuits, where electrons lack meaningful momentum, the overshoot and oscillation stem from the stray inductance inherent in the connecting wires. Steve tests this directly on Spintronics by linking a wound capacitor to a slack one without adding an explicit flywheel. Despite stray mechanical resistance damping the motion fairly quickly, the two sprockets clearly oscillate back and forth past their midpoint before settling: “Not for long because there's a decent amount of stray resistance in Spintronics as well, but it's nice to have my paper from 1997 validated.”

- High-pass filters use parallel flywheels (inductors) to bypass steady DC while allowing high-frequency oscillations through to the output. - Low-pass filters replace the flywheel with a torsion spring (capacitor), shunting fast wiggles and passing slow motion. - Connecting two matched capacitors loses 50% of total energy to transient oscillations driven by stray inductance, damping out via internal resistance.

[14:58](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=898s) Rectifiers, Logic Gates, and Mechanical Computing

To turn alternating current into direct current, electronics uses a full-bridge rectifier consisting of four diodes arranged in a diamond bridge. In a software simulation of Spintronics, Steve shows AC oscillation on one side converted into pulsating DC at the load, which can then be smoothed out using smoothing capacitors. While real electrical circuits require four diodes to accomplish full-wave rectification, Spintronics can do it with only two diodes. Steve explains the mechanical advantage: “The reason you can do it with Spintronics but not with regular electric circuits is because with Spintronics, it's really easy to change the direction of current, which it isn't in normal circuits.” Crossing chains over or routing them around opposing sides of a junction sprocket effortlessly reverses current direction without extra active semiconductor gates.

Steve moves on to computational elements, demonstrating how Spintronics functions as a mechanical digital computer: - **Set/Reset Flip-Flop:** Flipping a mechanical input switch triggers a volt-meter to register an active state; turning that switch off leaves the voltmeter displaced in memory until a secondary reset switch is engaged to restore it back to zero. - **XOR Logic Gate:** Turning either switch one or switch two independently routes mechanical power to register on an ammeter; activating both switches simultaneously creates opposing mechanical movements that cancel out, leaving the ammeter at zero.

Steve reflects that these basic elements could be chained together to build an entirely mechanical binary adder, echoing his past water-computer project. He signs off by thanking Paul from Spintronics and giving credit to Mr. Parkinson “for being just a brilliant physics teacher.”

[16:36](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=996s) Sponsor: Private Internet Access

Steve transitions into a sponsor segment for Private Internet Access (PIA), outlining how an unencrypted connection exposes internet browsing data. While SSL protects internal site content, ISPs, network administrators, and local governments can still catalog visited domain names to compile comprehensive tracking profiles. Steve praises PIA’s ability to bypass regional geoblocks, describing a trip to Spain where PIA allowed his family to stream a difficult-to-access service and avoid disruptive locale redirects on UK shopping websites. He emphasizes his preference for PIA's transparency: “But the reason I choose PIA is because all their software is open source.”

- PIA features servers distributed across more than 83 countries and covers all 50 individual US states. - The service maintains a strict no-logs policy backed by independent third-party audits and verified in court cases. - Steve offers an 82% discount via `piavpn.com/stevemould`, pricing the subscription at $2.11 per month with 3 free additional months. - A single account supports up to 10 simultaneous device connections and includes 24/7 customer support alongside a 30-day money-back guarantee.

## Notable quotes > [07:10](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=430s) An inductor resists the buildup of current, but once there is a current, it resists the reduction of current. — Steve Mould

> [01:41](https://www.youtube.com/watch?v=QrkiJZKJfpY&t=101s) Even quantum mechanics is just a model, and it's wrong in the sense that we know it's not a true description of the universe because it doesn't play well with relativity, but we still use quantum mechanics because it's insanely accurate under most circumstances. — Steve Mould

## Mentioned Steve Mould, Spintronics, Upper Story, Veritasium, Private Internet Access, Mr. Parkinson, Paul, Physics Education

## Video description Get 82% off Private Internet Access and 3 extra months free: https://piavpn.com/stevemould

Spintronics has mechanical resistors, inductors, transistors, diodes batteries and capacitors. When you connect them together with chains, they give a really good intuition for how circuits works.

Spintronics: https://upperstory.com/spintronics/

Veritasium's videos: https://youtu.be/bHIhgxav9LY https://youtu.be/oI_X2cMHNe0

You can buy my books here: https://stevemould.com/books

You can support me on Patreon and get access to the exclusive Discord: https://www.patreon.com/stevemould

just like these amazing people:

Frank Hereford Will Ackerly Brendan Williams Cameron Leigh Middleton Matthew Cocke Frederic Merizen Jeremy Cole Pavel Dubov Lizzy and Jack Alan Wilderland Joel Van der loo Glenn Watson Doug Peterson Paul Warelis John Zelinka Alnitak Grant Hay Heather Liu Marshall Fitzpatrik Lukas Biewalk JJ Masson Ben McIntosh Damien Szerszinski

Twitter: http://twitter.com/moulds Instagram: https://www.instagram.com/stevemouldscience/ Facebook: https://www.facebook.com/stevemouldscience/ Buy nerdy maths things: http://mathsgear.co.uk

[Source](https://www.youtube.com/watch?v=QrkiJZKJfpY)

中文简报

Steve Mould用机械链条复刻半导体电路机械链条系统能够复现电感反冲和晶体管开关等复杂电子现象，因为物理扭矩与链条速度直接对应于电压与电流。 Steve Mould展示了Spintronics的机械原理：利用物理链条、扭簧、飞轮和差速器来模拟电流、电压、电容器和并联节点。他将这些有形的机械装置与更深层次的物理概念联系起来，包括Veritasium关于能量传输的坡印亭矢量争论，以及他自己在高中时期发表的关于解决双电容悖论的论文。在整个视频中，Mould构建了可运行的机械高通与低通滤波器、基于晶体管的峰值电压检波器，以及触发器和异或门等数字逻辑元件。该演示突显了实体集总元件机械模型如何为理解复杂的电磁学原理提供直接且直观的途径。 Steve Mould展示了由Upper Story开发的机械电路系统Spintronics，该系统利用链条速度表示电流，链条张力表示电压，摩擦力提供电阻。 (0:00) Spintronics通过机械方式复刻先进的电子元件，利用扭簧模拟电容器，配重飞轮模拟电感器，周转轮系差速器模拟并联节点和直流电压变换，摩擦夹模拟晶体管。 (4:12) 与需要四个二极管的电气全波桥式整流器不同，Spintronics仅用两个二极管即可完成全波交流至直流整流，方法是通过在机械上交叉链条路径来改变电流方向。 (14:58) 该套件复现了真实的电子现象，例如电感反冲断开开关，以及通过LC瞬态振荡来解决经典的双电容能量悖论。 (7:10)

Source quotes

[7:10] Steve Mould: An inductor resists the buildup of current, but once there is a current, it resists the reduction of current. [1:41] Steve Mould: Even quantum mechanics is just a model, and it's wrong in the sense that we know it's not a true description of the universe because it doesn't play well with relativity, but we still use quantum mechanics because it's insanely accurate under most circumstances.

Original transcript

[0:00] - This is a mechanical circuit, [0:02] which is to say it's a collection [0:03] of mechanical components that together [0:05] are a really robust analogy for electric circuits. [0:09] You've got resistors, transistors, capacitors, [0:12] ammeters, batteries, inductors. [0:15] It's really incredible what you can build with this thing, [0:17] and for me, it's really interesting to see my intuitions [0:20] about electrical circuits in physical form. [0:23] I can feel voltages and see currents. [0:27] It's called Spintronics. There's a link in the description. [0:29] They're not paying me at all to make this video. [0:31] I want to be clear about that, [0:32] but when I saw the Kickstarter, [0:34] I just knew I had to make a video [0:36] about it because I love an analogy. [0:39] Let's have a look at really simple circuit. [0:41] So this is the battery and this is a resistor [0:44] and the two are joined together by a chain. [0:47] When I charge up the battery by pulling this string, [0:49] the battery then discharges its energy into the resistor. [0:53] What's really cool is you can feel the resistance [0:56] like this one, which is a hundred ohms, [0:58] has a certain amount of resistance. [1:00] This one, which is 200 ohms, has twice that resistance. [1:03] You can feel it in your hands and look, [1:05] when you put the 200 in there, the chain moves [1:08] at half the speed of when the hundred was in there. [1:11] So the speed of the chain is like the amps of a circuit. [1:14] You can connect two resistors in series like this [1:17] and the chain moves even more slowly. [1:19] You can also connect resistors in series like this. [1:21] That's a bit counterintuitive at first, [1:23] because it kind of looks like they're in parallel [1:26] but this is what a series looks like [1:28] in a Spintronics circuit. [1:30] It's a bit like the analogy of water flowing through pipes [1:32] that you often hear when talking about electronic circuits, [1:34] but instead of water being pushed through pipes, [1:37] it's a chain that's being pulled [1:39] and it's pulling on different components. [1:41] This is probably a good time to address [1:43] the Veritasium-shaped elephant in the room. [1:46] You probably saw his video where he talks [1:48] about how electronic circuits really work. [1:51] It's the one about the circuit [1:52] that's a light second in length. [1:54] It's a really important video, actually two videos [1:56] 'cause he did a followup that I recommend you watch [1:58] if you haven't already 'cause he clears a few things up. [2:00] They're important videos because it's important [2:02] to keep in mind what's really going on [2:05] beneath the models that we have [2:07] in our heads of how things work. [2:10] In the videos, he explains that the energy [2:13] in a circuit is carried by the electric [2:15] and magnetic fields that hug the wires [2:18] and components of the circuit, [2:20] and that kind of puts the lie to the hydrodynamic model, [2:24] doesn't it, because if you only have that [2:27] water flowing through pipes image in your head, [2:30] then you're thinking, okay, so I push water [2:33] at this end of the pipe, which pushes on the water [2:36] next to it, which pushes on the water next to it, [2:38] which pushes on the water next to it, [2:39] until this bit of water falls out the other end of the pipe, [2:42] and that bit of water can be used to do work. [2:46] And if that's the only model that you have, [2:48] then you might think, well, it's the same with electrons. [2:50] I'm pushing the electrons with a voltage [2:52] through this end of the wire, which pushes on the electrons [2:54] in front of it, which pushes on the electrons in front of it [2:57] until eventually these electrons at the end, [2:59] they can be pushed out of the wire and do work, [3:02] but that's not how it works. [3:04] The energy is carried by the electric and magnetic fields, [3:08] though to be fair to the electrons, [3:09] those electric and magnetic fields belong to the electrons, [3:13] so it is kind of fair to say that it's the electrons [3:16] that are carrying the energy. [3:18] It's just that they're not pushing on each other [3:20] in the way that the water in a pipe model [3:23] might lead you to believe. [3:25] But to make a broader philosophical point, [3:27] everything we know in science is just models. [3:30] Even quantum mechanics is just a model, [3:32] and it's wrong in the sense that [3:35] we know it's not a true description of the universe [3:37] because it doesn't play well with relativity, [3:40] but we still use quantum mechanics [3:42] because it's insanely accurate under most circumstances. [3:46] So you might have the water in a pipe model [3:50] of electric circuits in your head, but it's important [3:53] to know that underneath that, it's the electric [3:56] and magnetic fields that are carrying the energy. [3:59] But even that's a model, right? [4:00] I mean it's models all the way down. [4:01] But the point is when you're doing science, [4:03] you pick the model that is most appropriate [4:05] for the thing that you are doing. [4:06] This mechanical circuit model is good for intuition as well, [4:10] so let's explore some of those. [4:12] This is the Spintronics equivalent of a capacitor. [4:15] The more you try and turn it, the harder it gets to turn, [4:18] in the same way that the more you try and shove charge [4:21] into an actual capacitor, the harder it gets [4:24] and look, you can even see there's numbers on the dial [4:27] so you can see how charged up it is. [4:29] Actually this doubles as a volt meter, [4:31] so the numbers here are in volts. [4:34] A quick note on units, actually. [4:36] this says a hundred ohms, but actually [4:38] it's a hundred spin ohms, and this is spin volts. [4:42] What's really nice is they're all analogous [4:44] to the real things. [4:46] Like if we say that 10 meters of this chain [4:49] is equivalent to one coulomb of charge, [4:52] then all these other values like spin ohms [4:55] and spin volts and stuff written down here, [4:58] they're all equivalent then to the real thing. [5:01] Look what happens if I connect a resistor [5:03] and a capacitor to the battery. [5:05] You might think that these are in parallel, but look, [5:07] once the capacitor is charged up, the current stops flowing [5:10] through both the capacitor and the resistor. [5:13] That's because this is actually a series circuit. [5:15] So how do you achieve parallel circuits with Spintronics? [5:19] Well, you need junctions, [5:21] and actually junctions are more complicated [5:23] with Spintronics than they are in electric circuits. [5:27] Well, that's because they're incredibly simple [5:28] in electric circuits. [5:29] Look, here's a junction. [5:31] but this is a junction in Spintronics. [5:34] It's actually like a differential gear in a car. [5:37] If I hold this still, these two things move. [5:40] If I hold this one still, these two move. [5:44] If I hold this one still, these two move. [5:47] That means whatever voltage you apply [5:49] to the bottom sprocket will actually come out [5:52] of the top two sprockets, and vice versa for any pair. [5:56] So remember when I had these two resistors in series, [5:59] they span at the same rate, just like an electric circuit. [6:02] The current through two resistors in series [6:04] will be the same, but look. [6:06] When I put them in parallel, the smaller resistor [6:09] spins much more quickly and that's as we would expect. [6:12] We expect a greater current [6:14] to flow through the lower resistor. [6:16] Remember, these capacitors are also volt meters, [6:18] so if I put a volt meter in parallel with this resistor, [6:21] then I'm measuring the voltage across the resistor. [6:24] That's pretty cool, isn't it? [6:25] One thing that's nice about a junction [6:27] is you can feel the voltage. [6:30] If I hold the top sprocket of the junction with my fingers, [6:34] I can feel it pushing against me. [6:36] That's the voltage. [6:38] Another cool thing about junctions [6:39] is that you can use them as a transformer. [6:42] In this circuit, this resistor is attached [6:45] to the junction twice, and this free spinning component [6:48] is attached to the remaining sprocket. [6:50] Powering this with my hand, I only feel half the resistance, [6:53] but the resistor turns at half the speed. [6:56] That's mechanical advantage. [6:58] So transforming voltages in Spintronics [7:01] is actually a lot easier than in regular circuits. [7:05] You can easily do it with DC, [7:07] whereas with electrical circuits, [7:08] you need alternating current. [7:10] There's even something analogous to an ammeter, [7:12] a device that indicates the amps in a circuit, [7:14] how much current is flowing. [7:16] It doesn't give you a reading. [7:17] Instead it gives you a pitch. [7:19] All along this rim, there are these serrations. [7:23] If I scrape it with my nail, you can kind of hear it. [7:25] Under here, there's a ridge that is connected [7:28] to this gramophone amplifier. [7:30] So the faster this thing turns, the higher the pitch. [7:34] (meter playing high-pitched tone) [7:40] If I add it to this circuit here, [7:41] that gives you a sense of what the current is like. [7:44] If I put it against this smaller resistor, [7:46] the pitch goes up. [7:47] In series with a capacitor, when I turn the switch on, [7:50] you can hear how current flows quickly into the capacitor [7:54] but slows down as it gets full. [7:57] With this setup, I can charge and discharge [7:59] the capacitor with these two switches. [8:02] This is probably my favorite component. It's an inductor. [8:05] This is what an electrical inductor looks like. [8:07] It's a coil of wire and how does one of these things behave? [8:12] For me, it was one of those things [8:13] that wasn't super easy to get an intuition for. [8:17] An inductor resists the buildup of current, [8:20] but once there is a current, [8:22] it resists the reduction of current. [8:26] It's all mediated by magnetic fields, [8:28] but with this Spintronics analogy, [8:30] it's all about inertia and momentum, [8:32] which is really intuitive. [8:35] Because of these weights around here, [8:37] it's hard to get the thing spinning, [8:39] and once it's spinning, it's hard to stop it. [8:42] Connecting an inductor to a capacitor [8:44] really gives a nice intuition about both of these components [8:47] and how they might behave in a circuit. [8:49] Because it's hard to stop current flowing [8:51] through an inductor, it can cause damage [8:53] to your circuit if it's not designed well. [8:56] In fact, the same thing happens with Spintronics circuits. [9:00] This is a switch here and when I turn it on, [9:02] the inductor receives power. [9:04] It eventually spins up and then watch what happens [9:06] when I turn the switch off. [9:08] It destroys the switch. [9:10] If you have a problem like that in a circuit, [9:12] then you add a resistor in parallel. [9:14] So now when I turn the switch off, [9:16] the inductor can dump its energy through the resistor. [9:20] By the way, if there's nothing connected to the battery, [9:22] it spins like crazy, which is exactly what happens [9:25] if you connect the two ends of a battery together. [9:28] The battery will overheat because of [9:30] the immense flow of current through it. [9:32] There's a little seatbelt-type brake inside [9:34] to prevent damage to the Spintronics battery [9:37] If there isn't anything connected, [9:38] or if you make a circuit like this [9:39] and you realize that there's nothing attached [9:41] to this part of the junction. [9:42] Actually that illustrates something quite quirky [9:44] about Spintronics, which is that components [9:47] are by default closed circuits. [9:49] If you charge up a regular capacitor and then remove it [9:52] from the circuit, it holds its charge. [9:55] But look, if I charge up this Spintronics capacitor [9:59] and then let go, it discharges. [10:01] So unless a Spintronics component is connected [10:04] to something else, it's probably connected to itself. [10:07] This is really cool. It's a transistor. [10:09] Look, no current can flow through this bottom sprocket [10:13] because of these clamps clamping down on that rubber ring. [10:17] But if I apply a small voltage to the top sprocket, [10:20] it opens up these clamps and the bottom sprocket can spin. [10:23] You might know that you can use [10:25] a transistor as an amplifier. [10:26] So a small voltage applied up here [10:29] can lead to a large current flowing down here. [10:32] Here it is in a circuit. [10:33] This capacitor that's acting as a volt meter here [10:36] is showing the small change in voltage [10:38] that I'm supplying with my fingers. [10:40] This capacitor/volt meter is showing [10:42] the much bigger amplified voltage across the load. [10:45] You can even make a diode by connecting [10:47] a transistor to itself via a junction. [10:50] Look, it will spin one way but not the other, [10:53] just like a diode. [10:55] So let's see if we can build some interesting circuits. [10:57] This is a peak voltage detector. [11:00] So I can apply a voltage manually with my hand [11:04] and this capacitor/volt meter is showing me that voltage, [11:09] but this capacitor/volt meter is showing me [11:12] the peak voltage. [11:14] Because this pair of components as a diode, [11:18] the capacitor won't discharge when I let go over here. [11:23] What's brilliant about all this [11:24] is I can feel it as it happens. [11:27] It's really intuitive. [11:28] Spintronics actually comes with a dedicated diode component, [11:32] so look, here's a simplified version of the circuit. [11:35] The diode works by a simple ratchet mechanism. [11:38] You can turn it in one direction but not in the other. [11:41] This is a high pass filter, so I can send a voltage signal [11:46] into the circuit with my hand by spinning this thing. [11:50] I can do it at high frequency, [11:51] I can do it at low frequency, [11:53] and a high pass filter will filter out [11:55] the low frequency signals, [11:57] leaving the high frequency signals. [11:59] But how does it do that? [12:00] Well, this is an inductor and it's in parallel. [12:03] You can think of an inductor as a short circuit [12:06] for low frequencies and DC. [12:09] After all, DC is just very low frequency. [12:12] So if I try and turn this thing with a constant force, [12:15] once the inductor spins up, then it's basically [12:18] a short circuit and none of that DC signal [12:21] will get through to the ammeter. [12:23] An inductor is a bit like a circuit break [12:26] for high frequency signals. [12:28] So if I move this thing back and forth quickly, [12:31] then we do hear it at the ammeter [12:34] because it doesn't have much of an effect on the inductor. [12:37] To make a low pass filter, [12:38] you replace the inductor with a capacitor. [12:41] A capacitor is like a short circuit for high frequencies [12:44] but a circuit break for low frequencies. [12:47] This is an oscillating circuit if you ever need AC [12:52] for something in your Spintronics world. [12:54] Here's a fun story. [12:55] When I was 18, I wrote a scientific paper [12:59] and got it published in a journal. [13:01] That sounds more impressive than it actually is. [13:04] At the time, my physics teacher, Mr. Parkinson, [13:06] said something like, "You know that idea you had? [13:09] I could probably get it published [13:10] in Physics Education Journal." [13:12] They had this section for student submissions, [13:15] so that's what it was, but I still like to tell people [13:17] that I got a paper published when I was 18. [13:19] But the idea was this. [13:20] If you get a charged capacitor and you connect it [13:23] directly to an equivalent uncharged capacitor, [13:27] then half the charge will leave the charged capacitor [13:30] and end up in the uncharged capacitor. [13:32] They reach equilibrium with each other. [13:34] What's surprising is if you calculate the energy [13:37] of those two capacitors after you connect them, [13:40] you have exactly half the energy you had before. [13:44] But where does that energy go? [13:46] I had this idea that instead of thinking about capacitors, [13:49] think about tanks of water. [13:50] So you have one tank of water that's full [13:53] and another that's empty, and it's connected [13:55] at the bottom with a pipe. [13:56] When you open that pipe, water will flow [13:59] from the full tank to the empty tank until they're equal. [14:03] But actually because of momentum, there will be an overshoot [14:06] and the second tank will go slightly above halfway [14:10] and then it will relax down again [14:11] and it will oscillate backwards and forwards [14:13] until they have an equal amount of water in both. [14:16] But every time the water sloshes backwards and forwards, [14:19] it loses energy through friction [14:21] and that's where the energy goes. [14:23] Electrons in a circuit don't have momentum [14:25] in a meaningful way, but the wires between the capacitors [14:28] do have stray inductance, so I wanted to see [14:31] if the same thing would happen with Spintronics. [14:33] If I charge up one capacitor and connect it to another, [14:37] will they oscillate backwards and forwards [14:39] until they both have half the charge, and will it work [14:43] with just the stray inductance in Spintronics? [14:46] And look, there they go. [14:48] They do oscillate backwards and forwards. [14:49] Not for long because there's a decent amount [14:51] of stray resistance in Spintronics as well, [14:54] but it's nice to have my paper from 1997 validated. [14:58] What about a full bridge rectifier? [15:00] Can we make one of those? [15:01] A full bridge rectifier converts AC to DC with four diodes. [15:06] As you can see here, as I supply AC here, [15:09] you get DC at the load here. [15:12] It's not very smooth but you can fix that [15:14] with a couple of capacitors. [15:16] I only have two diodes, which is why I'm showing you this [15:19] in simulation, but luckily with Spintronics, [15:21] you can make a full bridge rectifier with just two diodes. [15:25] The reason you can do it with Spintronics [15:26] but not with regular electric circuits [15:29] is because with Spintronics, it's really easy [15:31] to change the direction of current, [15:34] which it isn't in normal circuits. [15:35] Depending on how you chain up a junction, [15:37] you can flip the direction of current flow. [15:40] What about computation? [15:41] Could we use this to build a computer of some kind? [15:45] Well, look, here's a flip flop. [15:47] When I turn this switch on, [15:49] it registers at this volt meter [15:51] but when I switch it off, the volt meter stays on. [15:54] This switch puts the volt meter back to zero, [15:57] and so I can flip flop between the two. [15:59] This is an XOR logic gate. [16:02] If switch one is on, we get a reading at the ammeter. [16:04] If switch two is on, we get a reading at the ammeter. [16:07] But if both switches are on, we don't. [16:12] So maybe we could use this stuff to build [16:14] a binary adder like I did with my water computer video. [16:18] If I ever get that working, I'll make a follow up, [16:20] but in the meantime, is there anything I've missed? [16:22] Is there a really cool circuit [16:24] that you would like to see built with Spintronics? [16:27] Let me know in the comments. [16:28] Thanks to Paul from Spintronics [16:30] for all his help with this video. [16:31] And thanks to Mr. Parkinson for being [16:33] just a brilliant physics teacher. [16:36] VPNs can be really useful. [16:39] I recommend the one I use, Private Internet Access. [16:42] They're sponsoring this video. [16:43] A VPN encrypts your web traffic [16:45] and hides your IP address. [16:47] Without a VPN, your internet service provider, [16:49] network administrator, and potentially your government [16:52] depending on where you live, can see and record [16:55] the websites you visit. [16:56] Probably not the content of the websites you visit [16:58] because that's encrypted with SSL usually, [17:01] but typically they can see the domain names [17:03] and build up a profile of you. [17:05] With Private Internet Access, none of that can happen [17:08] and so your digital life is safer and more private. [17:11] Here's another use case. [17:12] I was in Spain at the end of summer [17:14] and we wanted to keep watching a show [17:17] on a streaming service that is notoriously [17:19] difficult to get working abroad, even with a VPN, [17:23] and it worked flawlessly with Private Internet Access. [17:26] One thing I found quite annoying when I was in Spain [17:28] was trying to do any kind of home-based admin, [17:31] like trying to buy something to be delivered [17:35] to the UK, for example, was really hard, [17:38] 'cause websites were always redirecting me [17:40] to the Spanish version. [17:41] I can't speak or read Spanish. [17:44] If you have any kind of home admin to do while traveling, [17:47] it can be insanely useful to trick your device [17:50] into thinking you are at home. [17:52] When you connect your device to a VPN, [17:54] you can choose which country it appears to be in. [17:57] And with Private Internet Access, [17:58] you can choose from over 83 countries [18:00] and every single US state. [18:03] They also have a no log policy that's been tested in court [18:06] and recently verified by an independent audit. [18:09] But the reason I choose PIA is because [18:12] all their software is open source. [18:14] I'm not suggesting that everyone [18:15] who uses Private Internet Access [18:17] should audit the software on GitHub, [18:20] but it is out there so people have audited it. [18:23] It's a level of transparency that I value. [18:26] The deal on this one's really good. [18:27] If you go to my special URL, piavpn.com/stevemould, [18:32] you'll get 82% off. [18:34] That's just $2 11 cents a month [18:38] and you get three months free on top of that. [18:40] With one subscription, you can protect up to 10 devices [18:43] and since it comes with a 30-day money back guarantee, [18:47] and 24/7 customer support, you can try it out risk free. [18:52] The link is also in the description [18:53] so check out Private Internet Access today. [18:56] I hope you enjoyed this video. [18:57] If you did, don't forget to hit subscribe [18:59] and the algorithm thinks you'll enjoy this video next. [19:02] (upbeat music)