Switches in Series: The Hidden Logic of AND

In the first post in this series, we learned that a relay is a switch thrown by electricity — one circuit controlling another. That single device already hints at logic. But one switch by itself can only answer one question: on or off?

To build something that computes, you need switches working together. This post is about the simplest way to do that: wire two switches in series. It’s a circuit you could sketch in thirty seconds, but folded inside it is the first piece of formal logic a computer ever needs. By the end of this post, you’ll have seen AND — not as a diagram symbol, but as a physical fact about how current flows through wire.

The world’s most important job nobody remembers

Picture a telephone exchange in the mid-1890s. Rows of operators — mostly young women, headsets on — face long wooden panels covered in numbered jacks, crisscrossed with cords, bristling with signal lamps. A call arrives: a lamp flashes. An operator reaches across, plugs a cord from one jack to another, and two distant people can speak. When they hang up, she unplugs the cord and both jacks fall silent again. At peak hours she’s managing a dozen calls at once. Across a large city exchange, hundreds of operators, thousands of cords, tens of thousands of jacks — the largest switching machine humanity had ever built — and every connection made by human hands.

That human bottleneck bothered at least one Kansas City businessman. Almon Strowger, a funeral director, became convinced that the local telephone operator was routing calls meant for him to a competitor instead — perhaps on purpose, perhaps not. His response, patented in 1891, was to eliminate the operator entirely: the Strowger automatic exchange, the world’s first practical automatic telephone switch. A caller would dial a number digit by digit, and a bank of electromagnetic step-relays — close cousins of the device from Post 1 — would spin through their contacts, routing the call without any human intervention.

Strowger and the engineers who refined his invention over the following decades wired together thousands of relays in deliberate, calculated combinations. The fundamental question those combinations had to answer, over and over, millions of times a day, was exactly the question we’re going to answer now with just two switches and a lamp: when you have a chain of switches that current must pass through, what conditions have to be true for the circuit to close?

A single path, multiple gatekeepers

Take the simplest circuit possible: a battery, a wire, and a lamp. Now break the wire in two places and put a knife switch at each break. Call them Switch A and Switch B.

Current from the battery has exactly one job: travel from the positive terminal, through the wire, through the filament of the lamp, and back to the negative terminal. To complete that journey, it has to pass through the gap at Switch A and the gap at Switch B. There is only one road, and each switch is a gate on that road.

Work through the possibilities:

• A open, B open. Two blockages. The road is broken twice. Lamp off.
• A closed, B open. Current passes Switch A and immediately hits the open gap at B. Stopped cold. Lamp off.
• A open, B closed. Current might pass Switch B, but it never even gets there — Switch A is blocking the path before it. Lamp off.
• A closed, B closed. Every gate is open. The road is clear from battery to lamp and back. Current flows. Lamp on.

Three combinations out of four fail. Only one succeeds. And the reason is structural: a series circuit has a single path, and every switch on that path must cooperate. A single dissenting switch — anywhere in the chain — breaks the only road.

You don’t need a textbook to feel that logic. You feel it every time you need two keys to open a lock, or two signatures to authorize a document. Both, or nothing.

The truth table: AND in physical clothing

Let’s write that rule in the shorthand electronics uses. Treat a closed switch as 1 (conducting, true) and an open switch as 0 (not conducting, false). Treat the lamp as 1 when lit and 0 when dark. There are exactly four combinations of two switches — here they all are:

One row out of four produces a lit lamp. Every other combination leaves it dark. The lamp is the physical implementation of logical AND — the operation that is true only when every one of its inputs is true simultaneously.

This is the same four-row table printed on every AND gate datasheet in existence. Now you know where those rows come from: they are not a definition someone invented in a mathematics textbook. They are an observable fact about a wire with two breaks in it.

Notice, too, what the table says about asymmetry — or rather, the absence of it. The second and third rows both give 0, and they are mirror images of each other. Switch A closed with B open is the same outcome as B closed with A open. In series, position on the wire does not change the logic. It makes no difference whether Switch A is closer to the battery or Switch B is. The condition is the same: both, or the lamp stays dark. AND is symmetric.

Try it yourself — predict before you look

Below is a live simulation of two knife switches wired in series with a lamp. Before you touch anything, make a prediction about each of the four rows in that truth table. Which ones do you expect to light the lamp? Which ones do you expect to leave it dark?

Then work through all four combinations, one at a time, and check each row against the table above.

Three things to pay attention to as you play:

1. Open just one switch after both are closed. While both are closed and the lamp is lit, open only Switch A. The lamp goes dark instantly — even though Switch B is still closed, still cooperating, still offering its half of the path. One open switch anywhere in a series chain vetoes the entire circuit. That veto is not metaphorical; it’s physical.

2. Try every order. Close B first, then A. Close A first, then B. Both sequences end with the lamp lit, and the intermediate steps — where only one is closed — both leave the lamp dark. The truth table is symmetric in both directions. AND doesn’t care which input you set first; it only cares whether they’re both set together.

3. Turn your sound on and count the clicks. Each state change produces one click. Four clicks to walk the full truth table: two to close, two to open. In the relay computers of the 1940s, a roomful of switches clicking in rhythm was how engineers tracked a computation — a wrong pattern told you something was misbehaving before any readout confirmed it.

Why a series wire is worth understanding deeply

When the AND operation gets promoted to a diagram symbol — the flat-backed D shape on a logic schematic — something useful happens. Abstraction lets you think about hundreds of gates at once without tracking every wire. But something also gets left behind: the reason the rule is what it is.

AND is not arbitrary. It is the inevitable behavior of a single path with multiple gatekeepers — and that grounding matters for what comes later. Chaining ANDs is just adding more switches to the same series line. Three switches in series, all must be closed. $n$ switches, all $n$. The truth table expands to $2^n$ rows, but the lamp still lights on exactly one: the row where every input is 1.

Claude Shannon made this rigorous in his 1937 master’s thesis at MIT, now routinely called the most important master’s thesis ever written. Shannon showed — formally, with proofs — that any circuit of switches can be expressed as an equation in Boolean algebra, and any Boolean expression can be implemented as a circuit of switches. The series circuit, in Shannon’s notation, is simply the product $A \cdot B$: a multiplication that equals 1 only when both factors are 1. In his algebra, the telephone operators of the 1890s were evaluating Boolean expressions in real time, tens of thousands of times an hour, without knowing it.

Shannon’s insight will underpin every post from here to the end of the series. The reason it could work at all is that the switch — wired in series — already behaved like a mathematical operation before anyone wrote it down. We didn’t impose AND on the circuit. We read it off.

What you just learned — and what comes next

Add these to your mental toolkit from Post 1:

• Series wiring creates a single path. Every switch on that path must be closed for current to complete the journey.
• That physical rule is AND. True only when all inputs are true simultaneously.
• The four-row truth table is not invented — it is observed directly from the hardware.
• AND is symmetric. The order of switches on the wire does not change the outcome.

In Post 3 of Building a Computer from Relays, we’ll take those same two switches and wire them differently — this time in parallel, each on its own branch from battery to lamp. Current will have two roads to choose from, and the lamp will light if either switch closes. OR emerges from copper and wire just as naturally as AND did today, and the contrast between the two reveals something fundamental about what these operations actually mean.

Two posts in, two switch patterns, and already the backbone of logic is appearing without a gate symbol in sight. That’s the whole point of building up from wire: by the time Shannon’s algebra arrives, it won’t feel like mathematics handed down from above. It will feel like a language you already knew.

Ready to keep building? Every circuit in this series lives in the DigiSim Relay Lab — wire your own combinations, extend the series chain, and hear the clicks add up.