Campaign

A lamp that’s always on is a lamp you can’t turn off. Put a switch in the loop so the current has to pass through it, then close it.

A push button is a switch with no memory — it conducts only while your finger holds it down. Wire one in so the lamp lights while pressed and goes dark the instant you let go. Then hold it!

This rig worked yesterday. Today: nothing. Somewhere the loop is broken — find the gap and bridge it. Hover wires to see where the current stops flowing.

The kitchen and the hallway each get a lamp — and each lamp needs its OWN switch. Branches that share the battery but not each other’s fate: that’s a parallel circuit.

Ohm’s law time: V = I · R. A resistor in series steals some voltage from the lamp and calms the current down. Get the lamp into a gentle glow — between 0.1 W and 0.3 W.

Put two lamps in one loop — in SERIES — and something curious happens: the 9 volts get shared, and each lamp burns at a quarter of full power. Wire it up and see for yourself.

Two lamps, one battery — and BOTH must burn at full brightness. Chain them in series and each only gets half the voltage. There’s a better topology…

A doorbell: buzzer sounds only while the button is held. You’ve built this before with a lamp — now make some noise.

Three lamps in a chain — and none of them light. One filament has burned out and broken the whole loop, but every bulb LOOKS fine. Probe your way along the string, find the dead one, delete it and fit a fresh bulb.

This one arrived from the factory with TWO breaks in the wiring. Trace the loop, find both loose ends, and stitch them together.

A two-way switch (SPDT) doesn’t just break a circuit — it CHOOSES one. The common pin routes to the up throw or the down throw. Wire each throw to its own lamp and flip between them.

The classic hallway problem: ONE lamp, TWO switches — either one must toggle it. The trick: connect the two-way switches throw-to-throw with a pair of "traveler" wires. Electricians call this the veksel circuit.

Meet your instruments. An ammeter measures current and goes IN the loop (series). A voltmeter measures voltage and goes ACROSS things (parallel). Put both to work on this lamp circuit.

Three parallel resistors, one battery — but which branch hogs the current? Kirchhoff says the total splits by resistance. Prove it: wire your ammeter INTO the hungriest branch without disturbing the other two.

The lamp is dark, the battery is screaming. Someone added a wire that lets current skip the lamp entirely — a short circuit. Find the freeloader path and DELETE it.

A gadget needs exactly 3 V, but the battery gives 9. Build a voltage divider: two resistors in series split the voltage in proportion to their resistance. Feed the voltmeter 3.0 V.

Your level-16 divider has a problem: the moment a real gadget (that 1 kΩ load) hangs off it, the output sags. A loaded divider must be STIFFER than its load. Deliver 3.0 V — for real this time.

A potentiometer is a resistor with a movable tap. Wire one END and the WIPER in series with the lamp (that’s a rheostat), then drag the wiper slider until the lamp sits in the mood zone: 0.1–0.25 W.

A diode conducts in exactly one direction: from anode (the triangle) to cathode (the bar). Put one in series with the lamp and get the direction right — backwards, it blocks everything.

Two identical branches, but only one lamp lights. The dark branch has a resistor that failed open — it looks exactly like its healthy twin. Find it with the meter, replace it, and match the printed value.

Customer complaint: "the light switch rings the doorbell and the doorbell button turns on the porch light." The installer crossed the control wires. Untangle them: switch 1 → lamp, switch 2 → buzzer.

LEDs are greedy: above their ~2 V forward voltage they gulp current exponentially, and too much kills them. Give this one between 5 mA and 20 mA. Wire it straight to the battery and… well, you’ll see.

Two LEDs, one battery, and a fussy client: both must glow at the SAME safe brightness. You have two resistors — or do you even need both?

The spec demands the battery deliver 36 mA — but the parts bin only has 1 kΩ resistors, four of them. One alone gives 9 mA. Combine them until the numbers submit.

Stage lighting: three identical lamps, but the director wants full blast, half power, and a quarter glow — all at once, all from one battery. Series resistors are your dimmers.

A customer complains their lamp "barely glows". The wiring is fine — but someone fitted a wildly oversized resistor. Diagnose, then fix the VALUE, not the wiring.

Two lamps must both shine decently bright (≥0.35 W) — but the supply is fused at 150 mA. Straight parallel draws 200 mA; series is too dim. Find the middle path.

The Wheatstone bridge: two voltage dividers side by side, a bridge wire between their midpoints. When R1·R4 = R2·R3, the midpoints sit at the same voltage and the bridge carries nothing. Three arms are fixed at 1 kΩ — tune R4 until the bridge current dies (<100 µA).

A component of unknown resistance sits in the bottom-left arm. The top arms are 2 kΩ and 1 kΩ. Balance the bridge with your adjustable arm and the ratios will CALCULATE the mystery for you: R? · R2 = R1 · R4.

A relay is a switch thrown by an electromagnet: a SMALL coil current pulls a contact that switches a BIG load current. Drive the coil with the push button; put the lamp on the NO (normally-open) contact.

This lamp is rated for a gentle ~31 mA — but it is glaring, and running hot. One of the two ballast resistors has failed SHORT: zero ohms behind a healthy paint job. Find the one dropping no voltage and replace it.

The fuse is blown — that’s the symptom, not the disease. Somewhere a short is waiting to kill the NEXT fuse too. Find and remove the short FIRST, then click the fuse to fit a fresh one.

Main supply: 9 V. Backup battery: 6 V. Connect both straight to the lamp and they fight — current pours from the strong one into the weak one. Diodes make them take turns: main feeds the lamp, backup only steps in when main dies.

This rig has a fault: the test switch shorts the lamp when closed. Unprotected, the battery would melt. Fill the gap in the supply line with a fuse, then close the fault switch — the fuse must die so nothing else does.

Three LEDs, ONE resistor allowed. In series, they share a single current — but each takes its ~2 V toll first. Budget what’s left of the 9 volts and pick the resistor that lands all three between 5 and 20 mA.

This 12 V supply must feed a 5 V rail. A zener diode conducts backwards once its breakdown voltage is reached — clamping whatever is across it. Place the zener across the output, pointing the right way.

One 12 V supply, two clients: a 5 V rail and a 3.3 V rail, each read by its own voltmeter. Two zeners, two ballast resistors — build both regulators side by side. (A zener’s breakdown voltage is editable.)

A regulator is only as good as its budget. This zener rail feeds a variable load (the potentiometer). Drag the load heavier and heavier — find where regulation holds, and where it collapses.

The shop wants: doorbell position rings ONLY the buzzer, showroom position lights the lamp AND rings the buzzer. One two-way switch, two diodes, zero controllers. Route the current like mail.

The two supply rails arrive with UNKNOWN polarity — the pair of two-way switches flips them at will. Your motor must spin regardless. Four diodes, arranged as a bridge rectifier, always hand + to the top and − to the bottom.

Classic assembly-line blunder: the LED went in backwards, and diodes don’t forgive direction. Rip it out (Del) and fit a fresh one the right way round.

A main battery and a 7 V backup share the load through steering diodes. The lamp is on — so all good? It is dimmer than spec, and the backup is quietly doing ALL the work: the main battery is stone dead inside. Prove which source is carrying the load, then replace the dead one.

The adjustable supply can be cranked to lamp-killing voltages, and someone WILL crank it. Build the classic crowbar: a zener that slams the rail on overvoltage, dumping enough current to blow the fuse — sacrificing the supply line to save the lamp.

A capacitor stores charge like a tank stores water. Wire it to the battery through a resistor and watch the voltmeter: fast at first, slower as it fills — the RC curve.

Make an LED that fades in instead of snapping on: charge a capacitor through a resistor and let the LED drink from the capacitor. The bigger the cap, the slower the dawn.

The two-way switch chooses: mains power (up) or nothing (down). Add a capacitor across the load so that when the power drops, the LED rides through on stored charge.

A camera flash can’t pull huge power from a small battery — so it saves up. Charge the capacitor slowly through the resistor (switch up), then dump everything into the lamp at once (switch down).

An inductor is a flywheel for current: it hates change. Close the switch and the lamp doesn’t snap on — the current has to spin up through the coil first.

Opening a switch on a charged inductor is slamming a door on a freight train — the coil kicks back with a huge voltage spike. Give the current an escape lane: a freewheeling diode across the coil-and-lamp branch.

Same blackout, higher stakes: this time the light must HOLD through the outage — the win meter needs it continuously lit with the power cut. Your default capacitor won’t make it. Size it up.

One more blackout — but this supply room only stocks small capacitors. Good news: capacitors in PARALLEL add up. Build a bank big enough to hold the LED through the outage.

Two identical capacitors in SERIES: the 9 volts split evenly between them, and the pair together stores like HALF a capacitor. Close the loop and watch both gauges climb to 4.5 V — never 9.

Close the switch and this capacitor should fill to 9 V in a couple of seconds. Instead the resistor sits there warm, forever, and the tank never holds a volt. The capacitor has failed SHORT — a straight pipe to ground wearing a capacitor’s clothes. Prove it, replace it, and watch it actually charge.

Charge the capacitor, then swing it against the inductor: the tank empties into the coil, the coil overshoots and refills the tank backwards, and the current sloshes to and fro — an LC oscillator.

A button that stays pressed: wire the relay’s own NO contact so it feeds its own coil. Press START once — the relay grabs itself and holds. Add a switch in the coil path as STOP.

So far every control has meant "press = on". The relay’s NC (normally-closed) contact flips the logic: the lamp burns UNTIL the coil pulls the armature away. Wire the button to the coil and build your first inverter.

Bank-vault rules: the motor that opens the door runs only when BOTH key switches are turned. Two relays, their NO contacts chained in series — a hardware AND gate.

Light the lamp when the two switches DISAGREE — and only then. You solved this once with two-way switches; now build it from relays: each armature chooses a path, and the two paths must CROSS.

A doorbell that keeps a secret: the buzzer rings while the button is held, and a lamp LATCHES on to tell you someone came while you were out. The reset switch clears the memory.

The final build: a latching burglar alarm. The door button trips the relay; the relay must LATCH itself and hold the buzzer and lamp on until the guard opens the STOP switch. One wire is missing — the one that makes it remember.

Two conveyor motors must NEVER run together. Route each button’s supply through the OTHER relay’s NC contact: whoever energizes first physically disconnects the competition. Make the forbidden state unreachable.

This relay argues with itself: the coil is fed through its own NC contact, so pulling in cuts its own power. The simulator flags it — "keeps changing state". Rewire it into a proper latch: coil through NO, button as the starter.

Press the button: the relay should clack and the lamp should light. Instead — silence. The coil circuit is broken somewhere between the button and the coil, and the dropper resistor is the prime suspect. Find the open, replace it, and match the printed value — the coil needs real current to pull in.

Quiz-show lockout: first press latches YOUR lamp and freezes the other player out — even after release. Each relay’s single contact must do double duty: NO side holds its own coil and lamp; NC side supplies the RIVAL’s button. Host’s reset switch clears the round.

Build a lock from three two-way switches. The vault lamp lights ONLY for the secret pattern UP · DOWN · UP. Chain the switches so that every wrong throw parks the current on a dead stud.

The transistor: a whisper of base current commands a shout of collector current. Wire the push button to the base THROUGH a resistor and switch the lamp with milliamps. Never feed a base without a resistor — it has no self-control.

An emitter follower: the lamp sits under the emitter, and the emitter faithfully follows the base voltage (minus 0.7 V). Drag the potentiometer and dim the lamp smoothly — with almost no current from the knob.

The PNP transistor is the NPN’s mirror twin: its emitter lives at +, and it turns ON when you pull the base DOWN. Wire a resistor and switch from base to ground, and control the lamp from the high side.

A lamp that thinks: the photoresistor (LDR) is wired from the base to ground — in bright light its low resistance smothers the base. Add a resistor from + to the base, and the lamp will decide for itself when it gets dark.

One transistor multiplies current by ~100. Stack two — the Darlington pair — and you multiply by ~10,000: microamps in, a lamp’s worth out. Drive the pre-stacked pair through a HUGE base resistor and prove the feather can move the boulder.

Build a thermostat: the NTC thermistor drops resistance as it heats, lifting the base — the fan must kick in when things get hot. Add the bottom resistor that sets the tripping point.

One knob, two lamps, opposite moods: as the wiper rises, the NPN follower brightens the HIGH lamp while the PNP follower starves the LOW lamp — and vice versa. Wire both bases to the wiper and ride the see-saw.

Set a current ONCE — with a resistor in the reference leg — and let a second transistor copy it, no resistor needed. The trick: tie both bases together and short Q1’s base to its own collector. Twin transistors, twin currents.

The supply slider wanders from 7 to 15 volts, but the LED must hold a steady 9–15 mA through it all. Build a real current source: a zener pins the base, an emitter resistor turns that fixed voltage into fixed current. The LED upstairs never feels the storm.

This transistor switch should hold its lamp on, always. The lamp is dark, yet lamp and transistor both test fine. Work backwards: a transistor without base current is just a roadblock. Somewhere the base drive died — find the open part and replace it.

A "worked fine until I pressed the button" special. Inspect the base wiring: the button feeds the base DIRECTLY — one press and the junction becomes a fuse. Cut the direct wire and splice in a base resistor before testing.

The MOSFET switches like the relay did — but its gate is controlled by VOLTAGE, not current. Flip the motor on through the transistor and check what the control switch actually carries: nothing.

Your night light works, but WHERE is night? That’s a policy question. The potentiometer replaces the fixed pull-up: its setting decides how dark is dark enough. Prove control: at the SAME half-light, make the lamp choose ON — then re-dial and make it choose OFF.

The greenhouse must scream BEFORE the frost bites. Cold pushes the NTC’s resistance UP — put it under the base and its rising resistance lifts the base voltage until the Darlington pair wakes the buzzer. You supply the top resistor that sets the trip.

On/off cooling is crude. This fan should IDLE when cool, cruise at half over a warm chip, and howl at full when things get serious — a continuous curve, drawn by an emitter follower riding the thermistor divider. Add the bottom resistor that shapes it.

This lamp is picky: it burns only in the twilight BETWEEN too-bright and too-dark. Two thresholds guard it: the Darlington needs the sense node above ~1.4 V (some darkness), while the zener-fed "thief" transistor steals the base away once the node climbs past ~3.3 V (too much darkness). Walk the light through all three worlds.

Two light sensors, stacked into one divider, fight over the midpoint: sensor A pulls it toward +, sensor B drags it toward ground. The Darlington referee fires the relay — and the motor — only while A’s side is winning. Wire the midpoint to the referee’s base.

This "night light" salutes the sunrise — it burns all day and sleeps all night. Diagnose the divider: someone installed the LDR up top and the resistor below, inverting the whole policy. Swap the two (drag them) and restore darkness-sense.

A naive thermostat chatters — flipping madly at the threshold. This one doesn’t, because the relay itself has MEMORY: it grabs at 12 mA of coil current but only lets go below 4. Walk the greenhouse through a cold night and watch the dead-band work. (Everything is wired; the temperature slider is the story.)

The frost alarm from your greenhouse days — but now it sleeps through deep frost. Slide the temperature: nothing changes, ever. The NTC has failed SHORT, nailing the base to ground no matter how cold it gets. Prove the sensor is dead, replace it, and re-commission both worlds.

Two zones — the DOOR (push) and the WINDOW (toggle). Each trip must light its OWN zone lamp and set off the latched siren; the zone lamps must never lie about which zone fired. Diodes keep the stories straight; the latch relay keeps the grudge.

Control-room discipline: three fault switches, three indicator lamps, one shared horn — plus a TEST button that must light ALL lamps without honking the horn. Nine diodes, three per zone, form the matrix that keeps every path one-way. Welcome to diode logic, industrial grade.

Complaint: "the lamp blinks once when I connect the battery, then nothing." Someone wired a capacitor IN SERIES with the lamp. Caps pass changing current but block steady current. Get the lamp permanently lit.

Stage direction: flip the switch, count a beat, THEN the lamp rises. Build the delay: a resistor feeds a capacitor, the capacitor feeds the Darlington’s featherweight base — the lamp waits for the tank to cross ~1.4 V.

Cinema lights don’t snap off — they sigh. Press to charge the tank and light the lamp; release, and the follower drinks the capacitor dry so slowly the lamp takes seconds to die. Add the tank that makes the sigh.

Tank A holds 9 volts; tank B sits empty. Flip the switch and let A pour into B. Prediction time: where do they settle? (Spoiler in the goals — but notice something odd about the ENERGY when you get there.)

Stairwell rules: one press buys you light; the light decides when you’ve had enough. Put a fat capacitor ACROSS the relay coil — the button charges it instantly, and after release the coil drinks it for seconds before the armature lets go.

Slamming full volts into a big motor is how belts snap. Put the current-flywheel — an inductor — in series, and the motor must SPIN UP through a gentle middle before reaching full song. Size the inductance so the ramp is visible.

Tank A is charged. Between the tanks: a coil. Close the transfer switch and watch energy SLOSH — A empties through the inductor into B, overshoots, and comes sloshing back. Unlike the resistive transfer, almost nothing is lost. Catch the moments the goals demand.

This lamp demands more power than a 9 V battery can possibly push through it — 0.9 W is the physical ceiling. The heist: charge the capacitor in PARALLEL with the battery, then re-stack it ON TOP with the two-way switches. Nine plus nine is eighteen, and eighteen is a flashbulb.

Customer: "it takes a minute to charge and the flash is a polite cough." Diagnosis: someone fitted a 47 kΩ charging resistor (100× too slow) and a shrimpy 1000 µF tank (10× too small). Fix both VALUES — the wiring is fine.

A power-on delay: the capacitor slowly fills through its timing resistor until the transistor wakes the lamp. Except this one has been "warming up" since Tuesday. The RC node sits at zero, feeding nothing. Somewhere the timing chain is open — find it, fix it, and watch the delay actually end.

The trickiest timer: a pulse of FIXED length no matter how long the button is held. Secret: put the capacitor IN SERIES with the relay coil. The press slams a charging current through the coil — and as the cap fills, the current dies and the relay lets go mid-press. A bleed resistor re-arms the trick.

Three faults, one machine, no map. A gap starves something; a short smothers something; a diode faces the wrong way. Every station must run: both lamps bright, buzzer buzzing, battery calm.

Red, yellow, green — and NEVER two at once. No controller allowed: chain two two-way switches so the topology itself makes double-lighting impossible. Switch 1 picks red or "the rest"; switch 2 settles yellow versus green.

Two floors, two call buttons. A press must LATCH its floor lamp (the elevator remembers), and any latched call keeps the winch motor turning. The ARRIVAL switch wipes all calls at once. Two self-holding relays, one shared clearing path — and two diodes so the calls don’t gossip.

This beacon must NEVER go dark. Mains (the adjustable supply) does the daily work; the 9 V battery lurks behind a diode, awake but idle. Drag the mains to zero — brownout! — and the handover must be seamless. Honor rule: while mains is healthy, the battery may not leak a milliamp.

The yard light has standards: it burns only when it’s DARK **and** something MOVED recently. Darkness is judged by the pre-wired sensor stage; "recently" is a capacitor you charge with the motion button and let leak away. Two gates in series — both must say yes.

House-lighting console: ONE master knob sets the mood for three rooms, each with its own kill switch. An emitter follower does the muscle work — the pot only whispers. Wire the follower and hang the three switched branches off its emitter rail.

Full climate duty, two probes: the FROST probe at the seedlings arms a decisive relay heater; the ROOF probe drives the fan smoothly, harder as the glass bakes. In the mild middle — economy, nothing runs. Both circuits are yours to command through a real day’s weather.

The intern "improved" the demo board. Now the fuse is dead — and a fresh one will die the same way; something is starved; something is smothered; the LED faces backwards; and one resistor wears a value 20× too large. Five faults. Full restoration. Management is watching.

A commissioning bench: two lamp branches and a transistor stage fed by a reference divider. QA says ONE component on this board is counterfeit — failed short — and the third lamp is dead because of it. Five resistors, every one printing the right value. No guessing: measure, corner it, replace it. Surgeons don’t shotgun.

Everything you know, one machine. The light sensor ARMS it at night. The tripwire — fed only through the armed relay — FIRES the latched alarm: siren, beacon, and winch, all at once, immune to dawn and to the intruder fleeing. The master key alone stands it down. You wire the two runs that make it dangerous.

Empty bench. Full parts bin. A commissioning spec, like a real job ticket: a regulated 5-volt rail on the meter, an indicator LED at a healthy current, a motor at full duty, and a darkness alarm that knows when to shut up. Nothing is pre-wired. Build the whole station. Sign your work.

Meet the function generator: a battery that changed its mind twenty times a second. Wire it to the lamp and watch the light BREATHE. Then get out the probes — the oscilloscope in the multimeter panel draws what your eyes are too slow to see.

This gadget insists on DC — current in ONE direction only. Fit your diode into the gap and it becomes a half-wave rectifier: positive crests pass, negative troughs die at the gate. Watch the scope: half the wave simply… missing.

Your rectifier’s output is a heartbeat — flash, dark, flash, dark. Real electronics would faint. Add a reservoir capacitor across the lamp: it gulps charge at each crest and feeds the lamp through the gaps. The win meter demands the light NEVER falter for half a second straight.

A 2-volt whisper drives an LC circuit — and at ONE magic frequency, the tank answers with a ROAR, swinging far beyond the drive. Every push arrives exactly in rhythm, like a hand on a swing. Sweep the generator and find the note this circuit wants to sing.