Electrical Engineering Guide: Circuits, Power, Wiring and Safety

Introduction

When the Edison Illuminating Company switched on the Pearl Street Station in Lower Manhattan in September 1882, it did not just sell light. It sold a solution to a physics problem. Copper is expensive, current heats it, and every metre of conductor steals voltage from the load at the far end. Edison's direct-current network could only reach customers within roughly a mile of the generator before the losses became uneconomical, which is why the technology that eventually won — alternating current, pushed by Tesla and Westinghouse — was not a better idea about electricity so much as a better answer to the arithmetic of distribution.

That arithmetic is what this guide is about. Electrical engineering is unusual among engineering disciplines because almost everything in it can be derived from four quantities and two laws, yet the practical consequences are governed by tables, standards and safety margins that took a century to agree on. A wire that is electrically adequate can still be a fire hazard if it is installed in a hot attic. A circuit that computes perfectly on paper can still trip a breaker because the load is inductive. A battery rated at 100 amp-hours can deliver half of that at realistic currents. Understanding where the formulas end and the code begins is the difference between a calculation and an installation.

This guide is written for people who already have a reason to care: a homeowner planning a workshop circuit, a student preparing for a circuits exam, a maker sizing a 12-volt system for a camper, or an engineer who wants the practical counterpart to a textbook. It walks from the fundamental quantities through Ohm's law, series and parallel networks, capacitors, alternating current, transformers, batteries, signal levels, conductor sizing, overcurrent protection and grounding. Each section links to a calculator on this site that does the arithmetic, so you can move between understanding and computing without losing the thread.

The scope is deliberately practical and deliberately limited. It covers low-voltage and mains wiring as practised under the North American National Electrical Code (NEC), with notes where IEC practice differs, because the two standards disagree about units, conductor sizes and derating. It does not teach you to design switchgear, protection coordination for industrial plants, or anything requiring a licensed electrician's judgement. Where a decision can kill someone, this guide says so and stops.

The Four Quantities: Charge, Current, Voltage and Resistance

Everything in a circuit is bookkeeping about charged particles. Charge is the quantity itself, measured in coulombs. Current is charge in motion — one ampere is one coulomb per second — and it is the quantity that heats conductors, trips breakers and defines how much metal you need. Voltage is the potential difference that pushes the charge, measured in volts, and it is always a difference between two points: a "9-volt battery" has 9 volts across its terminals and no meaningful voltage at a single terminal in isolation.

Resistance is the opposition to that motion, measured in ohms. It is not a defect. It is the property that makes wires heat up in a toaster and the property that makes transmission lines waste energy. Every material sits somewhere on a spectrum from conductor (copper, aluminium, silver) through semiconductor to insulator (rubber, glass, air), and the spectrum is temperature-dependent: copper's resistance rises about 0.4% per degree Celsius, which is why a conductor's ampacity depends on the temperature rating of its insulation rather than on the metal alone.

Two derived quantities dominate practice. Power is the rate of energy transfer, measured in watts, and energy is power accumulated over time, measured in joules or — on your utility bill — in kilowatt-hours. A 1,500-watt space heater running for four hours consumes 6 kWh. That distinction matters more than beginners expect: a circuit's breaker is sized by current (power at a given voltage), but a battery's capacity and your electricity bill are sized by energy.

There is also a quantity that behaves unlike the other three: phase. Alternating current reverses direction periodically, so two voltages on the same wire can be equal in magnitude and still be different, because they peak at different moments. Nearly every confusing fact about mains electricity — why 240 volts is not simply "twice as dangerous" as 120, why a motor draws six times its running current at start-up, why a transformer only works on AC — comes from phase rather than from magnitude.

Ohm's Law and the Power Equation

Ohm's law is the entry point to the entire discipline. For a resistive element at a fixed temperature, the current through it is proportional to the voltage across it, and the constant of proportionality is the resistance:

The equation is trivial; its consequences are not. It says that for a fixed voltage, doubling resistance halves current. It says that for a fixed resistance, doubling voltage doubles current and therefore quadruples power, because power is the product of the two:

That squared term explains a great deal of electrical practice. It explains why long extension cords with undersized conductors get warm at the ends: the same current through a slightly higher resistance dissipates more heat. It explains why a 120-volt appliance fed at 110 volts may underperform while a motor fed at 110 volts draws more current to deliver the same mechanical power, which in turn makes it hotter. It explains why transmission lines run at hundreds of kilovolts: for a fixed power, raising voltage lowers current, and losses scale with the square of current. Raising transmission voltage by a factor of ten cuts line losses by a factor of a hundred.

The practical form of Ohm's law is a triangle you can solve three ways, and this site's Ohm's Law Calculator does exactly that: enter any two of voltage, current and resistance and read the third, with power derived in the same step. For loads where you know the power rating instead — a 1,200-watt kettle on a 120-volt circuit, say — the Electricity Calculator converts between watts, amps, volts and ohms and also handles the energy-per-hour view that utility billing uses.

A worked example makes the distinction between "rated" and "actual" concrete. A 60-watt incandescent lamp designed for 120 volts has a hot resistance of about 240 ohms, so it draws 0.5 amps. Measure that same lamp with a multimeter while it is cold and you will read something closer to 15 to 20 ohms, because tungsten's resistivity rises steeply with temperature. Ohm's law has not failed; the resistance is simply a function of an operating condition. This is why incandescent lamps draw a brief inrush current roughly ten times their steady-state current, and why dimmers and relays rated for incandescent loads are derated accordingly.

Series, Parallel and Kirchhoff's Laws

Real circuits are networks, and networks require two rules that hold regardless of what the components are. Kirchhoff's current law says that charge does not accumulate at a junction: whatever flows in must flow out.

Kirchhoff's voltage law says that if you walk a closed loop and add up every potential change, you return to where you started with nothing left over.

From these two statements, series and parallel behaviour follows mechanically. In series, the same current passes through every element, voltages add, and resistances add:

In parallel, the same voltage appears across every branch, currents add, and conductances add — which means the reciprocal of the total resistance is the sum of reciprocals:

The asymmetry matters. Adding resistors in series always increases total resistance; adding branches in parallel always decreases it, and the total can never fall below the smallest branch. That is why plugging more devices into a household circuit increases the current the breaker sees, and why a short circuit — a branch of nearly zero resistance — draws a current limited only by the wiring and the supply.

Resistor networks are also where the colour code lives. Through-hole resistors encode value and tolerance as coloured bands, and misreading the multiplier band by one position changes a value by a factor of ten. This site's Resistor Calculator decodes four- and five-band resistors and returns the tolerance range, which is the number that actually matters when you check whether a component is within specification. A 1,000-ohm resistor with a gold band is not "1,000 ohms"; it is somewhere between 950 and 1,050 ohms, and a circuit designed around the nominal value may behave differently at the extremes.

Two worked examples. First, a voltage divider: two 10-kilo-ohm resistors in series across 12 volts put 6 volts at the midpoint, but the moment you draw current from that midpoint the output sags, because the source now has a 5-kilo-ohm Thevenin resistance. Dividers are for sensing, not for powering. Second, parallel loads: four 100-watt lamps on a 120-volt circuit draw 0.833 amps each, 3.33 amps total, and each lamp sees the full 120 volts. Remove one lamp and the others stay equally bright — the behaviour that made parallel wiring the default for buildings and series wiring suitable only for things like old-fashioned fairy lights.

Capacitors, Time Constants and Stored Energy

A capacitor stores charge on two conductors separated by an insulator. Its capacitance, in farads, is the ratio of stored charge to applied voltage, and its behaviour is defined by the fact that current flows only while the voltage is changing:

Read that equation slowly, because it inverts the intuition built by resistors. A capacitor is an open circuit to steady direct current and a near-short to fast changes. That single property makes it a filter, a timer, a decoupling element and a memory cell. It also explains why capacitors are the components most often found failed in power supplies: they are the parts doing the continuous work of absorbing ripple.

Charging a capacitor through a resistor follows an exponential curve with a characteristic time constant:

After one time constant the capacitor has reached about 63% of the supply voltage; after five, about 99%. This is the basis of every "how long until the relay drops out" question, and of the debounce circuits inside mechanical switches. The stored energy is:

The squared voltage term is the safety warning. A 470-microfarad capacitor charged to 400 volts holds about 37.6 joules — enough to give a painful and potentially dangerous shock long after the equipment is unplugged, which is why power-supply repair procedures specify a bleeder resistor or a measured discharge rather than "wait a bit". Working with capacitors means verifying zero volts with a meter, not assuming it.

Capacitors also come with a coding problem of their own. Small ceramic and film capacitors mark their value in picofarads using a two-digit-plus-multiplier scheme, so a capacitor marked "104" is 10 followed by four zeros of picofarads, or 100 nanofarads. The Capacitor Calculator decodes these markings and converts between picofarad, nanofarad and microfarad, which removes the most common source of wrong-part substitutions in hobby electronics.

Capacitors are also the reason AC power systems have a "power factor" worth measuring. A capacitor draws current that leads the voltage by a quarter cycle; a motor draws current that lags by nearly the same amount. Put them on the same supply and they partly cancel, which is why industrial facilities install capacitor banks: not to save energy in the load, but to stop paying the utility for reactive current that does no work.

Alternating Current: RMS, Phase and Power Factor

Mains electricity reverses direction fifty or sixty times per second. A sinusoidal voltage is described by its peak value, but every rating you will read — 120 volts, 230 volts, 480 volts — is an RMS value, defined so that a resistor dissipates the same average power on AC as it would on an equal DC voltage:

For a 120-volt RMS supply the peak is about 170 volts, and the peak-to-peak swing is about 340 volts. This is why insulation and creepage distances are specified against peak, not RMS, and why a "120-volt" capacitor is marginal on a 120-volt AC line. It is also why a multimeter reading of 120 volts is not the number to use when computing stress on a semiconductor.

Phase becomes unavoidable as soon as loads are not purely resistive. Motors, transformers and fluorescent ballasts are inductive: current lags voltage, and the product of RMS volts and RMS amps overstates the power actually delivered. The ratio of real power to apparent power is the power factor, and for a balanced three-phase load the relationship is:

A 15-kilowatt three-phase motor at 400 volts and a power factor of 0.85 draws about 25.5 amps per phase. Ignore the power factor and you would compute 21.7 amps, specify a cable that runs hot, and wonder why the protection trips. Utilities in most of the world bill industrial customers for reactive power precisely because the current, not the power, determines how much copper the grid must carry.

Three-phase supply exists because it delivers constant power rather than pulsating power, and because it needs less conductor metal for the same capacity. The two common North American configurations are 120/240-volt single-phase residential service (two hot legs, neutral, with 240 volts between the legs) and 120/208-volt or 277/480-volt three-phase wye service in commercial buildings. Confusing the two is a classic and expensive mistake: a 208-volt-rated appliance on a 240-volt circuit is overstressed, and a single-phase load wired across two legs of a three-phase system sees a different voltage than the label suggests.

Frequency matters too, and not only for clocks. A 60-hertz motor run on 50 hertz runs about 20% slower and, because its impedance falls with frequency, draws more current and heats up. Conversely a 50-hertz motor on 60 hertz runs faster and loses torque margin. Appliances with universal motors and modern electronics often tolerate both; anything with a mains-frequency transformer or an induction motor usually does not.

Transformers: Turns Ratio and Why AC Won

A transformer is two coils sharing a magnetic core. Alternating current in the primary creates a changing flux, the flux induces a voltage in the secondary, and the ratio of voltages equals the ratio of turns:

Power is conserved (minus losses), so a step-up in voltage is matched by a step-down in current. This is the device that made AC distribution possible: generate at a few kilovolts, step up to hundreds of kilovolts for transport, step down to distribution voltage, step down again at the service entrance. Each transformation costs a small percentage in core and copper losses, and each is far cheaper than the conductor that would have been needed to carry low-voltage current over the same distance.

A transformer only works on alternating current, because it needs a changing flux. That is the whole substance of the AC/DC debate of the 1880s, and it is why Edison's DC network could not scale: without transformation, distributing power meant either generating it close to every customer or accepting enormous line losses. Modern high-voltage direct current (HVDC) transmission reverses this only at very long distances and very high power, where the efficiency of the line outweighs the cost of the converter stations at each end.

Transformer ratings are given in volt-amperes, not watts, because the limit is thermal and depends on current and voltage rather than on the phase relationship between them. A 10-kVA transformer can supply 10 kW to a resistive load, 8.5 kW to a load with a 0.85 power factor, and essentially nothing useful to a purely reactive load — while running at rated current in all three cases. Sizing a transformer by summing appliance wattages, a common DIY error, understates the required rating whenever motors are involved.

Two practical consequences follow. First, transformers have inrush current: switching one on can draw many times its rated current for a few cycles while the core saturates, which is why breaker selection for transformer circuits follows different rules than for resistive loads. Second, transformers hum at twice the line frequency, because magnetostriction occurs on both halves of the cycle — the 100- or 120-hertz buzz from a wall adapter is not a fault, it is the physics of the core.

Batteries and Runtime: Capacity, Peukert and Real Loads

A battery is specified by its nominal voltage and its capacity in amp-hours, which is charge, not energy. To get energy, multiply by voltage: a 12-volt, 100 amp-hour battery stores about 1,200 watt-hours, or 1.2 kilowatt-hours. The first and most common calculation is runtime at a known current:

A 100 amp-hour battery delivering 5 amps lasts a theoretical 20 hours. In practice it will not, for three reasons. First, capacity is quoted at a specific discharge rate, usually the 20-hour rate, and a battery discharged faster delivers less total charge. Second, the Peukert effect makes that loss systematic and predictable: a lead-acid battery has a Peukert exponent around 1.2 to 1.3, so discharging at ten times the rated current loses a meaningful fraction of capacity to internal resistance. Third, usable depth of discharge is limited — a lead-acid battery cycled to 100% depth of discharge will fail in a few hundred cycles, while limiting it to 50% roughly triples its life.

The Battery Life Calculator handles the arithmetic for mixed loads and includes the efficiency and depth-of-discharge adjustments, which is where most hand calculations go wrong. A camper with a 12-volt, 100 amp-hour battery running a 45-watt fridge, two 5-watt LED lights and a 10-watt phone charger draws about 5.4 amps; at a 50% depth-of-discharge limit and 85% inverter efficiency that is closer to 10 hours of usable runtime than the 18 hours the naive division suggests.

Temperature is the variable that surprises people most. A lead-acid battery rated at 100 amp-hours at 25 °C delivers roughly 80 amp-hours at 0 °C and around 65 at minus 18 °C, because the electrolyte's conductivity falls and the reaction kinetics slow. Lithium iron phosphate chemistries lose much less capacity in the cold but must not be charged below freezing, which is a protection issue rather than a capacity issue. Either way, a winter installation sized on the datasheet number will underperform, and the fix is capacity margin, not a bigger charger.

Charging deserves the same care as discharging. A lead-acid battery wants a three-stage profile (bulk, absorption, float) and will be damaged by sustained overvoltage; a lithium pack wants a constant-current/constant-voltage profile with a battery management system enforcing per-cell limits. Wiring batteries in parallel multiplies capacity but requires equal-length cables and matched internal resistance, otherwise the lower-resistance battery does most of the work and ages first. Wiring in series multiplies voltage and requires identical state of charge, which is why mixing an old battery with a new one in a series string is a reliable way to destroy both.

From Watts to Decibels: Signal Levels

Not every electrical quantity is about power delivery. In audio, radio and instrumentation, the interesting quantity is a ratio, and ratios spanning many orders of magnitude are easier to handle logarithmically. The decibel is that logarithm — for power, ten times the base-10 log of the ratio; for voltage or current into a fixed impedance, twenty times:

The factor of twenty rather than ten exists because power scales with the square of voltage. A doubling of voltage is therefore about 6 dB, a doubling of power about 3 dB, and a tenfold voltage ratio exactly 20 dB. These three conversions cover most practical audio work: a 6 dB gain doubles the voltage swing, a 3 dB loss halves the power, and a signal-to-noise ratio of 60 dB means the noise is a thousandth of the signal voltage.

The Decibel Calculator converts between ratios and decibels in both directions and between the power and amplitude conventions, which is where sign and factor-of-two errors concentrate. It is also the tool for the question that matters in any listening environment: how much louder is 85 dB than 70 dB? The answer is fifteen times the acoustic power and about five and a half times the perceived loudness, because hearing compresses intensity — the same logarithmic compression that makes the decibel the right unit in the first place.

Wire Sizing: AWG, Ampacity and Voltage Drop

Wire sizing is the section where electrical engineering stops being elegant and becomes a table lookup with safety consequences. Three constraints act simultaneously, and the binding one is whichever is most restrictive: the conductor must not overheat (ampacity), it must not drop so much voltage that the load misbehaves (voltage drop), and it must survive a fault long enough for the protection to clear it (short-circuit withstand).

North American conductors are sized by American Wire Gauge, a logarithmic scale inherited from wire-drawing practice in which larger gauge numbers mean smaller conductors, and every third step doubles the cross-sectional area. The diameter follows a closed-form definition:

That definition is why 10 AWG has roughly ten times the area of 20 AWG, and why the sequence 14, 12, 10, 8, 6 exists as the standard ladder for building wiring. Sizes above 4/0 are named in thousands of circular mils (250 kcmil, 500 kcmil) rather than in gauge, because the gauge scale becomes unwieldy.

The table below collects the physical and resistive properties of the conductors that appear in nearly every residential and light-commercial circuit. Resistance is quoted at 75 °C, the temperature at which most building wire is rated, and it is the number to use for voltage-drop arithmetic.

Ampacity is the second constraint, and it is not a property of the wire alone. It depends on the insulation's temperature rating, the ambient temperature, the number of current-carrying conductors bundled together, and whether the cable runs in free air, in conduit, or buried. NEC Table 310.16 gives the base values for copper and aluminium conductors at 60, 75 and 90 °C:

Notice that the 90 °C column does not license a hotter-running circuit. The 90 °C rating is available for derating calculations, but the termination — the breaker, the receptacle, the lug — is usually rated 75 °C, and the circuit's ampacity is the lower of the conductor and termination ratings. This is the single most misunderstood line in the code, and it is why a 90 °C-rated conductor cannot simply be run at its 90 °C current.

Small conductors carry an additional restriction: the overcurrent device protecting 14, 12 and 10 AWG copper is limited to 15, 20 and 30 amps respectively, regardless of what the ampacity table permits. The rule exists because small conductors have little thermal mass and can be damaged by a fault before a large breaker clears it. A 12-gauge conductor with 90 °C insulation may have an ampacity of 30 amps in free air, but the breaker in front of it is a 20-amp device.

The third constraint is voltage drop, and it is the one that most often forces a larger conductor than ampacity requires. For a single-phase circuit, the current travels out and back, so the drop doubles:

For three-phase circuits the geometry is different and the factor becomes the square root of three:

Both the NEC and IEC recommend keeping drop under about 3% on a branch circuit and about 5% overall from the service to the load. Those numbers are recommendations rather than hard limits in most of the NEC, but they are practical: a resistive load at 3% low voltage works slightly below rating, while a motor at 3% low voltage draws additional current to compensate and runs hotter, and a digital device may reset during a start-up surge.

A worked comparison shows how quickly voltage drop dominates. Run 20 amps through 12 AWG copper over 50 feet at 120 volts. The round-trip length is 100 feet, the resistance is 1.93 ohms per 1,000 feet, so the drop is 100 × 20 × 1.93 / 1,000 = 3.86 volts, which is 3.2% — already past the recommended 3% for a branch circuit. The same load at 15 feet drops only about 1.0 volt, comfortably inside recommendation. Same wire, same current, different answer, because distance is a variable that ampacity tables ignore entirely.

This is exactly the trade-off the Wire Gauge Calculator resolves: enter the load current, the one-way run length, the voltage, the conductor material and the termination temperature, and it returns the smallest standard conductor that satisfies ampacity, then upsizes until the voltage drop falls within 3%, and finally reports the matching breaker size. It is the calculator equivalent of the search that most people answer by trial and error. For the voltage-drop side alone — including three-phase circuits and mixed copper/aluminium runs — the Voltage Drop Calculator gives the drop in volts and percent against the NEC recommendations.

Metric practice, used across most of the world, sizes conductors in square millimetres from a preferred series (1.5, 2.5, 4, 6, 10, 16, 25, 35, 50...) and takes ampacity from IEC 60364-5-52 with installation-method tables rather than from a single NEC table. The two systems do not map cleanly: 2.5 mm² sits between 14 and 12 AWG, and 4 mm² sits between 12 and 10 AWG. A cable marked "2.5 mm²" is not a substitute for a specific AWG size without checking the actual ampacity of the installation method, and a cable marked "12 AWG" is not automatically acceptable in a jurisdiction that requires metric sizing for a given circuit.

Derating is the part of wire sizing that most DIY work skips and most inspections catch. Conductors bundled in a single conduit, cables run through a hot attic, and conductors in contact with insulation all carry less current than the table value. Bundling four to six current-carrying conductors typically costs 20% of ampacity; ambient temperature above 30 °C costs more, roughly 10% per additional 10 °C for common insulation classes. A circuit that computes to exactly 20 amps in a cool open basement can be non-compliant at 16 amps in a bundle in an attic.

Overcurrent Protection: Breakers, Fuses and Coordination

Every circuit needs a device whose job is to fail before the wiring does. Two technologies dominate: the fuse, which melts a calibrated element, and the circuit breaker, which trips a latch when either a thermal element (for overloads) or a magnetic element (for short circuits) actuates.

The thermal element responds to sustained overcurrent and is deliberately slow: a 20-amp breaker carrying 25 amps may take minutes to trip, which is desirable because motor start-up and inrush current are normal. The magnetic element responds to a fault in milliseconds, because a short circuit can deliver tens of thousands of amps. This dual behaviour is why a breaker protecting a motor circuit can be sized above the conductor's ampacity in specific, code-defined circumstances — and why copying that allowance to a resistive circuit is dangerous.

The inequality looks simple and is routinely violated. The load current must not exceed the protection rating, and the protection must not exceed the conductor's ampacity — after derating, after the termination limit, and after any allowance the code grants for specific equipment. A 30-amp breaker on 14-gauge wire satisfies the first inequality and violates the second, which is precisely the fire the code exists to prevent.

Breakers protect wiring, not people. A 20-amp breaker will not trip at the 30 milliamps that can stop a heart, which is why residual-current devices (RCDs in IEC terms, GFCIs in North American terms) exist as a separate layer. They compare current leaving and returning and trip on a mismatch of a few milliamps. Arc-fault devices add a third layer, detecting the characteristic signature of a parallel or series arc. Modern codes in North America require both in many residential locations; IEC practice increasingly requires RCDs on most final circuits.

Coordination — ensuring that the device nearest the fault trips first — is where protection becomes an engineering discipline rather than a lookup. If a 20-amp branch breaker and a 200-amp main breaker both see a 60-amp fault, both will eventually trip, and a building can go dark because of a fault on one circuit. Selectivity is achieved by choosing device types and ratings whose time-current curves do not overlap in the relevant range, which is why industrial installations use adjustable breakers with defined short-time settings rather than identical thermal-magnetic devices.

Grounding, Bonding and the Safety Chain

Grounding is the least intuitive part of electrical practice because it serves two different purposes that share one conductor. Equipment grounding provides a low-impedance return path so that a fault trips the breaker fast; system grounding fixes the relationship between the supply and earth so that voltages stay predictable. Bonding ties metal that could become energised to that same path.

The safety chain works only if every link is intact: a fault from hot to a metal enclosure raises the enclosure to line voltage; the equipment ground carries the fault current; the breaker sees a current high enough to trip in a fraction of a second; the touch voltage during that fraction stays low enough to survive. Break any link — a lifted ground, a painted joint, a plastic water fitting replacing a metal one — and the chain fails while everything still appears to work.

This is why the ground is never switched, never fused, and never used as a current-carrying conductor for normal loads. It is also why a "bootleg ground" (connecting neutral to ground at a receptacle) is dangerous rather than clever: it makes the test lamp look correct while removing the fault path's independence from the load current, and it energises the enclosure if the neutral opens upstream.

Grounding also interacts with the calculations in this guide in a way that surprises people. A ground rod does not clear faults. Earth has resistance measured in tens of ohms in typical soil, so a fault through earth alone would draw only a few amps at 120 volts — not enough to trip a 20-amp breaker, but enough to keep a deadly voltage on an enclosure indefinitely. The fault path must be a metallic conductor back to the source, which is what the bonding conductor provides.

Common Mistakes and How to Catch Them

The errors below account for most of the failed calculations seen in practice, and each has a check you can run in under a minute.

Sizing wire by ampacity and ignoring voltage drop. A 12-gauge conductor on a 20-amp circuit is code-legal and, at 150 feet, delivers about 108 volts instead of 120. Check the drop before choosing a conductor; the voltage-drop term is often the binding constraint on long runs.

Using the 90 °C column as the circuit ampacity. The conductor may be rated 90 °C, but the breaker lug usually is not. Use the lower of conductor and termination ratings, and treat the 90 °C column as an input to derating only.

Confusing amp-hours with watt-hours. A 12-volt 100 Ah battery and a 24-volt 100 Ah battery store different amounts of energy. Convert to watt-hours before comparing, and convert again before comparing to a load rated in watts.

Assuming a battery delivers its rated capacity at any current. The 20-hour rate is a rating condition, not a promise. Above it, Peukert losses shrink usable capacity; below it, capacity grows only slightly and self-discharge starts to matter.

Summing appliance wattages to size a transformer or inverter. Motor loads draw several times their running current at start-up, and reactive loads draw current without consuming real power. Size for the surge, and treat the power factor as part of the load.

Reading the resistor multiplier band one position off. A 4.7-kilo-ohm resistor and a 470-ohm resistor differ by one colour band. Decode with a calculator and check the tolerance — a 5% part at the edge of its band can be 5% outside the value the design assumes.

Treating a ground rod as a fault path. Earth does not reliably carry enough current to trip a breaker. Only the bonding conductor does.

Ignoring temperature and bundling derating. Tables are for reference conditions. An attic in summer is not a reference condition, and neither is a conduit with six current-carrying conductors.

Safety, Codes and the Limits of Calculation

Every number in this guide is a model, and models do not get inspected. The National Electrical Code and its IEC counterparts exist because electricity kills people in ways that arithmetic does not capture: contact with a live conductor, arc flash, thermal ignition of nearby materials, and the slow failure of a connection that was mechanically sound on the day it was made.

Three limits deserve explicit statement. First, codes are local. The NEC is adopted with amendments by state and municipality, and IEC-based national codes differ on conductor sizes, RCD requirements and permissible installation methods. A calculation that satisfies one jurisdiction may fail another. Second, equipment listings matter as much as calculations: a device listed for the application has been tested for the fault conditions the arithmetic only approximates. Third, and most important, work on mains wiring is licensed work in most jurisdictions, not because the arithmetic is hard but because the consequences of a mistake are not recoverable.

What this guide can do is make you a better client, a better student, or a safer hobbyist. If you are planning a circuit, you can arrive at a conversation with an electrician knowing why a 12-gauge conductor is being proposed instead of a 14, why a 20-amp breaker is not interchangeable with a 15, and why the run length matters as much as the load. If you are studying, the calculators turn the algebra into feedback. If you are building low-voltage systems, the same physics applies at safer magnitudes — with the caveat that a 12-volt battery bank can still deliver enough current through a shorted tool to cause a burn or a fire.

Frequently Asked Questions

❓ How do I size electrical wire for a new circuit?

✅ Start with the load current, then choose the smallest standard conductor whose ampacity at your termination temperature and after derating exceeds that current. Then compute voltage drop over the actual run and upsize until the drop is inside about 3% for a branch circuit. Finally confirm the breaker protects the conductor. The Wire Gauge Calculator performs all three steps in order.

❓ What is the difference between wire gauge and wire size?

✅ They usually mean the same thing: American Wire Gauge is the naming scale, and wire size is the resulting conductor designation such as 12 AWG or 4 mm squared. Ampacity is a separate property — how much current that size can carry under given installation conditions.

❓ Is 12 AWG wire good for a 20 amp circuit?

✅ Yes. Twelve-gauge copper with 60 C insulation has an ampacity of 20 amps, and with 75 C insulation 25 amps, and the code limits a 20-amp circuit's overcurrent device accordingly. If the run is long, check voltage drop, because 12 AWG at 20 amps exceeds 3% drop beyond roughly 45 feet of one-way distance at 120 volts.

❓ Why is voltage drop worse in low-voltage systems?

✅ Because drop is a fixed number of volts for a given current and length, and it is expressed as a percentage of the supply. Two volts lost is under 2% of 120 volts but over 16% of 12 volts. That is why 12-volt installations use much larger conductors for the same power than mains installations do.

❓ Does a higher temperature rating mean I can run more current?

✅ Not directly. The 90 C column is available for derating calculations, but the circuit ampacity is limited by the lowest-rated component in the chain, which is usually the breaker or receptacle termination at 75 C. Use the 90 C value only to compute derated ampacity, then apply the termination limit.

❓ Why does my inverter need to be larger than the sum of my appliance wattages?

✅ Because motors, compressors and pumps draw a start-up surge several times their running current, and because reactive loads draw current without consuming real power. Size for the surge and treat the power factor as part of the load, otherwise the inverter will trip on start-up or run hot continuously.

❓ How long will a 100 Ah battery run a 400 watt load?

✅ A 12-volt 100 Ah battery holds about 1,200 watt-hours, so a 400-watt load would last roughly three hours at full depth of discharge. In practice, limiting a lead-acid battery to 50% depth of discharge and allowing for 85% inverter efficiency brings usable runtime to about 1.3 hours.

❓ Why do capacitors hold a charge after the power is off?

✅ A capacitor stores energy in an electric field, and that energy has no path to dissipate once the supply is removed. The stored energy scales with the square of the voltage, so high-voltage capacitors can hold a dangerous charge for minutes. Always verify zero volts with a meter before touching the terminals.

❓ What is power factor and why does it matter?

✅ Power factor is the ratio of real power to apparent power in an AC circuit. It matters because the current — not the power — determines conductor and transformer sizing, so a load with a low power factor needs thicker cable and a larger supply for the same useful work. Utilities bill large customers for reactive power for exactly this reason.

❓ Why does a transformer only work with AC?

✅ Because it transfers energy through a changing magnetic flux. Direct current produces a steady flux, which induces no voltage in the secondary. This is the physical reason alternating current won the distribution debate: only AC can be stepped up and down economically.

❓ Can I use a 60 Hz appliance on a 50 Hz supply?

✅ It depends on the load. Resistive appliances and most switch-mode electronics tolerate both. Induction motors run about 20% slower and draw more current on 50 Hz, and transformers may saturate. Check the nameplate: equipment marked 50/60 Hz is designed for both, and equipment marked for one frequency only usually is not.

❓ Is a larger breaker always safer?

✅ No. The breaker's job is to protect the conductor, so a breaker larger than the conductor's ampacity is a fire risk rather than a safety margin. The correct relationship is that the load current stays below the breaker rating and the breaker rating stays at or below the conductor ampacity after derating.