Electrical Science curriculum

Conserved is the useful word: charge is never created or destroyed, only moved, so every circuit question is about where it went rather than where it came from. Only electrons move in a solid conductor; the protons stay fixed in the lattice.

It is an enormous amount of charge in everyday terms, which is why the elementary charge is such a tiny fraction of it. That mismatch is worth noticing: ordinary currents move vast numbers of carriers, each carrying almost nothing.

The hazard is not the voltage, which can reach thousands of volts, but that the stored energy is tiny — uncomfortable to a person and destructive to a semiconductor. That is why sensitive parts are handled with grounded straps and mats.

The inverse-square form means halving the distance quadruples the force, so intuitions built on linear relationships mislead badly here. It has the same shape as the gravitational law, but charge comes in two signs, so it can repel.

Per unit time is what makes it a rate rather than a quantity, which is the distinction behind every amp-hour and charge calculation. Through a cross-section matters too: current is what passes a point, not what is stored anywhere.

Because it is charge per second, an amp-second is a coulomb, which is the bridge between current problems and charge problems. Ammeters go in series for the same reason — they must carry what they measure.

The convention was fixed before the carrier was identified, so it points opposite to the electrons in a metal. It is not an error to correct: all the standard symbols, formulas and device markings assume it, so mixing the two is what causes trouble.

Energy per charge is what makes voltage the driver rather than the flow: it is the push available per unit moved, not the amount moving. That is also why a high voltage with negligible current can be harmless while the reverse is not.

Between is the whole point, and the source of most measurement mistakes: a single point has no voltage until a reference is named. A meter has two leads because the quantity itself has two ends.

The two are equal only when no current flows, so an unloaded battery reads high and disappoints under load. The difference is the drop inside the source, which means terminal voltage is a measurement of the source and its load together.

It rises as a cell ages and falls as it warms, which is why an old battery can read near its rated voltage and still fail to start anything. Testing under load is the only way to see it, since an open-circuit reading hides it entirely.

It is a property of the object rather than the material, since it depends on shape and size as well as substance. That distinction is what separates it from resistivity, and it is why a longer wire of the same metal resists more.

Length over area is the geometry factor: doubling length doubles resistance, while doubling the diameter quarters it, because area goes as the square of the radius. That square is what makes conductor sizing so sensitive to a small gauge change.

Loosely bound is a claim about the valence band, and it explains why the best conductors sit in a narrow group of metals. It also explains the temperature behavior: more lattice vibration means more scattering of the carriers already present.

No insulator is perfect — each has a voltage above which it conducts suddenly and often destructively. That breakdown limit, rather than resistance, is what a component's voltage rating is protecting, and it falls with moisture and contamination.

What makes the class useful is not the intermediate resistivity but that the resistivity is controllable, by doping, by light, by heat and by an applied field. Every device in the semiconductor chapter is one of those controls exploited.

The two signs come from different mechanisms: in a metal, heat scatters the existing carriers, while in a semiconductor heat frees more of them. That opposition is why a semiconductor can run away thermally and a resistor generally cannot.

They are opposite faults with opposite symptoms: an open passes no current and drops the full source voltage across the break, while a short passes very large current and drops almost nothing. Reading which one you have starts from that pair.

A schematic records connection rather than position, so a long wire and a short one look the same and physical layout carries no meaning. Reading one means following nodes, not tracing the picture as though it were a map.

Direction is defined by the force on a positive test charge, which is the same convention that fixes conventional current. Field explains action at a distance without contact, and it is the bridge to the magnetic chapter.

Electrons drift far slower than walking pace, yet a lamp lights instantly, because the field propagates at near light speed and the conductor is already full of carriers. The signal moves; the individual electrons barely do.

It is the same information stated the other way up, and it is the convenient form for parallel networks, where conductances simply add. Switching to it turns the awkward reciprocal-sum rule into ordinary addition.

Zero here means exactly zero rather than very small, which is what allows a current to persist indefinitely in a closed loop. The critical temperature is not the only limit: a strong enough field or current also destroys the state.

The bands are positional, so reading from the wrong end gives a plausible but wrong value — the tolerance band is the cue for which end is which. The multiplier band is an exponent, not a digit, which is the other common misread.

Tolerance says the value is somewhere in a band, so a circuit that only works at the nominal figure does not really work. Derating is the separate thermal margin: rated power assumes conditions that a cramped enclosure does not provide.

It is a definition of resistance as much as a law, and it holds only for ohmic devices. Treat it as three relationships rather than one formula, and identify which two quantities the problem gives you before rearranging anything.

The useful reading is that current is an outcome rather than a setting: a source supplies voltage and the load decides how much current flows. That is why adding load in parallel raises total current without changing the voltage.

This form is how resistance is measured in practice — apply a known voltage, read the current, divide. It also shows why a resistance reading on a powered circuit is meaningless: the meter's own source is no longer the only one.

This form needs no resistance at all, which makes it the right one for a source, a motor or any device whose resistance is not constant. The other two forms are this one with Ohm's law already substituted in.

The square is the point: current, not voltage, is what heats a conductor, and it does so disproportionately. This is the form to reach for whenever the current is common to the elements, which is to say in any series path.

Being joules per second makes it a rate, so a wattage alone says nothing about total energy consumed — that needs a duration. Confusing the rate with the quantity is what makes power and energy questions go wrong.

The distinction is about the direction of energy flow rather than the type of component, and it can reverse: a charging battery is a load, a discharging one a source. Identify it from which way current passes through the element.

The convention exists so that one set of equations covers both cases, with a negative result meaning the element supplies rather than absorbs. Reading that minus sign as an arithmetic mistake is the usual error.

Convert everything to base units before calculating, then convert once at the end — mixing milliamps with kilohms mid-problem is the single most common source of answers wrong by a factor of a thousand.

It follows directly from current being a rate, so this is the same relationship read backwards. It is also the arithmetic behind battery capacity, which is why an amp-hour is a unit of charge rather than of energy.

Every resistive element converts electrical energy to heat, so this is not a fault mode but the normal behavior, deliberate in a heater and a loss everywhere else. It is what sets current limits on conductors and components alike.

It is always below one for a real device, and what is missing has gone somewhere, usually to heat. Asking where the lost fraction went is more useful than the percentage itself, because it points at what to improve.

Maximum power is not maximum efficiency — at the matching point half the energy is dissipated inside the source. That is why the condition is sought in signal circuits and deliberately avoided in power distribution.

For a non-ohmic device the ratio still has a value at each operating point, so it is not that resistance vanishes but that no single number describes the device. A lamp filament, a diode and a thermistor are all in this class.

A straight line through the origin marks an ohmic device, and any curvature means the resistance changes with operating point. Reading the slope rather than a single point is what makes the curve more informative than one measurement.

The counterpart of the current-squared form, and the one to use when voltage is the shared quantity, which is to say across any parallel branch. Note that here a larger resistance dissipates less, which reverses the series intuition.

All three forms give the same answer, so picking the one matching your known quantities avoids an unnecessary intermediate step — and each intermediate step is a chance to introduce an error. Check the shared quantity first.

The trap catches people because the relationship for voltage and current is linear and the one for power is not. It is the reason an overload is so much more damaging than a proportional reading of the current would suggest.

Energy is what is billed and what heats a component; power is only the rate at that instant. A brief high-power event and a long low-power one can deliver the same energy, which is why duration belongs in every thermal judgment.

It is an energy unit written as a rate times a time, which is what makes it look unfamiliar. Because the hour is in it, dividing watts by a thousand is only half the conversion — the duration in hours has to come in too.

Convert to kilowatt-hours first, then multiply, and keep the rate's own units in view. The usual error is mixing watts with hours or kilowatts with minutes, which produces an answer wrong by a clean factor rather than an obviously absurd one.

Amp-hours measure charge rather than energy, so two batteries with the same rating and different voltages store different amounts. The rating also assumes a particular discharge rate, and drawing harder yields less than the label suggests.

It is a heat-dissipation limit rather than a performance figure, so exceeding it does not change the resistance — it destroys the part. The rating assumes free air at a stated temperature, which is why crowded enclosures need margin.

The wire is a resistor in series with the load, so the load receives less than the source supplies and the difference heats the run. It matters most on long runs at low voltage, where the drop is a large fraction of what is available.

For a given power, raising voltage lowers current proportionally, and the loss falls with the square of that current — so a tenfold voltage rise cuts the loss a hundredfold. That square is the whole reason for the transmission grid.

One path is the defining fact and everything else follows from it: the same current everywhere, voltages that add, and total failure if the path breaks anywhere. Identify the topology first, because the rules that apply depend entirely on it.

Shared voltage is the defining fact here, and it is why each branch behaves as though the others were absent. Adding a branch adds current without changing what any existing branch sees, which is why outlets are wired this way.

Adding in series always increases the total, because the current has more to get through in sequence. The largest resistor therefore dominates both the total and the voltage division, which is a fast way to sanity-check an answer.

The product-over-sum shortcut works for exactly two resistors; using it on three at once is a standard error. Apply it pairwise or use the reciprocal sum, and remember the result must come out below the smallest branch.

This is the free check on every parallel calculation: if the answer is not below the smallest branch, the arithmetic is wrong. The reason is that each added path gives current somewhere new to go, which can only ease the total opposition.

It follows from the reciprocal rule with identical terms, and it is worth recognizing because equal branches are common in practice. It also generalizes the previous check: with N equal branches the total is N times smaller, never merely smaller.

Work from the parts furthest from the source inward, replacing each recognized group with its equivalent and redrawing as you go. The redrawing is not optional neatness — it is what stops a group being misread at the next step.

Total resistance has to be found first, which is why the reduction step comes before any current calculation. Once the current is known it is the same everywhere in the string, so each individual drop follows from that one number.

Each branch is computed from the shared voltage and its own resistance, independently of the others, which is what makes parallel networks easy once the voltage is known. The branch currents then sum to the total drawn from the source.

It only holds while nothing significant is drawn from the output, because a load in parallel with the lower resistor changes the ratio. That loading effect is why a divider is fine as a reference and poor as a power supply.

Inverse is the word to watch: the smaller resistance takes the larger share, which is the opposite of the voltage divider. Expecting the two rules to work the same way is the usual source of an inverted answer.

It is charge conservation stated for a junction: nothing accumulates there, so what arrives must leave. Assigning a consistent sign to in and out is what makes it usable, and a negative result simply means a current runs the other way.

It is energy conservation stated for a loop: return to where you started and the net change must be zero. The discipline is to keep one direction of travel and one sign convention for the whole loop, then read the signs off the elements.

It is the systematic alternative to reduction, and it works on networks that cannot be reduced, such as a bridge. The unknowns are loop currents rather than branch currents, so a branch shared by two loops carries the difference of the two.

The dual of mesh analysis, and usually the lighter of the two when a circuit has more loops than nodes. Counting both before choosing is worth the few seconds, because it decides how many simultaneous equations you will have to solve.

It is a chosen zero point, which is why voltages are always relative and why moving the reference changes every reading without changing the circuit. Signal ground and earth are different ideas that often, but not always, coincide.

With one path, a break anywhere stops everything, and the full source voltage then appears across the break itself. That last part is the diagnostic: measuring across a suspected open shows source voltage rather than zero.

The surviving branches carry on unchanged, because each sees the same voltage as before, so the symptom is one load dead rather than all of them. Total current falls, which is the measurable clue that a branch has gone.

The short gives current a path of near-zero resistance, so it bypasses whatever it is across and the total resistance falls. Total current rises sharply, which is what overcurrent protection is there to catch.

Parallel plates effectively become one larger plate area, so the capacitances add exactly as series resistances do. Noticing that the rules are swapped relative to resistors is more reliable than memorizing four separate formulas.

The reciprocal rule here means the total is below the smallest, which feels wrong until you see that the plate separation has effectively increased. The voltage across the string divides in inverse proportion to capacitance.

The error is applying resistor rules to capacitors, and it is worth having a check rather than a memory: series always reduces capacitance and increases resistance, while parallel does the reverse. Test the answer against that.

Inductors follow the resistor rules, which makes capacitors the odd one out rather than a second pattern to learn. That holds only where the coils are not magnetically coupled; mutual inductance changes the result.

Its accuracy comes from detecting a null rather than measuring a value, so it does not depend on the meter's calibration — only on its sensitivity near zero. That is why the technique long outlived the specific instrument.

It works only for linear networks, and the discipline is in how the other sources are removed: a voltage source becomes a short, a current source an open. Getting that replacement backwards is the error that makes the method fail.

The value is that the rest of the circuit collapses to two numbers, so a changing load can be analyzed repeatedly without redoing the network. Find the open-circuit voltage and the resistance seen with sources removed.

It carries exactly the same information in current-source form, so neither is more correct — pick whichever makes the next step easier. The two share the same resistance value, which is what makes conversion between them trivial.

Because the equivalents share a resistance, converting is one division or multiplication, and it can simplify a network that neither reduction nor superposition handles neatly. It applies to the pair as a unit, never to the source alone.

Adding lamps in series raises total resistance, so current falls and every lamp dims — including the ones already there. One failing lamp also opens the path and extinguishes all of them, which is the practical reason this wiring was abandoned.

Each lamp gets the full supply voltage regardless of how many others are connected, so brightness stays constant as the count grows. What grows is the total current drawn, which is what the supply and its protection must be sized for.

Field lines always close on themselves, which is the structural difference from an electric field and the reason no isolated magnetic pole has ever been found. Cutting a magnet in half produces two magnets rather than two poles.

The field circles the wire rather than pointing along it, which is why a compass near a current-carrying conductor turns sideways. The right-hand rule ties its direction to conventional current, not to electron flow.

Coiling stacks each turn's field so they reinforce inside and largely cancel outside, which is what produces a bar-magnet pattern from a straight wire. Field strength depends on turns and current together, not on either alone.

The decisive advantage is that it can be switched and varied, which no permanent magnet offers. A ferromagnetic core multiplies the field by carrying flux far more readily than air, which is why the core matters as much as the winding.

It is the magnetic counterpart of conductivity: a high-permeability material offers flux an easy path, so the flux concentrates there. That is why a core guides the field and why a small air gap has a large effect on a magnetic circuit.

The strength comes from domains aligning together rather than from individual atoms, which is why the effect saturates once they are all aligned. Retained alignment after the field is removed is what makes a permanent magnet.

The lag means the material's state depends on its history, not just on the present field, so the magnetization curve encloses an area. That area is energy lost as heat every cycle, which is why it matters in any device driven by alternating current.

Flux is the total passing through a surface, so it depends on the area and its orientation as well as on the field. That orientation dependence is what a rotating generator coil exploits: the flux through it varies as it turns.

Density is the per-area figure, so a given flux squeezed through a smaller area gives a higher density — which is what a tapered pole piece is for. Confusing the two quantities is the usual error in a magnetic circuit calculation.

Rate of change is the whole content: a strong steady field induces nothing, while a weak rapidly changing one induces a great deal. That is why transformers need alternating current and why moving faster produces more voltage.

It is conservation of energy expressed as a direction: if the induced current reinforced the change, it would create energy from nothing. Practically it means induction always resists what is happening, which is what back-emf and braking both are.

It is the same law seen from the conductor's side: moving through a field changes the flux the circuit encloses. The voltage depends on the component of motion across the field, so moving along the lines induces nothing.

The output alternates because the coil's orientation reverses twice per revolution, so the alternation comes from the geometry rather than from anything in the driving motion. Peak output occurs when the coil moves across the field, not along it.

This is the converse of the generator, and the same machine can do both, which is why a motor generates when driven and a generator motors when fed. The force is perpendicular to both the current and the field.

A running motor is also a generator, and what it generates opposes the supply, so the net driving voltage falls as speed rises. That is what limits current at speed, and why its absence at standstill matters so much.

With no rotation there is no back-emf, so only the winding resistance limits current and a stalled motor can draw many times its running figure. This is why motor protection must tolerate a brief surge yet still act on a sustained one.

A coil's own changing current changes its own flux, which induces a voltage opposing the change — so an inductor resists current changes as a capacitor resists voltage changes. That duality organizes most of the reactance chapter.

Its definition is a rate relationship, which is why the unit looks awkward: it ties a voltage to how fast the current is changing rather than to the current itself. A large inductance means a small current change produces a large voltage.

When current is interrupted the field collapses very fast, and because induced voltage follows the rate of change, the spike can be many times the supply. This is what burns switch contacts and destroys transistors driving coils.

One coil's changing flux passing through another induces voltage there, which is transformer action and also unwanted crosstalk. How much links depends on geometry and on whether a core guides the flux between them.

It needs a changing flux, so it works on alternating current and not on steady direct current — which is the single fact that explains why distribution is alternating. The two windings are coupled magnetically, not electrically.

Voltage scales with the ratio and current scales inversely, because power cannot be created — a step-up in voltage is a step-down in available current. Expecting both to rise together is the misconception this relation removes.

The losses split by where they occur: copper loss is resistive heating in the windings and rises with load, while core losses occur in the iron and are largely independent of it. That split is why efficiency varies with loading.

Its value is isolation: a small, safe control current operates contacts carrying something much larger, with no electrical connection between the two. The coil is an inductor, which is why it needs protection against its own collapse spike.

It gives the collapsing field's current somewhere to go, so the energy circulates and decays instead of forcing a destructive voltage across the switch. Polarity is critical — reversed, it simply short-circuits the supply.

Amplitude and frequency describe the wave itself; phase only has meaning against a reference, which is why it is always quoted between two waveforms. Knowing which of the four a problem fixes tells you which of the others are free.

A higher frequency means a shorter period, and the arithmetic is one division either way. The common slip is a prefix one: milliseconds give kilohertz and microseconds give megahertz, so mixing them shifts the answer by a thousand.

It exists because one full cycle is two pi radians, so this is frequency expressed in the units trigonometry actually uses. Every reactance formula takes it rather than ordinary frequency, which is why the factor keeps reappearing.

It matters for insulation and component ratings, because a part experiences the peak even though a meter reports something lower. That gap between what is measured and what must be withstood is why both figures get quoted.

Symmetric is the condition: a waveform carrying a steady offset has a peak-to-peak that is no longer twice its peak. It is the natural figure to read from an oscilloscope, which displays the whole excursion rather than one half of it.

Defining it by heating is what makes it the honest figure for power, since heating follows the square of current and the average of a squared sinusoid is not the square of its average. That mismatch is the whole reason a separate quantity exists.

This factor applies to a pure sinusoid only — a square wave's value equals its peak, and a triangle wave matches neither. Applying the sinusoid factor to a distorted waveform is a standard error and an invisible one.

Because it is what determines heating and therefore useful power, it is the figure that matters for ratings. The consequence catches people out: the peak a circuit must insulate against sits well above the number printed on the nameplate.

It is meaningless for a single waveform, since it needs a reference, which is why questions always name two. Expressed as an angle it is a fraction of a cycle rather than an amount of time, so the same angle is less time at a higher frequency.

The terms are relative, so saying which quantity leads which is part of the statement rather than an optional detail. In a capacitive circuit current leads voltage and in an inductive one it lags, which the whole reactance chapter turns on.

It works because every quantity in a linear circuit at one frequency shares that frequency, so only magnitude and phase differ between them. Turning calculus into arithmetic is the payoff; the price is that it holds at one frequency only.

Power in a single-phase supply pulses down to zero twice each cycle, because voltage and current pass through zero together. That pulsation is precisely the deficiency three-phase exists to remove.

The spacing means total instantaneous power is constant rather than pulsing, so motors run smoothly and conductors are used more efficiently. It also produces a rotating field, which is what lets an induction motor start by itself.

A wye makes a neutral point available and a delta does not, which is the practical difference deciding where each is used. The same three windings can be connected either way, so the choice belongs to the system rather than the machine.

Which quantity carries the factor swaps between the two connections: in a wye the voltages differ while the currents match, and in a delta the reverse. Naming the connection before applying any factor prevents the standard error.

It is the part actually converted into heat, light or motion, so it is what the meter bills and what the generator must fuel. Only the in-phase component of current contributes, which is why power factor enters every practical calculation.

It does no work over a cycle, and it is not harmless for that: the current carrying it heats conductors and occupies capacity all the way back to the source. That is why large users are billed for it though it delivers nothing.

It is what conductors, breakers and transformers must be sized for, because they respond to current regardless of its phase. That is exactly why their ratings are given in volt-amps rather than in watts.

The three quantities add as vectors rather than as numbers, which is why apparent power is not the sum of the other two. Recognizing it as Pythagoras is what turns three separate definitions into one relationship you can use.

A low factor means a large current for the useful power delivered, so losses and conductor sizes rise without any extra work being done. It is bounded at one, and reaching it means voltage and current are in phase.

The decisive practical difference is that alternating current can be transformed and direct current cannot be directly, which is why transmission is alternating. Electronics mostly runs on direct current, which is why rectification is everywhere.

Diodes give one-way conduction, so the output is unidirectional but far from steady — rectification alone is only the first half of a supply. Smoothing is the second half, which is why the two topics always arrive together.

The positive and negative halves cancel exactly, which is why a simple averaging meter reads nothing on alternating current and why a separate measure was needed. Over a half cycle the average is not zero, which rectifier problems use.

A sine wave is the only one without harmonics, which is why it serves as the reference: every other shape is a sum of sines. That decomposition is why waveform shape matters for heating, interference and measurement alike.

They are generated by non-linear loads, which is why modern electronics creates them, and they cause heating and neutral currents out of proportion to their size. A meter that assumes a pure sinusoid misreads a distorted one badly.

For a pure sinusoid it is fixed, while a distorted waveform can run far higher — meaning large peaks hidden behind a modest reading. Equipment has to survive the peak while being rated by the averaged figure.

Rectangular form adds easily and polar form multiplies easily, so the practical skill is converting between them rather than preferring one. Series impedances add, which is why rectangular is the form for combining them.

Instantaneous power can even go negative, when voltage and current carry opposite signs, and that is energy returning to the source. Averaging that variation across a whole cycle is exactly what produces real power.

It works because the two kinds of reactance shift current in opposite directions, so they cancel. Correction is applied near the load to relieve the whole path back to the source, and over-correcting swaps a lagging problem for a leading one.

The capacitor holds charge between peaks, so a larger one and a lighter load both reduce the leftover variation. Ripple can never reach zero this way, which is why a regulator follows the filter wherever a steady output is needed.

The behavior follows from a capacitor needing a changing voltage to pass current: at zero frequency nothing changes and nothing flows. It is an opposition that stores rather than dissipates, so it produces no heat of its own.

The exact mirror of capacitive reactance, and worth learning as a pair rather than separately, since each is the other inverted. At steady direct current an ideal inductor is simply a length of wire.

This is why the two cannot simply be added: they act at ninety degrees to one another, which is what makes impedance a vector sum. It is also why reactive elements run cool while resistive ones get hot.

Net is the operative word: inductive and capacitive reactance oppose each other, so they subtract before combining with resistance. That subtraction is what makes resonance possible, at the frequency where the two cancel entirely.

It is Pythagoras again, the same relation as the power triangle, because both describe quantities ninety degrees apart. Adding resistance and reactance arithmetically always overstates the result, which is a quick way to catch the error.

Current has to flow before the voltage across a capacitor can build, which is the physical reason for the lead. The angle lies between zero and ninety degrees, reaching ninety only in the limit where the resistance vanishes.

The inductor opposes the change in current, so the current arrives late, which mirrors the capacitive case exactly. Most real loads are inductive because they contain motors and coils, which is why lagging power factor is the common problem.

At resonance the circuit looks purely resistive, so current is limited only by whatever resistance is present. That is why series resonance can produce voltages across the reactive elements far larger than the source voltage itself.

The dual of the series case, and the reversal is total: current circulates within the loop while very little is drawn from the source. That circulating current can greatly exceed the line current, which regularly surprises people.

Because it depends on the product, halving one value and doubling the other leaves it unchanged, and the same formula covers series and parallel resonance. Resistance does not appear in it — resistance sets the sharpness, not the position.

Opposing voltage change is the property to hold onto, since it explains smoothing, timing and decoupling alike. A charged capacitor also stays charged after power is removed, which is a genuine hazard in high-voltage equipment.

It does two jobs at once and they trade off: a material that raises capacitance may withstand less voltage. That is why a capacitor carries both a value and a voltage rating, and why exceeding the second destroys it outright.

That a resistance times a capacitance yields seconds is worth verifying once, because it makes the formula memorable. The constant sets the pace of every charge and discharge in the circuit, and both directions share the same value.

The curve is exponential, so it approaches the final value without ever quite arriving — which is why a practical settling time is quoted instead of a completion. Each time constant covers the same fraction of whatever is left.

The mirror of charging, falling by the same fraction each constant, so the familiar figures are the same numbers read from the other end. The shared constant is why a circuit that charges slowly also discharges slowly.

Note that resistance divides here where it multiplied in the capacitive case, so more resistance makes an inductive circuit faster and a capacitive one slower. Getting that inversion backwards is the usual error in this pair.

Reactive elements matter only while things are changing, so a steady-state answer often reduces to a simple resistive one. Identifying which regime the question asks about decides whether the reactances belong in the calculation at all.

The farad is enormous in practical terms, which is why real components are measured in millionths of one and smaller. Capacitance is fixed by geometry and dielectric rather than by the voltage applied, which is what makes it a property.

The dual of the capacitor throughout: a field rather than charge, current rather than voltage. Because the energy sits in the field, interrupting the current leaves it nowhere to go, which is what produces the inductive spike.

The square means doubling the voltage quadruples the stored energy, which is why a high-voltage capacitor is dangerous out of all proportion to its size. Energy is what does the harm, not the voltage reading on its own.

The same form with current in place of voltage, which continues the duality exactly. It is why a large current through a coil releases so much energy when interrupted, and why that energy has to be given a path.

High means selective, which is wanted in a tuned circuit and unwanted where a broad response is needed, so it is a design choice rather than a virtue. Resistance is what lowers it, because resistance is where the energy is lost.

It is set by the two quantities together, so a sharp circuit at a high frequency can still be broad in absolute terms. The half-power convention is used because it corresponds to a fixed fraction of the peak amplitude.

Every one exploits reactance varying with frequency, so which component sits where decides the type — the same parts rearranged give the opposite response. Naming where the output is taken from is how you identify one.

Half power corresponds to that fraction of the voltage because power follows the square, which is why the two figures look inconsistent and are not. It marks a gradual transition rather than a boundary; nothing is sharply cut off.

Carriers diffuse across the boundary and leave charged ions behind, building a field that opposes further diffusion until the two balance. That self-limiting barrier is what makes the junction one-way, and every device here is built on it.

Below the threshold almost nothing flows and above it current rises very steeply, so a diode is not a resistor and must not be treated as one. That steepness is exactly why a series resistor is needed to limit the current.

Blocking is never total — a small leakage always flows, and it grows with temperature. Breakdown destroys an ordinary diode and is the intended operating region for a zener, which is essentially the only difference between them.

Half the input is discarded, so the output pulses with long gaps and its average is low. Simplicity is the only advantage, which is why it appears in small supplies and almost nowhere that efficiency actually matters.

It inverts the negative half rather than discarding it, so the output pulses at twice the input frequency and is much easier to smooth. The cost is two diode drops in the conduction path rather than one.

Cell voltage is set by the chemistry, so a pack's voltage tells you how many cells sit in series rather than anything about its capacity. Nominal is an average across the discharge: a cell starts above it and finishes below.

The same rule as for any source, and the practical warning is about mismatch: paralleling cells at different states of charge drives current between them. A series string is limited by its weakest cell.

It works by absorbing the difference, which is why a linear regulator turns the surplus into heat and its loss grows with the input-output gap. A switching regulator avoids that by converting rather than absorbing.

Both respond to current over time rather than to current alone, which is why a brief surge passes and a sustained overload does not. The difference is resettability, and a breaker's mechanism gives it a defined interrupting capacity.

Tiny quantities change conductivity by orders of magnitude, which is what makes the process so powerful. Both types stay electrically neutral overall — the labels say which carrier is in the majority, not that there is a net charge.

It stays roughly constant across a wide current range, which is why it is treated as a fixed drop rather than as a resistance. It falls slightly as temperature rises, which matters wherever diodes set a bias point.

It is built to survive its own breakdown, which an ordinary diode is not, and it holds a nearly constant voltage there. It only regulates while current flows through it, so a series resistor is an intrinsic part of the circuit.

Its forward drop is set by the material's band gap, which is why color and voltage go together and why a blue one needs far more than a red. It has almost no reverse tolerance and always needs its current limited.

The two types are complementary rather than interchangeable: supply polarity and current direction both reverse between them. The very thin middle layer is what makes transistor action possible in the first place.

A switch is driven hard into one state or the other, spending as little time as possible in between, because that is where dissipation peaks. Saturation gives a low drop rather than zero, and that residue is the switching loss.

Here the device sits deliberately between the switching extremes, in the region a switch tries to cross quickly. It amplifies because a small input controls a larger output, not because it creates energy — the supply provides that.

It varies widely between devices of the same part number, and with temperature and current, so a design depending on a precise value is fragile. Good designs use feedback to make the result independent of it.

The insulated gate draws essentially no steady current, which is the fundamental difference from a bipolar device and why it dominates in switching. That same insulation is thin enough to be destroyed by a static discharge.

Latching is the key property: once triggered they stay on until the current falls near zero, so control means choosing when in the cycle to trigger. A triac handles both half-cycles where a thyristor handles one.

The ideal assumptions — enormous gain, no input current, no offset — are what make these circuits analyzable by hand, and real devices come close enough for most work. They amplify the difference, not either input on its own.

Feeding the output back to oppose the input makes behavior depend on the feedback components rather than on the device, which is what fixes the gain variability. The trade is deliberate: raw gain is spent buying predictability.

Both respond to light and they differ in the tradeoff: the transistor gives more output through internal gain, the diode responds faster and more linearly. Sensitivity against speed is the usual choice between them.

The response is deliberately large and non-linear, which is what makes it a good sensor and a poor resistor. The two types move in opposite directions, so the sign of the coefficient has to be checked before use.

It conducts hard above a threshold, clamping a surge by shunting it, so it is a sacrificial part that degrades a little with every event. Because it fails gradually and invisibly, a surge protector is not permanently effective.

Poles count the independent circuits switched and throws count the positions each can select, and the two are read independently of each other. Momentary against latching describes what happens on release, which is a separate axis again.

Voltage matters only as what drives current through the path's resistance, which is why a high-voltage static spark is merely startling while a much lower supply can kill. Duration matters too, since injury depends on current sustained over time.

The ranges are narrow, and the dangerous one is not far above the merely perceptible one, which is what makes casual contact so risky. They vary with frequency, path and individual, so published figures are guidance rather than a margin to work against.

Dry intact skin provides most of the protection, and it falls dramatically when wet, broken, or in prolonged contact — which is why the same voltage can be harmless one day and not the next. Internal tissue offers comparatively little.

Because the danger is current reaching the heart, the entry and exit points matter as much as the magnitude does. That reasoning lies behind the habit of working one-handed where practical, and behind attention to footwear and standing surface.

It is a different hazard from shock: no contact is required, and the danger is radiant heat, pressure and ejected material. That is why the protection is distance and rated clothing rather than insulation alone.

The two purposes are distinct and routinely confused: a low-impedance fault path lets protection operate quickly, while a reference sets the potential everything is measured against. Earth itself is a poor conductor and rarely the fault path.

Bonding removes a voltage difference between things a person could touch at once; grounding ties the system to a reference. Bonding does its job whether or not earth is involved, which is why the two are not interchangeable words.

Its job is to carry fault current back to the source at a low enough impedance that the protective device opens quickly. It normally carries nothing, so it can be broken and go unnoticed until the fault it exists for actually happens.

If outgoing and returning current differ, some is returning by another path, possibly a person, so the device acts on imbalance rather than on magnitude. That is why it protects where a fuse cannot: the leakage is far too small to trip one.

An arc can start a fire while drawing current well within a circuit's rating, so it is invisible to overcurrent protection. Detection works from the electrical signature rather than the amount, which is why it is a separate device again.

The device exists to stop the wire overheating, so the conductor sets the limit and a smaller load is simply protected more than it needs. Upsizing protection to stop nuisance operation removes the protection the conductor actually had.

A short is an abnormal low-resistance path producing very large current at once; an overload is the circuit doing its ordinary job for more load than it should carry. Protective devices answer each on different timescales, which is why they are time-current devices.

The principle is that the person exposed controls the isolation and verifies it personally, because a switch somebody else can operate is not isolation at all. Stored energy in capacitors, springs and raised parts has to be released as well.

The second proving is what makes it a test rather than an assumption: a tester that failed between the first check and the circuit would report exactly what a dead circuit reports. That silent-failure mode is the entire reason for the third step.

Ratings are specific to a voltage and degrade with use, damage and contamination, so inspection before each use is part of the protection rather than an extra. Equipment is the last line, after de-energizing and after distance.

Voltage is raised for transport and lowered again near the point of use, so the same power travels the chain at very different currents. Following that one variable explains why the equipment looks so different at each stage.

A transformer does not change power, only the voltage-current pairing, which is why raising voltage cuts current and therefore cuts loss. The same device performs both roles depending on which winding is driven.

Levels differ by country and by utility, so what is worth carrying is the pattern rather than the figures: highest for long transmission, intermediate for local distribution, lowest at the service. Always confirm the actual level on site.

The center tap makes two voltages available from one winding, one between either leg and neutral and double that between the legs. The neutral carries only the imbalance between the two sides rather than their sum.

These are deviations in the supply rather than interruptions, so equipment can misbehave while apparently powered normally — which is what makes them hard to diagnose. Each has a different cause and a different mitigation.