The Zener Diode as a Voltage Regulator

Every phone charger, every laptop brick, every USB port promises you a fixed voltage — 5 volts, 12 volts, whatever the label says. But the wall supply behind it sags and surges, the load keeps changing as your device wakes and sleeps, and a raw rectifier output ripples up and down at every mains cycle. Something has to take that messy, wobbling input and pin the output to one steady number. The oldest and simplest device that does exactly that is a single diode run backwards, on purpose, past the edge of a cliff that would destroy an ordinary diode. That cliff is reverse breakdown. In the last two articles you met the junction diode and its I–V curve: forward-bias it past the knee (about $0.7\ \text{V}$ for silicon) and it conducts; reverse-bias it and only a whisper of leakage flows — until the reverse voltage grows so large that the diode suddenly conducts hard backwards. For a normal rectifier that runaway means a puff of smoke. But a diode built for it — heavily doped, with its breakdown set to a chosen value — sits happily on that cliff edge, its voltage barely moving even as the current through it swings over a huge range. That flat, near-vertical breakdown region is a natural voltage clamp, and the diode that exploits it is the Zener diode. This article shows how the Zener holds an output constant, how a single series resistor turns it into a working voltage regulator, how to size that resistor so the diode neither starves nor cooks, and why the whole thing runs on algebra and Ohm's law you already own — no calculus, just current-sharing and a little circuit bookkeeping.

Key takeaways

A Zener diode is a diode designed to be run in reverse breakdown, where its voltage stays nearly constant at a chosen value $V_Z$ even as the current through it varies widely. That flat region is the whole trick.
Heavy doping makes the junction very narrow, so a few volts of reverse bias create an enormous internal field ( $\sim\!10^{8}\ \text{V/m}$ ) that frees carriers by quantum tunnelling (Zener breakdown, below $\sim\!5\ \text{V}$ ) or impact ionisation (avalanche breakdown, above $\sim\!6\ \text{V}$ ).
A regulator is just a Zener plus one series resistor $R_S$ . The resistor drops the excess voltage and limits the current; the Zener clamps the output at $V_Z$ . The load hangs across the Zener.
The circuit obeys three equations: $I_S=(V_i-V_Z)/R_S$ , $I_S=I_Z+I_L$ , and $I_L=V_Z/R_L$ . The Zener swallows whatever current the load does not, keeping the output fixed.
You must keep $I_Z$ inside a window. Below a minimum knee current the clamp fails; above $I_{Z,\max}=P_{Z,\max}/V_Z$ the diode overheats and dies. Choosing $R_S$ means keeping $I_Z$ in that window for every combination of input and load.
Board scope, honestly: the Zener as a regulator is core for CBSE, ICSE, JEE Main and Advanced, and A-Level, but NEET drops the regulator detail, AP Physics 2 carries no semiconductors at all, and IGCSE stops at the qualitative diode.
Key idea: a heavily doped diode held in its sharp reverse-breakdown region behaves as a constant-voltage source $V_Z$ ; with a series resistor to absorb the surplus and set the current, it clamps an output steady against changes in supply and load, provided the Zener current stays between its knee value and its power-limited maximum.

The core insight

Run a specially doped diode in reverse breakdown and its voltage stays fixed at $V_Z$ over a wide range of current, so a single series resistor — dropping the excess voltage and limiting the current — turns it into a voltage regulator that holds the output at $V_Z$ against supply ripple and load changes, as long as the Zener current never falls below its knee or rises above its power limit $P_{Z,\max}/V_Z$ .

Where this lives today

The humble Zener is everywhere a circuit needs a reference or a limit. As a cheap regulator it still sits in low-current corners of power supplies, bias networks, and battery chargers — anywhere a few milliamps at a fixed voltage will do. Reverse-connected across a signal line, a Zener (or a back-to-back pair) is a clamp that shears off spikes and protects delicate inputs from over-voltage — a front-line defence in surge and electrostatic-discharge protection. Precision "buried Zener" references, trimmed and temperature-compensated, are the quiet backbone of accurate voltmeters and analog-to-digital converters, holding a reference stable to parts per million. And the very tunnelling that Zener described in 1934 reappears in the tunnel diode and in the flash-memory cells that store data in your phone. For higher currents and tighter regulation, the Zener hands off to the transistor — the three-terminal device of the next article — which can be wrapped in feedback to build the precise, high-current regulators a bare Zener cannot.

The physical picture

Take the reverse-bias half of the diode I–V curve from the previous article and push it further than you dared before. In reverse bias the junction's depletion region widens and only a tiny saturation ("leakage") current flows, set by the trickle of minority carriers. Increase the reverse voltage and, for an ordinary diode, nothing much happens — until you reach a voltage where the current abruptly shoots up, almost vertically. Cross that in a rectifier diode and the uncontrolled current destroys it. A Zener diode is built so that this breakdown happens at a deliberately chosen, modest voltage $V_Z$ — anything from a few volts to a few tens of volts — and so that it can carry the breakdown current safely and repeatedly. Why does breakdown happen at all, and why can we choose where? The secret is heavy doping. Recall that the depletion region straddling a p–n junction is a zone swept clear of mobile carriers, and that its width shrinks when both sides are doped more heavily (there are more fixed ions per unit length, so fewer are needed to build the barrier). A Zener diode is doped far more heavily than a rectifier, so its depletion layer is extraordinarily thin — of order ten nanometres. Put even a few volts across ten nanometres and the electric field inside is colossal: $E \approx V/w \approx 5\ \text{V}/(10^{-8}\ \text{m}) \approx 5\times10^{8}\ \text{V/m}$ , over a hundred times the field that sparks air. Two mechanisms then flood the junction with carriers. The first is Zener breakdown proper, a genuinely quantum effect. The field is strong enough to rip valence electrons straight out of their covalent bonds — or, in band language, to let electrons tunnel across the forbidden gap from the valence band into empty conduction states. (Tunnelling is a quantum result we quote here, not derive; at this level it is a named picture, not a calculation.) This dominates in heavily doped, low-voltage devices, below about $5\ \text{V}$ . The second is avalanche breakdown: a stray carrier, accelerated by the field, smashes into the lattice and knocks loose an electron–hole pair, which are themselves accelerated and knock loose more — a chain multiplication of carriers. Avalanche dominates in lighter-doped, higher-voltage devices, above about $6\ \text{V}$ . Both give the same circuit behaviour — a sharp, near-constant-voltage breakdown — so by long habit every such diode is called a "Zener," even the avalanche ones. (Near $5.6\ \text{V}$ the two mechanisms overlap and their opposite temperature drifts cancel, which is why reference diodes are often specified there.)

Figure

The current versus voltage curve of a Zener diode: forward bias conducts past the silicon knee near 0.7 volts, reverse bias leaks only slightly until the sharp breakdown at minus V sub Z where the current drops nearly vertically while the voltage stays clamped at V sub Z

The payoff is in the shape of that breakdown segment: it is nearly vertical. A vertical I–V line means the voltage is the same whatever the current — precisely the definition of a constant-voltage source. Slide the current up and down that line and the voltage across the Zener barely stirs from $V_Z$ . If we can arrange for a load to sit across a Zener held in this region, the load sees a rock-steady $V_Z$ . The only remaining job is to feed the Zener from a messy supply without letting the current run away — and that is the work of a single resistor.

From a clamp to a regulator: the series resistor

A bare Zener across a supply would either see no breakdown (if the supply is below $V_Z$ ) or draw unlimited current (if above), because nothing in the loop limits it. The fix is a series resistor $R_S$ between the unregulated input $V_i$ and the Zener. The load $R_L$ hangs in parallel with the Zener, and the output is taken across that pair. The resistor plays two roles at once: it drops the difference between the input and $V_Z$ , and it limits the current into the junction. Everything the regulator does follows from Kirchhoff's laws applied to this three-element loop.

Figure

A Zener shunt regulator circuit: an unregulated input V sub i feeds a series resistor R sub S, then the node splits into a Zener diode in reverse to ground and a load resistor R sub L, with the series current equal to the Zener current plus the load current

Start with the loop from input to ground. The resistor and the Zener are in series across $V_i$ , and while the Zener is in breakdown the voltage across it is $V_Z$ . So the voltage across $R_S$ is the leftover, $V_i - V_Z$ , and the current the source pushes through the resistor is

I_S \;=\; \frac{V_i - V_Z}{R_S}. \qquad (1)

where: $I_S$ is the series (line) current out of the supply, $V_i$ is the unregulated input voltage, $V_Z$ is the Zener breakdown voltage, and $R_S$ is the series resistor. The physical reading: because $V_Z$ is fixed, the resistor sees a fixed voltage drop $V_i-V_Z$ for a given input, so $I_S$ is set entirely by the input and the resistor, not by the load. At the output node the series current splits between the Zener and the load. Kirchhoff's current law gives

I_S \;=\; I_Z + I_L, \qquad\text{with}\qquad I_L \;=\; \frac{V_Z}{R_L}. \qquad (2)

where: $I_Z$ is the current diverted through the Zener and $I_L=V_Z/R_L$ is the current the load draws at the regulated output $V_Z$ . Combining (1) and (2) gives the current the Zener actually carries,

I_Z \;=\; I_S - I_L \;=\; \frac{V_i - V_Z}{R_S} - \frac{V_Z}{R_L}. \qquad (3)

Equation (3) is the heart of the regulator. Read it as a shock absorber. The series current $I_S$ is nailed down by $V_i$ and $R_S$ . If the load suddenly demands more current (smaller $R_L$ , larger $I_L$ ), the Zener simply gives up the same amount ( $I_Z$ falls) so that the sum stays put — and because the total is fixed, the drop across $R_S$ is unchanged and the output stays at $V_Z$ . If the load eases off, the surplus current flows into the Zener instead ( $I_Z$ rises). The Zener soaks up the difference either way. That is why a shunt regulator regulates: the diode is a variable sink parked in parallel with the load, always adjusting $I_Z$ to keep $I_S=I_Z+I_L$ balanced at a fixed output.

The current window: where the regulator lives and dies

Equation (3) works only while the Zener is genuinely in breakdown, and that pins $I_Z$ inside a window with a floor and a ceiling. The floor is the knee current $I_{Z,\min}$ : near the top of the breakdown knee the curve is not yet vertical, so if $I_Z$ falls too low the "constant-voltage" approximation collapses and the output sags below $V_Z$ . The ceiling is set by heat. A Zener rated for a maximum power dissipation $P_{Z,\max}$ can carry at most

I_{Z,\max} \;=\; \frac{P_{Z,\max}}{V_Z}. \qquad (4)

where: $P_{Z,\max}$ is the diode's rated power and $V_Z$ its breakdown voltage. Beyond $I_{Z,\max}$ the junction overheats and is destroyed. So a working regulator must keep

I_{Z,\min} \;\le\; I_Z \;\le\; I_{Z,\max} \qquad\text{for every input and load it will meet.}

The two worst cases are opposite corners. The Zener current is largest when the load takes the least (load disconnected, $I_L=0$ , so $I_Z=I_S$ ) and the input is highest — that is the case that can blow the diode. The Zener current is smallest when the load takes the most ( $R_L$ smallest, $I_L$ largest) and the input is lowest — that is the case where regulation drops out. Designing $R_S$ is nothing more than keeping both corners inside the window, and we do exactly that in the worked examples below.

Go deeper · Derivation B · 2 minHow steady is steady? Line regulation from the Zener's dynamic resistance

Derivation BHow steady is steady? Line regulation from the Zener's dynamic resistance

The goal is to put a number on how much the output really moves when the input wobbles — because a real breakdown curve is not perfectly vertical; it has a small slope that separates a good regulator from a lazy one. We model the almost-vertical breakdown segment by a small dynamic resistance $r_Z$ (the reciprocal of the curve's steep slope, typically a few ohms) and show the input ripple is divided down by $r_Z/(R_S+r_Z)$ . (Scope note: the $r_Z$ small-signal model and the line-regulation ratio go beyond the board numericals — CBSE, ICSE, JEE Main and Advanced, and A-Level all stop at the current-window design of the previous section — so treat this as enrichment for insight, not exam drill.)

Step 1. Replace the Zener in breakdown by its small-signal model: a fixed voltage plus a small series resistance $r_Z$ , so a change in Zener current $\Delta I_Z$ produces a change in Zener voltage $\Delta V_Z=r_Z\,\Delta I_Z$ . A perfectly vertical curve would have $r_Z=0$ and never move; a real one has $r_Z$ of order a few ohms.

Step 2. For changes only (hold the load steady so the load draws its ripple through the same path), the input change $\Delta V_i$ sees $R_S$ and $r_Z$ in series to the output. The change in current is $\Delta I=\Delta V_i/(R_S+r_Z)$ , and this same current flows through $r_Z$ .

Step 3. The output change is the voltage this current develops across $r_Z$ :

\Delta V_{out} \;=\; \Delta V_i \cdot \frac{r_Z}{R_S+r_Z}. \qquad (5)

Because $r_Z\ll R_S$ by design, the fraction is tiny — the input ripple is attenuated by roughly $r_Z/R_S$ . With $r_Z=5\ \Omega$ and $R_S=200\ \Omega$ , a $2\ \text{V}$ swing on the input produces only $2\times 5/205\approx 0.049\ \text{V}$ , about $49\ \text{mV}$ , at the output — a factor-of-40 reduction. The same attenuation smooths mains ripple, which is why a Zener stage often follows a rectifier-and-filter.

Step 4. Read equation (5) as a design lever: make $R_S$ large (relative to $r_Z$ ) and the output holds tighter, but a larger $R_S$ also lowers the maximum load current the stage can supply. Good regulator design is the tension between these two, resolved by the current window of the previous section. Verified: $\Delta V_{out}=2\times 5/205=0.0488\ \text{V}$ and the line-regulation ratio $5/205=2.44\%$ — recomputed in verify_hs2-zener-voltage-regulator.py.

Limiting cases

case	result	meaning
$V_i < V_Z + I_{Z,\min}R_S$	Zener not in breakdown; $V_{out}<V_Z$	dropout — input too low, the clamp cannot form
$I_Z$ falls below $I_{Z,\min}$	output sags below $V_Z$	load too heavy or input too low — regulation lost
$I_Z$ rises above $I_{Z,\max}=P_{Z,\max}/V_Z$	junction overheats, diode destroyed	$R_S$ too small, or load removed at high input
load removed ( $R_L\to\infty$ , $I_L=0$ )	$I_Z=I_S$ — worst-case dissipation	the diode must survive the full series current
$V_i$ rises by $\Delta V_i$	$V_{out}$ rises only by $\Delta V_i\,r_Z/(R_S+r_Z)$	line regulation — input ripple divided down
ideal Zener, $r_Z\to 0$	$V_{out}=V_Z$ exactly, for all $I_Z$ in window	the perfect constant-voltage source
forward bias	ordinary diode, $\approx 0.7\ \text{V}$ drop	a Zener regulates only in reverse breakdown

Common trapA Zener regulates in reverse breakdown — not in forward bias, and its breakdown is not a fault

Two misreadings sink students here. The first is to wire the Zener like a rectifier diode, forward-biased. In forward bias a Zener is just an ordinary silicon diode with a $\approx 0.7\ \text{V}$ drop; it does nothing useful as a regulator. The regulating action lives entirely in reverse bias, past breakdown — so in a regulator the Zener's cathode goes to the positive side, the opposite of a rectifier. The second misreading is to think breakdown means the diode is broken. Reverse breakdown in a Zener is a designed, reversible, non-destructive operating region; the diode is meant to live there. What destroys it is not breakdown but excess current — exceeding $I_{Z,\max}=P_{Z,\max}/V_Z$ — which is a heat limit, not a voltage limit. Keep the current inside the window and a Zener will sit in breakdown indefinitely. A third, quieter slip: the output is clamped to $V_Z$ , the Zener's rated breakdown voltage, not to the $0.7\ \text{V}$ knee of an ordinary forward diode — do not confuse the two "knees."

Worked examples

Board level — a basic 10 V regulator (CBSE and ICSE pattern)

An unregulated supply $V_i=20\ \text{V}$ feeds a $10\ \text{V}$ Zener through a series resistor $R_S=200\ \Omega$ . A load $R_L=1\ \text{k}\Omega$ is connected across the Zener. Find (a) the series current, (b) the load current, (c) the Zener current, and (d) the power dissipated in the Zener and in $R_S$ . Assume an ideal Zener in breakdown.

Step 1. The output is clamped at $V_{out}=V_Z=10\ \text{V}$ . The series resistor drops the rest: $V_i-V_Z=20-10=10\ \text{V}$ . By equation (1), the series current is $I_S=(V_i-V_Z)/R_S=10/200=0.05\ \text{A}=\boxed{50\ \text{mA}}$ .

Step 2. The load sees the clamped output, so by Ohm's law $I_L=V_Z/R_L=10/1000=0.01\ \text{A}=\boxed{10\ \text{mA}}$ .

Step 3. The Zener carries the difference, equation (3): $I_Z=I_S-I_L=50-10=\boxed{40\ \text{mA}}$ .

Step 4. Power in the Zener: $P_Z=V_Z\,I_Z=10\times0.040=\boxed{0.4\ \text{W}}$ , so a $1\ \text{W}$ Zener is comfortable here. Power in the resistor: $P_{R_S}=(V_i-V_Z)^2/R_S=10^2/200=\boxed{0.5\ \text{W}}$ , so $R_S$ should be rated at $1\ \text{W}$ . As a check on the worst case, if the load is disconnected ( $I_L=0$ ) all $50\ \text{mA}$ flows through the Zener, raising its dissipation to $P_Z=10\times0.050=0.5\ \text{W}$ — still within a $1\ \text{W}$ rating, so this regulator is safe with no load.

Competitive level — choosing R_S for a range of load (JEE Main and Advanced pattern)

A regulated $12\ \text{V}$ output is to be produced from a fixed $V_i=24\ \text{V}$ supply using a $12\ \text{V}$ , $1\ \text{W}$ Zener. The load current varies from $0$ to $40\ \text{mA}$ . The Zener needs at least $I_{Z,\min}=10\ \text{mA}$ to stay in regulation. Find the range of series resistance $R_S$ that works, and choose a suitable standard value.

Step 1. Find the Zener's own ceiling from its power rating, equation (4): $I_{Z,\max}=P_{Z,\max}/V_Z=1/12=83.3\ \text{mA}$ . The Zener current must stay between $10\ \text{mA}$ and $83.3\ \text{mA}$ across all conditions.

Step 2. With a fixed input, the series current is $I_S=(V_i-V_Z)/R_S=(24-12)/R_S=12/R_S$ , the same for every load. The Zener share is $I_Z=I_S-I_L$ . The heaviest load ( $I_L=40\ \text{mA}$ ) drives $I_Z$ to its minimum; that minimum must not fall below $10\ \text{mA}$ , so $I_S\ge I_{L,\max}+I_{Z,\min}=40+10=50\ \text{mA}$ .

Step 3. The lightest load ( $I_L=0$ ) drives $I_Z$ to its maximum $I_Z=I_S$ ; that must not exceed $83.3\ \text{mA}$ , so $I_S\le83.3\ \text{mA}$ . The series current is therefore boxed into $50\ \text{mA}\le I_S\le 83.3\ \text{mA}$ .

Step 4. Translate to resistance through $R_S=12/I_S$ . The largest allowed resistor comes from the smallest current: $R_{S,\max}=12/0.050=240\ \Omega$ . The smallest allowed resistor comes from the largest current: $R_{S,\min}=12/0.0833=144\ \Omega$ . So any $R_S$ in $\boxed{144\ \Omega \le R_S \le 240\ \Omega}$ works. Choose a mid-range standard value, $R_S=200\ \Omega$ : then $I_S=12/200=60\ \text{mA}$ , so at no load $I_Z=60\ \text{mA}$ (dissipating $12\times0.060=0.72\ \text{W}<1\ \text{W}$ , safe) and at full load $I_Z=60-40=20\ \text{mA}$ (above $10\ \text{mA}$ , in regulation). The resistor dissipates $(24-12)^2/200=144/200=0.72\ \text{W}$ , so specify a $1\ \text{W}$ resistor. Timing note: about two to three minutes. The whole design is two inequalities — "smallest load fixes the max current, biggest load fixes the min current." Compute $I_{Z,\max}=P/V_Z$ first, write $I_S$ between $I_{L,\max}+I_{Z,\min}$ and $I_{Z,\max}$ , then invert $R_S=(V_i-V_Z)/I_S$ at both ends. Do not solve for $R_S$ directly from a single condition — examiners set the trap of a load range precisely to catch students who size the resistor for only one load.

Exam traps

Forward-biasing the Zener. The regulating action is in reverse breakdown. Draw the Zener with its cathode toward the positive rail; forward-biased, it is a plain $0.7\ \text{V}$ diode and regulates nothing. The circuit symbol's bent bar marks the cathode — get it the right way round.

Sizing $R_S$ for one load only. When the load varies, the resistor must satisfy both extremes at once: the diode must survive the no-load case ( $I_Z=I_S\le I_{Z,\max}$ ) and stay in regulation under full load ( $I_Z\ge I_{Z,\min}$ ). Solving only one inequality is the single most common lost-mark in JEE Zener problems.

Forgetting the power ceiling. $I_{Z,\max}$ is not a voltage limit; it is $P_{Z,\max}/V_Z$ , a heat limit. A " $12\ \text{V}$ Zener" says nothing about how much current it can take until you divide its power rating by $12\ \text{V}$ . Always convert the power rating to a current before checking the worst case.

Assuming the output equals the input's shape. The output is clamped at $V_Z$ , not at $V_i$ and not at the $0.7\ \text{V}$ forward knee. Under regulation, $V_{out}=V_Z$ regardless of how $V_i$ wobbles — that is the whole point — provided $V_i>V_Z+I_{Z,\min}R_S$ so the diode stays in breakdown.

Confusing Zener and avalanche as if the exam needs the mechanism. For the circuit, both give a constant $V_Z$ ; you rarely need to name which mechanism is at work. Do know that heavy doping + narrow junction + low voltage ( $<5\ \text{V}$ ) means Zener (tunnelling), and lighter doping + higher voltage ( $>6\ \text{V}$ ) means avalanche — but do not lose time deriving either; the tunnelling picture is quoted, never calculated, at this level.

Board scope. The Zener diode itself, its I–V curve, and its use as a voltage regulator are core for CBSE (Ch. 14), ICSE/ISC, JEE Main, JEE Advanced, and A-Level 9702. NEET carries semiconductors but drops the Zener regulator detail — know the diode and its breakdown, but the full $R_S$ -design numericals are not NEET material. AP Physics 2 carries no semiconductor electronics at all, and IGCSE 0625 stops at the qualitative ordinary diode and LED — neither expects the Zener regulator. IB has moved most electronics out of its core; do not assume it there. When in doubt, answer the physics; just do not tell a NEET or IGCSE student this is on their syllabus.

Origins · 1 minClarence Zener and how this was discovered

Clarence Zener

1905–1993 · Indianapolis and Pittsburgh, USA

Clarence Melvin Zener was an American theoretical physicist whose name is stamped on the diode even though the devices that carry it often work by a different mechanism. In 1934, while thinking about how insulators fail under strong electric fields, he proposed that an electron could be lifted straight across the forbidden energy gap by the field itself — a quantum tunnelling process now called Zener breakdown. That idea, published years before the transistor, became the theory behind low-voltage breakdown diodes when semiconductor engineering caught up in the 1950s; ironically, higher-voltage "Zener" diodes actually break down by avalanche, a mechanism others later clarified. Zener's restless career spanned Westinghouse and several universities, and his name also survives in metallurgy (the Zener–Hollomon parameter, Zener pinning) and in the theory of internal friction in solids. A rare theorist whose 1934 insight quietly powers billions of everyday power supplies.

History

1934

— Clarence Zener publishes a quantum-mechanical theory of electrical breakdown in insulators, the tunnelling mechanism later named after him.

1940

— Russell Ohl discovers the silicon p–n junction at Bell Labs, the structure whose reverse breakdown the Zener exploits.

1949

— William Shockley's junction-diode theory frames the forward, reverse, and breakdown behaviour of the p–n junction quantitatively.

1954

— Kenneth McKay shows that breakdown above about six volts is avalanche multiplication, not Zener tunnelling, sorting out the two mechanisms.

1950

s — Zener and reference diodes become standard catalogue parts, giving electronics its first cheap, stable voltage references.

Going further

Texts. NCERT Physics Class 12, Chapter 14 ("Semiconductor Electronics"), the special-purpose p–n junction diodes section · H. C. Verma, Concepts of Physics Vol. 2, "Semiconductors and Semiconductor Devices" · Boylestad & Nashelsky, Electronic Devices and Circuit Theory, the Zener-diode applications chapter (for the $R_S$ -design method in full) · Sedra & Smith, Microelectronic Circuits, for the small-signal $r_Z$ model.

Provenance. All values are LANE_SPEC anchors and standard device figures. Recompute: silicon knee $0.7\ \text{V}$ , Ge $0.3\ \text{V}$ ; Zener/avalanche crossover $\sim\!5$ – $6\ \text{V}$ , near-zero tempco $\sim\!5.6\ \text{V}$ ; internal field $E=5\ \text{V}/10^{-8}\ \text{m}=5\times10^{8}\ \text{V/m}$ ; $1\ \text{W}$ , $10\ \text{V}$ Zener gives $I_{Z,\max}=1/10=100\ \text{mA}$ . Board example: $V_i=20$ , $V_Z=10$ , $R_S=200\ \Omega$ , $R_L=1000\ \Omega$ $\Rightarrow$ $I_S=10/200=50\ \text{mA}$ , $I_L=10/1000=10\ \text{mA}$ , $I_Z=40\ \text{mA}$ , $P_Z=0.4\ \text{W}$ , $P_{R_S}=100/200=0.5\ \text{W}$ , no-load $I_Z=50\ \text{mA}$ , no-load $P_Z=0.5\ \text{W}$ . Design example: $V_i=24$ , $V_Z=12$ , $1\ \text{W}$ $\Rightarrow$ $I_{Z,\max}=1/12=83.3\ \text{mA}$ ; $I_L\in[0,40]\ \text{mA}$ , $I_{Z,\min}=10\ \text{mA}$ $\Rightarrow$ $I_S\in[50,83.3]\ \text{mA}$ , $R_S\in[144,240]\ \Omega$ ; pick $200\ \Omega$ $\Rightarrow$ $I_S=60\ \text{mA}$ , no-load $I_Z=60\ \text{mA}$ ( $0.72\ \text{W}$ ), full-load $I_Z=20\ \text{mA}$ , $P_{R_S}=144/200=0.72\ \text{W}$ . Line regulation: $r_Z=5\ \Omega$ , $R_S=200\ \Omega$ $\Rightarrow$ $\Delta V_{out}=2\times5/205=0.0488\ \text{V}\approx49\ \text{mV}$ , ratio $5/205=2.44\%$ .

Prerequisites. The Junction Diode and its I–V Curve — the forward knee, the reverse leakage, and the breakdown edge that this article turns into a tool. Everything here is the reverse-breakdown corner of that curve, put to work. Leads to. The Bipolar Junction Transistor, the three-terminal device that takes over where the Zener runs out of current and precision, and — as a university go-deeper seam only, never an exam dependency — avalanche photodiodes and photomultipliers, where the same reverse-breakdown avalanche is harnessed to multiply a photocurrent instead of to clamp a voltage.

Isn't "breakdown" the point where a diode is destroyed? Why run a diode there on purpose?

For an ordinary rectifier diode, reverse breakdown is fatal — but only because nothing limits the current, so it runs away and the junction cooks. A Zener diode is built for breakdown: heavily doped so it breaks down at a low, precise voltage, and packaged to shed the heat. In a regulator a series resistor limits the current, so the diode sits in breakdown safely and reversibly, indefinitely. Breakdown here is an operating region, not a failure. What actually destroys a Zener is exceeding its power rating — too much current, hence too much heat — not the mere fact of being in breakdown. Keep $I_Z$ below $P_{Z,\max}/V_Z$ and it lives happily on the cliff edge.

What is the difference between Zener breakdown and avalanche breakdown?

They are two different microscopic ways to flood the junction with carriers, but they look identical from the circuit's side — both give a sharp, near-constant breakdown voltage. Zener breakdown is quantum tunnelling: in a very heavily doped, very narrow junction, a few volts create a field strong enough (about $10^{8}\ \text{V/m}$ ) to tear electrons directly across the forbidden gap. It dominates below about $5\ \text{V}$ . Avalanche breakdown is a collision cascade: a carrier accelerated by the field knocks loose an electron–hole pair, which knock loose more, multiplying in a chain. It dominates above about $6\ \text{V}$ . By convention every such diode is called a "Zener," and for solving a circuit you almost never need to say which mechanism is at work.

Why does the Zener voltage stay so constant even as the current changes a lot?

Because the breakdown segment of the I–V curve is almost vertical. Once past the knee, a huge change in current corresponds to only a tiny change in voltage — that is what "nearly vertical" means graphically. Physically, in breakdown the junction supplies carriers extremely readily, so it can pass whatever current the external circuit demands without needing a bigger voltage to do it. The small residual slope is captured by the dynamic resistance $r_Z$ , typically a few ohms, so the voltage does creep up slightly with current: $\Delta V=r_Z\Delta I$ . But with $r_Z$ so small compared with the series resistor, the output barely moves. A perfectly vertical curve ( $r_Z=0$ ) would be an ideal constant-voltage source.

How do I choose the series resistor R sub S?

Work the two worst cases. First find the Zener's current ceiling, $I_{Z,\max}=P_{Z,\max}/V_Z$ , and note its knee minimum $I_{Z,\min}$ . The Zener current is highest when the load draws least (load off, so $I_Z=I_S$ ) and the input is highest — this must stay below $I_{Z,\max}$ , which sets the smallest allowed $R_S$ . The Zener current is lowest when the load draws most and the input is lowest — this must stay above $I_{Z,\min}$ , which sets the largest allowed $R_S$ . Any $R_S$ between those two limits works; pick a mid-range standard value and re-check both corners. The mistake to avoid is sizing $R_S$ for a single, typical load and forgetting that the diode must survive the no-load extreme and hold up under the full-load extreme.

What happens if the load is disconnected entirely?

Nothing bad — as long as the Zener can take it. With the load gone, $I_L=0$ , so the entire series current now flows through the Zener: $I_Z=I_S=(V_i-V_Z)/R_S$ . This is the worst case for the diode's dissipation, $P_Z=V_Z I_S$ , because the Zener is soaking up all the current the resistor delivers. A regulator must be designed so that even this no-load current stays below $I_{Z,\max}$ . This is precisely why you cannot size the resistor for the typical load alone: the moment the load is unplugged, the Zener inherits everything. If $R_S$ is too small, removing the load can destroy the diode.

Why not just use a plain resistor divider to drop the voltage?

A resistive divider gives a lower voltage but does not regulate. Its output depends on the load: put a load across the lower resistor and you change the divider ratio, so the output sags — by a different amount for every load. It also passes every wobble of the input straight through, scaled by the fixed ratio. The Zener beats this because its voltage is set by diode physics, not by a ratio: whatever current the load or input throws at the node, the Zener re-shares the current to hold $V_{out}=V_Z$ . That active current-sharing — impossible for a passive divider — is what makes it a regulator rather than an attenuator.

Does the Zener regulate against changes in the load, the supply, or both?

Both, within limits. Against supply ripple (line regulation): a change $\Delta V_i$ appears at the output only scaled down by $r_Z/(R_S+r_Z)$ , a factor of tens or more, so mains ripple and supply drift are strongly suppressed. Against load changes (load regulation): as the load current varies, the Zener adjusts its own current to keep $I_S=I_Z+I_L$ balanced at fixed output. The limits are the current window: if the load grows so heavy that $I_Z$ would fall below $I_{Z,\min}$ , or the supply falls so low that the diode leaves breakdown, regulation collapses; and if the load vanishes at high supply, the Zener must survive $I_Z=I_S$ . Stay inside the window and it holds against both.