Audited ·Last updated 31 Jul 2026·4 citations·Tier 2·0 uses

Voltage Regulation Calculator — Transformer, Load and Line Regulation

Work out percent voltage regulation for a transformer, or load and line regulation for a power supply — with the correct denominator shown both ways.

Voltage Regulation Calculator

Which regulation figure
No-load voltage for a transformer, minimum-load voltage for load regulation, highest-line voltage for line regulation. Measure it with the same meter you use for the other reading.
V
Full-load voltage for a transformer, maximum-load voltage for load regulation, lowest-line voltage for line regulation. This is the denominator all three published definitions use.
V
Line regulation only. How far the input was moved between the two readings — a sweep from 3 V to 40 V is a span of 37 V. It converts the percentage into the %/V figure datasheets quote. Ignored in the other two modes.
V
Regulation
6.87
(V_light − V_heavy) ÷ V_heavy × 100, the figure all three published definitions produce. Lower is better and zero is ideal; a negative value means the output rose under load, which usually means the two readings were entered the wrong way round.
Voltage change
0.642 V
Denominator used
9.348 V
Same figure on the other base
6.43
Reading the result
Transformer / generator voltage regulation: the output moves 0.642 V between 9.99 V at no-load and 9.348 V at full-load. Against the full-load voltage — the denominator all three published definitions use — that is 6.8678 %. Measured against the no-load voltage instead it would read 6.4264 %, which is the same circuit and the wrong convention. Check which base a quoted figure used before comparing two supplies. The figure is positive, which is the normal direction: loading the supply, or dropping the line, pulls the output down. At or above the 3 % expected of a good power transformer on a resistive load, so this is loose regulation. An inductive load will make it worse still — the published benchmark is a best case measured resistively. This is arithmetic on two measured voltages: it says nothing about why the output moved, and nothing about behaviour between the two points, which is rarely a straight line. Measure both voltages at the same temperature and with the same meter, because a 0.1 % meter error on a 1 % regulation figure is a tenth of the answer.

Background.

Voltage regulation measures how far a supply's output moves when you lean on it. Take a transformer whose secondary reads 9.990 V with nothing connected and 9.348 V at full load: the output has sagged 0.642 V, and expressed as a percentage of the loaded voltage that is 6.87 % regulation. The lower the number the stiffer the supply, and zero is the ideal that no real transformer reaches, because winding resistance and leakage inductance sit in series with the load.

The difficulty with this calculation is not the arithmetic — it is knowing which voltage goes on the bottom. Every published definition divides by the loaded, worst-case reading, not by the no-load reading and not by the nominal. Using the no-load voltage instead makes the same transformer look better than it is: the example above reads 6.87 % correctly and 6.43 % incorrectly, and nothing on a datasheet tells you which convention a quoted figure used. This page therefore computes both and labels the denominator it used, so the comparison you make between two supplies is a fair one.

Three different quantities share the name, and this calculator implements all three from their own sources. Transformer or generator voltage regulation compares no-load with full-load at constant primary voltage. Load regulation, the power-supply version, compares the output at minimum load current with the output at maximum load current. Line regulation holds the load fixed and sweeps the incoming supply instead, comparing the output at the highest line voltage with the output at the lowest. The arithmetic is deliberately the same shape in all three — that is what the published definitions say — but what you must vary while measuring is completely different, and mixing them up produces a number that describes nothing.

Line regulation gets one extra step, because semiconductor datasheets do not quote it as a plain percentage. They normalise it by the input swing and quote %/V: the LM317's data sheet gives 0.01 %/V typical and 0.04 %/V maximum, meaning the output moves at most four hundredths of a percent for every volt the input moves. Enter how far the input was swept and this page converts your percentage into that form so it can be compared with a datasheet directly.

One benchmark is worth carrying away. A good power transformer on a simple resistive load should come in under 3 % regulation; the worked example at 6.87 % is loose by that standard. And that 3 % is a best case — an inductive load makes regulation worse, so a transformer measured resistively will not repeat its figure driving a motor. The verdict beside the result applies that benchmark for you in transformer mode, and stays quiet in the other two, where the published guidance is simply that zero is ideal.

What is voltage regulation calculator?

Voltage regulation is the percentage change in a supply's output voltage between two named operating conditions, expressed relative to the more heavily loaded of the two. For a transformer, the conditions are no load and full load with the primary voltage held constant, and the number quantifies the internal impedance of the windings: current through the winding resistance and the leakage reactance drops voltage that never reaches the terminals, so the more current the load draws, the less voltage is left. For a linear or switching regulator, the same idea splits in two. Load regulation varies the current drawn while holding the input steady, and measures how well the control loop holds the output. Line regulation varies the input while holding the current steady, and measures how well the loop rejects what is happening upstream. All three are computed the same way — the difference between the two readings, divided by the heavier-loaded reading, times one hundred — and all three are ideally zero. The reason to keep them separate is that a supply can be excellent at one and poor at another: a regulator with superb line regulation may still sag badly when the load steps, because the two figures are set by different parts of the circuit.

How to use this calculator.

  1. Choose which of the three definitions you are measuring. The field labels change to match, because the two readings mean different things in each.
  2. Measure and enter the output voltage at the light condition — no load, minimum load current, or highest line voltage.
  3. Measure and enter the output voltage at the heavy condition — full load, maximum load current, or lowest line voltage. This is the denominator.
  4. For line regulation, enter how far you swept the input. A sweep from 3 V to 40 V is a span of 37 V.
  5. Read the regulation percentage, then check the denominator shown beside it to confirm the calculator used the reading you intended.
  6. Compare with the alternative-base figure before quoting a number against a datasheet, since the two conventions differ by roughly the regulation percentage itself.
  7. In transformer mode, check the verdict against the published 3 % benchmark — and remember it applies to a resistive load.

The formula.

Regulation % = ( V_light − V_heavy ) ⁄ V_heavy × 100 Line regulation (%/V) = Regulation % ⁄ ΔV_input

The arithmetic is one subtraction and one division, and the whole difficulty is in the division. Take the worked example: a transformer secondary measured at 9.990 V with the load disconnected and 9.348 V at what the source calls full load. The spread is 9.990 − 9.348 = 0.642 V. Divide by the full-load voltage, 9.348 V, and multiply by 100, and the regulation is 6.8677792041 %.

Divide by the no-load voltage instead and you get 6.4264264264 %. Same transformer, same measurements, a figure nearly half a percentage point lower — and both are numbers you will find quoted. The published definitions are unambiguous: Kuphaldt's transformer treatment, and Fiore's Equations 8.1 and 8.2 for line and load regulation, all put the smaller, more heavily loaded voltage on the bottom. This page therefore uses that base for the headline figure and reports the other one beside it, labelled, so that a figure copied from a datasheet can be checked for which convention it used.

The three modes differ in what is varied, not in the formula. For transformer regulation, the primary voltage is held constant and the secondary load current is swept from zero to rated. For load regulation, the input is held constant and the output current is swept from its minimum to its maximum. For line regulation, the load current is held constant and the input voltage is swept across its specified range. Vary two things at once and the number describes nothing.

Line regulation gets one further step. Datasheets normalise it by how far the input moved, so that the figure is comparable between parts specified over different ranges. The LM317 sheet gives line regulation as 0.01 %/V typical and 0.04 %/V maximum, with the footnote that it is 'expressed as the percentage change in output voltage per 1V change at the input', measured over 3 V ≤ (V_IN − V_OUT) ≤ 40 V. Enter the span you swept and this page divides your percentage by it. A 12 V output that moved from 12.000 V to 12.050 V across a 37 V input sweep is 0.4166666667 % overall, or 0.011261 %/V — right at that part's typical specification.

Rounding happens once, at the return boundary, and both classification bands read the unrounded percentage. That matters at the 3 % transformer benchmark: a regulation of 2.9996 % displays as 3.00 % but is genuinely under the mark, and the verdict beside it is decided on the unrounded value, so the two can never contradict each other.

What the number does not tell you: why the voltage moved, and what happens between the two points. A transformer's droop is roughly linear in load current, but a switching regulator's is not, and neither is a supply that hits current limit partway. Two readings give you two points; if the behaviour in between matters, sweep it. And measure both readings on the same meter at the same temperature — a 0.1 % meter error is a tenth of a 1 % regulation figure.

A worked example.

Example

The transformer measured in Kuphaldt's Alternating Current text, whose secondary 'spans a range of 9.990 volts at (virtually) no load and 9.348 volts at the point we decided to call full load'. The spread is 0.642 V. Divided by the full-load voltage of 9.348 V, the regulation is 6.8677792041 %. Divided by the no-load voltage instead it would read 6.4264264264 %, which is the same transformer described with the wrong convention — and the gap between those two numbers, 0.44 percentage points, is bigger than the entire regulation budget of a decent supply. The source's own benchmark is that a good power transformer on a resistive load should come in under 3 %, so this one is loose, and the verdict says so. Two caveats travel with that judgement. The benchmark is measured resistively and an inductive load makes regulation worse, so this figure is a best case. And two readings describe two points, not the curve between them: the droop of a transformer is roughly proportional to load current, but nothing in this calculation checks that.

input Voltage Span37
regulation Typetransformer
lower Voltage9.348
upper Voltage9.99

Frequently asked questions.

Do I divide by the no-load voltage or the full-load voltage?
The full-load voltage — the more heavily loaded of the two readings. Kuphaldt's transformer treatment and Fiore's Equations 8.1 and 8.2 for line and load regulation all use that base. It matters more than it looks: the worked example on this page reads 6.87 % correctly and 6.43 % on the wrong base, a difference bigger than the whole regulation budget of a good supply. Because quoted figures rarely state their convention, this page prints both numbers and labels which one it used for the headline.
What is a good voltage regulation percentage?
For a power transformer driving a simple resistive load, under 3 % — that is the benchmark stated in the source this page implements. Below that the secondary is holding up well; above it the windings have enough impedance to matter. Two qualifications. The 3 % figure is a best case, because inductive loads make regulation worse, so a transformer that measures 2 % on a resistor bank will do worse on a motor. And it does not transfer to regulators: a linear regulator is expected to be one or two orders of magnitude better, with the LM317 specified at 0.1 % typical load regulation.
What is the difference between line regulation and load regulation?
What you vary. For load regulation you hold the input steady and sweep the output current from its minimum to its maximum, which tests how well the supply holds up under load. For line regulation you hold the load steady and sweep the input voltage across its specified range, which tests how well the supply rejects what is happening upstream. The arithmetic is identical — Fiore gives both as (V_max − V_min)/V_min × 100 — but they are set by different parts of the circuit, and a supply can be excellent at one and poor at the other. Measure them one at a time; vary both together and the number means nothing.
Why does my datasheet quote line regulation in %/V instead of a plain percentage?
So that parts specified over different input ranges can be compared. A percentage measured over a 5 V sweep and one measured over a 37 V sweep are not the same test. Dividing by the span normalises it: the LM317 datasheet's footnote defines its figure as 'the percentage change in output voltage per 1V change at the input', and quotes 0.01 %/V typical and 0.04 %/V maximum over 3 V to 40 V of headroom. Enter your sweep width in the input-span field and this page converts your percentage into that form.
My answer came out negative. What does that mean?
That the output was higher under load than off load, which is the opposite of what almost any supply does. Two explanations. Usually the two readings have been entered the wrong way round, which the verdict on this page says explicitly rather than just showing a minus sign. Occasionally it is real: a supply with remote sensing that over-compensates for cable drop, or a switching converter with deliberate positive load-line shaping, genuinely produces a small negative regulation figure. Check the wiring before believing it.
Is this the same as voltage drop in a cable?
Related but not the same. Voltage drop is computed from the conductor — its resistance, the current and the run length — and predicts a loss before anything is built. Regulation is measured at the terminals of a source and describes the source's own internal impedance, including everything inside it. In a real installation the two add up: a transformer with 3 % regulation feeding a run with 3 % drop leaves the load with 6 % less than nominal. If it is the cable you are sizing rather than the source you are characterising, a voltage-drop calculation is the tool you want.

References& sources.

  1. [1]Tony R. Kuphaldt, 'Electric Circuits II — Alternating Current' (Lessons in Electric Circuits, Volume II), Workforce LibreTexts, §10.6 'Voltage Regulation'. Primary source for the transformer definition and the quality benchmark. States: 'The measure of how well a power transformer maintains constant secondary voltage over a range of load currents is called the transformer's voltage regulation'; that its worked secondary 'spans a range of 9.990 volts at (virtually) no load and 9.348 volts at the point we decided to call full load'; and that 'a good power transformer should exhibit a regulation percentage of less than 3%' on a simple resistive load, with inductive loads producing worse regulation. NOTE: the section's own arithmetic is rendered as an image which could not be read reliably, so this page does not quote the source's printed percentage — it applies the stated formula to the source's stated voltages and obtains 6.8677792041 %. Retrieved 2026-07-29.
  2. [2]James M. Fiore, 'Operational Amplifiers and Linear Integrated Circuits: Theory and Application', Engineering LibreTexts §8.2 'The Need for Regulation'. Primary source for the two power-supply definitions. Equation 8.1: 'Line Regulation = (V_max − V_min)/V_min × 100', where 'V_max is the load voltage produced at the maximum AC line potential and V_min is the load voltage for the lowest AC line potential', and 'the perfect line regulation figure would be 0%'. Equation 8.2: 'Load Regulation = (V_max − V_min)/V_min × 100', where 'V_max is the largest load voltage produced, and V_min is the minimum load voltage produced. These points usually occur at the minimum and maximum load currents, respectively', and it 'would ideally be 0%'. Independent of the Kuphaldt source and agreeing with it on the denominator. Retrieved 2026-07-29.
  3. [3]Texas Instruments, 'LM317 3-Pin Adjustable Regulator' datasheet, SLVS044Z, September 1997, revised April 2025. Source of the %/V form of line regulation and of the benchmark figures quoted on this page. Features list: 'Line regulation: 0.01%/V (typ)' and 'Load regulation: 0.1% (typ)'. Electrical Characteristics: line regulation 0.01 typical, 0.04 maximum, unit %/V, at TJ = 25 °C over 3 V ≤ (VIN − VOUT) ≤ 40 V; load regulation for VO ≥ 5 V, 0.1 typical and 0.5 maximum, unit %VO, over IO = 10 mA to 1500 mA. Footnote (2) defines the measure: 'line regulation is expressed as the percentage change in output voltage per 1V change at the input'. PDF datasheet; text extracted locally because the fetch tool could not read it. Retrieved 2026-07-29.
  4. [4]James M. Fiore, 'Operational Amplifiers and Linear Integrated Circuits: Theory and Application', Engineering LibreTexts, Chapter 8 'Voltage Regulation' (chapter landing page listing §8.1–8.3). Consulted for the surrounding treatment of linear regulators that gives the two equations their context — in particular that the two figures characterise different parts of a supply and are measured with different quantities held constant. Retrieved 2026-07-29.

In this category

Embed

Quanta Pro

Paid features are coming later.

  • All 977 calculators remain free
  • No billing is enabled
Coming soon