Microvolts to Millivolts

Reading Biosignal Amplitudes in µV and mV

Biopotentials recorded from the body span roughly two decades of amplitude, so one paper or device manual will happily quote brain activity in microvolts and heart activity in millivolts on the same page. Converting microvolts to millivolts is what lets you put those numbers on a single scale before you compare them.

Conversion factor: 1 µV = 0.001 mV, so divide microvolts by 1,000 (or multiply by 0.001). A 45 µV alpha burst is 0.045 mV; a 2,500 µV EMG burst is 2.5 mV; the 1,000 µV that defines an ECG calibration pulse is exactly 1 mV.

Why the Unit Changes Between Signals

The brain works in microvolts

Scalp EEG is the summed post-synaptic activity of cortical neurons, attenuated by cerebrospinal fluid, skull and scalp before it reaches the electrode. Background rhythms land around 10–100 µV, so microvolts keep the numbers in a readable 1–100 range.

The heart works in millivolts

The myocardium depolarises as one large, synchronous mass close to the chest electrodes, so a QRS complex typically reaches about 0.5–5 mV — roughly 50 times an EEG rhythm. Writing that as 500–5,000 µV is correct but clumsy, hence millivolts.

Muscle and eye straddle the border

Surface EMG runs from about 50 µV at rest-level activation to several millivolts during strong contraction, and EOG covers roughly 50–3,500 µV. These are exactly the signals whose literature switches units mid-paragraph.

Evoked responses sit below both

An auditory brainstem response wave V is only about 0.1–1 µV — 0.0001–0.001 mV — which is why it is buried in the ongoing EEG and only appears after hundreds or thousands of sweeps are averaged.

Educational reference: the amplitudes below are typical published ranges for teaching and instrumentation work. They are not clinical criteria and nothing here interprets a recording from a real person.

Converting a Figure From a Paper or Device Manual

Type the amplitude you are reading

Enter the microvolt figure exactly as printed — a display sensitivity of 70 µV/cm, a peak-to-peak 3,500 µV EOG deflection, or a 0.5 µV evoked-response peak. A comma or a dot both work as the decimal mark, and spaces are ignored.

Read the millivolt value as you type

Both fields are live, so the mV side updates on every keystroke — no convert button. Results carry up to 8 decimals, which is enough to keep a sub-microvolt evoked potential from collapsing to zero, and very small or very large results switch to scientific notation automatically.

Swap when the manual quotes millivolts

ECG gain, thermocouple-style sensor outputs and amplifier input ranges are usually written in mV. Press the swap button (↔) to run mV → µV instead, or just type into the millivolt field — the direction follows whichever box you use.

Copy the bare number into your log

Each field has its own copy button, and it copies the digits only — no unit, no thousands spaces — so the value drops straight into a spreadsheet column or an analysis script without cleanup. Everything is computed in the browser; nothing you type is sent anywhere.

Need a different rung on the ladder? Both unit menus are searchable and cover all twelve voltage units, from gigavolts down to nanovolts and picovolts — handy when an amplifier datasheet reports input-referred noise in nV rather than µV.

Typical Biosignal Amplitudes Side by Side

Put the common recordings on one scale and the reason for the two units becomes obvious: from a brainstem response to a QRS complex is a factor of several thousand.

Signal	Typical amplitude (µV)	Same value (mV)	Note
Auditory brainstem response (wave V)	0.1 – 1 µV	0.0001 – 0.001 mV	Only visible after averaging
Pattern-reversal VEP (P100)	≈ 5 – 16 µV	0.005 – 0.016 mV	Cortical, larger than brainstem responses
Scalp EEG background	10 – 100 µV	0.01 – 0.1 mV	Always quoted in µV
EOG (eye movement)	50 – 3,500 µV	0.05 – 3.5 mV	Roughly 20 µV per degree of gaze
Surface EMG	50 – 5,000 µV	0.05 – 5 mV	Crosses the µV/mV boundary
ECG QRS complex	≈ 500 – 5,000 µV	0.5 – 5 mV	Always quoted in mV
ECG calibration pulse	1,000 µV	1 mV = 10 mm	Standard gain 10 mm/mV

What the Converter Gives You at This Scale

Either box drives the other

Type the µV figure from a montage sheet or the mV figure from an ECG manual — both fields are editable and update live, and the swap button flips the pair when the source changes.

Sub-microvolt values stay readable

Up to 8 decimals keeps a 0.1 µV brainstem peak from rounding away, and results automatically fall back to scientific notation once they run past that range.

Clean numbers, nothing uploaded

Copy takes the digits alone — ready for a results table — and every conversion happens in your browser, so recording-related figures never leave the device.

Biosignal Amplitude Questions

Why is EEG measured in microvolts but ECG in millivolts?

Because the sources are physically different. The heart depolarises as a single large muscle mass close to the recording electrodes and produces roughly 0.5–5 mV at the body surface. Scalp EEG is the summed activity of cortical neurons, spatially smeared and attenuated by the tissue between cortex and electrode, so it arrives at about 10–100 µV — around fifty times smaller. Each field simply picked the prefix that keeps everyday values between 1 and 100 instead of writing 0.05 mV or 3,000,000 µV.

What does the 1 mV calibration pulse on an ECG strip mean?

It is the amplitude yardstick for the whole trace. At standard settings the recorder injects a known 1 mV (1,000 µV) step and prints it as a rectangular mark 10 mm tall — the 10 mm/mV standard gain — typically 0.2 s wide at the usual 25 mm/s paper speed. Once you see that mark, every 1 mm small square on the vertical axis equals 0.1 mV, i.e. 100 µV. If the gain was halved to 5 mm/mV or doubled to 20 mm/mV, the pulse height changes with it and the millimetre-to-microvolt scale must be recalculated before any amplitude is read.

How big is 50/60 Hz mains interference compared with the signal itself?

Enormously bigger, before the amplifier does its job. Capacitive coupling to the mains puts a common-mode voltage on the body that is measured in volts, not microvolts: the IEC 60601-2-25 common-mode rejection test deliberately drives 10–20 V rms at line frequency into the inputs and requires the resulting artefact to stay within 1 mV (10 mm) referred to the input. That test level is around ten million times an evoked-potential peak. What keeps the recording usable is the differential front end — a common-mode rejection ratio of roughly 80–110 dB at 50/60 Hz, a driven-ground / right-leg-drive circuit, short well-matched leads, and low, similar electrode impedances. A notch filter is the last resort, not the first.

How many microvolts is an evoked potential, and why does it need averaging?

An auditory brainstem response wave V is roughly 0.1–1 µV (0.0001–0.001 mV); a cortical pattern-reversal P100 is larger, commonly around 5–16 µV. Either way the response is smaller than the ongoing EEG it rides on, which is 10–100 µV — the brainstem response can be a hundred times below the background. Averaging works because the response is time-locked to the stimulus while the background is not: averaging N sweeps improves the signal-to-noise ratio by about √N, so 2,000 sweeps buys a factor of roughly 45. That is why brainstem protocols stack thousands of sweeps while a cortical VEP, starting from a bigger response, needs far fewer.

Does electrode–skin impedance change the amplitude I measure?

Far less through attenuation than through noise. A modern biopotential amplifier has an input impedance in the megohm-to-gigohm range, so a 5 kΩ electrode feeding a 10 MΩ input loses about 0.05 % of the signal — invisible next to the biological variability of the measurement. The real damage is elsewhere: high impedance raises thermal noise and turns the lead into a better antenna for mains hum, and unequal impedances across the montage break the symmetry the differential amplifier relies on, so common-mode interference leaks through as a differential artefact. That is why labs prepare skin to reach low impedance — historically a target of about 5 kΩ for wet-gel EEG — and why matching impedance across channels matters as much as the absolute value.