EMG
Reads voltage between two surface electrodes (or fine-wire intramuscular electrodes), bandpass-filters 20–500 Hz to capture muscle action potentials, and digitises the result for envelope detection or single-unit classification.
Surface electrodes pick up the time-varying sum of motor-unit action potentials. An instrumentation amplifier with CMRR >80 dB rejects mains and movement-induced common-mode noise. A bandpass filter (20–500 Hz) cuts ECG bleed-through (<20 Hz) and motion artefacts. A 16-bit ADC samples at 1–2 kSPS; an envelope detector or RMS computer turns the bursty signal into a smooth proxy for muscle effort.
In plain terms
Like a stethoscope on a swarming beehive — the chip filters out the room hum and amplifies the chorus of individual motor-unit firings in a single muscle.
Why designers use it
- Prosthetic control — myoelectric arms read EMG from residual limb muscles to drive grip and elbow flex.
- Rehab and sports science — visualising muscle activation ratio between agonist and antagonist guides recovery protocols.
- Biofeedback for chronic pain or pelvic-floor therapy — the patient sees muscle activation in real time and learns to modulate it.
- Gesture interfaces — wristbands like the (now-discontinued) Myo and ongoing CTRL-Labs / Meta neural wristband prototypes use forearm EMG to control devices.
Best for
- Prosthetics
- Rehab
- Sports science
Key specifications
- Operating limits: Check the exact manufacturer's datasheet (A family name does not establish voltage, current, temperature or timing limits.)
- Pin assignment: Match the complete part and package code (A similar name or function does not guarantee the same wiring.)
- Mechanical fit: Use the exact package drawing (Check pad layout, dimensions and viewing direction before building.)
When not to use it
- Where individual motor-unit precision is required (clinical neuromuscular diagnosis) — surface EMG is too coarse; needle EMG and a clinical-grade amp are needed.
- Through wet skin or sweat — electrode-skin impedance becomes erratic; pull dry-electrode designs with active shielding.
Common mistakes
- Forgetting to clean and lightly abrade skin under electrodes — high contact impedance turns CMRR into garbage and the trace becomes unreadable mains hum.
- Placing electrodes parallel to (instead of along) the muscle fibre direction — signal amplitude drops 50 % and the SNR collapses.
Where you will find it
- An Ottobock Michelangelo myoelectric prosthetic hand reads EMG from two residual forearm electrodes through a Texas Instruments ADS1294 24-bit ECG/EMG AFE: the chip's high CMRR keeps the signal usable even when the user runs a vacuum cleaner in the same room, and the on-chip processor classifies wrist rotation vs. grip in <30 ms.
- A Delsys Trigno Avanti wireless EMG sensor used in the NIH-funded Knee Osteoarthritis Initiative captures 16-channel EMG from leg muscles during walking trials: each pod uses an Analog Devices AD8232 single-lead biopotential AFE, transmitting 2 kSPS data over 2.4 GHz to the lab-side receiver.
- An EMOTIV EPOC X EEG/EMG headset uses a Texas Instruments ADS1299 8-channel 24-bit AFE to read scalp EEG and facial EMG: facial-EMG channels detect blinks and jaw-clenches that the platform's BCI software uses as discrete commands in research applications.
A short history
Surface EMG uses electrodes on the skin to detect electrical activity associated with muscle contraction. In the tutorial's sensor module, amplification and filtering condition these small signals for a microcontroller. The resulting signal can let an experiment respond to muscle flexing, for example by controlling an LED or motor.