Surface EMG Electrode Layout for Flexible Printed Patches

JASPER certifications: ISO 9001, ISO 13485, IATF 16949, and ISO 14001.

A flexible patch can hold two electrodes or sixty-four. Channel count by itself says little about whether either layout will answer the engineering question. For conventional bipolar surface electromyography (sEMG), SENIAM recommends a 20 mm center-to-center inter-electrode distance. High-density research arrays use much tighter pitches for a different purpose. The choice also changes the trace pattern, substrate mechanics, connector, acquisition electronics, and validation burden.

This article helps an OEM turn a muscle and channel concept into controlled artwork. It does not prescribe clinical placement, diagnose a condition, or establish that a patch is safe or effective for a medical use. Those conclusions belong to the legal manufacturer’s finished-device program.

1. Start with the Measurement, Not the Patch Outline

A surface EMG electrode layout is the controlled spatial arrangement of sensing electrodes, reference or ground features, conductors, and interconnects used to record electrical activity at the skin over a target muscle. The design includes more than visible contact discs. Pitch, exposed area, orientation, anatomical datums, adhesive openings, trace paths, tail exit, and pin assignment all affect the implemented montage.

Before drawing EMG electrodes, define the measurement in one sentence. “Record bilateral biceps activity during a controlled elbow-flexion task” is actionable. “Measure muscle signals” is not. The first statement identifies a muscle, side, task, and comparison. It gives the physiologist, electronics engineer, mechanical designer, and printed-component manufacturer the same starting point.

SENIAM publishes location and orientation recommendations for 30 individual muscles. Its guidance is a useful anatomical baseline, but a manufacturer cannot select the final site from a muscle name alone. The customer must still resolve the study or device population, nearby muscles, tendon and innervation-zone boundaries, motion, skin preparation, repeat-placement method, and analysis goal. Mesin and colleagues showed why location matters: sEMG variables can change when the detection system moves across a muscle.

The component boundary matters just as much. A printed patch supplier can build registered conductive features, dielectric openings, adhesive laminations, tails, and connector interfaces to an approved drawing. The supplier cannot prove finished-system signal validity from a continuity check. The acquisition front end, common-mode behavior, sampling, filtering, software, cable management, human factors, and intended-use evidence remain part of system validation.

2. Choose the Layout Family Before Choosing Pitch

Three layout families cover most early discussions: a conventional bipolar pair, a sparse multi-channel arrangement, and a high-density grid. They answer different questions. Selecting an array because it appears more advanced can add channels, connector bulk, routing crossings, and analysis work without improving the required output.

The 20 mm and sub-10-mm values are not competing rules. They come from different measurement architectures. SENIAM’s recommendation addresses conventional bipolar sensors. Yang et al. Reported a high-density design objective in a particular 2023 metal-polymer array study. An OEM should not copy either number until the montage, anatomy, front end, and analysis plan match its scope.

A custom printed electrode array earns its complexity when spatial information changes a decision or algorithm. If a single bipolar channel answers the question, a dense array may be the wrong construction. It is also a poor choice when the target is too small for reliable skin-surface isolation, the required site cannot accept an adhered patch, clinical practice requires repositionable discrete electrodes, or a rigid active sensor must sit directly at the electrode. In those cases, discrete gel electrodes, a textile sleeve, a rigid active module, or another sensing method may fit better.

3. How to Set Surface EMG Electrode Layout Geometry and Pitch

3.1 Define the measurement purpose and anatomical boundary

The layout team should freeze four inputs first: target muscle or muscle group, output variable, movement or posture, and intended population. The output may be activation timing, relative amplitude under a controlled normalization method, spatial distribution, conduction analysis, or a feature set for classification. Each output places different demands on electrode size, pitch, channel count, and repeatability.

A body map should show named anatomical landmarks and avoid zones. These are not decorative placement illustrations. They form the coordinate system for the patch drawing and application instructions. For a bilateral study, the map should also say whether layouts are mirrored or identically oriented. If the patch can be applied upside down or shifted to a neighboring muscle without a clear visual or mechanical cue, the drawing is incomplete.

Good signal: the requirement names the muscle, task, anatomical landmarks, expected patch rotation, nearby interference sources, and body-size range.

Red flag: the RFQ supplies only a skin outline and asks the manufacturer to “optimize the EMG position.”

3.2 Decide whether the system needs a pair or an array

A bipolar pair produces a differential channel from two detecting surfaces. A sparse layout may repeat that architecture over several muscles or positions. A high-density array records many spatially adjacent sites, then forms channels in the acquisition or processing system. These are system architectures, not merely artwork variants.

Channel count should trace to a requirement. For each channel, state what information would be lost if it were removed. That question often exposes redundant sites. It also reveals whether the connector and acquisition hardware have been selected before array artwork begins. A 32-electrode patch with a 16-input front end needs an explicit multiplexing or channel-selection strategy; the printed layout cannot solve that mismatch later.

The CEDE project addresses montage, electrode type and configuration, location, orientation, analysis, and interpretation as linked decisions. The same discipline belongs in an OEM design review. Freeze the channel naming convention early—such as CH01+, CH01−, and REF—and carry it unchanged through the artwork, connector drawing, schematic, firmware, data files, and validation report.

Good signal: every electrode has a channel role, polarity or coordinate, anatomical purpose, and matching acquisition input.

Red flag: electrode count was chosen to fill the patch, while the algorithm and analog front end are still undefined.

3.3 Set exposed area, shape, and center-to-center pitch together

Inter-electrode distance is measured between the centers of conductive areas for a conventional bipolar pair. The exposed contact diameter or dimensions must therefore appear beside the pitch. “20 mm spacing” is ambiguous if the drawing does not say center-to-center, edge-to-edge, or aperture-to-aperture.

Electrode area affects spatial averaging and interface behavior. Pitch affects the sampled volume, correlation between neighboring sites, and sensitivity to spatial changes. Smaller features permit denser sampling but tighten print registration, aperture alignment, connector density, and inspection requirements. Larger features may ease contact and manufacturing while averaging over more tissue. There is no free dimension.

For printed EMG electrodes, define at least:

• exposed electrode shape and X/Y dimensions;
• center coordinate and center-to-center pitch;
• conductor area under the interface material;
• adhesive and coverlay opening dimensions;
• registration tolerance between conductor, dielectric, adhesive, and die cut;
• minimum edge distance to the patch perimeter;
• acceptance method for missing print, contamination, pinholes, and aperture shift.

Good signal: the drawing links pitch and exposed area to a named montage, then gives tolerances the printing and converting processes can inspect.

Red flag: a literature figure was scaled to fit a different body area, changing pitch and electrode area without re-validating the signal model.

3.4 Orient the detecting axis and control the reference strategy

For a conventional pair, SENIAM advises placing the bipolar sensor around the recommended location and aligning it with the relevant muscle-fiber direction. Orientation should be encoded as a drawing datum or keyed patch feature. The phrase “parallel to fibers” is insufficient if the production artwork has no proximal/distal marker or anatomical arrow.

An array needs two axes. Label rows and columns, define the origin, and state which axis follows the expected fiber direction. If the patch stretches, the system team should decide whether coordinates refer to the released part, the applied part, or an image-registered on-body state. This distinction matters when spatial pitch is part of the algorithm.

The reference or ground electrode is a system-level choice. Its site, area, lead path, and connection to the front end can affect common-mode behavior and motion sensitivity. It should not be placed automatically in leftover artwork space. Document whether it is integrated into the patch or applied separately, which tissue and motion conditions it sees, and how the user identifies it.

Good signal: artwork carries a keyed orientation mark, array origin, row/column labels, and explicit reference connection.

Red flag: the patch can rotate, mirror, or swap channel polarity without either the user or software detecting the error.

4. Build the Stack and Routing Around Body Motion

4.1 Treat the skin interface as a controlled layer stack

A wearable EMG patch is a stack of coupled interfaces. The exact materials depend on intended use and process capability, but the functional order should be explicit.

Conceptual stack, skin side to outer side:

BODY / SKIN
│
├─ Electrode–skin interface
│  └─ wet gel, hydrogel, dry conductive surface, or another validated interface
├─ Skin-contact adhesive pattern
│  └─ openings aligned to each exposed electrode; vent and edge geometry as specified
├─ Printed electrode/contact layer
│  └─ conductor or electrochemical surface selected for the acquisition chemistry
├─ Flexible or stretchable substrate
│  └─ PET, polyimide, TPU, elastomeric carrier, or another qualified film
├─ Printed traces and optional crossover / shield features
├─ Dielectric or coverlay
│  └─ exposes electrodes and connector pads while protecting routed conductors
├─ Tail transition, stiffener, and connector
│
ACQUISITION ELECTRONICS

Material names do not prove system performance. Ag/AgCl, silver, carbon, gold, PEDOT:PSS, hydrogels, dry interfaces, PET, polyimide, and TPU each describe part of a construction, not its finished behavior. Ink formulation, cure, thickness, surface treatment, substrate, adhesive opening, sweat exposure, motion, and front-end input all matter.

The 2023 Yang et al. Study illustrates this coupling. Its metal-polymer array used a tannic-acid/PEDOT:PSS/polyvinyl-alcohol electrode and patternable liquid-metal circuitry. Those materials enabled one highly stretchable research construction; they are not a blanket specification for production wearable medical electrode patches.

For a medical device, biological evaluation is also construction-specific. FDA guidance applies ISO 10993-1 through a risk-based assessment that considers material, tissue contact, and duration. A supplier declaration or familiar polymer name cannot replace that assessment. Formulation changes, processing residues, inks beneath openings, adhesives, release liners, packaging, and cleaning or sterilization—if any—belong in the legal manufacturer’s evaluation.

Good signal: the controlled bill of materials identifies every direct and indirect skin-contacting layer, formulation revision, supplier, processing condition, and allowed substitute.

Red flag: “medical-grade adhesive” or “biocompatible TPU” appears without a defined material, contact category, duration, and finished-device rationale.

4.2 Map strain before routing conductors

Skin stretches in more than one direction, while a tail and connector introduce stiff zones. Put those facts on the body map before routing. The useful mechanical drawing identifies low-strain islands, principal stretch directions, crease lines, patch edges prone to lift, and likely cable pull directions.

A flexible film can bend without tolerating much in-plane stretch. Polyimide, PET, TPU, and elastomeric carriers therefore should not be compared by the word flexible alone. The Yang study cites a roughly 3 GPa modulus for a commercial polyimide array and discusses serpentine geometry as a way to improve conformal behavior relative to skin. Frontiers researchers later compared open-mesh, closed-mesh, and island-bridge geometries on gold-coated polyimide, reinforcing that pattern geometry and substrate mechanics must be evaluated together.

A straight conductor crossing a crease concentrates strain. A serpentine can distribute extension, but tighter curves consume area and change line resistance, conductor width, print fidelity, and spacing. An island-bridge layout can protect electrode sites while allowing motion between them, yet the bridge roots become critical fatigue locations. These are coupon questions before they become body-worn failures.

Good signal: the CAD overlay includes strain directions, rigid islands, minimum bend radii, no-bend zones, and the intended applied shape.

Red flag: all traces take the shortest Euclidean path to the connector, regardless of body motion or tail stiffness.

4.3 Route differential channels as a signal path, not decorative lines

The electrode and trace form a high-impedance input path before the signal reaches the amplifier. Keep that path short where the architecture allows. Route the two sides of a differential channel with comparable geometry and environmental exposure. Avoid avoidable asymmetry in length, neighboring conductors, crossover count, and motion. Do not assume equal DC resistance alone means equal motion response.

A routing review should address conductor width and spacing, printed resistance limits, crossover construction, dielectric coverage, shielding or driven-reference strategy, guard concepts where applicable, connector contact resistance, and cleaning/contamination controls. Whether a shield helps depends on the front end and return path. A floating decorative shield can add capacitance without solving interference.

IPC-2223 is relevant when the interconnect is designed and procured as a flexible or rigid-flex printed board. It is not an EMG placement or clinical-performance standard. Screen-printed conductive traces may require process-specific design rules beyond an etched-copper flex standard. The drawing should name the applicable fabrication standard and then state the actual conductor, dielectric, registration, and test limits.

Good signal: a schematic-to-artwork review checks every channel from exposed electrode through connector pin, with test points or test nets defined.

Red flag: a vendor receives only a vector pattern and must infer polarity, shield connection, acceptable trace resistance, and crossover rules.

4.4 Design the connector transition as part of the wearable

The connector is often the thickest and least compliant region. Place it where body movement, clothing, and cable load are manageable. Define the tail exit angle, tail length, conductor orientation, stiffener, bend radius, mating connector, pin numbering, insertion depth, exposed-contact finish, and strain-relief method before releasing electrode positions.

A zero-insertion-force FPC interface, board-to-board connector, crimp, snap, conductive adhesive joint, or permanently attached lead each changes assembly and use. A detachable connector may support electronics reuse but adds mating wear and a rigid transition. An attached cable removes a mating interface but can transmit cable motion into the electrode patch. No connector is neutral.

The common failure chain is mechanical first and electrical second:

Cable tug or repeated body flex
→ strain concentrates at tail/stiffener boundary
→ conductor or printed crossover develops a microcrack
→ channel resistance becomes intermittent during motion
→ differential balance changes and artefact rises
→ algorithm rejects the channel or misclassifies the task

A bench continuity measurement with the part lying flat will miss this chain. Test the transition while flexed, loaded, and connected to the intended mating hardware.

Good signal: the drawing defines a controlled flex zone, no-bend zone, pull direction, connector datum, and dynamic acceptance test.

Red flag: connector selection waits until the electrode artwork is frozen, forcing the tail through the patch’s highest-strain region.

5. Release a Drawing Package and Validation Matrix

5.1 EMG electrode array design drawing checklist

A quote-ready package should let physiology, electronics, mechanical, manufacturing, quality, and regulatory reviewers discuss the same revision. The checklist below is deliberately more detailed than a picture of electrode circles.

Vector artwork or Gerber data should accompany the dimensioned PDF, not replace it. The PDF holds datums, tolerances, notes, revision, and acceptance criteria. The native electrical file holds exact geometry. A channel list should independently map every electrode ID to connector pin and acquisition input; comparing that list with the schematic catches errors that a visual review misses.

5.2 Validate from coupon to complete system

Validation should progress through increasingly representative specimens. A single “prototype test” combines too many failure sources and makes root-cause work slow.

ASTM D3330/D3330M can support controlled peel comparisons for pressure-sensitive constructions, including a 180° method on a specified panel. It does not simulate skin movement, sweat, hair, application technique, wear duration, or removal from a person. If it is used, record the method, panel or substrate, dwell, conditioning, width, rate, angle, and failure mode. Then run a separate representative-use protocol.

For medical electrical equipment, IEC 60601-2-40:2024 addresses basic safety and essential performance of electromyographs and evoked-response equipment. A passive printed component is not independently “IEC 60601-2-40 compliant” merely because it connects to such equipment. Applicability, collateral standards, risk controls, and evidence are decided in the finished-device standards plan.

Component and assembly testing should therefore be specified in layers. Supplier inspection can confirm dimensions, visual quality, continuity, resistance, registration, and agreed mechanical coupon results. The customer system team must confirm signal acquisition, motion artefact, reference behavior, algorithm output, safety, usability, and regulatory acceptance.

Good signal: each test has a specimen revision, setup, condition, sample rationale, acceptance criterion, raw-data format, and named owner.

Red flag: “passes EMG test” appears as one line with no front end, bandwidth, placement, motion, or acceptance definition.

6. A Six-Step Development Process

Step 1 — Freeze the body map and channel requirement

Mark the target muscle, landmarks, expected fiber axis, motion, patch envelope, avoid zones, and cable direction. State the measurement purpose for every channel. Resolve whether the system needs a bipolar pair, selected sites, or dense spatial sampling. Keep the map under revision control.

Step 2 — Build the electrical and mechanical architecture together

Select the electrode interface concept, channel count, reference approach, acquisition connector, substrate class, and rough stack. Overlay the strain map on the electrode and trace plan. Reject layouts that force a stiff transition across a crease or require unsupported channel density.

Step 3 — Use coupons to retire material and geometry risks

Print conductor and dielectric coupons for line resistance, registration, bend/stretch behavior, and interface compatibility. Build tail/connector transition coupons separately. When adhesive comparison is needed, use a controlled method such as ASTM D3330/D3330M for screening, then retain representative on-body evaluation as a different question.

Step 4 — Build an integrated alpha patch

Release controlled artwork and a channel map. Inspect electrode apertures, registration, routing, and connector assembly. Verify every electrode-to-pin path, then exercise the tail and connector while monitoring continuity. Record deviations before changing the design.

Step 5 — Validate with the production-intent acquisition chain

Connect the intended front end, cable or reusable module, firmware, filters, and analysis pipeline. Use a defined protocol for placement, skin preparation, posture, movement, and repeat application. Compare channels and failure modes; do not report only the best trace.

Step 6 — Lock the sample-approval package

Approve the drawing, native artwork, bill of materials, pin map, golden sample, test methods, acceptance limits, packaging, labeling, and change-notification rules together. Electrode patch prototype development is complete only when the approved sample can be traced back to that package.

7. Eight Red Flags That Should Stop Release

• No named muscle or task — the manufacturer cannot infer physiology from a patch outline.
• Pitch copied without scope — 20 mm bipolar guidance and sub-10-mm HD studies answer different questions.
• No anatomical orientation mark — rotation or mirroring can silently change the montage.
• Reference electrode treated as spare area — its location and connection belong to the front-end architecture.
• Skin-contact materials hidden behind generic labels — biological evaluation needs formulation and process control.
• Straight traces cross the peak-strain zone — flat-table continuity will not expose dynamic failure.
• Connector and pin map are still initial — artwork cannot be released while channel order is moving.
• Component inspection is called system validation — continuity, peel, or visual acceptance cannot prove finished-device performance.

8. Project Inputs and the Next Engineering Step

A useful first review does not need final production artwork. It does need enough information to expose the unresolved decisions.

Send these inputs:

1. Target muscle area and anatomical landmarks;
2. Measurement purpose and movement/task;
3. Bipolar, sparse, or HD-array concept;
4. Electrode count, tentative exposed dimensions, and pitch;
5. Patch envelope, body curvature, and stretch directions;
6. Wet/dry interface and adhesive concept;
7. Tail exit, mating connector, pin map, and acquisition input count;
8. Intended contact duration and target markets;
9. Expected prototype and annual volume ranges; and
10. Customer-owned validation criteria.

Release Ownership Matrix

Sample State Matrix

9. Frequently Asked Questions

What is a practical starting point for bipolar surface EMG electrode spacing?

SENIAM recommends 20 mm center-to-center inter-electrode distance for its conventional bipolar sEMG placement method. Treat 20 mm as a documented starting point, not a universal optimum. Confirm the target muscle, electrode area, nearby muscles, orientation, front end, movement, and analysis before freezing artwork.

Should a wearable EMG patch use a bipolar pair or an electrode array?

Use a bipolar pair when one defined muscle channel answers the measurement question. Choose a sparse or high-density array only when spatial information, multi-muscle coverage, channel selection, or the algorithm requires it. More electrodes add routing, connector, acquisition, processing, and validation work.

Does a smaller inter-electrode distance always improve surface EMG?

No. Smaller pitch can increase spatial sampling density, but it also changes sampled volume, channel correlation, feature size, registration tolerance, trace density, and connector requirements. Sub-10-mm values reported in HD-sEMG research are architecture-specific and should not replace conventional placement guidance without system validation.

How should printed EMG electrodes be oriented over a muscle?

For conventional bipolar placement, align the detecting axis with the relevant muscle-fiber direction and use a defined anatomical location, following a muscle-specific source such as SENIAM where applicable. Put the proximal/distal direction and a keyed orientation mark on the controlled patch drawing and application method.

Where should the reference electrode go?

There is no universal leftover-space location. Select the reference or ground site with the acquisition-front-end design, target muscle, movement, cable path, and common-mode strategy. Document whether it is integrated or separate, its contact area, anatomical location, connection, and expected motion exposure.

Which substrate is best for a flexible EMG patch?

No substrate is best for every patch. PET and polyimide suit many flexible circuits but tolerate limited in-plane strain; TPU and elastomeric systems can accommodate more motion with different printing, dimensional, and connector constraints. Choose from the body strain map, conductor system, stack, process, and validation protocol.

What must be validated after the printed patch passes continuity testing?

Validate the connector under load, channel balance, front-end noise and saturation, crosstalk, reference behavior, motion artefact, repeat placement, representative wear/use conditions, software processing, and user application. Medical devices also need the applicable safety, usability, biological, risk-management, and regulatory evidence at finished-device level.

What files are needed to review a custom surface EMG electrode layout?

Provide a dimensioned drawing, native vector or Gerber artwork, anatomical placement map, channel and pin map, schematic or input definition, layer stack, material/interface concept, mechanical strain map, connector specification, intended contact duration, target markets, volumes, and initial acceptance and validation criteria.

Technical References

• Source: FDA Use of ISO 10993-1 Biological Evaluation Guidance. Accessed 2026.
• Source: ISO 10993-1:2025 Biological Evaluation of Medical Devices. Accessed 2026.
• Source: ISO 14971:2019 Medical Device Risk Management. Accessed 2026.
• Source: FDA Design Control Guidance for Medical Device Manufacturers. Accessed 2026.
• Source: SENIAM Surface Electromyography Sensor Placement Recommendations. Accessed 2026.
• Source: ISEK Standards for Reporting EMG Data. Accessed 2026.
• Source: IEC 60601-1 Medical Electrical Equipment General Requirements. Accessed 2026.
• Source: IEC 60601-1-2 Electromagnetic Disturbances Requirements. Accessed 2026.
• Source: ASTM D3330 Pressure-Sensitive Tape Peel Adhesion. Accessed 2026.
• Source: IPC-6013 Qualification and Performance for Flexible Printed Boards. Accessed 2026.
• Source: ISO 10993-23 Irritation Testing for Medical Devices. Accessed 2026.
• Source: FDA Applying Human Factors and Usability Engineering Guidance. Accessed 2026.
• Source: Mesin L, Merletti R, Rainoldi A. “Surface EMG: The issue of electrode location.” Journal of Electromyography and Kinesiology (2009). Accessed 2026.
• Source: Yang S, Cheng J, Shang J, et al. “Stretchable surface electromyography electrode array patch for tendon location and muscle injury prevention.” Nature Communications 14, 6494 (2023). Accessed 2026.
• Source: Consensus for Experimental Design in Electromyography (CEDE) project; ISEK EMG Reporting Standards. Accessed 2026.
• Source: U.S. Food and Drug Administration. “Use of International Standard ISO 10993-1.”. Accessed 2026.
• Source: “A comparative study of flexible electrode design on the performance of flexible wearable electronics.” Frontiers in Nanotechnology (2025). Accessed 2026.
• Source: ASTM International. ASTM D3330/D3330M, “Standard Test Method for Peel Adhesion of Pressure-Sensitive Tape.”. Accessed 2026.
• Source: International Electrotechnical Commission. IEC 60601-2-40:2024, “Particular requirements for the basic safety and essential performance of electromyographs and evoked response equipment.”. Accessed 2026.
• Source: SENIAM. “Placement and fixation of the sensor,” “Inter Electrode Distance,” and “Sensor Locations.”. Accessed 2026.
• Source: “Design, Fabrication and Evaluation of a Stretchable High-Density Electromyography Array.” Sensors 24(6), 1810 (2024). Accessed 2026.