Bendable Electronics: Flex, Rigid-Flex and Stretchable Circuits

Bendable electronics are circuits built on a flexible substrate — usually polyimide instead of rigid fiberglass — so the board can fold, curve or flex as part of the product rather than sitting flat inside a box. They exist in four practical forms: rigid boards, flex circuits, rigid-flex hybrids and stretchable printed electronics.

Choosing among them is a mechanical decision as much as an electrical one, and it is made early, because it changes enclosure design, assembly and cost.

Flexible construction is rarely chosen because bending looks impressive. It is chosen because it removes connectors and wire harnesses, fits electronics into a curved or very thin housing, survives repeated motion in a hinge or wearable, and cuts assembly labor by replacing several parts with one folded circuit.

Every one of those is a reliability and cost argument, which is why flex shows up in phones, cameras, hearing aids, medical wearables and automotive interiors long before it shows up in a marketing headline.

Two more distinctions matter in a quote. A flex-to-install circuit is bent once during assembly and then stays put; a dynamic flex circuit bends thousands or millions of times in service. They are different designs — different copper, different bend radius, different stack-up — and confusing the two is the single most common cause of field failures in bendable electronics.

Per square inch, a flex circuit costs more than a rigid board and rigid-flex costs more again. The comparison that matters is not board versus board but assembly versus assembly.

When a flex circuit deletes two connectors, a wire harness, a labor step and a common field failure, it frequently wins on total cost even though the bare board line item went up. Ask your contract manufacturer to quote both architectures at your target volume before deciding — the crossover point is product-specific.

The three technologies solve different problems. A single-sided flex circuit is a cable replacement: it removes connectors, saves space and takes a fixed bend. Rigid-flex fuses rigid board sections to flex layers so an assembly can fold into a housing with no interconnects at all. Stretchable circuits, built on elastomer with serpentine conductors, exist for skin-worn and soft-robotics applications where the substrate has to deform, not just bend.

Cost climbs steeply across that list, and so does the number of vendors capable of building it well.

Phones and foldables, cameras, hearing aids and other tiny consumer devices, medical wearables and disposable sensor patches, automotive displays and interior controls, and industrial equipment where a circuit must follow a curved surface or move with a hinge.

Not inherently. A correctly designed flex circuit often improves reliability by eliminating connectors and solder joints, which are the usual failure points. Flex fails when it is bent tighter than its design allows, flexed dynamically with the wrong copper, or stressed at a stiffener transition.

Bare-board cost is typically several times higher, and the gap narrows with volume. Judge it at the assembly level: rigid-flex removes connectors, cables, assembly labor and test steps, so the finished-device cost difference is much smaller than the board quotes suggest, and sometimes negative.

Yes. Several fabricators offer low-quantity flex and rigid-flex prototypes with lead times measured in one to three weeks. Budget for at least two spins: the first one usually teaches you something about how the circuit actually folds inside the enclosure.

Most flex circuit failures happen at the interface, not in the middle of the bend. Terminations concentrate stress, and assembly processes designed for rigid boards handle thin polyimide badly. Choosing the termination method early changes stack-up, stiffener placement and even the panel layout your fabricator can offer.

Add a stiffener under every connector and every component pad, keep traces out of the transition zone between stiffened and unstiffened areas, and never route a bend across a plated through-hole. These three rules eliminate the majority of field cracking reported on first-generation flex designs.

Two decisions made in the first week of layout determine whether a bendable circuit survives: the substrate and adhesive system, and where the copper sits in the stack.

Polyimide is the default film because it tolerates soldering temperatures and repeated flexure; PET is cheaper and used for single-use printed sensors that never see a reflow oven. Adhesiveless laminates cost more but flex better, because there is no soft adhesive layer to shear at the bend.

Record the failure mode, not just the cycle count. A crack in the middle of a bend means the radius or copper choice is wrong. A crack at a stiffener edge means the transition needs relief. An intermittent open at a connector means the assembly process, not the circuit, is the problem. Those three signatures lead to entirely different fixes, and teams routinely respin the wrong thing.

Ask the fabricator to nest your outline on their standard panel before the mechanical outline is frozen. A five millimetre change to a tail length can improve panel yield by a full row of parts, which at production volume outweighs almost every other saving available in the design.

Flex and rigid-flex are not commodity purchases. A shop that produces excellent rigid boards may subcontract the flex work, which means the engineering support you need during layout is one company removed from the people building your part. Ask directly whether the flex layers are fabricated in house, how many rigid-flex jobs the shop runs a month, and whether they will review your stack-up before you release.

The assembler matters just as much. Thin polyimide needs dedicated carriers through reflow, hand-loading a flex panel warps it, and pick-and-place programs written for rigid boards will not hold position on an unsupported film.

Confirm that the assembler has run flex before and ask to see the carrier design they intend to use. Prototype builds hide these problems because they are hand-assembled with unlimited care; the first production run is where an unqualified process reveals itself as a yield number.

Finally, treat the two vendors as one decision. The fabricator sets the stiffener and coverlay geometry, and the assembler depends on it for fixturing. When they are chosen independently and never speak, the gap between them becomes your rework cost.

Work with LA NPDT: if you are moving from here to execution, start with our CAD engineering services or talk to us about design optimization.