Bendable Circuit Boards: Materials, Structure and Uses

A bendable circuit board is not simply a thin rigid board. It is a different construction, built from materials chosen for fatigue resistance rather than stiffness, and its behaviour is governed by rules that have no equivalent in ordinary printed circuit work. Getting those rules right is what allows a circuit to be folded into a camera module, wrapped around a battery or cycled thousands of times inside a hinge, and getting them wrong produces a product that passes every electrical test and then cracks in the field.

What Makes a Circuit Board Flexible

The substrate is the starting point. Polyimide film is the workhorse material, chosen because it survives soldering temperatures, holds its mechanical properties across a wide temperature range and resists repeated bending far better than any glass-reinforced laminate. Polyester film is cheaper and is used in high-volume, low-temperature products such as simple membrane assemblies, but it cannot be soldered to directly and it degrades at temperatures that polyimide ignores.

The conductor matters just as much as the dielectric. Standard electrodeposited copper is brittle because of the way it is grown, and it fractures after relatively few bending cycles. Rolled and annealed copper has a grain structure that accommodates deformation, and it is the standard choice for any flex circuit that will be bent more than a handful of times. The difference is not subtle; the same geometry can differ by an order of magnitude in flex life depending on which copper is used.

Structure and Layer Build-Up

A single-layer flex circuit is a copper foil bonded to a polyimide film and covered by a protective layer, with the coverlay opening only where the pads are. The coverlay is usually a polyimide film with an adhesive, though adhesive-less constructions bond the coverlay directly to the copper and are preferred where the circuit will be flexed repeatedly, because the adhesive is the weakest mechanical link in the stack.

Multilayer flex circuits bond several single-layer constructions together with adhesive, and they are inherently stiffer and less tolerant of bending than a single layer, so the bending region is normally kept to a single layer while the rigid, populated areas use multiple layers. Where the assembly must also carry components that need mechanical support, stiffeners of FR-4, aluminium or stainless steel are bonded to the back of the flex circuit. That combination is known as rigid flex, and it is what allows a single part to replace a board plus a connector plus a cable harness.

Bend Radius and Copper Orientation

The bend radius is the single most important number in a flex design. As a rule of thumb, a static bend, applied once during assembly and never moving afterwards, should use a radius of at least ten times the total thickness of the circuit. A dynamic bend, where the part will flex repeatedly in service, needs something closer to twenty times the thickness, and the calculation should be based on the thinnest part of the stack rather than the finished thickness at a stiffener.

The direction of the traces relative to the bend line is the second rule. Conductors should run perpendicular to the bend line, so that bending puts them into tension along their length rather than across it. A trace that runs parallel to the bend line and sits near the outer surface of the curve is the most likely place for a crack to start, and a wide trace in that orientation is worse than a narrow one because it cannot distribute the strain. Where a wide conductor must cross a bend, splitting it into several narrow traces is the standard remedy.

Layout Practices That Extend Flex Life

Curves are better than corners. A trace that turns through a right angle creates a stress concentration at the corner, and rounding that corner, or replacing it with a gentle arc, removes it. This is why flex layouts look different from rigid ones, and why a rigid layout copied directly onto a flex substrate is usually a reliability problem.

Vias should be kept out of the bending region entirely, because a plated barrel is rigid and the surrounding dielectric is not. Pads and component footprints belong in the flat areas, and the transition from a populated area to a bending area should be gradual, with the coverlay edge or the stiffener edge kept clear of the bend. Our component tolerance and reliability notes describe how the mechanical stress at those transitions is assessed.

Electrical Advantages Beyond Flexibility

Flex circuits also solve problems that have nothing to do with movement. A three-dimensional assembly can be built from a single flex circuit that folds into place, removing connectors, solder joints and the assembly operations that go with them. Weight falls, because the polyimide and copper stack is thinner and lighter than an equivalent rigid board. Impedance can be controlled in the same way as on a rigid board, and the thin dielectric layer often makes it easier to achieve a stable characteristic impedance.

Thermal performance is one area where flex is genuinely worse. Polyimide is a poor conductor of heat compared with aluminium or the copper areas of a rigid board, so flex is a poor choice for carrying significant power. Where a flex circuit must handle current, the practical answers are wider copper in the flat regions, thermal relief through a stiffener, and keeping the heat-generating components on a separate rigid section of a rigid flex design. Our thermal management article describes how those paths are evaluated.

Where Bendable Boards Are Used

Consumer electronics use them wherever space is scarce: camera modules connect to the main board through a flex tail, display panels carry their driver circuitry on a flex extension, and folding phones rely on a flex circuit that must survive hundreds of thousands of cycles. Wearable devices use flex to route a sensor along a curved surface where a rigid board simply cannot go.

Medical devices take advantage of the thin profile and the ability to guarantee flex life, using flex circuits inside catheters and handheld instruments where the circuit must follow a curved housing. Automotive applications use them behind instrument clusters and inside lighting assemblies, where the flex path absorbs the vibration that would fatigue a soldered connector. Industrial and aerospace products use rigid flex to reduce the number of connectors in a system, because every connector is a potential intermittent failure.

Assembly and Handling

Assembly is similar to rigid board work with a few additions. Panels need to be supported during printing and placement, so flex circuits are normally supplied on a carrier or with stiffeners already bonded in the areas that will be handled. The parts are baked before assembly in the same way as any moisture-sensitive package, and the reflow profile must respect the polyimide and the adhesive system.

Handling after assembly is where damage usually occurs. Flex circuits should be supported while they are bent into position, and any forming operation should be done once, deliberately, at the specified radius rather than by hand. Testing should include a continuity check on the nets that cross the bending region, and for products where the part will move in service, the qualification programme should include a flex cycle test at the expected radius and temperature. Our design release checklist covers the items that should be confirmed before the design is released.

FAQ

How many times can a flexible circuit be bent? A static bend, made once during assembly, can be very tight and permanent. A dynamic bend is limited by the copper and the radius, and the qualification limit is normally set by a flex cycle test rather than by a theoretical figure.

Do flexible circuits need a stiffener? Only where components are placed or where the circuit must be pressed into a connector. The rest of the circuit should be left flexible, and stiffeners should be kept clear of the bending region.

Can flexible circuits carry high current? They can carry moderate current using wider copper in the flat areas, but the polyimide substrate does not spread heat well, so high-power designs normally move the hot components onto a rigid section.