Rigid Flex PCB Development: Where the Two Boards Meet

A rigid flex board is not a flexible circuit with a stiffener glued on. It is a stack in which rigid and flexible layers are laminated together in one construction, so that a single part can carry connectors and processors in one area and fold into a housing in another.

Why the Construction Exists

Small products rarely have room for a flat board and a separate cable. Folding the interconnect into the board removes a connector pair, a cable and the labour to fit them, and it removes the two joints that a cable would have introduced.

The alternative, a flat rigid board with a flexible tail, solves the same problem for a simple case. A true rigid flex extends the construction further, because the rigid areas can be on both sides of the bend and can carry several layers each.

How the Stack Is Built

The flexible layers run continuously through the whole board, and the rigid layers are added on the outside of the bend area. Where the board must bend, the stack is reduced to the flexible core plus its cover layers, and the rigid material is removed.

The transition between the two regions is called the bend area boundary. Every design rule that matters is applied there, because the stiff material ends abruptly and the flexible material must carry the mechanical and electrical load by itself.

Where the Bend Belongs

The bend should be in the flexible section and nowhere else. Placing a bend within a few millimetres of the rigid boundary concentrates the stress at the point where the copper is least supported, and that is where a crack begins.

The bend radius should be generous relative to the thickness of the flexible stack, and the copper in the bend area should run perpendicular to the fold line rather than along it, so that the conductors are not asked to stretch.

Copper in the Flex Area

Rolled annealed copper is used in the flexible layers because it withstands repeated flexing far better than the electrodeposited foil used on rigid boards. The difference is in the grain structure, and it is not interchangeable.

Conductors in the bend area are usually wider than the minimum and covered on both sides by an adhesive layer and a cover film. Solid copper planes are avoided in the bend, because a plane has no room to move and will crack at the first fold.

Stiffeners and Their Purpose

A stiffener is a piece of rigid material bonded to a flexible area to support a connector, a component or a test point. It does not turn the area into a rigid layer; it simply prevents the flex from deflecting where a part is attached.

The stiffener has to be placed so that it does not end at the same point as the bend, and its thickness has to be consistent with the assembly. A connector mounted on an unsupported flex will fail at the joint rather than at the cable.

Adhesives and Cover Layers

The flexible core is covered by a film that insulates and protects the conductors. It may be a separate cover layer bonded with adhesive, or a photoimageable coverlay, and the two have different resolution and different mechanical behaviour.

The adhesive flows during lamination, so the finished thickness of the flexible section is not simply the sum of the nominal layers. The thickness that the fabricator achieves is what the bend radius calculation has to use.

Design Rules at the Boundary

At the transition, the copper should be routed with no change of layer and no via within the bend region. Vias are hard points that cannot flex, and a via placed in a bend will crack the plating or delaminate the stack.

Where a layer change is unavoidable, it belongs in the rigid section, well away from the moving area. The same principle applies to large copper features, which should be relieved or cross hatched where the stack must flex.

Manufacturing Considerations

Rigid flex is more difficult to build than either a rigid or a flexible board alone. The materials are laminated in several press cycles, the registration between the flexible and rigid layers must be controlled, and the panel has to survive processes designed for a flat board.

The laser cutting or routing that frees the flexible area is a separate operation with its own tolerance. The design should allow for the edge quality that the process produces and keep conductors away from the cut line by a margin the fabricator specifies.

Electrical Behaviour

A rigid flex board is a transmission path across several different constructions, so impedance changes between the rigid and flexible areas. Where a high speed signal crosses the boundary, the geometry has to be adjusted to keep the impedance continuous.

The flexible section also has a different dielectric constant and a thinner stack, both of which change the impedance, and the cover layer adds a material that the rigid part of the design does not include. Simulation of the flexible section is worth the effort where the data rate is high.

Testing and Reliability

The bend area is tested mechanically as well as electrically. A sample is folded through the specified radius for a defined number of cycles, and the resistance of the conductors is monitored for the increase that indicates a crack forming.

Thermal cycling is applied to the whole assembly, because the rigid and flexible materials expand differently and the interface between them is where a delamination appears. The coupon used for these tests should include the same boundary geometry as the product.

Choosing Between Flexible and Rigid Flex

A flexible circuit with a stiffener is the simpler and cheaper answer where the board only needs one bend and the component count is low. Rigid flex is justified when several rigid areas have to be connected, when the layer count is high, or when the product cannot tolerate connectors.

The decision is usually made on space and reliability rather than on cost, because the rigid flex route is almost always the more expensive one to build. The flexible multilayer construction is the intermediate case, and the layer count is described under the general stackup guidance.

Process Control and Verification

On a design of this kind, bend is the item that decides how the rest of the board is arranged. Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed.

FAQ

Can a rigid flex board be folded after assembly? Yes, if the bend radius and the number of cycles are within the design. The fold should be made once during assembly rather than repeatedly in service.

Does the flexible area need its own impedance control? Where a high speed signal crosses it, yes. The geometry differs from the rigid section and has to be adjusted to hold the same impedance.

Where should a stiffener be placed? Under the component or connector that needs support, kept clear of the bend area, and dimensioned so that the outline of the rigid section does not define the board edge at the fold.