Rigid Flex Manufacturing Guide

A rigid flex board combines rigid sections that carry the components with flexible sections that fold to connect them. The result replaces a cable and a pair of connectors with a single continuous circuit, which removes a set of contact interfaces and the failures that come with them, at the cost of a much more demanding fabrication process and a set of design rules that are unlike those of a normal multilayer board.

Construction in Outline

The board is built as one stack in which the flexible layers run through the whole panel and the rigid layers are added on the outside of the areas that must stay stiff. After lamination, the rigid areas are milled away from the flexible sections to a controlled depth, exposing the flex that was always there.

The flexible core is usually a polyimide film with an acrylic or an epoxy based adhesive on both sides, though adhesive free constructions that build the copper directly on the polyimide are also used where the flex must be thin or must bend many times.

The number of flexible layers is limited by the bend requirement. A single layer folds easily, two layers bend to a larger radius, and beyond that the assembly is usually described as a flex with rigidised areas rather than as a board that folds, because the flex cannot be formed into a tight curve. The layer count is therefore settled with the mechanical requirement in front of both the designer and the fabricator.

The Flex to Rigid Transition

The flex to rigid transition is the region where the rigid material stops and the flexible material continues. It is the most sensitive part of the design, and most field failures of a flex assembly originate within a few millimetres of it.

The copper in the transition should be covered by a continuous coverlay rather than left exposed, and the coverlay should extend into the rigid area where it can be bonded and anchored. A coverlay that stops exactly at the transition leaves an edge that lifts under repeated bending.

Plated holes must be kept at a defined distance from the transition, because a hole through the bend region is a stress riser and the plating is brittle. The usual rule is to keep vias and pads outside the bend area by a margin that grows with the number of layers and with the bend radius.

Adhesive Flow and Bonding

Bonding is the step that holds the stack together, and the adhesive has to do two jobs at once. It must flow far enough to fill the cavities between the rigid and flexible layers without voids, and it must not flow so far that it reaches the bend area where it would change the mechanical behaviour.

Adhesive flow is controlled by the layup, by the pressure and by the temperature ramp of the lamination cycle. Too little flow leaves voids and delamination that appear at test or in the field, and too much flow produces a resin rich region at the transition that makes the flex stiff and brittle.

Where the adhesive meets the flex, the surface preparation matters, because a polyimide film does not bond easily to a plain resin. Plasma treatment or a chemical etch before lamination is used to raise the surface energy, and the treatment has to be applied consistently or the bond strength will vary across the panel.

Bend Radius and Bend Area

The bend radius is the property that determines whether a design will survive repeated folding. The usual working figure is that the minimum radius is a multiple of the total thickness of the flex, and the multiple grows as the copper thickness and the layer count increase.

Copper in the bend area should be as thin as the current allows and should run perpendicular to the bend line where possible, because a trace that runs along the bend direction is strained over a long length rather than over a narrow one. Where a trace has to cross the bend, it should do so at a right angle.

The bend area should be free of plating, of solder mask edges and of anything that can crack. The copper should be covered by coverlay and the coverlay should be bonded over its whole area, since an unbonded coverlay behaves as a separate layer and takes the strain in a way that the design did not anticipate.

Manufacturing Steps and Their Risks

The sequence is a lamination of the flexible layers, a lamination of the rigid layers, a drilling and plating cycle that reaches through both, an imaging and etching cycle, the milling of the rigid material at the flex areas and a finish. Each step has to protect the flex that will be exposed at the end.

Drilling through a mixed stack is more difficult than drilling a normal board because the drill meets materials with different hardness. The hole wall quality varies, and the plating that follows must be thick enough for a flexible substrate, which is more compliant than a rigid one.

The removal of the rigid material at the flex area is done by controlled depth milling, and it is the step where a single error destroys the board. The depth has to be set from a measurement of the laminated panel rather than from a nominal figure, and the operator should confirm the first panel before the run continues.

Assembly and Handling

A rigid flex assembly is handled more than a normal board. It should be supported at the rigid areas during assembly, and it should never be folded to a radius smaller than the design minimum, including during packing and shipping.

The reflow profile has to suit the whole assembly, and the flexible material loses moisture that has to be baked out beforehand. A flex that has absorbed water will blister during reflow, and the blister appears at the interface that is hardest to inspect.

Where the assembly is folded in production, a forming fixture should set the radius rather than a hand operation. The fixture makes the fold repeatable and takes the judgment out of the hands of an operator who cannot see the copper inside the stack.

Practical Rules

Keep the bend area free of vias, thick copper and adhesive, and control the transition with a coverlay that is bonded continuously. Design the radius as a multiple of the stack thickness and confirm it with the fabricator before the artwork is released.

Record the layup, the lamination and the forming radius with the build records and the defect history, and use the flex processing rules and the flexible material data when the stack is being defined.

FAQ

Why is the flex to rigid transition so critical? It is where the stiffness changes, so the bending strain concentrates within a few millimetres of it. Most field failures of a rigid flex assembly start there.

What controls adhesive flow? The layup, the lamination pressure and the temperature ramp. Too little flow leaves voids, and too much makes the transition stiff and brittle.

How is the flex exposed at the end? By controlled depth milling of the rigid material after lamination. The depth is set from a measurement of the actual panel, since a single error destroys the board.