Why Foldable Phones Demand More From Flexible PCBs

The biggest story in phones this year is the foldable "three-way race." Huawei and Samsung have been at it for years, and in the early hours of September 10, 2026, Apple finally joined the party with its first foldable iPhone (internally referred to as iPhone Fold/Ultra) — a book-style inward fold, a roughly 7.6–7.8 inch inner display when unfolded, a titanium body, and a marketing pitch built around "no crease, high reliability."

All three companies are tackling the same hard problem in different ways: once a screen can fold, how does the circuitry running through the hinge fold along with it — and keep working reliably for years? While most of the conversation focuses on display quality, hinge feel, and body thickness, few people notice the real bottleneck: a sliver of flexible circuitry, thinner than a human hair, hidden inside the hinge, that has to survive well over a hundred thousand open-close cycles.

The circuit board in a foldable phone isn't like a regular one

A regular phone's motherboard is a "rigid PCB" — a hard, flat board. Once components are soldered on, it never deforms again. Engineers mainly worry about whether it fits and sits securely.

Foldable phones can't work that way. Power and signal lines need to cross the hinge, and that section of circuitry has to fold along with the screen — not just once during assembly, but tens of times a day, adding up to well over a hundred thousand, sometimes two hundred thousand, folds over the phone's lifetime, all without signal loss or broken traces. This kind of circuit that can bend repeatedly without failing is called an FPC (Flexible Printed Circuit). Foldable hinges typically use a "rigid-flex board" — part rigid, to hold the chips, and part flexible, dedicated to following the fold.

With a normal circuit board, engineers mostly ask "does it fit, does it run stably." With the flexible section in a foldable phone, there's one more question that matters just as much: will it still be intact after a hundred thousand folds?

Challenge 1: The materials have to survive repeated stretching

A regular PCB is built on fiberglass — rigid and brittle, and it cracks the moment you try to bend it. Foldable phones use polyimide (PI) film instead: a plastic film that tolerates temperatures up to 400°C and can stretch repeatedly without breaking, typically only a few tens of microns thick — not much thicker than plastic wrap.

Copper foil matters too. The rolled copper foil used in regular flex circuits already has good ductility for dynamic bending, but the hinge zone of a foldable phone needs it custom-made thinner and finer to resist fatigue cracking under repeated flexing. Industry experience suggests that using the wrong copper foil type can lead to trace cracking after as few as 50,000 folds; switching to a more flexible-grade process is what gets a design past the 100,000-fold mark.

Challenge 2: The routing has to be redesigned from scratch

On a normal board, traces just take the shortest path. But in a zone that bends repeatedly, straight-line routing is a bad idea — every point along a straight trace bears the same stretching stress when it bends, and it tends to fail first at the weakest point.

That's why traces in a foldable hinge zone are usually laid out in a wave-like "serpentine" pattern, giving the trace some slack to stretch and absorb stress instead of taking it head-on. It's the same reason a straight piece of string snaps easily under repeated pulling, while the same string coiled into a spring shape can take it over and over.

Vias (the small holes that connect different layers of a circuit board) also have to be kept out of the bend zone — a via is inherently a structural weak point, and stress tends to concentrate there during bending, leading to cracked or delaminated pads.

Challenge 3: You can't just guess the bend radius

The bend radius is basically "how tight the fold is." The smaller the radius, the more stretching stress the material takes on for the same number of folds — and the easier it is to fail.

(Figures compiled from publicly available industry design references; actual values vary by manufacturer.)

This is also why early foldable phones tended to show a noticeable crease down the middle — the smaller the radius, the less visible the crease, but too small a radius shortens the circuit's usable life. Engineers have to balance "looks crease-free" against "the circuit survives." That balance is exactly what makes Apple's "no crease" pitch a genuinely hard engineering problem — it means finding a smarter structural trade-off, not just cranking the radius up or down.

Three companies, three approaches

The three companies' fold formats and marketing emphasis differ, but the underlying problem is the same: how to make a circuit that bends repeatedly keep working — and stay visually clean — over years of use.

Why this is hard to mass-produce cheaply

Making a single flexible circuit that survives a hundred thousand folds isn't the hard part. The hard part is doing it at scale, where every single unit has to hit the same reliability bar without the cost spiraling out of control.

That involves material selection (which grade of PI film and copper foil), routing design (how to lay out serpentine traces that are both space-efficient and fold-resistant), stiffener design (reinforcing the chip areas without compromising flexibility in the bend zone), and finally running tens or even hundreds of thousands of robotic dynamic-bend test cycles to confirm yield before mass production can even start. That's part of why foldable phones are expensive — beyond the screen and hinge, the R&D and testing cost behind this invisible flexible circuit is a real line item too. It's also why the market broadly expects Apple's foldable iPhone to land in the premium pricing tier.

Next time you see a marketing line like "supports 200,000 folds," you'll have a better sense of what's behind that number: a sliver of a circuit board, thin as gossamer, that got there through engineers repeatedly tuning materials, routing, and bend radius.

FAQ