Flexible printed circuits have emerged as a standalone industry, complete with dedicated purchasing teams.
Perhaps I’m being a bit bold, but I believe that not only have flexible printed circuits and assemblies (FPC/A) matured into their own industry, but they also have enabled a density of electronics that rigid PCBs cannot achieve on their own.
What trends might I see that support this notion? As my colleague Jay Desai has pointed out in past columns, thin is in. And flex assemblies deliver on that promise with high reliability. FPC/A has its own stable and large supply chain in the US, Japan, Taiwan, Korea and China. OEMs have dedicated FPC/A commodity managers. In fact, OEMs now expect FPC/A companies to do both flex and flex assembly – beyond just surface mount technology. Leading FPC/A companies know how to engineer and build in 3D, enabling and supporting the flexible interconnect and packaging solution from OEM concept design to mass production and functional test. Finally, we’ve been doing this for a decade now; this isn’t new. It’s mature, it’s automated, it’s laser-guided, high-technology FPC/A in mass volumes.
OK – time to pinch myself. From an engineer’s perspective, how much of this is visionary, and how much of this is actual, developed materials, process and applications that meet the current and future feature density of smartphones and tablets? Over the next year, I’ll answer those questions using evidence from new materials characteristics, examples of innovative applications of current and new technologies, and results of both characterized process and simulations that highlight the latest interconnect and packaging advantages FPC/A provides.
How did FPC/A get to where it is today? Flex circuit and assembly technology has grown into its own industry largely from its success as a critical component to smartphones and tablets. Portable devices have transformed not only the way the world communicates and interacts with data, but also the way we think of flex.
FPC’s core ability to bend, fold and twist thin interconnect traces with high mechanical reliability enabled unique applications to emerge from simple board-to-board jumpers to high cycle-count dynamic constructions needed to produce thinner, stylish clamshell devices in mass volumes.
As dynamic applications dwindled and the dominant design moved to the iconic candy-bar form-factor – a non-dynamic application with larger, more power-hungry displays – why didn’t the complex-FPC/A market diminish as well? FPC/A manufacturers engineered and produced subassemblies and functional modules that could be placed improbably close to the surface and corners of a device, and connect those modules reliably to the main board, all while increasing available volume for bigger batteries and additional features.
A material difference. Dielectric materials have seen significant, progressive decreases in thickness over the past few years, from 25, to 18, to 12µm for current production volumes, and as low as 6µm for development work. To mass produce such thin materials, key suppliers made significant improvements in properties such as tensile modulus, moisture absorption, coefficient of thermal expansion, and dimensional stability.
Improvements in the thickness and ductility of copper foil, combined with the reduction in dielectric thickness, have enabled flex layers as thin as 32µm, or even 24µm. When compared to the thinnest mass-producible PCBs, the difference is 38µm – roughly double the thickness. Roll-to-roll equipment advancements greatly improved processing these materials and expanded the capability to make finer and finer traces at high yield.
Making new connections. Dimensional stability improvements and multilayer registration have enabled continued advancements in miniaturization and density. Components as small as 01005 (0.41 × 0.20mm) for passives and I/Os as dense as 0.40mm pitch BGA for packages are being assembled on FPC in production. Other assembly technologies such as anisotropic conductive film (ACF) bonding offer small-pitch, connector-less, flex-to-flex, flex-to-board, and flex-to-glass over small-pitch, large-area interconnect solutions. Designers now may place multiple functions onto a single flex, instead of connecting discrete modules to a main board.
As more modularization happened at the wafer package level, more packages with unique heights needed to be incorporated into devices. OEMs needed help designing outside the box. Integrated FPC/A design and early supplier involvement provided flex modules that we built, assembled, pre-bent, tested and sent to contract manufacturers to be crammed into nonlinear, non-rectangular spaces in the device, allowing OEMs to pack an amazing amount of electronics into smaller devices.
If you think about it, FPC/A manufacturers gave OEMs a reliable alternative to embedding components into PCBs, by maximizing the use of space outside the rigid board. This represents the fundamental change in thinking around the value of flex. Flex isn’t just for flexible features anymore.
What’s next? Future electronics devices won’t lose functionality. For high-end devices, even more functionality must be crammed into the same or smaller spaces. For lower-end devices, functionality has to be further integrated in order to reduce cost. Higher component density, better signal integrity at higher frequencies, and thinner overall circuits are needed, and at the same or lower overall device cost.
How might the FPC/A industry respond given the new, 3D way in which we think about flex? How might FPC/A companies use new materials and processes in both flex fabrication and assembly to provide mechanical and electrical performance solutions for the next generation of electronic devices?
I believe low-loss dielectrics and continued improvements in copper-trace structure may answer the higher-density, higher-frequency, and higher signal-integrity challenges. I believe that single- and double-sided assembly on flex with direct chip integration by flip chip or other circuit-level packaging technologies may answer some of the space and cost challenges with wafer-level packaging.
And I believe FPC/A will be used in future product designs more equally with rigid PCBs, enabling the next generation of electronics products.