Printed circuit boards (PCBs) used in most electronic products are basically rigid plates to connect circuitry. But demand for flexible PCBs (or flexible circuits) is rapidly increasing thanks in large part to the burgeoning wearable device market.
Perhaps the largest segment of that market is the healthcare industry where wearable devices will be used to collect all varieties of physiological data for diagnosis and study, as well as personal health use.
Already wearables are available to monitor heart rate, blood pressure, glucose, ECG, muscle movement, and more.
Those wearable devices present a number of difficulties for PCB designers that rigid boards do not. Here are some of those problems and what designers can do to alleviate them.
Need for 3D design
While every PCB is actually three-dimensional, flexible circuits allow the entire assembly to be bent and folded to conform to the package that the product occupies. A typical assembly is shown in Figure 1. The flexible circuitry is folded so that the rigid PCBs fit in the product package, occupying minimal space.
There is a lot more to the design, hence the additional challenges, than just connecting the rigid boards. Bends must be precisely designed so boards line up where they are intended to mount, while not putting stress on the connection points. Up until recently, engineers used “paper doll” models to simulate the PCB assembly. Now, design tools are available that provide 3D modelling of the rigid-flex assembly, allowing quicker design and much greater accuracy.
By definition, wearable products must be small and unobtrusive. In the past, a medical “wearable” such as a Holter monitor included a fairly large external device with a neck strap or belt mount. The new wearables are small and attach directly to the patient with no or few external wires. They collect a variety of data and can even process some analyses.
An unobtrusive device attaching directly to the patient dictates flex circuitry and very dense layouts. In addition, the board shapes are often circular or even more unusual shapes, calling for clever placement and routing. For such small and densely-packed boards, a PCB tool that is optimised for rigid-flex designs makes handling odd shapes much easier.
The point of using flexible circuitry is to be able to shape the final assembly by bending the flex. This presents a number of problems that are not encountered on rigid boards. Bending produces stresses that do not occur with rigid boards. Most PCB tools have tools that allow you to optimise the flex circuitry. To avoid problems with bending forces, here are four tips when designing flex:
Don’t use 90° bends on traces: The corners of traces endure more bending stress than straight paths. To avoid delamination problems over time, use straight paths or if traces must change direction, use curves or piecewise-linear curves rather than anything approaching 90°.
Stagger traces on double-sided flex: When traces are run on top of each other on double-sided flex circuits this causes uneven distribution of the tension. Instead, traces should be staggered. This also improves the flexibility.
Use teardrops to improve strength and yield: The flexibility of the substrate can lead to delamination over time if not controlled. Instead of circular pads, teardrop pads can be used to add additional material, providing greater strength to the pad to prevent delamination. The teardrops also provide greater tolerance for drilling.
Support your pads: The copper on a flexible substrate is more likely to detach than on a rigid board because of the bending. In addition, the adhesion of copper to the substrate is not as good as on an FR4 PCB. Fabricators suggest through-hole plating and anchor stubs for SMT mounting pads. They also suggest reducing coverlay openings as much as possible.
The map of the PCB layers, or stackup, is critical when using rigid-flex techniques. Ideally, your PCB design software has the capability to design your stackup including both the rigid and flexible parts of the assembly.
Figure 2 shows a stackup illustration with both rigid and flexible sections. The number of layers and different materials used add to the complexity of the design. Therefore, it is important to design the stackup carefully.
One of the biggest challenges with rigid-flex designs is qualifying multiple manufacturers. After the design is completed, all aspects of the design must be communicated to the board fabricator so that it will be properly manufactured.
However, the best practice is to choose one or more manufacturers early in the design and collaborate with them to ensure your design matches their fabrication requirements as the design progresses. Collaborating with fabricators is simplified by using standards. In this case, IPC-2223 is the vehicle for communicating with your fabricators.
Once the design is complete, the data package must be assembled to hand-off to be manufactured.
While Gerber is still used for standard PCBs in some companies, when it comes to the complexities of rigid-flex, it is highly recommended by both PCB software tool vendors as well as fabricators that a more intelligent data exchange format be used. The two most popular intelligent formats are ODG++ (version 7 or later) and IPC-2581, both of which clearly specify layer requirements.
The vast majority of PCBs in the world today are basically rigid plates to connect circuitry. The most common basic characteristic of the human body is its flexibility and movability. Mixing the two creates challenges that can only be solved with rigid-flex PCB designs.
Mark Forbes is director of marketing content at design tool firm Altium