To Design For Rigid/Flex Printed Circuit Boards

Combining all of the aspects of a flex circuit with a rigid board that is making full use of HDI techniques is one of the breakthroughs of our time. The stacking connectors for board to board or the typical flex circuits are bypassed. If you’ve ever tried to connect a flex circuit to a stacking connector, you know that’s a bottleneck in the process - blindly positioning the flex connector over the mating connector can be fiddly to the point of destroying the connectors. Now what?

Rigid/Flex projects remind me of Digital/Analog projects. The best of both worlds and also the worst of both. Just for starters, if the team is taking this route, you know that they are serious about holding things together with all possible integration. Both of these technologies are well understood on their own though the rigid camp is larger and better understood at this point.

Flex Circuits On Their Own

Flexible printed circuits (FPCs) require more than a change of materials from their stiffer cousins. Additional tolerance has to be designed into the data. A key reason being the manufacturing process with all of the different types of material stacks. For the most part, a flex will also have a rigid section where the connector is mounted. The stiffened area could also be extended to host the ESD protection, an LED or microphone; we’re flexible.

Figure 1. Image Credit: Hirose - A printed connector with staggered pins. Capturing this footprint manually was no picnic. Thankfully, they are downloadable these days.

A good example of this is a Zero Insertion Force (ZIF) connector that can be printed on the flex. With that approach you have a stiffener at the tail end of the FPC that provides a backstop for the pins. It slides into a mating connector and a lever on that connector can be pressed down to lock the flex tail in place. I prefer these to the stacking connector method whether it’s a pure flex or a rigid-flex. Assembly is easier with ZIF connectors.

An Example of a Rigid-Flex Use Case

Contributing to wearable technology provided us with some interesting outlines. At one point, there were circular islands where we glued on a stiffener in the middle of the flex. Small colonies of components clustered over the stiffener islands while circuitry passed by on all sides. The rigid zones were built into the Augmented Reality helmet where it passed by the ears among other locations. Tour an FPC factory or visit their booth, the applications are widespread. I, for one, can’t seem to get away from them.

Something like eye tracking required a more sophisticated approach. A 12 layer board isn’t a flex no matter what you use for the dielectric material. I’ve only read about the semi-flexible boards. That sounds cool. The thing you get with Rigid Flex is all of those extra layers where you need them.

Then, there’s the flex core jutting out to get busy with a specific collection of nets. The configuration I’ve seen most is a three layer flex inside of an 8 or 10 layer board. It’s not uncommon to have an odd number of layers starting with a single sided flex for ultimate flexibility.

Figure 2. Image Credit: Cadence - Note the radius where the flexible core extends from the rigid section. This along with a bead of epoxy act as stress relief for the flex tail.

The Advantage of Rigid Flex

The polyimide core stack-up opens the door for components on both sides of the rigid area. Most of the time, the form factor is going to be “as small as you can make it”. This isn’t a low-tech solution so you know that the problem you’re solving is going to be complex. Enter the ball grid array (BGA) and all of its little constituents. Smaller BGA’s support the bigger ones, though they may wind up on a different piece of rigid material elsewhere in the rigid-flex board.

Sticking with the A/R theme, one of the use cases was an antenna flex extension where location and orientation of the antenna was part of the overall product outline. The radio chip was on the rigid/flex board so the remote antenna would be part of the flex appendage. Special cases like that might also get their dedicated layer of EMI shielding.

This EMI suppression material has to be soldered to the ground mesh that is selectively filled in for the purpose. A small number of slots need to be cut in the coverlay so as to expose those filled areas since the EMI film goes on last. These things need to be accounted for in an extended number of artwork layers.

Being a Good Neighbor is a Universal Goal

Coexistence is always front of mind, especially early on. Once there is a working solution, you can try removing the safeguards to see if it still meets the design criteria. That’s what we call lean design as the parts count reduces over time. In practice, it’s just as likely in the early goings that the design will want more filters or some other improvements that add to the parts list.

If you’re already familiar with the FPC process, then you know that the transition from the stiffened area to the flex area is one of the pain points. The same holds true for exiting a rigid zone. The polyimide spans the entire rigid zone plus any flex excursions out to their destinations. Those destinations can have an entire new board with the same construction as the primary rigid zone. Otherwise, the usual type of stiffener and connector options are in play.

Figure 3. Image Credit: Cadence - The options are endless with segmenting electronics over rigid-flex designs.

What you don’t get to have is a 4-layer board in one location and a 10-layer in another. It’s all being baked as the same layer cake so the lamination process would be the same for all multilayer rigid sections. It’s the far end of the flex tails where we can deploy the usual assortment of flex geometries. We usually picture a connector but it could be any sort of component mix that can be implemented on a single-sided FPCA.

Routing Controlled Impedance Lines on a Rigid Flex PCB

This scenario comes up quite often. We want to extend some differential pairs from the rigid zone out across a flex zone. We have to assume a 3-layer flex with ground mesh on the outer layers and signals inside the faraday cage. That’s the table stakes for controlled impedance. Having the signals in the center of the flex stack-up reduces stress on the signals as they occupy the centerline of the flex stack. A two layer approach stretches and compresses those traces where the flex has a bend region.

We want to maintain the impedance from inside the rigid zone to outside. The way to do that is to continue the mesh on the outer layers of the polyimide wherever those traces may go. Beyond that controlled impedance routing region, it’s more likely that we would go with a solid ground plane on those outer flex layers. We’re not maintaining flexibility but we are maintaining the reference planes above and below the transmission lines.

If you want to cut down on connectors with their inherent failure modes and assembly challenges, a rigid flex might be the way forward. They take more time to floorplan and get them to conform to the fabricator’s limitations but the results in assembly can be their saving grace. The increased reliability is just icing on the cake. You can be rigid and flexible at the same moment.