Rigid-flex Stackup: It’s a 3D World

Estimated reading time: 3 minutes

Z-zero founder Bill Hargin has been studying stackup design techniques for years. He developed the company’s PCB stackup planning software, and he wrote The Printed Circuit Designer’s Guide to… Stackups: The Design within the Design. 

In this interview, Bill shares his thoughts on designing rigid-flex stackups, the challenges they bring, and what rigid board designers need to know about designing stackups in 3D. “Flexperts” Mark Finstad of Flexible Circuit Technologies and Nick Koop of TTM Technologies also offer insight into the many tradeoffs that rigid-flex designers face. 

You literally “wrote the book” on stackup design techniques. What are some of the unique challenges designers face with rigid-flex stackups? 

As I often say, the mechanical world and the electrical world are at odds with each other. We wouldn’t need to concern ourselves with signal integrity if the physical/mechanical worlds weren’t involved. 

The rigid-flex structure carries a lot of advantages—reliability, dynamic flexure, and the ability to get things done in tight spaces. But there’s the additional burden of needing to manage a mechanical world that has additional dimensions or “degrees of freedom.” A rigid stackup, for example, can easily be viewed in 2D, and that world is relatively easy for most electrical hardware engineers to understand and manage. You have impedance, frequency, and loss and then physical and electrical parameters that drive them. When flex substacks come into the picture, the fact that the flex portion needs to bend turns it into a 3D mechanical engineering problem that takes some time to learn. This is where you really need to lean on a flex fabricator’s expertise. Toward that end, I decided to poll a few “flexperts,” including Mark Finstad from Flexible Circuit Technologies and Nick Koop from TTM. Now I’ll weave their thoughts in with some of my own. 

Mark emphasized that rigid-flex stackups have to blend both rigid and flexible base materials into the final product, and that these materials all stretch and shrink at different rates due to processes like etching and lamination. Additionally, flex materials can stretch and shrink just due to temperature and humidity changes. To further complicate things, these materials typically don’t move at the same rate in the X and Y axes. Most manufacturers have a good history with many materials, so they know how much to scale their artwork and programs to account for these dimensional instabilities. Also, laser direct imaging and smart laser and mechanical drills help with a lot of these issues, but the need to use those processing tools does come at a cost. 

What do rigid board PCB designers need to understand about rigid-flex design and stackups? 

Hargin: I don’t know if this will catch on, but I’ve started referring to rigid-flex design as “designing in 3D” or “3D design.” I doubt that someone who’s been designing rigid-flex boards (in 3D) for 20 years would learn much from my brief commentary here, so I’ll throw out a challenge to the rest of the designers who may benefit from this. 

Along with HDI, rigid-flex PCBs have been trending up in recent years. Younger engineers would do well to start adding rigid-flex design skills to their repertoire in order to open up career opportunities. Maybe start with the rigid-flex chapter in my book on stackups and branch out from there. For signal integrity and EMI control, we like to have nearby reference planes for the current return path. But that additional 

plane layer reduces the bend radius, so there are lots of new tradeoffs to make when you’re designing for flexibility. 

There are some important differences relative to rigid PCBs that you should consider when designing a rigid-flex project. You can connect flex cores using a loose-leaf or bonded approach (Figure 1).

Click here to continue reading this interview which appears in the September 2023 issue of Design007 Magazine.