Nick KoopFlat or folded, flex circuit designs provide superior impedance performance across a wide spectrum of signal speeds.

Question: Will impedance change when the flex portion of a rigid-flex is bent into place during assembly?

The short answer is no.

When a circuit is bent, the distance from a single-ended or differential pair to the reference planes remains constant, as does the distance between traces in a pair. There is no compression of the dielectric or movement of signal lines. As a result, the impedance remains constant as well. The performance as modeled in the flat condition will continue when bent.

The same cannot always be said about discrete wiring if the pairs are not within the same insulating sleeve. They can move, causing variation in signal integrity. In fact, it is one of the reasons designers are opting for flex and rigid-flex. Compared to discrete wiring and cables, flex provides several advantages when it comes to impedance.

Flex provides superior signal skew performance relative to wiring. This is because of the precise nature of a print and etched circuit pattern. Cutting, stripping and crimping or soldering of wires to connectors is inherently less precise than a flex circuit pattern. Twisted pair wire has inherent skew, which compounds the problem.
Testing has shown flex designs reduce skew by a factor of 10 compared with traditional wiring techniques.

Flex materials have great dielectric properties, with dielectric constant (Dk) values in the 3.3 to 3.6 range. The lower dielectric constant means the target impedance can be hit with wider traces on flex than with rigid laminate of the same thickness. Flex can also work in the GHz ranges. At high data rates, the need for low loss may require substitution of low-loss flexible coverlayer and bond-ply materials. A large number of high-speed flex designs are used in a range of demanding applications today. Your manufacturing partner can assist with material selections to make this work.

Eliminating connectors to join circuits brings additional advantages. First, there is the amount of space saved. Connectors take up physical space in the end-product. In many applications, there is not enough room for connectors. Rigid-flex can eliminate the need for connectors and allow a designer to create the best form factor possible.

Second, connector I/O consumes real estate on the circuit itself. By making direct connections, that space is not needed, and can be used for additional functionality or eliminated to reduce overall size.

Third, elimination of connectors and associated wiring can help meet overall weight targets. Flex solutions eliminate redundant insulation and shielding, as well as excess copper weight by precise sizing of signal lines.

Fourth, signal integrity is improved by eliminating reflections and discontinuities created by solder joints and the connectors themselves.The signal can be routed directly from the originating component right to the receiving component with impedance control the entire way.

Fifth, flex solutions are more reliable than a cable and connector solution. It is well-known that product failures are typically found at solder joints and connection points. By eliminating these failure mechanisms, product reliability can be substantially improved. In one example, a redesign from a rigid printed board with wire harnesses and connectors to a rigid-flex with installed connectors improved the mean time between failures (MTBF) five times to over 2 million hr.

In some cases where high impedance is needed along with maximum flexibility, cross-hatch shielding can be employed for reference planes. In these situations, a mesh pattern is created in the planes that runs at a 45° rotation relative to the prevailing direction of the signal traces. This method can create increases of 20% or more in impedance values, depending on the density of the mesh pattern. Dense mesh patterns provide the greatest shielding function, but result in limited gains in impedance. Sparse mesh patterns are not as effective in blocking certain frequencies of noise, but provide large increases in the impedance values. As a result, 100Ω impedance values can be achieved with only 2 mil dielectric separation, making for very flexible parts.

One challenge in rigid-flex when coupled with impedance needs is the limited availability of low-loss prepreg available in a no-flow form. Most low-loss rigid material is not available in a no-flow prepreg. No-flow prepreg is needed by the rigid-flex manufacturer to bond rigid and flexible layers without bonding them in the flex region, making the flexible portion of the part stiff.

In many cases, this can be worked around. PCB manufacturers suggest how to use higher performance materials in the high-impedance layers, while permitting standard materials where flex and rigid layers meet. Also, in many cases there is not much concern, as the impedance signal is traveling down a through-hole to an impedance-controlled flexible layer. In this case, the dielectric performance of the rigid laminate and prepreg are of less importance.

When designing for impedance, often the conductor width in the flex region will be wider than in the rigid region of a part for the same impedance. Reason: The Dk for flex materials is much lower than for rigid material. In these cases, we recommend the transition to the narrower conductor take place approximately 0.050" to 0.100" into the rigid section to avoid a potential mechanical stress riser in the flex portion of the part.

Another consideration is the number of reference planes in the rigid and flex regions. Often, designs have top and bottom planes in the rigid section, creating a stripline construction, but then transition to a single reference plane microstrip construction in a two-layer flex zone. There is nothing wrong with that, but the designer should be aware of the performance differences and account for them, as needed. Some cases call for use of an additional flex layer to have a stripline in the flex zone as well.

Nick Koop is senior field applications engineer at TTM Technologies (ttm.com), vice chairman of the IPC Flexible Circuits Committee and co-chair of the IPC-6013 Qualification and Performance Specification for Flexible Printed Boards Subcommittee; This email address is being protected from spambots. You need JavaScript enabled to view it.. He and co-Flexpert Mark Finstad (This email address is being protected from spambots. You need JavaScript enabled to view it.) welcome your comments.

Submit to FacebookSubmit to Google PlusSubmit to TwitterSubmit to LinkedInPrint Article