Flex PCBs are a wild card during assembly, but arrays may ease the process.
As you complete your new flex or rigid-flex design, thoughts naturally turn to the next steps. No circuit is complete until the components are placed and it is installed into the next higher-level assembly. Maybe the flex is like a cable and just has connectors on both ends. At the other end of the spectrum, it could be chock-full of BGAs, passive and active SMT components, and maybe even through-hole devices.
In any case, the next step is assembly. Will parts be placed manually one-by-one, or are pick-and-place or other automated processes planned? These considerations may drive how you want the parts delivered. In some cases, parts are fully singulated and shipped. This is the default unless something else is specified. In other cases, parts are arranged in an array and shipped in a pallet or sub-panel.
Generally speaking, shipping as single units is the lowest cost per part. This is because the manufacturer can optimize parts per panel without considering array borders, tooling holes and fiducials. Array layouts usually have fewer total parts per panel than individual part layouts. This is especially true of flex and rigid-flex, which often have appendages that complicate how parts may be laid out in the larger manufacturing panel for best utilization. From a total cost of ownership point of view, however, the array may result in the best total cost of PCB and assembly.
In all cases, I discourage defining the array in terms of size and number of parts. It tends to be a more efficient process if you work with your supplier to determine the specifics of the array. The best scenario is to define the minimum and maximum physical size of the array you can handle in your equipment. You might even limit how many parts per array. Then let the manufacturer lay out the array to balance its design with panel utilization. They can provide you with a CAD file of the array for your assembly operation.
Flex brings a wildcard to the assembly process that rigid boards do not. That is, they are flexible. This means that the circuit will not stay flat during assembly unless it is supported. While this can be managed in a hand solder situation, it is a problem for automated processes like pick-and-place and reflow. This will tend to drive use of arrays. When this is the choice, there are things to keep in mind to be successful.
If the design is rigid-flex, array designs can follow similar rules of thumb as rigid PCBs. But remember the array may not be as stiff or dimensionally stable as a rigid PCB array. This is because there are flex regions where the rigid material has been removed. The flex regions will be naturally flexible, which can lead to some sag in the array. It may also mean that as arrays get larger, there may be more of a variable scale effect, making screening paste more challenging. Consider a fiducial at each part to help mitigate this issue.
Like any array, the more parts per array, the more potential for X-outs or reject parts, so there can be a diminishing return to large multiple arrays. Also, thinner flex and rigid-flex may bow enough to cause automation problems.
Type 1-3 flex designs bring a few other challenges. First, they are very flexible by design; they will need support if delivered in arrays. Often, these flexes have stiffeners in portions of the circuit, especially at connector and component locations.
You can take advantage of the stiffener by creating an array support or a frame. It can be bonded around the perimeter like a frame, or it can fully support the parts but only bond it in the regions where a stiffener is needed. In these cases, we make a stiffener array that matches the part array and is selectively bonded where stiffeners are wanted. The rest remains unbonded for support under the parts. In either case, when the parts are cut out, the excess stiffener array material falls away. This is a popular strategy, and your supplier can help you work through the specifics for your part.
For flex and rigid-flex, array tab placement is also a bigger consideration. Make sure there are tabs for each rigid section to adequately support the part and avoid any twisting or dislocation when applying the pressure of pasting or component placement. If the flex has a long tail that does not require any components, it still needs tabs to keep it from flailing about.
V-scoring is not usually used with flex and rigid-flex, as the flex materials do not snap like FR-4, so it can be difficult to break cleanly.
On rigid-flex, the tabs can be cut and cleaned like rigid boards. With flex, we often see a knife or a side-cutting wire cutter tool to cut through the thin flex material. Flex tabs can also be laser-cut for a clean and accurate edge.
An alternative to arrays is getting parts delivered individually. Then, design fixtures for one or more parts to be placed for assembly. These can be reusable, making them cost-effective in many cases. The advantage is potentially better unit pricing and the knowledge that all parts on the fixture are good; no X-outs.
There is no right or wrong when it comes to single parts or arrays. The key is ensuring you get the advantage you want at a sensible total cost.
Nick Koop is director of flex technology 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 suggestions.
