Flexible Circuit Assembly: What Changes On A Flex Board
Assembling components onto a flexible circuit is not the same job as assembling them onto a rigid board, even though the equipment and the solder are the same. The substrate bends, absorbs moisture, changes size during processing and cannot support the placement head on its own. Each of those differences requires a change to the process, and most of the defects that appear on a flexible assembly can be traced to one of them being ignored.
This article explains what changes, why it changes, and what the process has to add to compensate.
The material is the starting point, because everything else follows from it.
The Material Sets The Rules
The base film is normally polyimide, sometimes polyester for low temperature work, and its glass transition is far below that of the epoxy glass used for rigid boards. A lead free reflow profile peaks at a temperature where the film is softening and expanding, so the assembly has to be designed and fixtured for a substrate that is moving while the solder is molten.
The copper is part of the same picture. A flexible circuit that will be bent in service uses rolled annealed foil, which is more tolerant of flexing than the electrodeposited foil of a rigid board, and the thin dielectric means the whole assembly is lighter and more conductive of heat than its rigid equivalent. These are advantages, but they also mean the thermal behaviour of the panel in the oven is different and the profile cannot simply be copied.
<img src="https://www.gopcba.com/wp-content/uploads/2026/09/24-2.jpg" alt="Flexible circuit mounted on a carrier fixture for assembly” />
Pre-Bake And Moisture
The first process addition is a pre-bake. Polyimide absorbs moisture from the air, and when that moisture is heated rapidly it turns to steam inside the laminate. The result is delamination between the layers, which is the most common serious defect on a flexible assembly and is difficult to repair.
The standard practice is a vacuum bake at a moderate temperature for several hours before assembly, followed by assembly within a defined window. The temperature and duration depend on the thickness of the material and on how it has been stored, and the window matters as much as the bake: a circuit that sits on the bench for a shift will have absorbed moisture again before it reaches the printer.
Carrier Fixtures And Handling
A flexible circuit cannot be fed through a placement machine on its own. It is mounted on a carrier fixture, a rigid plate of aluminium or a synthetic stone material, held flat by high temperature tape or magnetic clamps. The fixture does two jobs. It gives the machine a flat surface to place against and prevents the circuit from being pushed out of position by the nozzle, and it provides thermal mass that changes how the panel heats in the oven.
The second job is easy to overlook. The rigid areas of a rigid flex board heat differently from the flexible areas, and the fixture can be designed with different thermal characteristics in different regions to compensate. A carrier that is simply a flat plate, without that consideration, can shift the process window enough to matter on a board where the two regions are close together.
<img src="https://www.gopcba.com/wp-content/uploads/2026/08/High‑Speed_PCB_Manufacturing_Guide.jpg.webp" alt="Underfill dispensed around a device on a flex circuit” />
Scaling And The Breathing Effect
A flexible circuit changes size during its own manufacture. Etching, plating and coverlay lamination all leave residual stress, and the release of that stress as the material relaxes produces a dimensional change that is small but not uniform. On a board with pads for a fine pitch device this is enough to cause misplacement or bridging, and the effect is worse on the flexible areas than on the rigid ones.
The answer is not a single global offset but a local one. Fiducials are placed per region, and the machine calculates the actual offset for each before placing. A supplier that builds the same part repeatedly will measure the shrinkage of each batch and keep a compensation database, and the first article is inspected in three dimensions so that the parameters for the rest of the batch can be corrected. The design of the part determines how much of this is needed, since a layout with local fiducials makes the compensation straightforward.
Stiffeners And Reinforcing The Joint
A solder joint on a flexible substrate is subject to bending every time the product is handled or assembled, and a joint that is repeatedly flexed will eventually crack. The countermeasure is a stiffener, a piece of polyimide or rigid material laminated behind the component area so that the local region does not bend. The stiffener turns the joint area into a stiff island, and where it is placed correctly the solder joint sits at or near the neutral axis of the bend instead of at the surface where the strain is greatest.
The stiffener has to be designed into the part, because adding one afterwards changes the thickness and the assembly stack. The same is true of the connector area, where a stiffener is what allows a zero insertion force connector to be plugged without deforming the substrate.
Underfill And Corner Bonding
For a large device such as a ball grid array, or for a connector, solder alone is not a sufficient mechanical attachment on a substrate that flexes. The process answer is to add material: an underfill applied by capillary action around the perimeter of the device, or a corner bond applied to the four corners. Both transfer mechanical load from the solder to the adhesive, and both require the correct quantity of material in the correct place.
Underfill is an epoxy dispensed after reflow and cured, and its job is to spread the strain that would otherwise concentrate in the outermost joints. Corner bonding is quicker and cheaper and is used where the device is not large enough to need full coverage. The dispensing and bonding materials have to be selected for the application, and the dispense path has to avoid the gold fingers and the test points, which is a layout consideration as much as a process one.
Assembly Order On A Rigid Flex Board
A rigid flex board has three regions: rigid, flexible, and the transition between them. The usual strategy is to populate the rigid areas first, where the components are denser and the process window is wider, and then to populate the flexible areas with the whole board held on a fixture that compensates for warpage. Doing it the other way round risks damaging the flexible area while working on the rigid one.
One rule is worth following on the layout. Large components should not be placed within about two millimetres of the transition between a rigid and a flexible region, because the resin flow and the thickness change during lamination make that area poor for coplanarity. Resistors and capacitors can be placed closer, but an integrated circuit in that band will cause trouble. The protection applied to the finished assembly also has to be chosen for a part that flexes, since a rigid coating will crack where the board bends.
Verification
Verification on a flexible assembly relies on inspection that can see under the components and on measurements of the results that matter. Three dimensional optical inspection finds placement errors, and X-ray confirms the joints that cannot be seen. Beyond the joints, the properties that decide whether the part will survive are the flex life and the adhesion of the stiffeners and the underfill, and those are verified on samples rather than on every piece.
gopcb assembles components onto flexible and rigid flex circuits, with pre-bake, carrier design, local scaling compensation and underfill as standard parts of the process rather than optional extras. The result is an assembly whose process matches the substrate rather than one that assumes a rigid board and compensates afterwards.
FAQ
Why does a flexible circuit need a bake before assembly? Because the polyimide absorbs moisture, and that moisture turns to steam during reflow, which delaminates the layers.
What is a carrier fixture for? To hold the circuit flat and rigid during printing, placement and reflow, and to provide thermal mass that smooths the temperature profile.
Is underfill always needed? No. It is used where a device or connector is large enough that the solder joints would otherwise see significant strain as the substrate flexes.
