Measuring Flatness and Coplanarity on Flexible Circuits

The flatness of a flexible circuit matters at the interfaces where it has to make contact. A flex tail that plugs into a connector must be flat so that all of its contacts touch their counterparts, a flex that is bonded to a display must lie flat against the glass, and a flex that carries a component must be flat enough that the component can be placed and reflowed. Measuring that flatness is more difficult than measuring it on a rigid board, because the part is thin and can be flattened by the measuring method itself.

Optical three dimensional profilometry is the method that has become standard for the purpose. It measures the height of a surface without touching it, using a technique that is fast enough for production and accurate enough to resolve the few micrometres that matter at a connector interface. Understanding how it works, and where it struggles, is what makes the measurement meaningful.

Why Flatness Matters on a Flex

The contacts of a zero insertion force connector are spring loaded and have a limited compliance, typically a few tens of micrometres. If the flex tail is not flat, the high points touch and the low points do not, and the contact resistance at the low points rises or the connection is intermittent. The problem is particularly acute on a fine pitch connector, where the contact force per pad is small.

At a display bonding interface the requirement is different but equally strict. The bonding process presses the flex against the glass, and a non flat flex produces an uneven bond line with regions of low adhesion. Those regions are where delamination begins, and the failure appears after thermal cycling rather than at the time of bonding. The same applies to a flex that carries a component: a pad that is not coplanar with its neighbours produces a joint with insufficient solder or with a void.

<img src="https://www.gopcba.com/wp-content/uploads/2026/09/172.png" alt="Optical profilometer measuring a flex connector tail” />

How Optical Profilometry Works

The common technique for this application is structured light or fringe projection. A pattern of fringes is projected onto the surface and viewed from an angle, and the distortion of the fringes is converted into a height map. The measurement is fast, covers a field of view of centimetres, and produces a full three dimensional image rather than a profile along one line.

The alternative is confocal or focus variation microscopy, which scans the focal plane through the surface and records the height at which each point is in focus. The lateral resolution is higher than fringe projection and the technique works on steep slopes and on surfaces with varying reflectivity, at the cost of a smaller field of view and a longer measurement time. Where the requirement is a measurement of a few pads at a connector, confocal is often the better choice; where the requirement is a whole tail or a whole panel, fringe projection is faster.

Problems Specific to Flex

The first problem is that a thin flex is not rigid. Clamping it flat introduces the very deformation the measurement is trying to detect, and letting it lie unsupported allows it to curl in a way that is not representative of the installed condition. The fixture has to hold the part in a way that reproduces the product, and it has to be repeatable so that successive measurements can be compared.

The second problem is reflectivity. A flex with a bare copper pad and a glossy coverlay presents a surface with reflections that vary by orders of magnitude, and a single exposure setting will saturate on the copper and be too dark on the coverlay. Multi exposure measurement, in which several images at different exposures are combined into one height map, is the usual solution. Where the surface is translucent, light penetrates the coverlay and reflects from the copper beneath, which produces a height that corresponds to the copper rather than to the top surface. That is a useful measurement in itself, but it must be interpreted correctly.

What to Measure and How to Report It

The measurement should be defined in terms of the feature whose flatness matters. Coplanarity of the contact pads at a connector, bow of the whole tail, twist, and the height of a component pad above a reference are all different measurements and require different handling of the data. Reporting a single number for the whole part is usually less useful than reporting the coplanarity of the pads that have to make contact, with the maximum deviation from the fitted plane as the result.

The reference plane also has to be defined. A plane fitted to the whole tail is not the same as a plane fitted to the contact pads, and the two give different results. The convention should be stated on the drawing and used consistently in production, because a change of convention changes the measured value without changing the part. Where the part is measured in a fixture, the fixture geometry should be recorded with the result so that the number is reproducible.

Using the Measurement in Production

For production, the measurement is most useful as a trend rather than as a limit. Tracking the coplanarity of a sample from each panel shows whether the lamination and the profile process are drifting, and a gradual change is easier to act on than a sudden failure at the limit. The sampling rate should be increased after any change to the material, the stiffener or the profile process.

The measurement is also the right tool for diagnosing a rejection. Where a batch fails a continuity test at a connector, the height map shows whether the pads were coplanar and whether the tail was flat, and that often identifies the cause immediately. Building the measurement into the qualification of a new flex design, alongside the mechanical tests, produces the baseline against which production data can be compared. It is the flexible circuit equivalent of the dimensional checks that a manufacturing tolerance document defines for a rigid board.

Additional Considerations for This Build

Practical attention to flatness measurement pays for itself here, because it decides whether the finished board behaves as the drawing intended. Where the requirement is not stated on the fabrication drawing or in the assembly notes, the shop has to assume a default, and that default is rarely the value the design was simulated with. Stating flatness measurement explicitly, together with the tolerance that applies, removes the assumption and keeps the result predictable from batch to batch.

Process Control and Verification

On a design of this kind, connector tail is the item that decides how the rest of the board is arranged. Running a first article through the same checks as the production panel confirms that the two agree, and that comparison is the cheapest form of process control available at prototype stage. Keeping a sample from the panel turns a dispute into a measurement, because the same coupon can be re-examined by both parties without rebuilding the batch.

Related reading: our fabrication notes, board quality and design release notes cover the same ground.

FAQ

Can a flex be measured while it is loose? It can, but the result depends on how it happens to be lying. A fixture that reproduces the installed condition gives a much more meaningful number.

Why does a glossy coverlay cause problems? Because the reflection saturates the sensor on some areas and is too weak on others. Multiple exposures across the range usually solve it.

What is the tolerance for a connector tail? It comes from the compliance of the connector contacts, typically a few tens of micrometres, and it should be obtained from the connector datasheet rather than assumed.