Chip Tombstoning and Paste Print Balance

Tombstoning is the failure where a two terminal chip component stands on one end after reflow instead of lying flat. It is a visible defect, easily detected, and it is caused by an imbalance in the forces acting on the two terminations during melting. Because it is a balance problem, the fix is usually to remove the imbalance rather than to change a single setting. This article covers the causes and the controls.

The Mechanics of Tombstoning

When the paste on both pads melts, the surface tension of the two molten fillets pulls the terminations toward the pads. If the two forces are equal, the component stays flat. If one side melts first, or if one side has more alloy, the force is unbalanced and the component rotates about the termination that is still solid.

The rotation happens quickly, within the time the alloy is liquid, and the component ends up standing on one end. The solder on the lifted side has often formed a ball that did not wet, which makes the defective part easy to identify and also explains why the defect is not always repairable by reheating.

Paste Volume Imbalance

An unequal paste volume on the two pads is the most common cause. The imbalance can come from a partially clogged aperture, a stencil with uneven tension or an aperture design where one pad receives more paste than the other. It can also come from a paste transfer that varies with the position on the board, so the defect appears at one location rather than randomly.

Because the volume difference needed to cause the defect can be small, the print has to be controlled tightly. Measuring the deposit volume on both pads of a component, rather than on the board as a whole, is the way to identify the cause. The measurement approach used for solder paste inspection applies, with the data grouped by component rather than by board.

Pad Geometry and Copper Balance

The pads should be symmetrical and connected to the same amount of copper. If one pad is connected to a plane and the other is not, the two sides heat at different rates and one side melts before the other, which starts the rotation. The asymmetry often comes from a thermal relief that was applied to one pad only, or from a trace that connects to one pad and not the other.

The pad dimensions also affect the forces. A pad that is longer on one side gives the molten fillet more area to pull on, and a pad that is wider holds more paste. The design should use identical pad geometry and identical copper connections for both ends of the component, and the routing should be checked for asymmetry before the board is released.

Placement Offset and Component Fit

A component placed off centre sits with more of its termination over one pad, which changes the force balance even if the paste volumes are equal. The acceptable offset depends on the pad size, and a small component on a small pad has less margin than a large one. The placement accuracy required is therefore a function of the component size rather than a general machine specification.

The component itself can contribute. A termination with a different metallisation on the two ends, a part that is not symmetrical or a part that has been stored so that one end is oxidised will behave differently on the two sides. Incoming inspection rarely checks this, and it is one of the reasons a tombstoning problem can appear suddenly with a new component batch.

Reflow Heating Rate and Atmosphere

The heating rate controls how quickly the alloy melts on both sides. A fast ramp melts the paste on the two pads at slightly different times, which increases the chance of rotation, while a slower approach to the melting point allows both sides to liquefy at nearly the same moment. Increasing the soak time before the peak is often the most effective single change.

The atmosphere also matters. Reflow in nitrogen improves wetting and reduces the surface oxide on the alloy, which reduces the tendency for the fillet on one side to ball up instead of wetting. Where a product has a persistent tombstoning problem on small components, the atmosphere is one of the variables to review, along with the profile described in reflow oven profile verification.

Design and Process Controls

The controls are cumulative. Keep the paste volume equal on both pads by controlling the print, make the pads and their copper connections identical, place the component centrally, and slow the approach to the melting point. On a component that is inherently marginal, a reduction in paste volume can help, because a smaller fillet exerts less force even if it is slightly unbalanced.

The gopcb assembly group reviews the pad and copper design of small chip components during the layout review, because the geometry decides the margin available to the process. Where a design has an unavoidable asymmetry, the profile and the paste volume are adjusted on the first article and the settings are recorded, so that the margin does not have to be rediscovered on every production run.

Acceptance and Its Evidence

A result that cannot be reproduced is not a result, and reproducibility should be demonstrated rather than assumed. The tooling, the material and the profile form one system, and a change to any of them should be assessed against the other two before it is released.

Handling between operations is part of the process, and the damage it causes is often attributed to the operation that preceded it. A change that is not recorded is a change that cannot be explained when the result moves, which is why the record is part of the process.

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

Checks Before Release

Where two operations share a tolerance, the allocation between them should be explicit rather than left to whichever is measured first.

FAQ

Can a tombstoned component be reflowed again to fix it? Usually not. The lifted end often has an oxide that prevented wetting, and a second reflow reproduces the same balance problem. The component should be replaced.

Is tombstoning caused by the placement machine? Only partly. Placement offset contributes, but paste volume and thermal asymmetry are more often the dominant causes.

Why does it happen only on some components? Because the margin depends on the component size and mass. A small, light part rotates with a smaller imbalance than a large one.