Common Soldering Defects and How the Process Prevents Them

Most soldering defects are not mysterious. They appear in recognisable forms, they have a limited number of causes and the causes are almost always upstream of where the defect becomes visible. The difficulty is not in understanding the mechanism but in refusing to treat the symptom: a bridge that is cleared by hand, a tombstone that is reset with a soldering iron and an open joint that is reinforced with solder all restore the board and leave the process exactly as it was. This is a survey of the common forms and what actually produces them.

Tombstone and the Thermal Imbalance

A tombstone is a small passive component that has been lifted onto one end during reflow, held there by the surface tension of a molten joint on one side while the other end has not yet melted. The immediate cause is always that one termination reached the liquidus before the other, and the question is why.

The reasons are usually thermal or geometric. A pad connected to a large copper area heats more slowly than one connected to a thin trace, so the two ends of the same component see different temperatures. A pad that receives more paste than its neighbour takes longer to melt and exerts a different force. A component placed slightly off centre begins melting on the loaded side rather than symmetrically. The reflow profile that is too fast removes the time the assembly needs to equalise, and a component of very small thermal mass is the first to be affected because it has the least capacity to absorb the difference.

The correction is therefore in the profile or in the footprint rather than in the placement machine, and the check is to compare the thermal path of the two pads before adjusting the machine’s behaviour.

<img src="https://www.gopcba.com/wp-content/uploads/2026/08/1-1.png" alt="soldering defect inspection on an assembled circuit board” />

Bridging and the Volume of Solder

Bridging occurs when the alloy from two adjacent joints joins into one, and it appears most often at fine pitch devices and at connectors where the pins are close. It follows from too much alloy in the gap: an excessive paste deposit, a stencil opening that is too wide, a pad that is wider than the lead or a solder mask web that is too thin to contain the molten material.

The distinction to draw is between a bridge that forms during reflow and one that forms as the board leaves a solder wave. The first is a printing and design issue and is corrected in the stencil or the pad geometry. The second is a process issue and responds to the direction of travel, the wave shape and the separation of the board from the wave, and the two require different investigations even though the defect looks the same.

Clearing a bridge by hand is a repair that carries a real risk. It removes alloy from the joint it belongs to as well as from the bridge, and the resulting joint is smaller than the design intended even when it looks complete.

Insufficient Paste and the Joint That Never Formed

Insufficient paste produces a joint that is present but weak, or in the worst case an open circuit that looks like a joint. It is the most dangerous of the common defects because it is invisible: the component is in the right place, the fillet may be adequate on one side and the electrical test may pass on the day it is performed.

The causes are the mirror image of the bridging causes. A stencil opening that releases poorly, a paste that has thickened on the stencil, a support that leaves the board flexing during printing, an aperture that has partially blocked as the paste dried on its walls. Each produces a deposit that is smaller than intended at a specific position, and the pattern of the shortfall identifies which of them applies.

The measurement that catches this is the paste inspection taken before placement, and the reason it is performed on the deposit rather than on the joint is that the deposit is where a correction is still a parameter change.

Voids, Beading and Discolouration

A void is an empty space inside a joint, most often seen beneath a large thermal pad or inside a barrel. It follows from flux that has not been able to escape while the alloy solidified, and it is produced by too much paste, an insufficient soak or a pad whose opening does not give the vapour a route out. Some voiding is normal and some is a defect, and the acceptance limit belongs in the specification rather than in the inspector’s opinion.

Solder beading appears as small spheres of alloy on the board surface beside a component. It frequently follows a ramp that is too fast, which causes the paste to spatter before the flux has had time to hold it together, and it can also follow a deposit that has been disturbed after printing.

Discolouration and a dull, grainy appearance indicate that the joint did not reach a proper reflow temperature, or that it was disturbed while solidifying. The joint may still be electrically continuous, and it will fail differently from a correct joint when it is thermally cycled.

Preventing Rather Than Repairing

The reason to classify a defect accurately is that the classification determines where the correction is applied. A defect that originates in the stencil, the pad geometry, the paste, the support or the profile cannot be corrected at the reflow oven or at the repair bench, and a process that responds by reworking boards has converted a manufacturing cost into a labour cost without reducing the defect rate.

The preventive measures are unglamorous and they work. The stencil is designed from the pads and the packages rather than reused. The paste deposit is measured after printing and the parameters are adjusted when the measurement moves. The profile is measured on the actual assembly rather than inherited. The support beneath the board is checked rather than assumed. And the defects that do occur are classified by mechanism and fed back to the stage that caused them. Our SMT assembly lines apply those controls, the measurements are held under quality management, the paste and profile work runs through process control, and the electrical confirmation follows under PCBA testing.

Reading a Defect Pattern

The pattern of the defects is more informative than any individual defect, because a systematic cause produces a systematic result. A defect that appears at one position on every board is a design or a stencil issue, since a random process does not select one position repeatedly. A defect that appears at the same position on every second panel is a stencil, a support or a fixture issue, because the panel geometry repeats. A defect that appears across the whole board and grows during a run points to a consumable: the paste changing with time on the stencil, a cleaning interval that is too long or a stencil that is progressively blocking.

A defect that appears randomly across positions and boards points in a different direction. Random appearance suggests material variation or equipment stability rather than geometry, and the investigation moves to the incoming material, the feeder behaviour and the condition of the machine rather than to the stencil and the design.

A defect that changes with the batch rather than with the position suggests that the material has changed, and the traceability record is then what makes the comparison possible. The batch record and the defect record held together are what turn a pattern into a cause rather than a hypothesis.

Why Rework Hides the Problem

Rework is legitimate and necessary; a defect that cannot be reworked is scrap. The problem is not the rework itself but the effect it has on the evidence. When the boards are corrected individually, the number that reaches the customer is adequate and the defect rate that the process is actually producing is never recorded. The process continues unchanged, the rework labour becomes a permanent part of the cost, and the defect rate is invisible until a batch is large enough that the rework capacity cannot absorb it.

Recording the defect, the position and the mechanism before the board is corrected costs very little and preserves the information the process needs. A defect log kept alongside the rework station is one of the cheapest quality instruments available, because it converts an invisible trend into a visible one and gives the investigation a starting point that no amount of inspection after the fact can provide.

FAQ

What causes a tombstone? One termination reaching the melting point before the other, which is a thermal or geometric imbalance rather than a placement error.

Why is insufficient paste more dangerous than a bridge? Because it is invisible and may pass an electrical test while leaving a joint that will fail under thermal cycling.

Can a defect be corrected at the oven? Only if its cause lies there. A stencil, pad or paste problem is fixed at the stage that produced it, not after reflow.