Attempts at defect correction may be futile if surface tension isn’t accounted for.

When a PCB exits a wave soldering unit, it sometimes appears that the process did not proceed quite as planned, as solder bridges may be evident. Believing we had our process in order, we ask, How did this happen?

Inadequate solder drainage conditions are the primary cause of bridging. These conditions stem from several sources: One of the most common occurs when residual oxides from the solder remain at the exit of the solder wave because of lack of flux activity – e.g., because of a wrong composition or distribution.

Assuming the solderability requirements of the parts to be joined are met, the joints will be wetted by solder as they contact the wave. During this contact, all parts in the solder wave are now bridged by solder. At this stage, solder bridging is normal. The amount of solder on a joint during separation from the wave can hardly be influenced by the soldering conditions, as the amount of solder on a joint depends mainly on the joint design in combination with the surface tension of the solder, according to the Young-Laplace equation.


Mono-stable solder joints. A mono-stable solder joint is one that never will create a solder bridge in a stable wavesoldering process. The design in combination with the surrounding joints makes it impossible to create solder bridges between such joints during separation from the solder wave. Hence the label mono-stable; there is only one stable situation. Even relatively large deviations from the optimal process settings will have no effect on such joints. Note that a good design will produce sound joints without unwanted solder bridging; i.e., solder quality starts with design quality.

Bi-stable joints/bridges. There are layouts where the solder might produce sound joints without solder bridging, or create a bridge with the same process settings for apparently no reason. Nothing was changed in the soldering process, but for an “unclear” reason, solder bridges occasionally occur, always in the same area or layout. This might be explained by the fact that some joints, because of their design, demonstrate bi-stable behavior. This phenomenon can be demonstrated by the surface tension behavior of soap models with a cubic metal frame. These also show bi-stable behavior in a full symmetrical model.

Trials to improve bi-stable performance by changing soldering process parameters might prove frustrating if one does not recognize them. One often blames the soldering process because it is difficult to understand that this can happen in a stable process.

When the PCB exits the soldering machine, flux still should be sufficiently active to remove the oxide film continuously forming on the solder. This flux should be transferred from the PCB material to the wetted joints.

As joints are wetted with solder, no flux can be present. For large metal surfaces, the supply of flux is usually insufficient in this stage, leading to soldering defects such as icicles (peaks) and flags.

These defects on large metal surfaces can be avoided; one method is by applying solder resist over these metal surfaces. In that case, however, it is necessary to build heat barriers into the related metal surfaces at the soldering side if plated through-holes are used. Moreover, the apertures (spots) in the solder mask will have to be sufficiently large to permit sufficient joint wetting during soldering. During soldering, the solder mask also acts as a heat barrier, and this may impede the soldering process (hole filling).

Design requirements imposed on a multilayer board are even stricter than for a typical two-sided PWB. It is necessary to build in a heat barrier – e.g., by narrowing the conductor at the terminal spot – particularly in the case of conductor linked between the intermediate layers. This prevents heat from being prematurely extracted from the joint to be formed. Use of larger holes also has a positive effect on hole filling.

Ed.: This will be Gert Schouten’s last column, as he recently retired after years of soldering process engineering. We thank him for his many contributions and wish him well!

Gert Schouten is a senior engineer at Vitronics Soltec (vitronicssoltec.com); gschouten@vsww.com.