Conventional leaded components and wave-soldered surface mount devices are fixed in position before any soldering starts, by insertion/clinching and glue respectively. For reflow soldering, however, it is only the surface tension of the solder paste that holds components onto their pads. Consequently the devices actually float on molten solder for a short period before the temperature decreases and the solder solidifies, during which time they are totally in the control of surface tension forces. These act throughout the solder volume, whether component lead/termination to pad, between pads, or between leads.
The effect of surface tension forces is particularly dramatically seen in this video clip, which shows the reflow of a BGA. First the solder on the pads melts and wets the pad and the ball surface; then the whole ball becomes molten. The separation between BGA body and the board reduces twice, first during the molten paste stage and again when the ball melts. This ‘double drop’ has often been used as an indicator that the BGA had completed reflowed.
A layout fault related to surface tension is when a second pad or a via hole is in close proximity to a pad, and able to drain solder away from the intended location (Figure 1). This problem of solder drainage is particularly acute with small holes, where capillary action produces a high negative pressure, but can be prevented by the application of solder resist.
Sometimes surface tension effects will lead to ‘satellites’ or ‘beading’ (Figure 2). These are solder balls associated with chip devices, usually caused by solder particles being present underneath the components prior to reflow. The mechanism of solder beading during reflow is shown on this video clip.
Although commonly the result of excessive slump due to an incorrect profile, or of excessive placement pressure, solder beading may also be caused by too much solder on the pads or incorrect solder mask. It is always worthwhile checking that your pad designs comply with your assembler’s recommended practice.
This is the phenomenon where the solder travels up the component lead rather than adding to the volume of the desired joint. It can exist at all levels of severity up to open circuit.
Possible reasons include:
The key factor to remember is that surface tension forces drive solder to those places which are hottest, and that the only way to guarantee a single joint of the right profile is by having both pad and component termination at the same temperature when paste reflow is initiated.
[ back to top ]
Nearly all components float to some extent during reflow. This phenomenon is increased when components have tin-lead coated terminations as the component surface is molten under its protective oxide before this makes contact with the solder mass.
Surface tension attempts to make the free surfaces of a liquid as small as possible to reduce the free surface energy.This is why surface tension will tend to pull every solderable surface to the middle of its pad, as long as all the pads for a given component are of exactly the same size and shape, and provided that there is sufficient solder volume. This feature is very useful since it corrects any minor misalignments incurred during the placement process. This ‘self-alignment’ is effective with chips, MELFs and SO-packages, but with multi-leaded components there is hardly any self-alignment since the devices are heavier and less mobile, the forces on separate leads are averaged, and there is usually a smaller solder volume relative to the size of joint.
However, should there be a lack of symmetry between the pads of a component, the pad with the larger area, the one with more capillarity, will try to pull the wetted surfaces of the component towards itself whilst it is free to move on the molten solder. Surface tension can also pull components onto the wrong pads: it all depends on having the right solder volume to ‘float’ the part and a sufficient imbalance of the surface tension forces on it.
The potential for component movement is affected by:
Given that floating will probably occur, it is important to make sure that the paste print is correctly aligned, that parts are correctly positioned during placement, and that the pad design for each component is symmetric.
Author: Martin Tarr
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 2.0 Licence.
Terms and conditions apply.