News

The process itself cannot compensate for poor solderability.

When the soldering process is converted to Pb-free, one may often find solder connections where hole filling is not complete. This was not much of a problem when boards were soldered with common SnPb alloys. It sometimes occurred due to incorrect process settings, such as insufficient flux. Now, with Pb-free soldering, these problems seem to be part of the process as they cannot be reduced even with optimal settings.

The main problem with Pb-free soldering is that the solder solidification temperature is rather high, commonly 217°C for SAC alloys or even 221°C with Sn100C alloy, so the process window has become rather small. In Pb-free soldering, the top side of the joint must at least come to a temperature of 217° or 221°C to achieve good hole-fill, while for SnPb37 this temperature was only 183°C. This is the bottleneck in Pb-free soldering.

These minimum temperatures are necessary, since below these temperatures the solder is no longer liquid. Only liquid solder can fill capillaries, providing that the solderability of the capillary surfaces is good.

Good solderability is a prerequisite for the soldering process, since the process itself can never compensate for poor solderability. When we speak of solderability, it is not only the solderability of the surfaces involved; the thermal solderability must also fit the process. In this area, Pb-free solder makes higher demands.

Can the process compensate for these higher thermal demands? We have three possibilities in terms of process settings that we can manipulate to bring improvement: 1) a higher preheat temperature, 2) a longer dwell time in the solder and 3) a higher solder temperature.

If a higher preheat temperature is available for the flux and the board, this option should be used first. Every degree temperature rise that is put in by preheating is that much less that would have to be provided by the solder wave. The total thermal demand for the soldering process is just the combined input of the preheat stage and the wave; thus, a higher preheat temperature assists the process. (Keep in mind that this solution will, in general, not be especially effective for solder joints with massive heatsink parts on the component side. Such joints also caused problems in SnPb soldering.)

There are two reasons why longer dwell times are frequently not the best solution. The first is that as soon as the solder is "starving" in the hole, it directly improves heat conductivity to the surrounding area and the lead or component to which the lead is connected. This means that the heat will drain rather quickly from the joint to that surrounding area. Once the solder is solidifying one must once again replace the lost melting heat energy, putting it into the joint to get the solder liquid again. This turns out to be nearly impossible.

The second reason why longer dwell times may not help is that the flux may be exhausted due to the longer dwell time, resulting in hampered solder flow, more bridging or even webbing and spikes.

A higher temperature will provide faster heat transfer. This might be a good option, provided that the flux can handle this. With this possible solution, however, flux is often the weak link.

Even with these solutions, one is often faced with a side effect that constitutes a serious drawback. Due to the higher thermal load during the soldering process, the mechanical tension on the copper barrel in the PTH may become so high that the barrel may crack. Quite often, a contributing factor is poor plating quality. Outgassing from the base material will escape through the crack and into the solder joint, creating blowholes or craters.

Although we have some options in the soldering process, there are also serious restrictions that cannot be prevented by the process. We need the cooperation of the flux suppliers to meet these new challenges.

In fact, Pb-free soldering often needs a redesign of the joint in view of the new thermal demands. Methods to measure these demands are introduced and described in Chapter 3.4 of Soldering in Electronics.¹ Although most users are not pleased with this solution, it is the only way to improve solder joint quality.

Keep in mind that tweaking the process can never compensate for poor solderability, whether this comes from unsolderable surfaces, a poor thermal design, or both.

With Pb-free soldering, the design needs review with regard to these aspects, especially in terms of thermal solderability. That responsibility falls on the designer of the board. Remember that 217°C (or 221°C) versus 183°C can make a world of difference. This is something that process engineers are well aware of, and something that they need to make board designers equally aware of.

References

1. R.J. Klein Wassink, Soldering in Electronics, State Mutual Book & Periodical Service Ltd., February 1997.

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