Wave soldering can easily be converted to a lead-free process. However, assemblers will need to modify existing equipment and tightly control process parameters.

Besides a higher melting temperature, the biggest difference between lead-free solder and tin-lead solder is that lead-free alloys can seriously corrode components of the wave soldering machine. This is no small concern: Failure of a solder pot due to corrosion could severely injure an operator.

In wave soldering equipment manufactured before the advent of lead-free solder, the solder pot, valves, ducts, impellers and other parts in contact with molten metal are usually made of untreated 300-series stainless steel. Unfortunately, in a high-volume operation, lead-free solder can cause these parts to fail in as little as 6 months.

Normally, stainless steel is protected against corrosion by a thin surface layer of chromium oxide. This layer is impervious to most materials, including standard tin-lead solder. "However, lead-free solder, which has a high concentration of tin, will attack and dissolve the natural protective coating on stainless steels," explains Matthew J. O'Keefe, Ph.D., associate professor of metallurgy at the University of Missouri-Rolla. "When this layer is gone, wetting occurs. Once wetted, it's only a matter of time before the underlying material dissolves into the molten solder bath."

For lead-free wave soldering, untreated stainless steel pots can be replaced with pots made of grey cast iron, ceramic-coated stainless steel or nitrided stainless steel. Nozzles, ducts and other parts can be made of titanium, ceramic-coated stainless steel, nitrided stainless steel, or stainless steel or carbon steel coated with Melonite.

The latter is produced by a salt-bath nitriding process. The part goes through a quenching process, a polishing operation and then a second quenching operation. The coating consists of a compound layer and a diffusion layer. The compound layer is made of Fe3N, which is a hard, chemically stable compound that is highly resistant to corrosion. The diffusion layer is Fe4N, which improves the metal's fatigue strength.

"Based on corrosion resistance alone, titanium is by far the best material," says O'Keefe. "However, the cost of a wave solder machine with an all titanium solder unit would be double the cost of a regular unit."

Although lead-free solder quickly wets grey cast iron, graphite flakes within the metal slow the rate of corrosion to a low level. "Cast iron will corrode at a rate of 0.25 millimeter per year when exposed to molten tin at 300 C," says O'Keefe. At that rate, a solder pot with walls 10 to 12 millimeters thick would have a service life of several years.

Steel coated with Melonite or other nitrides will last significantly longer than uncoated steel. "Nitride coatings do not protect stainless steel forever," O'Keefe warns. "They only delay corrosion. From field experience, it's not unreasonable to expect a service life of 3 to 5 years from parts protected with a nitride coating. The key...is to avoid scratching the coating. Once the coating is damaged, corrosion of the substrate will accelerate."

Equipment corrosion can also be controlled by varying the composition of lead-free solder, says Masahiko Ikeda, Ph.D., professor of materials science and engineering at Kansai University (Osaka, Japan). "The most aggressive solder is tin-copper, but the rate of erosion can be slowed by the addition of nickel," he says. "However, both tin-copper-nickel and tin-silver-copper alloys erode 304 stainless steel more rapidly than tin-lead solder."

Phosphorus is sometimes added to tin-lead solder to prevent oxidation and reduce dross during wave soldering. Adding phosphorus to tin-copper-nickel or tin-silver-copper solder alloys will slow their rate of erosion to that of tin-lead solder. However, phosphorus will also increase the ability of lead-free alloys to leech copper from the circuit board.

"Any reduction in the thickness of traces, pads and through-holes can reduce the reliability of the board," warns Ikeda. "In addition, increasing the copper content of the solder bath beyond the specified maximum can increase the incidence of defects, such as bridges and incompletely filled through-holes."

The Process Window Narrows

Besides corrosion, a major concern with lead-free wave soldering is temperature, says Keith Howell, product manager for wave soldering equipment at Speedline Technologies Inc. (Franklin, MA). Because lead-free alloys melt at temperatures 30 to 40 C higher than tin-lead solder, the difference between the preheat temperature and the solder wave temperature can be an issue.

Raising the preheat temperature to narrow this difference must be done carefully. Excessive preheating can unintentionally reflow topside surface mount components, but inadequate preheating won't maximize the activity of the flux.

"Solderability will be affected if the components aren't hot enough to wet in the limited time they are exposed to the wave," adds Howell. "In addition, extreme thermal excursions at the point of contact with the solder wave can shock delicate components, causing permanent damage that may not be evident during testing."

The amount of preheating varies with board density and can be determined by standard thermal profiling methods, says Howell. "The amount of heat provided by the preheater at a particular setting should be monitored to ensure repeatability of the thermal profile," he says. "Infrared heaters are subject to more variability than convection heaters."

Because of the slower wetting time of lead-free alloys, dwell time in the solder bath will increase by 1 to 1.5 seconds compared with a typical tin-lead process. Assemblies also must be cooled more quickly after passing through the wave, to prevent problems with fillet lifting.

Even though solder pots have built-in thermocouples for closed-loop control, Howell advises assemblers to monitor bath temperature periodically with an external instrument. "Periodic checks help maintain calibration of the thermocouple," he says. "More importantly, dross build-up around the thermocouple insulates it, effectively slowing the feedback loop and causing temperatures to swing higher or lower than necessary."

Solder contact time and uniformity can be measured in several ways. The simplest is to run a temperature-resistant glass plate through the wave. "The best results are achieved when the plate is fluxed and preheated like a circuit board to promote good wetting," says Howell.

Ideally, the wave should leave a uniform, rectangular patch of solder on the glass. Contact time can be calculated by dividing the length of the rectangle by the conveyor speed. If the wave solder machine does not provide closed-loop control of the conveyor, a handheld tachometer can be used to measure conveyor speed.