Wave soldering myths, Part 2: peel-off is more critical than ever; care must be taken to attain the best possible setup.

Last month we reviewed some of the rumors surrounding the wave solder process's transition to Pb-free alloys--the ones that weren't true. This month we look at some of the buzz on the subject that is true, or at least partially true, depending on the circumstance. To gain a better understanding of the stories on the street, we consulted with some well-respected experts on the topic: Jim Morris and Richard Szymanowski of Electrovert.

1. Pb-free Solder pots need analysis more frequently than SnPb solder pots. Fact. Sn-rich Pb-free alloys erode copper faster than their SnPb counterparts. This means that copper from PCB pads and through-hole leads will be picked up faster than in SnPb solder pots. As copper content increases in the solder bath, the fluidity of the solder decreases; this can cause more soldering defects.

Furthermore, the RoHS limit for lead is 0.1%; if more lead is present in the bath, the final assembly will not be RoHS-complaint. It is generally suggested that solder pot analysis be performed more often than with Pb-free than with SnPb solder pots. Current guidance is to begin the Pb-free wave solder process with pot analyses approximately every 8,000 PCBs, track the results, and adjust sampling frequency accordingly.

2. Stainless steel solder pot components need to be treated to protect them from Sn-rich corrosion. It depends. Most Sn-rich alloys do attack stainless steel, but all attack them at different rates. SAC305 and 405 are known to be particularly aggressive. The attack rate depends on the alloy constituencies of both the solder melt, the stainless steel itself, and the temperature of the melt. The best way to determine if protection is required is to discuss your process with your wave solder equipment manufacturer.

According to Morris, newer alloys such as those with low silver contents or small amounts of nickel show much less corrosive attacks on unprotected stainless steel.

3. PCB designs with high thermal mass and poor thermal relief that are difficult to solder with SnPb may be nearly impossible to solder with Pb-free alloys. Fact. This is one of those areas where the engineering community anticipated the challenges would worsen, but I don't think any of us knew how much worse until we began soldering and saw it with our own eyes!

Hole fill is one of the biggest challenges. Most SnPb solder pots operate roughly 120[degrees]F above the freezing temperature of solder. Compare that to Pb-free solder pots that operate at about 70[degrees]F above the freezing point, and couple it with alloys that don't wet as readily as SnPb. Not only will the solder wick up the barrel more slowly, an internal ground plane that is not properly connected can stop it dead in its tracks.

Ways to improve hole fill include running longer contact times and deeper immersion depths, both of which can impair peel-off mechanics and create more solder bridges. Given this scenario, it would not be unreasonable to update DfM guidelines to add increased solder bridges to the list of potential defects that poor internal ground connections can cause. In the days of SnPb, we could overcome bridging problems with hot gas debridging knives. Running Pb-free now brings us to another rumor regarding the process:

4. Hot gas debridging knives do not work with Pb-free solder. It depends. To me, this is one of the biggest downsides of going Pb-free with wave soldering. I have always regarded debridging knives as the "great forgivers." Engineers could crank up the wave (to get hole fill on thermally challenging designs, to climb up into the apertures of selective solder pallets or to overcome shadowing on densely populated layouts) and the knife would efficiently blow away all those pesky bridges before the board exited the solder process. We just can't count on that anymore. In some cases, such as when using selective pallets, the knives may be able to provide some relief to solder bridging, but it is not likely in mainstream production. This is because the wider throats on the wave nozzles that provide more contact also create a longer distance between the point of separation and the knife itself. Given that Pb-free alloy runs closer to the freezing point than does SnPb alloy, by the time the knife can act on the bridges, they are either frozen solid or close to it. To get good debridging action, the knife would need to be turned up to temperatures and pressures that could adversely affect the reliability of the final assembly.

We know that Pb-free solders have lower fluidity than SnPb, and that the drainage is worse, so in many cases, we can expect more solder bridges. When we lose the effectiveness of our debridging tool, we have to rely solely on peel-off mechanics to remove the bridges. Because peel-off is now more critical than ever, it is important to take advantage of the best possible setup. According to Szymanowski, the three things to look for when setting up the wave are:

* Smooth flow at the point of peel-off. Look at the peel-off region as a PCB exits the wave. Turbulence should be as minimal as possible. As we rely on surface tension to snap the solder bridges apart, we want it to be as uniform as possible, both in magnitude and direction.

* To get the smoothest flow possible, run the lowest lead clearance and lowest effective pump speed possible. Note that this contradicts some of the options to improve hole fill such as deeper immersion depths, so a tradeoff may be required.

* Just enough backflow so that when no PCB is crossing the wave, a mere trickle flows over the back gate, but when the PCB does enter the wave, the leading edge initiates typical backflow. This will push accumulated dross off the wave surface and provide fresh, oxide-free metal for soldering.

Note that all three tips require watching the wave. Our eyes are probably the most useful diagnostic tool we have in wave soldering.

What is this month's lesson learned? The alloy is the cornerstone of the wave soldering process, and when it changes, many process considerations change with it. Last month we mentioned that the wave-soldering process has been around for 50 years, and until recently, 99% of all wave soldering processes in the world used SnPb. From the primary considerations of hole fill and solder bridges to the ancillary process concerns of maintenance and hardware protection, we need to look at everything the alloy change can affect. Unfortunately, investigating every possible main and side effect is hardly a viable option in today's business environment, so we need to rely on our suppliers to do much of the legwork. Many thanks to Jim Morris for guidance on solder pot corrosion and Richard Szymanowski for tips on optimizing peel-off mechanics.