Repent, for you too are guilty of process mistakes.

“I’m very well acquainted with the seven deadly sins. I keep a busy schedule trying to fit them in.” – Warren Zevon

We have all heard of the seven deadly sins that will, supposedly, lead one to H-ll. There are also the deadly sins of SMT – but there are more than just seven – and they can make your assembly process a “hell on earth.”

During the course of our assembly process audits and troubleshooting work, we tend to see trends in the types of problems. In other words, a lot of people are making the same mistakes. The resulting process problems wreak havoc, with the impact on assembly yields ranging from 5 to 20%. In addition to this direct cost, there is also financial loss from time spent reworking and repairing, the one corrective action by QC, Engineering and Management, and, of course, “do-over.”

Suffice it to say, in my 30 years in this industry, I have yet to see the perfect assembly process. We’ve seen some good ones and some sloppy ones. Six Sigma is, generally, a good thing. However, we’ve seen it (mis)applied with parameters so loose as to render it useless. ISO 9000 is not an indicator of quality. A company can be ISO 9002-certified and still build crap – albeit well documented crap. In the meantime, we continue to do audits and point out the wayward ways, and we’re even presenting (preaching) workshops on the Deadly Sins of SMT.

RoHS, in all its confusing varieties – EU, China, Korea, California and others yet to come – has spawned problems and errors experienced by many practitioners. Let’s take a look at the most common ones: the “Four Horsemen of the Pb-free Apocalypse.”

Uncontrolled mixed assemblies. This is likely one of the prevalent problems in Pb-free manufacturing. Interestingly, it is also a problem for those using SnPb processes. While this problem makes itself known during reflow, it can occur in a number of places.

Even the most meticulous BoM scrubbing and supposedly bulletproof relationships with component distributors are not guarantees of a resulting Pb-free product and, hence, RoHS compliance. At the component level, we’ve seen many occurrences of components identified by the manufacturer or distributor as Pb-free (on packaging or electronically by date-code) that are not, and vice-versa. From an RoHS compliancy perspective, SnPb-finished parts slipping into the process results in noncompliance. From a process standpoint, this becomes a “conspicuous” problem, predominantly with area arrays configured with SAC, SnCu or other Pb-free balls reflowed in a non-RoHS product with a SnPb profile. If you can’t trust your distributor, who can you trust? If you want to be absolutely certain, you have to audit your incoming components. The best Pb-free processes use methodologies like XRF to screen components.

Contaminating an otherwise Pb-free process is something you don’t necessarily have to leave up to your component supplier. If you think the stories of mixed materials – the wrong solder paste alloy being used or SnPb solder bars accidentally being put into the solder pots of a Pb-free wave-solder machine – are urban legends, welcome to the Big City. Again, careful and meticulous logistics combined with monitoring the process (read: XRF) will prevent noncompliant assemblies from slipping into RoHS-land.

Stencil design. While some Pb-free alloy solder paste formulations demonstrate different rheological characteristics at ambient temperatures during printing, the real differences become apparent during reflow. Pb-free alloys particularly behave differently; they tend to wet less. This means, for a given volume of printed paste, as determined by the stencil aperture dimensions and stencil thickness, the molten Pb-free metal will spread less. This is not all bad: Sometimes bridging will be abated. On the other hand, in certain cases, the old aperture will result in an insufficient joint. Some people have been spooked by exposed pad areas after reflow, which, for most applications, has no negative implications. But we still want the solder to go to the appropriate places for proper joint formation.

In general, a stencil used to print Pb-free solder paste will incorporate less aperture reduction. Aperture shapes that previously worked well for SnPb alloys may not be optimum for Pb-free. For example, many of us use the home plate aperture for ceramic chip capacitors; an inverse version of this pattern is gaining acceptance. Experimentation ensues; stay tuned, particularly to printer and solder paste manufacturers for the latest and greatest.

(It should be noted that a major cause of reflow soldering defects, both for SnPb and Pb-free alloys, is improper aperture geometries. But that’s a sin for another day.)

Inspection. Since Pb-free solder wets differently, the resulting joints may not always look like their SnPb counterparts. Hence, inspection criteria have to adapt. Inspectors have to be retrained and certified to determine whether proper joint formation is present. It all sounds logical, but it is surprising how many false negatives are subjected to unnecessary touchup and rework. Inversely, it is still possible to have a legitimate cold Pb-free solder joint. Either situation can result if inspection and quality personnel have not been reeducated with the gospel of IPC-A-610D, and (hopefully) some real examples of their own products. Forgive them, for they know not what they do.

Wave Soldering. At the wave-solder machine, there’s ample sinning going on – too much to even scratch the surface. Besides the aforementioned intermixing of alloys, the actual machine configuration may be at fault. For example, bear in mind that the Pb-free solder wave is going to be 5 to 30°C higher than a SnPb wave – about 255 to 260°C. Accordingly, the assembly temperature as it exits preheat has to be higher if thermal shock is to be avoided and complete hole-fill achieved. Many waves have inadequate pre-heaters. Improper flux selection is another serious misdemeanor.

One word of advice: Think hard about trying to convert an existing SnPb stoked wave solder machine to a lean, green, Pb-free machine. There is a lot more than meets the eye. Besides the solder pot heater and pot materials, there are thermal switches, pumps, floats (remember, Pb-free is less dense than leaded alloys) and myriad other little “gotchas.” To avoid problems, talk to an expert on your specific machine, model, and option configuration; do not trust generalities.

Brothers and sisters, many other mistakes and grievous errors are being made – perhaps not quite fire and brimstone stuff, but certainly affecting yields and efficiency. Remain diligent and keep learning. And remember, we’re all in this together. End of sermon.

Phil Zarrow is president and SMT process consultant with ITM Consulting (itmconsulting.org).

• Prev
• Next