More tips and tricks to help assemblers in a pinch.

Continuing last month’s discussion…

Undercured solder mask. Although this is not often noted, it is still occasionally reported. I think it occurs more frequently than we realize, but we just don’t have the awareness to check for it. Fortunately, if undercured mask creates wave-soldering issues, frequently they can be remedied by simply completing the cure cycle – either in a bakeout step or by exposure to ultraviolet light, depending on the type of mask on the PWB.

Blowholes. Blowholing issues are often due to pinholes or other irregularities in the PTH plating, and can be very difficult to address. In SnPb processing, simple bakeouts often removed enough moisture to permit acceptable solder joints to be formed. From experience, Pb-free soldering has proven far more difficult to resolve. In extreme cases, bakeouts under nitrogen or in vacuum chambers have been required to mitigate the blowholing to acceptable levels. Notice I said mitigate, not eliminate. This phenomenon is tough to get a handle on.

Copper erosion. This is one of the single largest wave solder issues associated with the Pb-free transition. Research has found alloy composition, solder temperature, dwell time and flow rate all factor into the erosion equation. And although we haven’t pinpointed the specific factor yet, we do know that there’s something about the copper itself that dictates its propensity to erode. If a trace, knee or annular ring is eroded, there’s no fixing it. Once you pull the copper off, you just can’t put it back. Fortunately, research also has found that lowering the silver content and/or including nickel in the solder alloy can help limit erosion rates.

Having reviewed plenty of data on the topic, I would recommend using a low-erosion alloy for rework processes that require long dwell times. Assuming the erosion rate of any specific electrodeposited copper is set at the board fabrication stage and cannot be modified once the plating is applied, the best advice I can offer is to adhere to workmanship standards for hole fill, and not try to exceed them. Whatever the hole-fill requirement is – 50% or 75% – do what it takes to hit that target. In a high-erosion process scenario, going for 100% barrel fill can result in 0% knee thickness, which is a hiddendefect not detectable by nondestructive methods.

How do assembly engineers determine if defects are rooted in their process or their fabricator’s? Process of elimination is a great start. Typically, a date code and vendor symbol are located somewhere on the board – usually the secondary side or on a breakaway. The date code is a four-digit number indicating the week and year the board was fabricated. If the assembly issues can be correlated with a specific date code or fabricator symbol, then we’re off to the races. Isolate the offending material, and investigate the problem on the poor materials while running production with the acceptable ones. It sounds simplistic, and it is, but it sure can cut troubleshooting time. Never assume all the PWBs on a line came from the same lot or even the same vendor, especially in high-volume production.

I spend much time and energy preaching the tenets of good quality, and I continue to believe it is better to practice proactive control rather than reactive. But I am a pragmatist, and I learned early on that my goal as a process engineer was not to make the most perfect solder joints possible, but to make the most money possible. That goal is achieved by making the most solder joints that meet workmanship standards in the shortest possible time for the lowest possible cost. I continue to believe the best way to achieve this is inherently proactive, but I understand that is not always possible. Therefore, I have offered these tips and tricks to help get through in a pinch, but in no way do I endorse or recommend them as compensatory tools to permit compromising overall fab quality for purchase price.

One final opinion on quality issues with PWB fabs: Everybody makes occasional mistakes; processes naturally drift, and quality escapes happen to even the best suppliers. Over the years, my measure of a good supplier has evolved to consider not just how many good or bad products they ship, but how they react when a bad product is received and a complaint is registered. When you identify fabrication issues, give the supplier a fair shot at correcting it. If they do not respond in a manner commensurate with your expectations, consider the number of fabricators available and work with those who are willing to meet your service needs. We all know how competitive the assembly business is; the fabrication market is no different. And like assembly, if customers are willing to pay for quality, they can get it.

Chrys Shea is an R&D applications engineering manager at Cookson Electronics (cooksonelectronics.com); chrysshea@cooksonelectronics.com. Her column appears monthly.

• Prev
• Next