ROLLING MEADOWS, IL – With a traveling classroom on its doorstep (AKA the rework bus), the front office at BEST Inc. looks like any other office. But then you round the corner. Like Oz behind the curtain, the bustling building is much larger than the front desks imply. And even at 7:30 a.m., ample staff works diligently as if they’d been there for hours.

On Aug. 14, BEST opened its suburban Chicago doors to local SMTA chapter members and about 50 industry professionals from around the country. President Bob Wetterman’s spacious repair/rework facility provided the backdrop (and tasty breakfast casseroles) for a classroom-style educational symposium, with speakers from the host firm, OK International and VJ Electronix, as well as five hands-on demonstrations presented by on-the-job technical staff.

Paul Wood, applications manager at OK, discussed a rework process for land grid arrays. “Deposits of solder paste can make or break an application,” Wood explained. “Placement control for height is critical.” Furthermore, solder paste is considered a must for heat dissipation into a PCB from the (center) ground pad, he said. LGA is now considered complex with multiple layers of I/O, and it’s “all about the volume of solder paste.”

LGA rework is difficult, Wood added, because of uneven solder deposition; some pads are different sizes. One should cover the center of the pad 60% to 80%, and I/O pad printing can be 80% to 100%, Wood said.

Next, BEST operations manager Ray Cirimele shared secrets of Pb-free rework in a presentation dubbed “Pb-free Rework for Dummies.” In a political move worthy of election, he quickly added that in no way were the onlookers “dummies.”

“Hand-soldering is a big deal,” Cirimele opened. “Pb-free doesn’t wet the same as SnPb.” Also, it’s a “common misconception that Pb-free solder processes require higher temperatures.” Automated processes do require higher temperatures, he added. But, hand soldering may not require those higher temperatures.

The problem, according to Cirimele, is assemblers are accustomed to SnPb, which means good wetting. When Pb-free doesn’t wet well, the response is to increase heat or pour a gallon of flux, he said.

Pb-free (SAC 305, for example) has a liquidus of 217ºC, while SnPb’s is 183ºC. A typical tip temperature is 315ºC (600ºF), and higher isn’t necessary, he said. Typical peak flow temperature is between 245ºC to 260ºC for Pb-free, so assemblers need to “be efficient to get the heat of the tip in better,” said Cirimele.

What affects thermal transfer? Temperature; mass (reservoir/wattage); contact area, and time. Thermal transfer has similar dynamics as a reflow oven, he said; an increase in one parameter affects others.

“A tip at 600º F is already hot enough. Increased temperature will speed the process, but it has its risks,” such as a lifted pad or board damage, he explained.

Power/wattage is important, and physical size of a tip is the amount of mass for heat transfer; with contact area, “bigger is better. Overkill to a certain extent works,” he said. “You want to select the largest contact area for the pad.” The bigger the contact area, the better the transfer of heat into the connection.

“For 90% of soldering people, contact area is the biggest issue,” he stressed. “They don’t like to change tips because they usually have a favorite one.” Preheating can help, he said, but isn’t needed for most SMT soldering opportunities. There should be a topside board temperature of about 150ºC, Cirimele suggested. “600ºF works great for soldering, but it draws heat out quickly.” In other words, the tip can’t keep up. In this case, when preheating, one should preheat the entire assembly, he stressed. Preheat temperature ranges mentioned were 100ºC for SnPb and 150ºC for Pb-free. He warned that many preheaters lack that precision.

In the third presentation, Non-Contact Solder Removal for Rework, Don Naugler of VJ Electronix discussed solder scavenging. Solder scavenging consists of SMT rework (removing the component at the failure site), site dressing and replacing the component.

One of the challenges with Pb-free site dressing is that manual techniques require high tip temperatures, Naugler said. “Avoid site damage, lifted pads and solder mask degradation,” he said. “Improve repeatability and maximize post-dressing solderability.”

A basic (manual) scavenger requires tooling changes, but the dedicated scavenger head has a built-in heater. Several factors for controlling scavenging exist for the non-contact solder scavenger. For example, the hot gas temperature. One should deliver sufficient energy to melt the solder, while protecting the board and adjacent components, said Naugler.

The scavenger gap between the tip and the board should be 0.005" to 0.007" for BGA, he continued. “Often there is a manual gap setting.” However, static sensing can define a plane and compensate for board taper. Dynamic height sensing is continuous and requires breaking the vacuum line and adding a pressure sensor.

Another factor for control is the rate of scavenger motion: manual versus motorized.  When you have a motorized system, you can “set [the rate] in the software and forget it.” Things to consider when controlling the rate of motion include protecting the boards and solder mask; preventing over-travel; pre-heating the “dummy row,” the extra row of free space, if there is one, and conditioning versus throughput.

To explain motion control, Naugler used the example of a tip diameter of 0.125" and a pass speed 0.1" – 0.5" per second. “The system calculates the number of passes automatically,” he said.

“You’re working with tight spaces on a cellphone board,” for example. With a 0.002" gap off the board and a 5 mm CSP site, scavenging time is very short. Naugler presented video clips of the scavenging process; indeed removal was quick and slick.

In that instance, Naugler recommended preheating the board. “There’s no room for error,” he explained. “The rate can be increased with higher conditioning temperatures.”

Another important factor involved with dynamic height sensing is air versus nitrogen, he said. “Nitrogen does help; it produces a much cleaner joint and higher rates (50% higher is possible).”

Nitrogen cleans the entire surface more effectively in a Pb-free, no flux situation, according to Naugler. What surprised him is that flux is not required when using nitrogen. In fact, one loses vacuum capacity with flux when it clogs tubes. “You need a higher level vacuum and a smaller diameter tube for chip scale packages,” he said.

Scavenging is a safe, repeatable, automated solder removal method with high consistency. The noncontact process protects boards, he said.

But there are tradeoffs, he warned: It impacts throughput and quality, and “standards are not well defined for solder height; interpad uniformity, and intrapad uniformity. What could we live with in terms of volume of solder?”

After the three sessions, BEST presented five visual demonstrations around its facility – a day-in-the-life of a rework operation – providing a glimpse into what the staff does in its multi-tasking workplace. First, technician Norman Mier demonstrated methodologies for plastic and ceramic package reballing in accordance with IPC-7711 using BEST’s EZReball. The stencil is the same size as the BGA and the process is simple: Add flux; put the BGA on top of the stencil; put it in the oven; flip it over, and peel the corners to remove the stencil.

Second, instructor Kris Roberson showed techniques for Pb-free soldering of fine-pitch PLCCs using SAC 305. He presented an easy rinsing technique to move seamlessly between Pb-free solders and Pb-bearing ones.

Next, BGA supervisor Hung Hoang presented enhanced imaging software for measurement of voids, roundness calculations and ball sizes. He mentioned, when measuring the roundness of balls, 95% is the target.

PACE technicians also demonstrated bottomside heating, profiling and the impact of heavy ground planes. And finally, Harris Towne of MicroCare exhibited PW2 PowerClean II, for cleaning after Pb-free rework. When rinsing components, he suggested, spray the cleaner, scrub the component with a sable-hair brush and then spray again to rinse away any flux residue leftover.