Issue link: https://iconnect007.uberflip.com/i/1124788
JUNE 2019 I SMT007 MAGAZINE 51 worse?" Because everyone's used to a certain microstructure, intermetallics, and coalescence. Moreover, we're taking a process that takes six minutes for reflow and turning it into less than 10 seconds. The next question is, "Are we getting the same dynamics?" about the physi - cal operations taking place in reflow compared to our machine, which is why we've taken the time to conduct a two-year study to do just that. Matties: Do their customers care that it's going through this process versus a reflow? Barbini: Initially, the questions come up, so yes, they do. Matties: How do you satisfy those questions? Barbini: We're doing evaluations. Currently, we have two demo centers in the U.S. and one in South Korea. We're doing a lot of evaluations to prove to the customer that we can do it in an automated fashion with quality. Matties: Going back to what you just said ear- lier, it's typically a 10-minute cycle versus 10 seconds. Barbini: We're using the AREA laser. Tradi- tional spot lasers aim at a spot where it applies all of the energy and solders one lead at a time. That's visible, right? But we're not doing that. We're taking that energy and distributing that over an area. It could be 80 by 80 mm or as small as 600 by 300 microns, and the uniform density of energy will solder all of those com- ponents equally in that area. By doing that, if you're giving an 80 by 80 mm, there are a handful of applications that are much smaller than that—such as sensors or automotive or medical applications—so we can process that very quickly. Matties: It sounds like this has a real opportu- nity to move into mass production. Barbini: It does, and in terms of taking a process for mass reflow and turning that into a selec- tive laser, we can do that by using the same process development tools, such as recorders and thermal coupling The same process win- dows to build up like you would a recipe in an oven that you would normally have for your materials. But the big difference is we're going from six minutes down to 10 seconds plus transport time. Matties: Is there dwell time for machine startup and come up to temperature? Barbini: No. For a reflow oven, you would turn the machine on in the morning. It would take about 30–150 minutes to get ready with the potential for using a nitrogen atmo- sphere as well. Laserssel's machines have pre- heat options, but our observations are such that we don't have to use preheat. Nitrogen is not required in even the most challenging of applications. These two main setup differ- ences compared to mass convection reflow significantly reduces the cost of ownership. In addition, forced convection ovens remain on throughout the shift or day. Compared to Laserssel's machine, the laser is only on when the soldering process requires it. The end user is only operating a laser for a couple of sec- onds and only when the product is in there. Meanwhile, a reflow oven is on all day long no matter what. Matties: So, the cost of energy is substantially lower. Have you done any studies on that? Barbini: Correct, and we have. Right now, Laserssel has characterized a 60–75% reduc- tion compared to mass reflow. The footprint is also going to be smaller than an oven. If The big difference is we're going from six minutes down to 10 seconds plus transport time.