Cleaning with Data

Estimated reading time: 7 minutes

And somewhere between those two clear in-points is where decisions get made, yea or nay, regarding whether cleaning is a value-adder or not. As the world of miniaturizations continues to grow, and devices continue to shrink, employing bottom termination components that are growing in popularity and creating cleaning challenges, the general consensus of where that line between clean and no-clean is moving, drifting gently in the direction of more cleaning, rather than less cleaning. And cleaning is what we do.

We evaluate popular soldering materials from 10 or 20 soldering companies. We have a database with decades of data, which we share the data with the individual solder materials company alone, so they get that feedback for future development purposes. In fact, some companies choose to put us in their product development cycle where we’ll get several experimental formulas, and that’s simply another data point for them to use as they make their final down selects. Sometimes, even we are surprised that we had an early look at a new product. We have all been to a show like this, and someone’s introducing a new paste, saying, “Oh, we need to get some of this new product to test,” and they say, “No, remember that XYZ product from last year?” “Yes, that’s this one that we had tested months and months before.” That’s a common part of our business. We provide materials to get that cleaning done, not the hardware. We don’t make cleaning machines.

Matties: But now you’re in the data business.

Forsythe: Well, we’ve been in the control business for many years because it needed doing. The cleaning machine companies didn’t really see that as in their lane because it was a process control game, rather than a process set point, which is more what the machines are about.

Matties: Right. Here’s the temperature.

Forsythe: Exactly, and that’s what they’re doing, and they’re doing a great job. So, we’ve always been in the process control end of the pool, and it’s just in these last couple of years where we’ve realized that under the roof of process control, we’re touching loads and loads of data. If we can develop a delivery system, and then record it, it’s a value-adder for our customers. This evolution has gone on over the last 36 months, or so.

Matties: Well, what you’re creating, though, is software that presumably could work for any solution, right?

Forsythe: Right, fundamentally there’s a sensor in the system that’s doing the right sensing, the concentration measurements. The latest systems we have introduced are pretty much product-agnostic, though, while there are certain technologies that won’t work, most of the technologies from most of the people in the market will successfully measure the concentration.

So, whether that will be true across the board or not is unknown at this point, but with the latest sensor technology, it is known. Yes, we might be willing to do that for this sort of stuff as well, because we think this data delivery system is something that customers need and want, and the tooling that we’ve put in place to capture that data and relay it is, we think, a raw winner for our customers.

Matties: Are you programming all your software in-house?

Forsythe: We’re doing it in-house in Nashville, Tennessee, where our headquarters and R&D center is.

Matties: So, it’s all under your control. That’s makes a big difference, doesn’t it?

Forsythe: Well, in the beginning it’s hard. But once you figure things out and move forward, it also allows you to be very hands-on. Our approach to things is to try to understand the inner workings of hidden mechanisms, because that allows you to advance the technology at a rapid rate. How do you get to the next generation if you don’t really know what the other one did? So, we are believers and investors in bringing this sort of product development in-house for our customers’ mission-critical needs. If they have a mission-critical problem, rest assured that we also have a mission-critical problem, and we both win if we can help solve it quickly with robust, reliable control systems and rich data sets.

Matties: And it was a wide-open gap.

Forsythe: Yes, and while there are others that do it a bit, we tend to be the ones who do it the most.

To read the full version of this article, which appeared in the March 2018 issue of SMT007 Magazine, click here.

• < Previous Page