Towards a Data-Driven Cleaning Environment

Estimated reading time: 4 minutes

In an interview with SMT007 Magazine, Tom Forsythe, executive vice president at KYZEN discusses the most critical challenges of the cleaning process, a new approach to testing the cleanliness of PCB assemblies, and how KYZEN is helping customers develop a more data-driven approach to cleaning.

Stephen Las Marias: What are the critical challenges when it comes to cleaning?

Tom Forsythe: There are really two critical challenges today: The first is the evolution of our industry, in that bottom-termination components (BTCs) are becoming very popular for a host of performance reasons, and they present challenges both in cleaning, or in the case of no-clean components, reliability because of the challenges of ensuring that the no-cleans are fully cured underneath the BTCs. That’s a big challenge these days because what’s going on is that everyone doesn’t respectfully recognize outside the high-reliability world, and the industry is sorting out just how it is we can solve this problem. I think over the next year or so, there will be changes that that will be generally accepted to try to improve that opportunity.

The second challenge is more on the confirmation, certification, or validation aspect. Historically, over the last 10 years anyway, the industry has largely relied on an operational basis on a particular test equipment known as Resistivity of Solvent Extract (ROSE). Last fall, the industry standards groups spoke out on the fact that ROSE is not very compatible with no-clean residues and has simply been relied upon as the only method for assuring cleanliness. Right now, the industry is trying to decide just what that means. Is there a new method that’s necessary? So, there’s a great deal of work going on. Over the next year or two, there will be very meaningful change in that test and validation space, which is a little bit unknown right now.

Las Marias: Is there still this confusion between no-clean and the regular fluxes?

Forsythe: It depends on the application. Here’s a little bit of an extreme example. A doctor tells you, “You need a pacemaker.” Should a pacemaker, an electronics device, be clean? Also, do you think that your pacemaker should be made with soldering technology that was invented in 1970? Then, consider—mind you, this is just an extreme example for the sake of discussion—if you want to use a modern soldering material, what type of modern soldering material has been developed over the last five, 10 or 15 years? Aren’t they mostly no-clean?

Remember, no-clean is really the wrong name. No-clean really means low residue, low reactivity. But whether it is low enough for any particular application is the decision of the person assembling the board. I don’t have a percentage handy, but I can say a large majority of our customers have been running no-clean for 10 years.

Las Marias: Does that mean the question, “How clean is really clean?” depends upon the application?

Forsythe: There is a bit of that, to be sure. Over the last many years this ROSE testing has been around, and the old military standard for cleaning was 10 micrograms per square inch of equivalent ionic residue level. Ten is a great number. We had very good customers who were happy with a 10 in those days.

But it wasn’t a matter of how clean is clean enough; it was a matter of how clean could it get. Because those people who’ve wanted 1 instead of a 10, they weren’t doing it because it was a good round number, they want it to be as clean as they could possibly get.

That cleanliness, surely, is helping them with reliability. Because if I am removing contaminants, I am certainly not hurting reliability. I think the whole idea here is what the performance profile is of the device. As we expect our devices to last longer and longer, cleaning and making sure these things are going to last that long becomes a valuable part of that equation.

Therefore, we have a few customers saying, "Imagine going to the doctor and being asked, 'How surgically clean do you want us to make these cuts? Do you want really clean, or are you fine with so-so?'" I think I know the answer to that. Well, what about your car? "Well, so-so is okay. I am good if it will give me five or six years; I'll just get a new one."

The reality is, as consumers, we are quite fickle. We want things that are very inexpensive, and we want them to last forever. The reality is, we have also been impacting this decision from the technology we put inside things for a long time. The BTCs, densities, line spacings and things like these are driving up the whole risk factor of the contaminates. These are the challenges that people are dealing with. While cleaning never completely went away, it kind of went over to the specialty area.

In general, and this has been coming gradually for some time, more people are cleaning. And we believe it is okay for the consumer. Because the incidental cost of cleaning is quite high. With a circuit card, we’re talking pennies. From a production side, the incidental cost is very modest, yet the reward will be very successful. The assembly business is a business of pennies. Every penny counts. So, if we can help in reliability with a few cents a card, it’s a gain.

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