Issue link: https://iconnect007.uberflip.com/i/1379105
JUNE 2021 I SMT007 MAGAZINE 17 days most cleaning materials are water-dilu- table, but you need the appropriate concentra- tion of cleaning agent to get the job done. Half isn't good. Double isn't any better, usually. You need mechanical energy, and you also need time. You don't necessarily need a light year's worth of time, but you need time, and the vari- ables are interconnected: if you use more heat, oen you can clean more quickly. Same goes for more mechanical energy. Likewise, if the temperature is low, you probably need more time. If the mechanical energy is low, you also probably need more time. Optimizing a cleaning process is balancing those parameters to get something cleaned well in a reasonable time. Because if you don't get it clean, it doesn't mat- ter how long it takes, no one will do it. If you don't get it clean, it doesn't matter what the product costs, because no one will buy it. Now, granted, we oen employ accelerated life tests to get a more comprehensive picture of reliability. In the cleaning business, you start by trying to clean some parts and then you look at them. ere's sophisticated test- ing oen to do beyond visual inspection, but it always starts with looking. Johnson: ere are, however, parameters to pay attention to, job-by-job, design-by-design, to get it right. Forsythe: Well, it's not generally quite that fluid. It turns out that assemblies all have their chal- lenges, but most clients have a cleaning process that runs the same way for just about every- thing they make. ere are exceptions to that. For example, there are parts and assemblies where water is bad, and those operations use the solvent cleaning products. ere are some edges to the box, but by and large, most peo- ple run an in-line or a batch cleaner and most everything they make uses the same process. It's not always true, but it is most commonly true. From that perspective, it's much like the wave solder machine or the reflow oven. If I'm using the same paste, I don't change the reflow generally from product to product. I might, but it's certainly not a draconian change. Obvi- ously, my setup on my pick-and-place changes, depending upon the components. ere are different devices and components involved, so I must do it differently. ere's always a little adjustment to every process, but on average, I think a lit- tle less than there are from product to product. ere are extremes, it's true. But the old 80–20 rule—if you're in "80" land, you're tinker- ing a lot less with the clean- ing process than you are with anything else on your floor. Johnson: I'm hearing that it's a pretty forgiving process window. Forsythe: If you do the pro- cess design work right up front, that is true. Now, if you just magically dream up numbers out of thin air, who knows? If you design and develop your process before going into production, and say, "Okay, this represents what we do. Let's work on this," by and large, they are robust processes. ere are automated control systems, and we sell some of those, where people will set it next to the machine and it monitors the concentration so it's always operating at the ideal concentration. It turns out that makes a lot of things better. Cleaning machines are generally pretty good at holding the desired temperature, just like wave solder machines and reflow ovens. ey're