Windex is great for windows, but not so much for PCBs.

Ed.: Filling in for vacationing Phil Zarrow this month is Dr. H.J. Zapfardt, well-known (but antagonistic) assembly process guru.

We’ve all heard it professed that there are no stupid questions. What a load of effluent. If you believe that, then you probably believe that politically embroiled (i.e., caught up in their own lies and incompetence) government officials leave high-paying jobs “to spend more time with their family.” A heckuva lot of inane questions are out there, and don’t think the government has a monopoly on them; hang around our industry long enough and you’ll be inundated. Ordinarily that wouldn’t be so bad, but stupid questions usually connect to goofy concepts and really bad practices.

A few months ago, this question came to several consultants, as well as on an email forum: “I tried cleaning a PCB assembly with window cleaner. It did a marvelous job. Are there any conductivity issues that would arise if such a solution is used to remove flux residue?” Upon being presented with this conundrum, a number of thoughts crossed my mind:

• Holy crap, are you nuts? Why the heck would you ever even think of doing something like that?
• What kind of mess do you have on your board, although it appears “marvelous?”
• What product are you building? I want to be sure I don’t buy it.

Having calmed down and pondered the question a bit more, I thought: Holy crap, you are nuts! Why the heck would you ever think of doing something like that?

Of course, for anyone who has worked in electronics assembly, particularly the commercial, non-military sector, the answer is obvious. Politely stated: economics. More realistically, there is Lean manufacturing, and there is what some consultant friends of mine have identified as “anorexic manufacturing,” or Lean gone extreme. To those I would add “Cheap-Ass Manufacturing” (C-AM). Don’t think for a nanosecond this poor chap is alone.

Why would you not want to put Windex on your assemblies? First, who knows what’s in it? Yes, likely a good dose of alcohol (yum!). Maybe some chlorine and ammonia – you can smell it. All these may have an eventual impact on the materials and insulation resistance of the circuits. We have no idea how well this concoction performs, but my guess is that subjected to the usual cleanliness tests like ionic contamination, SIR and others, you would find it does not really do a “marvelous” job, particularly compared to solvents specifically designed to clean flux residues. But our inquiring mind doesn’t do such testing. After all, you can’t buy Omegameters and megohmmeters at K-Mart.

Among those running around manufacturing assemblies on a frayed shoestring, he’s not alone. Another scary example of C-AM was the manual stencil printing observed at a small manufacturer once upon a too-long-ago time. It was literally that: printing with a stencil. No printer, no machine, not even a frame. The PCB was positioned between taped-down shims with a stencil foil placed on it. The operator held the stencil in place with one hand, and used the other to manually scrape a squeegee. Hey, DEK, Ekra, Speedline, AMI, Panasonic: You have it all wrong. Who needs to spend big bucks for a stencil printer with all those whistles and bells just to push solder paste on a board? Accuracy and repeatability? Uhhh, what’s that? Want to guess what they were using for a squeegee? I’ll give you a hint: You can get them (cheap) at the auto-supply store, right next to the Bondo. I can’t tell you who and where this shop was and what product they were building, however, because they had me sign a nondisclosure agreement.

At another C-AM practitioner, I observed an older aqueous cleaner used to clean the assemblies. They also pointed out an in-house water recirculation system for the cleaner. In fact, they’d had it for a number of years. But it wasn’t hooked up to the cleaner, the water supply or anything. It was sitting in the corner, occupying space, collecting dust. But they had one of them newfangled recirculation systems. Picked it up for a good price at an auction, don’t ya know. Looks mighty impressive. Don’t really see the point of hooking it up. Yes, they were dumping the water, with effluent, right into the ground. I’ll pass on the coffee, thanks. Do you have any bottled water?

Other examples of C-AM: using a toaster oven to reflow boards; picking and placing by hand with tweezers (preferably zircon encrusted); using a range-oven for curing adhesives; adapting an old pen-plotter into an adhesive dispenser; soldering through-hole components over a stationary solder fountain; and, of course, using a dishwasher as a batch aqueous cleaner. Then there are the “screw-it-up-yourself” projects like homemade wave solder pallets and other swell ideas. With all these innovative and cost-saving approaches, it’s absolutely amazing equipment manufacturers stay in business. Have pity on the end-user of whatever it is they’re building, and hope to hell you’re not flying or riding in it.

By far the most prevalent application of C-AM is in board singulation. Up until this point in the manufacturing process, whether good or questionable practices, equipment and procedures were used, a number of companies falter here. Most would agree a router is the best way to go for most applications. But routers are expensive. A less expensive alternative is depanelers that use high-speed rotating, cutting disks, known affectionately as “pizza cutters.” Using these devices properly entails keeping the blades sharp and properly aligned, having the proper scoring on the PCB and keeping components at least 5 mm away from the score. Not much to ask. Suffice it to say, a lot of people do this with improper parameters (and resulting cracked ceramic capacitors). But some don’t even get that “advanced.”

A while back, a client asked me to determine the cause of cracks in ceramic capacitors on their assemblies. They were sure the cracks were thermally induced, and sent along copies of the wave-solder and reflow profiles. The cracks were not thermally induced, however, but had the telltale indication of mechanical stress. When I asked about the singulation process used in their Taiwan facility, I received photos of a worker manually depaneling boards by breaking them at the scoring over the edge of a workbench. Perhaps board singulation is considered a martial art. Of course, we’ve seen paper cutters and dikes used for depaneling, too – and lots of resulting cracked caps.

Innovation is great and being cost-effective can be rewarding. Whether you draw inspiration from “thinking outside the box” or ingesting chemicals, don’t get ridiculous. Cover the consequences. Getting your manufacturing equipment from Office-Max, Pep Boys and Wally Mart and having Mr. Clean on your technical staff may seem like forward thinking, but really, it’s pretty stupid.

Dr. H.J. Zapfardt is a member of the National Academy of Arrogant Engineering and an inductee of GITs International Hall of Fame. He has been a highly solicited advisor to industry and government, or basically anyone who will put up with him. A popular speaker in the SMT “rubber chicken circuit,” he was cited as “a personality to avoid.” He has written more than 300 articles and textbooks, including the highly acclaimed A Children’s Garden of Flux. Dr. Zapfardt holds a doctorate in metaphysics from the University of Bonghit. He can be reached at groucholives@aol.com.

• Prev
• Next