-Daryl

First, in the 1990s I was on an engineering team for modem design in which DAA isolation was critical for the proper operation of the product, both from the safety and noise points of views. These were high volume products and were running at 1 million+ annual production rates. So they had to be right when they hit the production line.

We were designing some new products in our R. and D. laboratory and our technician had just built some new prototype Modems of which I was the designer. We started preliminary testing and the products were not operating properly. The DAA (which is comprised of several analog circuits) was operating in a completely unpredictable manor. I measured component values (which were not correct) and came back a day later and the same component would read a different value. The technician concluded that the boards were not cleaned properly so he hand-carried the boards back to a professional production cleaning station for extensive re-cleaning. I retested the DAA component values and they were different, but still did not read properly. We battled this strange problem for a week or two, when we finally started testing raw PCBs. We were successful in being able to duplicate resistances between pads on blank circuit boards using standard solders, chemicals and techniques. We then contacted the circuit board manufacturer and requested them to visit us to resolve this problem. They concluded that they had a chemical problem at their company. They corrected the problem, but never told us exactly what they did wrong the first time.

I learned some good lessons from this. First, always receive prototype boards from the production facility as not all circuit boards are created equally. Keep the prototypes as golden samples. Seal them, tag them, log them and keep them under lock and key. These golden samples could be critical evidence if litigation were to come to fruition.

The next lesson I learned, which was surprising to me was that quality circuit board manufacturers go through extensive testing on the circuit boards to confirm that their boards will not react abnormally to cleaning chemicals or standard manufacturing processes. So buyer beware, when you receive PCBs from some Far-East suppliers, you could be at risk.

So how is this related to the Voltage gradient issue? Contamination (of the aforementioned nature) and corrosion are related as they both involve electrochemical reaction. They can occur immediately or years later.

I am a consultant and don’t have extensive cleaning resources at my disposal, thus I have to be a little inventive on how I clean and protect circuit boards. First, my process is home-grown and has an element of danger to it, so I’m not encouraging anyone to do this, thus I’m simply explaining what I do.

First, I use an industrial vacuum-cleaner placed outside of my lab in which the exhaust is not vented through the unit’s motor brushes. Thus, if explosive gasses passed through the vacuum, they would not be pronged to ignition. I have the vacuum intake routed through my wall so it’s a convenient and fixed part of my lab.

My first phase is to extensively clean assembled PCBs with 91% alcohol. I do not try to dry the PCBs with towels. This is because residue from the dried alcohol can be retained under components and once dried, can contribute to corrosive effects. I instead use this powerful vacuum-cleaner to remove the alcohol which (before dried) will contain contaminates. Once completely dried with the vacuum-cleaner, I repeat this process. The PCBs are then tested and then covered with clear finger-nail polish, which I have been told is very close to the same chemical composition as conformal coating. This process has proven itself to be a very effective protective coating with circuit boards operating properly years into the future.

I hope this helps.

David Quinn
484-225-8663
Principal, QED.