Issue link: https://iconnect007.uberflip.com/i/679419
48 The PCB Design Magazine • May 2016 better equipped to deal with it. Often the spac- ing of components, board finish and distance to ground planes can be optimised for corrosion resistance. Once the design is optimised for corrosion resistance, the conformal coating will often be even more effective in ensuring corrosion free operation. Stress tests (thermal shock, for ex- ample) are not intended to be used on final as- semblies for product validation. Failure mecha- nisms may well be different, and may be missed by stress tests. Product validation testing should mimic end use environments as closely as possi- ble. (Here at Electrolube, our testing is designed to push products even further than in actual us- age scenarios.) Fact 2: If you assume that a conformal coat- ing will be necessary from the very beginning of the process, you will not go far wrong. Plan for this, design for this and test for this. You can bet your life that a coating will be required and by factoring this into the design process, it will have the knock on effect of simplifying the manufacturing process and you will have a happy team. Fact 3: We would all love a universal all-sing- ing, all-dancing coating but life is not like that. Conformal coating is always a compromise and there is no fixed solution, so plan ahead and bear this in mind. Why not have two or three inter- nally approved and tested materials at your fin- gertips? This will allow you the freedom of choice and will ensure that you can select the best one for the environment in mind, rather than the one that is used the most or the operators prefer using; after all, the most familiar or popular is not always the best for your application. Fact 4: Technical specifications and stan- dards give a warm fuzzy glow, but the use of a MIL spec conformal coating does not in itself guarantee the coating will perform well in your application! The coatings are tested on a flat, bare FR-4 substrate, without solder mask, flux chemistry or the effects of multiple thermal ex- cursions. A key tip is to obtain the MIL docu- ment and familiarise yourself with the test con- ditions and properties tested and remember the standard is a guide, intended to prevent truly unsuitable coating materials from being consid- ered by the end user. The MIL spec doesn't real- ly distinguish between materials or explore the areas of use for which they might be suitable. Fact 5: Make sure you check the boilerplate language contained within your drawings to ensure it is appropriate and relevant. For ex- ample, MIL-I-46058C was declared inactive for new designs in 1999. However, it still remains "active" due to the boilerplate language that re- quires a MIL-I approved coating that has been propagated onto new drawings produced since 1999. I'm very keen to learn if any of the recom- mendations and hot tips offered in my columns over the past few months have been put into effect by readers, as it would be great to have some feedback from you and hopefully some good news about transforming your process. If you have found my design tips useful and have achieved some success through their implemen- tation, let's hear about it! Sharing your experi- ences with fellow readers—if you are at liberty to do so, of course—can only have a positive impact on our industry, so spread the word. In the meantime, I'll do some "coating" of my own and pull on an extra layer of clothing! Roll on summer. PCBDESIGN Phil Kinner is technical director For Electrolube's Conformal Coat- ings Division. COATINGS: FIVE ESSENTIALS FOR DESIGNERS " Failure mechanisms may well be different, and may be missed by stress tests. "