Issue link: https://iconnect007.uberflip.com/i/1441485
90 DESIGN007 MAGAZINE I JANUARY 2022 cuit traces to be closer than 0.63 mm from the edge of the board. e act of routing the board edges results in smearing the epoxy resin over the cut glass fibers or the reinforcement mate- rial, sealing off the innerlayers. A designer might specify coating board edges where such sealing does not occur, such as with V-scored, punched, or sheared edges. One area where designers can really help production is to specify where coatings are optional or "don't care" in the engineer- ing drawing. It is best practice to specify the areas that need to be coated and the areas that don't as well as the "don't care" areas to help the coating process run as smoothly as pos- sible. Another helpful tactic is placing connec- tors and components that must not be coated along one edge of the assembly to simplify the conformal coating application process. is may allow dip coating to be explored as a potential alternative methodology, speed- ing up application times and reducing costs. Also, avoid large arrays of discrete compo- nents, which can pose a huge coating chal- lenge due to the high levels of capillary forces present. e net result is oen areas of no coverage/protection on the board as well as areas of excessive thickness that are prone to stress-cracking, de-lamination, and other coating defects. Similarly, tall components present chal- lenges of their own by the creation of shad- owed or hard to reach areas. Splashing is another associated problem. e trick is to avoid placing tall components next to "must- coat" components to avoid this eventuality. 2. How is coating thickness specified? When specifying the thickness of coat- ing on an engineering drawing, the industry practice is to measure coating on a flat, unen- cumbered area of the assembly, and not on items like component leads. As many mod- ern assemblies are very component-dense, it is oen difficult to find a flat, unencumbered area of the assembly. Consequently, it is a common and accepted practice to use witness or process control coupons for such measure- ment. Designers will need to know that the "nomi- nal thickness" measured on flat, unencum- bered areas or witness coupons will have no relation to the thickness of coating achieved on the corner of a discrete, or the leads of ICs, where thickness might be on the order of 1 micron or less, with a nominal thickness of 25 or 50 microns. In a recent study performed by the IPC's "Conformal Coating—State of the Industry," there were some very eye-opening results for many folks who assumed they were getting 25 microns everywhere. Again, the key takeaway is that nominal coating thickness is a process indicator only; the actual coverage on leads and components is far more important to the reliability of the assembly and this should be understood by all parties. e goal is the greatest degree of coverage possible, and cross-sections should be performed to understand how the appli- cation process is delivering this critical-to- success parameter. e witness coupon will only tell you if your process has changed sig- nificantly. 3. Why is coating cycle time important? e coating cycle time is a critical require- ment in Lean manufacturing to ensure the production takt time can be met to maintain a balanced production line. It is advanta- geous for the coating cycle time to be as fast as possible to ensure any accumulated pro- duction can be cleared as rapidly as possible, should any previous operations suffer a stop- page. If we look at a selective coating machine as an example, depositing stripes of coating material that can be programmed to stop and start on demand, the coating stripes can be deposited to create a pattern, avoiding areas such as switches, connectors and test points which must not be coated to prevent interfer-