Issue link: https://iconnect007.uberflip.com/i/701250
76 The PCB Magazine • July 2016 But how in the world could it be possible with so many variables and so much uncertain- ty? What are the challenges to get this done? The following are a few gleaned from my expe- rience in this field. 1. Relentless pursuit of improvement There is a need to pursue perfection when dealing with on-board impedance testing. Cop- per plating distribution, prepreg thickness, lam- ination flow, etching compensation and line formation, both on coupon and on board, can be exhaustively analyzed by process engineers and controlled in the production. It might sound basic, but there is a need to set patterns or controls all over the production steps, so that one may find a trend. Boards cannot be etched or developed or even plated in a random way. There must be a standard definition of how boards flow into each and every line in the fac- tory, otherwise data will be irrelevant. Once, during a presentation to a customer who was very concerned about impedance con- trol, I showed him some results that were quite good at that time. There was a difference in cop- per thickness between the coupon and all the points of impedance control on the board of about 5−7 microns, but there was a pattern in the deviation. We had never achieved such good results, but the customer looked at us and said, "It needs improvement." I started laughing, and for my surprise and amazement, I learned one of the most important lessons a process engi- neer could ever learn. He said, "What is data for if not for improvement?" When it comes to impedance, every single product is different, and although there are some rules to be followed, it is not guaranteed that the results will be the same. 2. Cross-section analyses Analyze 15 or 20 cross-sections. 3. AOI AOI became a critical process during fabrica- tion. Every single panel with impedance con- trol needs to go through this inspection, and defects that once were imperceptible will start flourishing in the factory. It dramatically changes the flow of the boards, and also makes a huge impact on opera- tors' training and understanding of process. As lines' integrity are a major concern on impedance, focusing on the impedance lines must be a common practice. Also, it is impor- tant to state that AOI does not guarantee im- pedance and it is not a measurement inspection machine, so the process as a whole must be un- der control. 4. TDR equipment One of the real challenges is finding equip- ment that can fit on-board testing and coupon testing and still be robust enough to be used in the production line. There are many suppliers of impedance equipment in the market, some of them with breakthrough promises but still unknown, and others that have been well-established for years that are trying to keep up with the new trend of on-board testing. When choosing the equipment, one needs to keep in mind that bandwidth and rise time are important characteristics for on-board test- ing. Basically, all TDR equipment can test cou- pons, because they are about six inches or so in length. However, it is not uncommon to see lines in circuits of just about one-inch length with impedance control that cannot be tested with the same "standard" equipment. The bandwidth is one of the characteristics that will determine how short the trace can be. Simply saying, the higher the bandwidth and the lower the rise time, the shorter the line to be tested. Another aspect that has changed is that, THE EVOLUTION AND REVOLUTION OF IMPEDANCE CONTROL IN PCB PRODUCTION " There must be a standard definition of how boards flow into each and every line in the factory, otherwise data will be irrelevant. "