Issue link: https://iconnect007.uberflip.com/i/1166358
16 PCB007 MAGAZINE I SEPTEMBER 2019 particular performance requirements that are in any way demanding will go to the board shop and say, "Here's my design. What should I use?" That isn't how standards should work at all. You should be able to say, "Here's what I'm designing and the performance I need. I need to choose within this exact category of material. Any material meeting that standard that will do the job for me," and you can't say that right now. Matties: You mentioned you're starting to identify the product by application, which is a help to the designers, but it's not nec- essarily identified by the stan- dard because there isn't a perfor- mance standard. Goodwin: It has nothing to do with the standard; it's about recognizing the need of our end customers. Morgan: For example, a customer comes to Mark and says, "We're building a 5G network here. What kind of material should we use?" Mark pulls a datasheet out and says, "Here are the ones that fit that space." Or the customer says, "I want to do some thermal management on one of my boards. What should I use?" and Mark, again, gives a sheet to them with all of the products that meet that category, and that's the way it's nice to see it moving. Otherwise, you end up with so many products that you can't narrow it down. Goodwin: I just scrolled down to the detail of IPC slash sheets, and it's interesting. The head- lines are not even performance. The headlines are woven glass, then the resin system, flame retardant, and fillers, and we get to things like Tg, CTE, etc. All of the performance stuff is lower down the hierarchy of characteristics than the other stuff. It's completely the wrong way around, in my opinion. Morgan: To the analogy of the smartphone again. Who cares whose chip is in there? All you care is that it works. If someone says, "I know about and come up with a general stan- dard. Then, you do the same for other areas. Regarding low loss, what does that matter? We don't care about the Tg. Okay, let's put it at 120 minimum. What about Dk? We want to have 4 as a maximum. You do that as a first step, and then you start to match products into those categories. Next, you'd work down those categories later to make them more specific to particular industries. As a designer, you'd probably be able to find a very nice group of materials or specification stan- dards that would meet your exact requirements. Goodwin: That's true because when you talk to a PCB company, the only thing they care about is if they comply with the IPC slash sheet number the customer has called out. But it becomes joining the dots. Managing it with this limited number of specifications would be a huge step forward. Morgan: It's also not very efficient because some people overspecify, of course, because they find it the easiest. Someone says, "What's the best laminate to use?" "Oh, polyimide is the best one. I would use polyimide for every- thing." It's entirely wrong, but that's what the military has been doing for years. They thought it was the best because it's soldier-proof in the field and we'll use that for everything. Honest- ly, I sympathize entirely with people trying to choose materials now. With all of the new ma- terials that we have put into this old formula, they just add another line. Goodwin: Or worse still, IPC is adding catego- ries to accommodate new materials and largely not because of their performance. It's because somebody's achieving that performance in a different way (i.e., resin chemistry). At least one slash sheet number is being discussed be- cause of that, and that's nuts, in my opinion. Morgan: This classification based on basic chemistry is broken. Anybody who has any