Issue link: https://iconnect007.uberflip.com/i/1219242
32 DESIGN007 MAGAZINE I MARCH 2020 Next, it goes to the supply chain, and they say, "Cross that off. No, change that," and/or, "You can't buy that because it's obsolete. This won't work," and then they kick it back. Then, the engineer has to go back and change it; it's a horrible cycle. With SiliconExpert, we thought, "What if we brought that data into the design phase?" If engineers have the data, they can make the right decisions. It's all about helping them to work smarter, not harder. When you know a part is going to go obsolete in three years, and you have a five-year product life expectancy, or there's low stock so they can't buy enough, then you're going to make decisions based on this data. Shaughnessy: Sure. It does seem like a lot of designers are forced, because of the time-to- market constraints, to make decisions early on when they don't necessarily have all of the data. Banton: Agreed. However, sometimes, that's be- cause the data is not made accessible to them. We were interested in working with Silicon- Expert because they have the data engineers need—it's what they do. Give it to the engi- neers and let them be able to access it when they make their selections. Recently, we conducted a little experiment where we said, "Here are four parts based on this standard data, as well as the electrical func- tionality and price. Which one of these would you pick?" Every engineer picked the cheapest one. Their thought process was, "We're going to save the company some money." Afterward, we said, "What if we told you that the cheapest is out of stock, there won't be any more for six months, and it's going obsolete in three years?" Their answer then changed, and they picked the one that was a little bit more expensive. Ulti- mately, we asked ourselves the question, "How can we help an engineer have the information they need to make a cost-driven decision?" Shaughnessy: Are your customers expecting you to help them make cost-aware design de- cisions? Banton: Yes. As you mentioned, they have schedules to meet. They don't have time to scour the web to figure this stuff out. They need the data packaged and delivered to them in a way where they can continue to design, but also have that additional information avail- able. That is what we focus on. How can we auto- mate all of this and take away tasks that engi- neers get stuck in that really aren't productive? Software can do that much faster, so the engi- neer can focus on what they really need to be thinking about. Does this product work? Is it going to accomplish our goals? Shaughnessy: Right, that's what it's all about. This could also help designers and engineers avoid over-constraining their boards and mak- ing them reliable but more expensive than nec- essary. Banton: Right, and, obviously, we're very fo- cused on the software. Customers don't have much time, and they're being asked to do a tremendous amount. Some engineers will take a known-to-be good design or circuit and keep using it. Even though you may be able to shave some space off, if you do that, you have to requalify it, and that's when simulation and analysis really help there. Capacitors are a pretty common and basic example of over-en- gineering. You'll have a design, and on the last page is just a pile of capacitors. Shaughnessy: The shotgun approach. Banton: Exactly. Sometimes, they just spray capacitors everywhere; often, they're pretty cheap, so it's not that big of a deal, but if you think about the volume you might be produc- ing and the board space you might save, then those can become meaningful if you can quan- tify that ahead of time. What if you had some software that can help you identify the opti- mal placement for capacitors? We have a prod- uct that will help you optimize decoupling ca- pacitor placement based on your design. It can run thousands of scenarios very fast and come back to you with the ideal way to place them.