Freedom CAD: Navigating the Unpredictable Design Marketplace

Estimated reading time: 15 minutes

Matties: So when you look at the marketplace, what do you think the most important message is to a potential customer or the market in general?

Miller: Relative to our business, you don't get many “at-bats” anymore. If you miss a window, you miss a timing, or you miss a cost target, it's a very unforgiving marketplace. From my perspective, the key is to do your homework and pick the right partners and make sure that they're really good at what they do, because you can't afford for them not to be. I see a lot of companies that hem and haw and say, “Well, you guys are expensive and we don't want to pay it.” That's fine, but you're also going to run the risk of not knowing whether they’re going to be able to execute and get what you want. At the end of the process, did they deliver on what they said they were going to do? If they don't, that's an expensive mistake.

Matties: That goes to the next thought that I have. I often hear that designers might over-design just to cover their ass, and when you over-design that adds cost.

Miller: Yeah, and that's why we have a dialogue with our customers’ CAD managers or engineering managers, because as a service company we're working for our customer—they're giving us direction and we're responding to it. There's probably better control of scope creep when you're working within the company because their people are there and are under their watchful eye. When you're working remotely, like we do with many of our customers, there can be more churn because we're doing what the customer asks. And this is where we try to add value to our customers, by saying, “We're happy to do this; just make sure we're doing what you want as a company.” You can see these cycles of the perfect engineering, it never ends.

Matties: You have to say “done” at some point. When you have a customer, do you offer a design review service? Say someone designs this board internally and they just want that sanity check. They just send it over and say, “Hey, tell us what you think.”

Miller: We actually do, and we have some customers and some CMs that actually use that service. A key philosophy for us is we've invested in Valor as a tool to help us check our work, so when a designer completes a design it goes to our Valor QA department, which is independent of the designer. The designers have to do an audit themselves. Our designers use a quality checklist to go through their designs and make sure everything's done. The CAD tools will allow you to design in flaws or manufacturing defects. The high-performance CAD tools have DRC checks so that if something's not connected or fails to meet a design rule you can see that.

Matties: I'm thinking more of the definition of optimum design rather than the functional “it works.”

Miller: That is two different things. That's a little harder to do as an afterthought, because you really need to understand the design intent. From a design review standpoint, we do offer DFM analysis to customers where we'll bring in their data package, we'll run it through Valor, and we'll look at it for DFM issues. If they go to fabrication with this design, will it end up having some manufacturing problems?

Matties: Don't a lot of the manufacturers offer this as part of the front-end service anyway?

Miller: Some of them, yes. Some of what they're doing is really only looking at the board data as it pertains to their ability to manufacture it. We're also looking at some of the other documentation that goes along with the board data, like the manufacturing drawings, the pick-and-place drawings, all the information, that's part of the deliverables package we're also checking, not just that the Gerber file is right. We're looking at the entire documentation package.

We're focused on the board only, but as it pertains to the board and the board data that then gets propagated into the other manufacturing processes. To talk about the DFM or the design intent of a particular design, you really have to then go back to the original intent. What are you guys doing? What is the intent of this board? Where does it fit in the system? What are you looking for? We can go through and do a design review, and like I said, we've done a few of those, but that's a whole different animal than doing a DFM analysis, because you've really got to get into the best components for that particular solution. Some of those types of requests actually happen because of component obsolescence.

Matties: We do a lot of surveys of designers and one of the things that we’ve learned and is repeated time after time is exactly what you're talking about. The designers come in late in the process because system designers are including the circuit designers, then they have a small box that they have to fit this real estate functionality in that doesn't fit, and it's just an incredible challenge.

Miller: It is. That's a consistent observation in my mind. The layout portion of the design tends to be victimized by the time schedule. It's like, “OK, now we're ready for layout. Now we've got to get it done in the shortest amount of time because we've used all that runway in trying to get the electronic design perfect. Now we've got to try to figure out how to make the highway system work on the board.”

Matties: It's such a prevalent complaint. It seems unbelievable that we're not doing concurrent engineering with the designers.

Miller: This, again, goes back to the relationship you have with customers. We have a lot of customers that we work with in a concurrent engineering mode because we've earned the right and the opportunity to do that. They'll come to us and say “Hey, we have a schematic. It's 60–70% done. We're still working with marketing to figure it out, but we're behind schedule and we need to get this going.” So we'll probably start doing some of the library work or some of the floor-planning work around the main core component and the DDR. That's well-known, so we know that this chip and these devices are going to need to be connected. Let's get started on that. That high-density area oftentimes tends to drive the stack-ups. The mechanicals are roughly known at that point. Today, just to fit everything that needs to be put on boards is often a challenge. We very rarely have the luxury of a lot of real estate to work with. You're pushing things together just to try to make things fit.

Matties: That goes back to optimum design; you don't have room to throw on unneeded components.

Miller: Right, and we need to be aware of what new technologies are out there that are evolving, because there are different ways to solve problems. People can think traditionally, but we tell them to try to stay current with manufacturing technologies and board manufacturing changes and improvements, because this a core philosophy for us. Like buried capacitance, there are times when you just don't physically have enough space on the board. We have experience with those types of things. We integrate with Insulectro as a board materials provider to stay current with them on what new materials Isola is coming out with as options. For instance, Rogers has long lead times on their laminate materials, so we're making recommendations on alternate materials when we need to satisfy schedules and material technologies. Another example is a company called eSurface that we've been working with that does an additive process for board fabrication. Their process enables finer features, thinner lines and spaces and more consistency than the traditional PCB fabrication copper etching process can achieve. Those are some of the things that we're trying to stay current with technology and differentiate with so we're not just a connect-the-dots company.

Matties: My expectation is that when I come to you, I'm coming to a fountain of knowledge because of your visibility into so many different designs.

Miller: And markets.

Matties: I would think that it's a great advantage. You mentioned earlier about the cost, which some people think is expensive, but it seems to me that when you look at the cost vs. the value, it actually would be on par because of that knowledge. Maybe an advantage, even.

Miller: It comes down to the individual customers. I think that the majority of our customers would tell you that we're a very competitive company. That we're good value to them, because they get high quality and they get the work done consistently well. But all customers are different. Just like selling printed circuit boards, you can sell printed circuit boards on a price per square inch, and to some people it's way too much and other people are very happy. It all depends on the technology, value and service.

Matties: Some final thoughts. What advice would you give a potential customer or the industry about design?

Page 2 of 3

• < Previous Page
• Next Page >