Libraries: A Must-have for Design
Last month, I-Connect007 was invited to attend a session of the Orange County Chapter of the IPC Designers Council (DC). Even though I have been an IPC member for over half a century (yes, almost since vacuum tubes dominated design), this was my first DC event.
The Orange County Chapter is the largest IPC DC in the U.S. They have a very active group of 50–70 PCB designers, associates, and other experts attend these regularly scheduled “lunch ‘n learn” events. Everyone in attendance gets the chance to have lunch together; talk about the industry, the economy, and even politics; and hear outstanding speakers make educational presentations on a wide variety of PCB-related topics. The few hours spent by those in attendance was quite valuable, including meeting with colleagues and building or renewing relationships. Many locals who attended could still get back to the office afterward.
Primarily designers were at this event but there were also others in the industry, including suppliers. It has been my opinion for quite some time that increased discourse between suppliers and designers is a major plus for the industry. In the past, designers would sometimes wish for a product to meet their needs, and suppliers might have a product that had not been commercialized due to their not realizing that it might be needed by those designing the next generation of devices.
The meeting started with an introduction by Scott McCurdy, the president of the Orange County Chapter of the IPC DC and the director of sales and marketing for Freedom CAD Services Inc. He provided an update on the council’s activities and previewed the day’s presentations as well as thanked the various industry organizations and press for their support.
This meeting included two excellent presentations. The first one was by Natasha Baker, the founder and CEO of SnapEDA, who gave a very informative presentation on the mistakes even the best engineers make as well as how and why to build better libraries. Without stating it directly, she clearly showed the high value of SnapEDA library available to circuit designers and used etfby over one million engineers.
The second presentation, which was also interesting and providing valuable information for designers, was by Terri Kleekamp of Mentor, a Siemens Business, titled "Practices for ECAD Library Development."
After the event, Natasha Baker was kind enough to sit down for an interview. I caught up with her to discuss the importance of libraries in design, current best practices, and how SnapEDA is enabling a new generation of engineers to bring their ideas to life more quickly and efficiently.
Overall, if you are involved with the design segment in any way and have not done so, find out where the nearest IPC Designers Council meets near you and seriously consider attending one of these sessions.
Dan Feinberg: I’m at the IPC Orange County Designers Council session with Natasha Baker from SnapEDA. You gave a very interesting and informative presentation to the group. What does the “EDA” in SnapEDA stand for?
Natasha Baker: Hi, Dan. EDA stands for electronic design automation. It’s a term coined in the 1980s, which encapsulates both computer-aided design (CAD)—which originally referred to PCB layout in the electronics industry—and computer-aided engineering (CAE)—which referred to the front-end design (schematic capture and simulation).
Feinberg: That’s a great name, and I love your logo.
Baker: Thank you.
Feinberg: A lot is going on in this industry, and there are tremendous advances with everything. What has been happening with SnapEDA in the last six to eight months? What are the things you’d like people to know about most?
Baker: The big news is that we’ve hit one million unique engineers that use SnapEDA each year to speed up their designs. Since SnapEDA works with nearly every PCB design tool. It gives engineers an instant productivity boost without them needing to switch tools.
Other than that, we’ve been working with many new component vendors, which means that we’ve been launching tons of new PCB footprints, symbols, and 3D models on SnapEDA. We’ve also soft-launched a new integration with a large distributor this year to allow engineers to download SnapEDA models right from their website. We continue to listen to users and ship new features and improvements every day.
Feinberg: That’s good. I can just tell by your expression and demeanor that you’re excited about some of the things that are happening.
Feinberg: This spring, one of our upcoming magazine themes is going to be “everything starts with design.” You made a comment that I agreed with. Your indication was that when you started the company, it started with an idea. You’re saying that when companies have an idea, then they go to design.
Baker: Absolutely. What makes me so passionate about what we do is that every design starts with an idea and the desire to bring something useful into the world. It’s easy to lose sight of that when, as engineers, we’re so focused on the design process. A lot of what we’re doing at SnapEDA is enabling people to bring those ideas to life more quickly. We help engineers iterate more quickly, and realize their ideas much faster so that they can spend more time on coming up with new ideas, innovating, and optimizing their designs. Ultimately, this means that more and better ideas are brought into the world, helping to increase the rate of innovation.
Feinberg: You mentioned earlier that when you have an idea, it goes to design, and design is the jumping off point. I also loved your comment that libraries get no glory. Can you expand on that a little bit?
Baker: Yes. I think it all comes back to where the value is. Ultimately, the value of an idea is what the end product accomplishes. The value is not in the product or the design itself, and it’s certainly not in the library. It’s in making something that does something useful.
These days, a custom PCB is not always required because there are so many ways to accomplish ideas using off-the-shelf technology. However, once that solution needs to scale, a custom PCB is needed to reduce costs or achieve different form factors.
Entrepreneurs understand this principle very well because they need to—they’re working with fixed runways. I’ve seen so many startups that built casings around a Raspberry Pi or Arduino in the early phases of their companies and sold their initial products like that. Only once they proved out the concept did they build out custom PCBs to shrink the size of their product and reduce costs or make other improvements.
The point is the further you get away from the value that this idea brings into the world, the less value people see in it intrinsically, and that’s just natural. But it means that even among electronics designers, the glamour is placed on the design instead of the libraries.
But that’s risky because libraries are a fundamental gating factor in the design process. Bad libraries can cause delays in the design process, and unnecessary prototype iterations that add further delays and cost. Requiring engineers to make every library from scratch is also completely inefficient, especially when that eats up days of time and when other engineers around the world have already done that same work. Further, libraries require a particular skill set that is completely distinct from design work. Engineers need to know to look for common “gotchas,” like mirrored components in datasheets, and that only comes from having the keen eye of analyzing these datasheets every day, all day. It’s not easy work, especially for complex connectors.
Of course, big companies have always known this. They have dedicated PCB librarians on staff to ensure the quality of their designs, and to operate more efficiently. This makes sense since large companies can afford to compartmentalize roles, and generally, are more focused on de-risking projects and optimizing yield.
But design teams are getting smaller, and more smaller companies are emerging, building electronics. And that usually means their budgets smaller. Many of these companies can’t afford librarians and fall prey to the fallacy that libraries aren’t of value. It isn’t until they spend thousands of dollars in hours creating a footprint from scratch, or until they make an error on a footprint. Then, they have to spend weeks on reworking a board or thousands of dollars on new boards. But even then, the human brain will still struggle with attributing value to the libraries because they’re just so far removed from the end result. It’s a bit of a conundrum.
Feinberg: Are the designers your customers?
Baker: Yes. Although our platform is free, we have a very popular service called InstaPart where designers can request any symbol and footprint and get it in 24 hours. Our service supports Altium, Eagle, OrCad, Allegro, Mentor PADS and DXDesigner, KiCad, PCB123, CircuitMaker, and SolidWorks PCB.
Feinberg: It seems suppliers’ practices are changing. For example, I was at the AltiumLive event last summer talking with some designers, and as an ex-supplier to the fabricators, we wished that the fabricators would come and talk to the suppliers when they needed something. One of the things I asked some designers at the AltiumLive event was if they were starting to talk to suppliers about what they needed. You may be surprised by things you don’t think are available.
One time, I had a big customer come to us and say, “I wish we could do this,” and we said, “We can do that.” They asked, “What product would do that?” I responded, “We haven’t brought it out because we haven’t seen a need for it.” They had wished they’d know because they had to design around a weakness in the products that were available to them. They also asked if we could make it for them and why we didn’t talk to them more often. So, how do you see design best practices changing?
Baker: Let’s start with the communication between engineers, vendors, suppliers, and I’ll give you a little story. We have a huge community of engineers—over a million users on our platform—and we receive thousands of emails and chat support inquiries each month where people ask questions about our data or suppliers’ parts. But we only receive one or two phone calls every week or two, and we have a 1-800 number where anyone in the world can call.
I bring that up because engineers want information and they want to give feedback. But they don’t want to pick up the phone to get it. The internet is enabling more communication between all these parties because people can contact a company online through text-based channels. For example, a lot of companies have chat bots now or live chat.
The trend that we’re seeing in the design world is that engineers want self-serve information. Sometimes, we as engineers don’t feel like talking to human beings. Instead, we want to get the information ourselves quickly and efficiently, or we want to communicate via asynchronous channels because it’s more efficient and we can keep doing things while we’re waiting for a response. Believe me, if you give engineers the opportunity to ask questions or provide feedback via text-based channels, they will be very vocal. We’re not shy about things we believe should exist.
Feinberg: You’re seeing that this is what people want and following that trend. Where do you think you’re leaning into these best practice changes?
Baker: We see ourselves as the focal point between the design engineer and various vendors and suppliers in the industry. For example, for component vendors and distributors, it’s a huge pain point to support all of these smaller customers. But at SnapEDA, we love hearing from designers, no matter how small their company, because it enriches our product and communication with them. Engineers can write to us and say, “I can’t find the datasheet for this component. You don’t have it on your site.” We’ll reach out to the component manufacturer and manage that relationship for them. The engineer does not pay for that. We do it because it helps us enrich our platform with the design data that another engineer might need, and it allows us to help the designer. We see ourselves as a helper between the engineer and the thousands of different vendors out there.
Feinberg: I can see that what you provide is valuable. I haven’t done any design stuff since vacuum tubes.
Baker: Oh, wow (laughs). It’s been a while.
Feinberg: We used to take a chassis, cut holes in it, and put tube sockets in it. You didn’t know how much space you were going to need, and you had no idea when you started. It’s totally different now. I’m just amazed.
Baker: And that’s a perfect example. Our business model wouldn’t have worked 20–30 years ago. Because to just give away stuff that we’re spending millions of dollars making made no sense back then; it barely makes sense now (laughs). I’m joking, but we’ve been able to do it because we have the scale. We have so many engineers using us on the web. We can say to our partners, “We have a massive community.” But 20–30 years ago, how would you get one million engineers in your city? You couldn’t. That’s the scale that only the internet can enable.
Feinberg: You are being kind; that was actually over 50 years ago. Do you find that the engineer and the design segments are growing in the United States and North America?
Baker: Definitely, but we’re seeing a big shift away from huge companies being the only ones designing electronics. One startup company that uses our platform makes electric airplanes. They have good funding, but still, they’re a company with a small team that didn’t exist two years ago. We’re also seeing companies doing satellites and space electronics. Overall, it’s the shift from complex design being done by larger companies to many smaller companies and/or multiple smaller groups within large companies. By the way, all of the top technology companies—such as Google, Amazon, and Facebook—have electronics teams. Even Netflix, Adidas, and Nike are making circuit boards. Because of this, I think electronics designers are going to be the next tech superstars.
Feinberg: There’s no doubt that Asia is where all the volume stuff is being made, but look at the design stuff that’s happening within 100 miles of here. There are probably 1,000 companies designing stuff that we don’t even know about right now, and look at some of the amazing stuff that’s coming out of Israel right now.
Baker: Definitely. Design is happening everywhere, and there are some awesome technology startups in Israel.
Feinberg: How do you monetize what you do since you’re giving away so much for free?
Baker: As a team of 20 engineers, it is extremely important to us that SnapEDA always remains free for other engineers. We make money by working with component vendors and distributors and helping them drive the demand for their products. It’s a win-win.
Feinberg: Very good. Thank you so much, Natasha.
Baker: Thank you.