Judy Warner: James, please share a little bit about your academic background and how you learned to design PCBs.
James MacKinnon: In my junior and senior year at the University of Florida I had some classes where we were required to do some fairly simple PCB design. Basically, if you don't pass those classes, then you don't graduate. And since it's an EE-type class, they want you to create a PCB. They were simple circuits for junior design. However, for senior design they want you to create something pretty complicated. So that's how I first got familiar with PCB design and Altium Designer® .
Judy Warner: Sounds like they actually taught some kind of fundamentals of board design at your university. That's pretty rare.
James MacKinnon: Yeah. I guess "fundamentals" is a good way to put it. In that junior class, we had to layout a very simple circuit. It had a simple microprocessor, maybe some I/O lines. In my case, it was a motor controller, so it's just some PWM lines. They taught us how to use the program, and how to set up a simple two-layer board and do basic routing. For senior design we got into some of the more complicated stuff like high-speed routing, complex analog designs, impedance matching, and things like that. They spent several lecture periods going over different aspects of board design, and I learned a lot there, but it wasn't really until I got my master's degree that I really learned a lot about serious board design and the broader scope of tools inside Altium .
Since Altium was being taught to people at the bachelor level, I think that's sort of why it was also used in my master's, because I did both my bachelor's and my master's at University of Florida, so a lot of materials and learning carried over. The Electrical Engineering Department had licenses, so it's what everyone used. For my master's program, I worked at a lab called CHREC, that's the Center for High-Performance Reconfigurable Computing, and one part of that lab was basically in charge of creating a single-board computer for doing space-based payload processing—so something that you'd see on a CubeSat or a small satellite.
Judy Warner: Oh, that's an exciting application to work on.
James MacKinnon: Yes it was. The team that I was on developed a board that had a Xilinx chip on it. It was a pretty complex design with high-speed routing, DDR memory, differential pairs, impedance matching, all that fun stuff. So I was able to work under a guy that did the first draft of that board and he taught me a lot of the ins and outs of the software. Because Altium has so many advanced features, sometimes it helps to have someone show you how to do different things, rather than learning the hard way.
Judy Warner: Yes, absolutely.
James MacKinnon: He was a Ph.D. student, and he taught me to use mostly everything in Altium , and then I was able to apply those skills as I progressed in my master's program. I went from modifying other people’s designs to creating my own designs. We wanted to make the single-board computer that we developed more reliable, so I wound up redesigning the entire power subsystem in order to make it more radiation-hardened. You can run into a lot of issues when you try to design electronics for space. In particular, you have radiation effects, which can affect your in different ways. You can buy that are radiation-hardened, and they can help protect against these effects, but what we were trying to accomplish was to design and build a board that would otherwise be super-expensive, and keep it within the budget of a university trying to build their own small satellite.
Judy Warner: And radiation-hardened are very expensive...
James MacKinnon: Exactly, they're ridiculously expensive. And not only are they expensive but if you get, say, a processor that's radiation-hardened, it's very low-performance. You can't really do a lot of processing, so you tend to have a lot of them, and they're spread out all over a large spacecraft and they draw a lot of power. All of that is a nightmare when you're trying to design a CubeSat—something that fits in your hand essentially.
So we wanted to design a board that people could use in a CubeSat that would have the reliability of something you'd buy from, say, Lockheed-Martin, but have a lot more performance so you can do cool things in space. In our case, we had a situation where we were taking in image data from an image sensor and processing it on board, and because we were doing processing on board, we were able to send more interesting data products down to the ground. Because with CubeSats, your big limitation is your radio, and you can only send so much data to the ground, so if you can do your processing on board, you can reduce the amount of data that you need to send down, whether you're actually generating data products in orbit, or if you're doing data compression or whatever.
That’s why we thought that this board would be a great addition to that, especially at the university level. So, on the first cut of that board we used a high-performance Xilinx chip as the main processor that isn’t radiation-hardened and surrounded it with radiation-hardened . We called this a hybrid-design. In this way you spend money on the radiation-hardened that you absolutely need while keeping the consumer (COTS) in place to get the performance. By intelligently combining them into a single board, you can get a really high-performance processing board while also getting the reliability of a radiation-hardened board.
So that was basically my main task while I was there. During the first cut of the board, some of the power system wasn't as radiation-hardened as it could be, so I was able to basically redesign the board to use better power supply . We wound up taking the board to Brookhaven National Labs where they have the relativistic heavy ion collider, and we were able to actually use time in NASA's lab to blast our board with radiation and test it.
Judy Warner: Oh, that's cool!
James MacKinnon: It was really cool. So we took this board, we tested it, and it worked out pretty well. After we designed that board, we wanted to design an even smaller board, so I designed a credit card-sized processing board that has very similar capabilities to the larger board, but is smaller and more low-power, and I started. At this point Altium became a big help—we were using micro vias, we've had a lot more layers, everything was closer, part fitment was a huge deal, so being able to look at the 3D model of the board as I was designing it was extremely helpful. We were using a high-speed gigabit transceiver, so being able to use the built-in impedance calculators and differential pair routers was crucial.
There were a few other boards I designed that were sort of just minor break-out boards, debug boards, that sort of thing. But those two single-board computers were the big things that I worked on in college.
Judy: Sounds like a very rich experience. What became of that board?
James MacKinnon: That board is now licensed through a company called Space Micro, which is actually based out of San Diego. So they sold the board we designed. They have the first-generation of that board.I believe Space Micro is now in the process of licensing the smaller version, which I built about a year-and-a-half ago. That's still going through radiation testing and things like that.
So I guess it would make sense for me to explain the lab a little bit. The way it works is it's an NSF-funded research center where industry partners will buy memberships into the lab. These memberships will fund grad students, and what you get for your membership money is any research that the lab does; it becomes something that you can use at your company. So Space Micro is a member, for instance. They fund particular projects. So every year there's a conference, and we show what we worked on in the previous year. And then we propose projects, and then all the industry members get together, they talk to the faculty, and they figure out what sort of projects they wanna work on that would help the company or the agency. So it's kind of funny that I spent so much time on the student side of that, and now I’m on the industry member side of that.
Judy Warner: Is that how you got connected to NASA?
James MacKinnon: Yeah, so that's what I was getting at. In order for this stuff to be actually qualified to go into space, we had to work really closely with NASA. And the branch where I work now is called the Science Data Processing Branch, and they develop something called the Space Cube. Now, a Space Cube is probably one size larger than what you'd want in a CubeSat. It's small, it's low-powered, but it's not something that's gonna fit on a CubeSat. So NASA, being a member of the lab said, "Well, there's a hole in the market for something that's high-performance, high-reliability, but smaller than what that branch at NASA provides.” So they were able to steer the lab to develop this board. The entire time the board was under development we were able to get really high-quality design reviews from NASA engineers, and in the process of doing that we built a really good relationship between the lab and NASA, which eventually led to me working there.
Judy Warner: Oh, that's great. I love hearing that. So many seasoned designers are worried that there won’t be replacements to fill their place once they retire. Your story is an encouraging one.
MacKinnon: Yeah, I think that problem is even more amplified in the aerospace industry, because it's relatively small. You tend to meet everybody in the industry fairly quickly after you go to a handful of conferences. NASA in particular, I think, realizes this, and that's why they're trying to get a pipeline of students to be able to do this.
Judy Warner: So you seem to have had a really rare but wonderful experience, particularly in your master's program, where you got to do hands-on, complex designs. But not everybody has that opportunity, so I guess what I'm asking is where do you think the next generation of designers are going to come from? It seems like NASA, for instance, is seeing the problem and then partnering with organizations like CHREC.
James MacKinnon: I think that's definitely one way—and I've noticed in the case of the lab there are companies that get some good research out of the lab—but basically everybody winds up getting a job at one of these industry partners. One of the guys I worked with wound up at Space Micro. So they use this to recruit people. However, a pretty small percentage of people will have the opportunity to go to a lab like that. One thing I've noticed and have been trying to get involved with here in Baltimore is a rise in amateur electronics. Within twenty or thirty minutes from my house, there are three Hackerspaces. I've been to one a few times and they're always laying out little debug PCBs for our Arduino shields or Raspberry Pi adapter boards. I know a guy that built boards to put on his drone. So I think there's a really cool home brew movement building up with electronics designers. Getting people involved early helps.
One thing that might have helped is getting more involved at the high school level. I never worked with electronics at the high school level. It would be beneficial if there were some way to get people into Hackerspaces or even high school based Hackerspaces, or something where there's hardware that people can play around with to spark their interest. Eventually they're gonna get to a point where they need something they don't have, and they start going down the road of, "Let me try to lay this out myself and get out the acid etch chemicals and see what I can do,” and then ultimately move on to designing cool little boards that they have fabricated. Especially now that it’s so cheap to get boards made. It’s really driving these amateurs in a good way.
Judy Warner: We see that too, James. In fact, one of my roles here is to help sponsor teams, universities and even maker and hackerspaces—and they do amazing work in these places.
James MacKinnon: I think that's the way to do it. I mean, the main reason I use Altium Designer is because it was there at my university. Since I learned on Altium Designer first, I continued to become more proficient. It’s a while before you design PCBs that are so hard you really need to use the more complex Altium Designer features. I'm really happy that I learned Altium Designer because now that I can't use the free license at the university anymore, I've used Kicad a few times for home projects and it's not nearly as good. [laughs]
Judy Warner: It is a luxury to have an Altium Designer license at home for certain! When you were a kid, as you look back, do you see that you had technical tendencies?
James MacKinnon: Yeah, luckily I had a father that was really good at enabling that. He taught me how to solder at a really young age, and he was always giving me stuff to take apart, and I was messing around with circuits. Even just making LEDs light up at a young age was pretty cool. So I think that helped in my decision of where to go. I've always had an interest in electronics, but I sort of had a weird path to college anyway. Right after high school, I didn't go to college for five years. I didn't know what I wanted to do with my life. I sort of knew I was into technical-type stuff, but I actually worked at Sears for five years.
Judy Warner: That's so funny, I know someone that did the exact same thing, worked at Sears and ended up in this industry.
James MacKinnon: I worked there for a while and eventually I saw the writing on the wall and I thought, "Man, I need to do something with my life!” I'd always been into electronics and had played around with Arduino and that sort of thing. Nothing very serious, but enough to get my feet wet. So when I finally made the decision to go back to school I was thinking, "You know what, I've always liked electronics, let me try electrical engineering." I was leaning towards robotics so I went back to community college initially and I always wanted to go to University of Florida. I'm originally from Jacksonville,so University of Florida was always somewhere I wanted to go.
I got my associate's degree and was able to transfer to UF, and that's where opportunity and a lot of luck came together. I just happened to get referred to the right professor to the lab manager of CHREC. I'm so happy I had the opportunity to do all of that, and now I'm at NASA.
Judy Warner: That's a really good story. Since you grew up in Florida, were you inspired by space when you were a kid?
James MacKinnon: Definitely. From Jacksonville, you could see the shuttle launches.
Judy Warner: Wow, I’m jealous!
James MacKinnon: I remember getting pulled out to the backyard and being like, "Oh, look at that." And you can see it, because the SRBs on the shuttle make huge trails behind them. So even if you're far away, you can clearly see them. Where they launch, at Kennedy, was about three hours away, but you could see them because of the amount of junk that comes out of the back of those rockets.
Judy Warner: What a great experience to have as a technically inclined young boy.
James MacKinnon: Yes, and that is why I grew up wanting to work at NASA. I never really knew a path to NASA, and it kind of fell in my lap, so I'm definitely thankful for that.
Judy Warner: I'm sure once you were exposed to the opportunity, you worked extremely hard and made the most of those opportunities.
James MacKinnon: You’re right. I guess it's a little bit of both, right? Hard work and luck.
Judy Warner: James, you mentioned that you're currently working on a mid-size CubeSat, and that's what you're designing boards for. Is that the singular application you're working on right now?
James MacKinnon: No, I work on a lot of things. Life as a civil servant at NASA, you split your time among a lot of different projects. The branch that I work in specializes in data processing. That can include a lot of things, such as creating the processing boards to do the processing in the first place—whether you build data processing pipelines in FPGA or anything of that nature. Usually the board work that I do is either sitting on review panels for other boards or I'm building debug boards—or even boards that hang off actual flight hardware in order to facilitate debugging. Let's say you want to SSH into your machine, you're going to need to build some sort of breakout for that, which includes creating a small debug PCB board that has an ethernet-fi on it which hooks up to the flight-rated connectors that we have access to and can hook up to a router and talk to.
As far as designing flight hardware, I haven't done that yet. I don't think I'm high enough on the food chain yet for them to let me do that just yet. For now, the boards I design are mainly helper boards. They might include some high-speed...because routing an internet-fi and ethernet-fi can be tricky, especially a gigabit-fi. So I was able to use my high-speed digital experience there to have that work.
Judy Warner: How long have you been at NASA?
James MacKinnon: About a year and a half now.
Judy Warner: That's not long at all, which is why you are starting out with the less rigorous designs, I assume.
James MacKinnon: No, not long at all, I recently graduated. That's why the single-board computer (the one that I modified and the one that I designed) is very fresh in my mind. Two of the boards I designed then are flying on the space station right now. Then next year another flight is planned. You tend to design the stuff and then wait a while for it to actually launch.
Judy Warner: Well, you sound like you're on the track to an exciting career! Do you have any technology heroes?
James MacKinnon: I guess I'm going to have to go with my dad. Because he's an engineer as well, and I've always been pushing myself to try to be better than him. [laughs] So it makes me try as hard as possible. I mean, he's a design engineer and I got a lot of my skills from him, and if it wasn't for him I wouldn't be in the field I am today, and I'd probably still be selling lawn tractors at Sears. [laughs]
Judy Warner: What kind of engineer is he?
James MacKinnon: He is an electrical engineer.
Judy Warner: And what kind of work does he do?
James MacKinnon: Well, he's worked at a few different places, but right now he works at a place that does industrial automation. So he designs machines to automate the construction of pipe fittings and valves and things like that for the Navy, so there's a lot of big machinery that handles that and he tries to automate it as much as possible.
Judy Warner: He sounds like he’s been a very good mentor and dad.
James MacKinnon: Yeah, definitely.
Judy: James, thank you very much for taking the time to share your path to NASA. It’s been a delight to hear your story.
James: You’re very welcome and thank you.
About the AuthorFollow on Twitter Follow on Linkedin Visit Website More Content by Judy Warner