How does signal integrity engineering compare to other EE fields? What are open-source resources worth these days? What makes for a good work life for an engineer? Learn this and more in this Engineer Spotlight!
Jason Ellison started down the path to becoming an electrical engineer because someone told him it was "fun and easy if you're good at math." In this interview with AAC's Mark Hughes, Ellison—a Senior Staff Signal Integrity Engineer at Amphenol ICC—describes how his career has grown from these beginnings into the rewarding and diverse work of signal integrity engineering.
We'll also talk about Jason's free high-speed signal analysis software related to IEEE P370—Electrical Characterization of Printed Circuit Board and Related Interconnects at Frequencies up to 50 GHz—that's designed to help with PCB design.
All About Circuits (AAC): How did you become an electrical engineer?
Jason Ellison (JE): Oh, that's a very funny story, I think. I was in seventh grade, chewing gum in homeroom with a friend. Back then, we would take the foil wrappers around our thumbs, twist the tip, and pop them to try to get them stuck in the ceiling. While we were doing this, my friend was like, "Hey man, you're good at math, right?" I'm like, "Yeah, I'm pretty good." He said, "You know, my brother just became an electrical engineer. He said it's pretty fun and easy if you're good at math. Maybe you should think about becoming an electrical engineer." I was like, "Maybe I'll do that."
That's really how it started. And then I went to an open house at Penn State Wilkes-Barre and they had a really nice introduction to electrical engineering where they had us build a circuit and do an analysis. I also enjoyed circuits during physics in high school. So that seed was planted and just kind of led into electrical engineering.
AAC: So what do you like about being an engineer?
JE: It's very challenging. As a signal integrity engineer, it really offers almost all aspects of electrical engineering. There's statistical electrical analysis, linear systems, circuit design, and electromagnetism, to name a few.
I've worked at other places before where it was more like you would either do only Verilog or schematics for power designs and nothing else. You would get into a small niche, if you will. But, with signal integrity, it really is whatever you want to make of it. And that's really exciting.
With signal integrity, it really is whatever you want to make of it. And that's really exciting.
Even as far as embedded systems, you could do that with microcode. Almost everything I've learned in college, I could apply in my career.
AAC: What about the profession do you wish was different?
JE: I actually think it would be nice if it was more of a day job. I work on connectors primarily and am based in Harrisburg, Pennsylvania. Harrisburg is known as the connector capital of the world, but our customers are mainly in China and California. So, the travel can be extreme, especially if you're on projects that are high visibility.
Some people travel 75% of the time and I used to travel 25%. That just wasn't good for family life. Now I'm around 10% and it's ok—and I changed careers for that.
It would be nice if we were able to use more teleconferencing instead.
AAC: Do you think it's possible for engineers to telecommute more in the future?
JE: I don't know. I said I wished for more telecommuting, but I also see a lot of value in facetime. You get more communication through seeing someone in person.
A lot of times people are very passionate and might be very stressed about their job because the pressures placed on them by their management. So they might be really wired and stressed during calls. When you're in person, that stress seems to lighten up, and it allows you to build a rapport.
AAC: Is there a particular project that you've worked on (even going back into college) that you're particularly proud of or that still excites you today?
JE: I was exposed to automatic signal integrity testers in my first job and I always thought they could be done better. I had an opportunity in my second job to design and build one. I really didn't know how to do it. But I accepted the opportunity.
Long story short, that system was deployed all over the world and it's still used today. I called it the Sentry because it tested products before they went out the door.
AAC: You mentioned earlier that you've been able to apply almost everything that you learned in college. What would you say you've learned after your graduation that allowed you to be the engineer that you are today?
JE: I think it is the value of networking. I found that the theory and engineering were the easiest parts. The truly successful engineers need to branch out and network to find people to help them.
When you're working on things that don't necessarily have a solution, you need people around you to help with brainstorming. They might bring a whole different perspective since you can look at signal integrity problems, or really I guess any problem, in more than one way.
So I think I learned that it isn’t necessarily the book smarts or the engineering knowledge that makes you more valuable. It's the people that you interact with that make you more valuable.
...it isn't necessarily the book smarts or the engineering knowledge that makes you more valuable. It's the people that you interact with that make you more valuable.
AAC: Speaking of who you interact with, I first heard about you through some of the software that you and your colleagues David Banas and Davi Correia offer for signal integrity engineers. How did you come to meet Davi and David?
JE: I used to work with Davi at one of my employers and I ran across David for almost the same reason I want to give my code away. I worked for a company that didn't have money for an IBIS-AMI modeling tool. I Googled IBIS-AMI and found the group that defined it was a dot org.
And I said, "There has to be something on here to use IBIS-AMI without buying a tool." And they had a list of all the software they used to build the specification. David’s was one of them. I downloaded it and had trouble getting it installed because I'm not really a python guy at the moment. I contacted David, and he worked with me for about a year to get me started.
Last year, we—the IEEE P370 group—announced that we had the de-embedding software available for download through Git lab. Davi heard about it. After he picked his jaw off the ground, he said, "I have this idea for a bootcamp and I can give away some code I’ve worked on, as well. Do you know anyone else that we can work with?" After some thought, I said, "Oh yeah, David Banas! He's great!"
And plus, I selfishly wanted to meet him, because I had been working with him for so long through email. I hoped I could shake his hand and pick his brain. So that's how that came to be.
AAC: Let's talk a bit about the project in question, which is free simulation and analysis software to help engineers design their next PCB—and it's open source. Can you tell us how you got involved?
JE: I actually wrote that code as part of a thesis. I just didn't feel right using a commercial tool as part of the educational experience, since it seemed like it was something that was relatively simple. I was part of an IEEE working group that was trying to identify fixture electrical requirements for the de-embedding up to 50 gigahertz.
So, I thought, "Hey, maybe I can give what I've done in grad school to the community as part of the standard, and it would help us develop these specifications." And the chair of that task group said, "I think that's a great idea."
AAC: Making such a project open source is an interesting decision. What was your philosophy for giving this away for free instead of trying to monetize it?
JE: At that time, we really had no intention of [monetizing it]. I guess I may have said we can give it away. But really, in the beginning, it was more a means of getting those fixture electrical requirements.
I remember being a young engineer struggling to get tools. And, truly, the math behind the de-embedding process isn't terribly well known. So I felt like it was a really great service to the community. Giving the code away enables young engineers that need something—and if they're in a smaller company, they may not have access to the commercial tools.
Giving the code away enables young engineers that need something—and if they're in a smaller company, they may not have access to the commercial tools.
AAC: Are you hoping that this movement will spread or are you actively trying to spread it?
JE: I do hope that someone grabs the code and improves it. It's definitely not as good as some of the commercial tools available and there's some theoretical aspects that I haven't completely figured out yet. I'm working with some people to try to get through that—but if there's just somebody who has the same interests as me and can pick it up and fix it, then I can learn something new. I would really like that.
I don't really think of it as a movement necessarily so much as I just kind of wanted to contribute to the community as a whole.
AAC: Do you have any advice for readers who may be interested in the open source community?
JE: Try to improve the code and ask questions. The beauty of open source is the community, and authors enjoy collaborating with their users. As more people get involved, the product gets better without a price increase.
AAC: Is there anything else that you'd like All About Circuits readers to know about your code, or any advice you have for new engineers?
JE: Well, you need the RF toolbox through the code but if you look around you can easily get around some of the functions that are in there and use it in Octave. It would be great if someone ported it over to Python because I really think everything's going to Python. There are many products that have gone away from other coding languages. Now they are almost exclusively going to Python. It's just so powerful and there's a huge community behind it, as well.
Since this interview, Jason has improved the code and has released a version that is compatible with Octave at https://gitlab.com/IEEE-SA/ElecChar/P370.
Featured image used courtesy of Erik Buer [CC BY-SA 4.0], altered from original.