Partnerships between electronics manufacturers and institutions of higher education are helping to prepare the next generation of engineering for real world challenges; this effort presents many benefits for both sides.
While at APEC this year, I became aware that Littelfuse was working with the University of Alabama (U of A) on the development of SiC power semiconductor devices during an APEC meeting with Littelfuse and Sujit Banerjee, CEO of Monolith Semiconductor in which we discussed university partnerships with industry. University of Alabama students are getting hands-on industry training from Littelfuse.
Why is industry working with universities? Universities can take risks that companies cannot. What company is going to dedicate an application engineer to a project for six months that might not have a payoff?
U of A research has resolved some tough problems for Littelfuse, and in exchange the students get solid real-world experience.
How did this partnership happen?
When Andy Lemmon was a PhD student at Mississippi State University in 2011, he worked a little bit with Kevin Matocha, PhD and Kiran Chatty, PhD, ultimately two of the Founders of Monolith Semiconductor, who were then working for a SiC start-up named SemiSouth right across the street from Mississippi State University. A few years later, Matocha and Chatty left SemiSouth and formed Monolith Semiconductor.
In 2014, Professor Lemmon gave a talk at APEC on module characterization work that his team at U of A was doing. Kevin Matocha was sitting in the front row and after the talk, Lemmon went over and shook Matocha’s hand, who mentioned that they had started a Silicon Carbide startup, and needed some help with device characterization. He asked if Lemmon was interested. Well, as a new assistant professor, anything that had the hint of potential collaboration, was extremely attractive. Lemmon later was introduced to the CEO of Monolith, Sujit Banerjee, PhD. A brief time after that, Lemmon had a research agreement to help Monolith do some device modeling and characterizations.
U of A leadership
I recently spoke to Professor Andy Lemmon from the University of Alabama. He is in his fifth year as an assistant professor in the Electrical and Computer Engineering Department there, shaping his students’ careers as well as their lives. Lemmon commented that “…mentoring graduate students is something that is really, really exciting, I suffer no delusions that the flow of the information is one way. I know that these guys teach me every bit as much as I'm teaching them.”
In a university environment, “students will have an inquisitiveness” remarked Lemmon, “they don't have enough experience to know which questions not to ask. And sometimes they ask questions that kind of challenge our preconceptions and make me kind of drop back and think, well why is that the way that this is always done?”
Lemmon told me that students appreciate the industry experience. About two years ago he taught the senior design course sequence, in which they were learning how to execute a project. Lemmon was perfect for this role since he was full-time engineer in industry for 10 years before he became a professor. I commented that some of my best professors, at NYU and Brooklyn Polytechnic University which I attended, were those that worked in the electronics industry.
U of A students participating in these joint projects will gain invaluable real-world experiences and opportunities not typically available in traditional university undergraduate programs. These include:
- Interfacing with industry partners through project update meetings (both in person and remotely) and even daily email exchanges– a critical aspect of any team-oriented project in industry.
- Exposure to practical project considerations that are sometimes overlooked in a university research lab setting.
- The opportunity to peek outside the research lab setting and see mainstream industry practices and trends. This benefits the students by not only exposing them to a broader knowledge base, but it also puts into perspective some of the research they are doing.
- The opportunity to network with an extensive list of industry experts, professors, and other students/peers.
- A chance for employment with the industry partner upon graduation.
- The U of A faculty has an opportunity to sharpen their skills and stay up-to-date on current industry practices.
From Littelfuse’s perspective, benefits of a partnership with the U of A include:
- The opportunity to publish content of a more technical nature than the typical editor articles.
- Insight into very innovative research that may one day give Littelfuse an advantage over competitors.
- Leveraging a team of very knowledgeable and capable engineers to perform a project that may not be necessarily justifiable to allocate resources to within the company.
- U of A students can be “interviewed” over a long period of time for future employment opportunities.
- Some members of Littelfuse may have the opportunity to participate in the educational process (e.g. sit on thesis/dissertation committees, etc.)
How does this partnership help Littelfuse manage the risk associated with emerging technologies?
Levi Gant, a former Applications Engineer at Monolith Semiconductor and now a Technical Marketing Engineer at Littelfuse, told me that SiC is an emerging technology, and because of that, some designers still have hesitation in implementing it into their new designs, in spite of all this technology’s good points and potential. He felt that the main reason was because there was not much in the way of long-term reliability studies. Littelfuse is attempting to ease some of these hesitations by showcasing SiC devices in real-world applications that may appeal to those interested in using this technology in their designs.
Monolith Semiconductor SiC MOSFETs (Image courtesy of Monolith Semiconductor)
Since Littelfuse’s resources were limited in using SiC technology and designing with it, the help of university resources would be advantageous in accomplishing such research. A university partnership would offer some unique perspectives into market trajectories and new designs that industry may be pursuing. One project that Lemmon’s group had worked on with Littelfuse came from a professional relationship that the U of A had with an applications group in the Data Center power supply arena. This effort guided and framed the parameters for the deliverable, which was a 10 kW power converter that Lemmon’s group designed for Littelfuse. This design became a talking point at several shows and in presentations.
Professor Lemmon told me that university labs are pretty uniquely positioned to take on high risk research, and it kind of boils down to the fact that our charter is one of education. “At the end of the day, I am training students to be successful as engineers in the industry, and if I take on a research program that's a high risk, high payoff type of program, sometimes the research program isn't able to meet its objectives, due to the inherent risk that were acknowledged up front. However, if the outcome of that program is that I've got a highly trained and highly skilled graduate student, who matriculates through the program– then I'm still successful. And my student can still be successful, even if the technical objectives of the particular research endeavor aren't able to be met.”
“So, a university is uniquely positioned to take on high risk technical challenges, and we endeavor to fulfill every aspect of the program that we can, but at the end of the day, our work product is a graduate student who is knowledgeable and who's equipped to be successful in industry.”
“And so, from my standpoint, that's my charter. And that's what my primary objective is. Whenever we take on a high-risk program with an industry sponsor, industry collaborator, we always try to be upfront with scoping the risks and sizing everything up at the outset, so that we know we have consensus of the expectations.”
“But what's not generally nonnegotiable, is that my primary objective is to train the student. I can be successful at that, even if the risks of the project prevent the attainment of the established technical objectives. The industry collaborator may be looking for the answer in terms of whether the technology ready. Or if there is some blocking item that needs to be addressed before this technology is ready for the market.”
What are the types of functions that the university typically performs?
Lemmon gave his inputs as to the faculty standpoint. On the technical side, we get to involve the students in very practical, hands-on design exercises, where we have realistic constraints.
“We can involve them in looking at budget constraints, as well as time and schedule constraints. Where they are able to see the way a project is run in industry. And this is not pie-in-the-sky research. This is not something that is being done under the auspices of a million dollar, federally funded program, where there's no real deliverables. This is industry sponsored research where we have actual deliverables where we have timelines and the types of constraints that you would find if you were working in industry.”
(Image courtesy of University of Alabama)
“Now, what that said, it is research. We do have the anticipation of schedule slippage and what happens when we meet unanticipated challenges. But that's really no different than what you would have in industry. It's just perhaps the acceptance of risk is a little bit higher in this case, than you would if you were in product development, for example. But that's just one side, the technical and practical side. But there's also other very important kinds of ancillary benefits that the students receive, such as interacting with the customer.”
“The student learns what the appropriate type of interpersonal communication is expected when they are dealing with your customer. Even things like email etiquette or what the protocol is for presenting research progress on a teleconference. These are nontechnical, but very important kinds of soft skills that they learn by virtue of interacting with our customers; these are things with which I have conversations with my students.”
“They also learn what a nondisclosure agreement is. They learn about what it means to maintain your client's confidential information, and to be a good steward of the confidential information that we have access to, as a trusted partner of Littelfuse.”
Levi Gant from Littelfuse had his inputs here as well. He said that “when a student finishes a bachelor's degree, they're likely to be able to accomplish a lot of engineering tasks, but if you give them a project, they could get you a good paper design.”
“But even just based on their bachelor's degree, they may struggle in the practical aspects such as ordering a PCB from a board house, and doing the actual debugging, as well as doing design trade-offs. They also have to interact with the project lead person and give them updates. So, it's almost like they would learn these skills eventually in industry, but being able to go through a project and a program like this with a mentor professor, this is half of their job to keep teaching the students.”
“It really accelerates the process, and refines them, and in the end they have the skills and they're even better than they would be if they would have learned these things in industry; it really puts them on a fast track to getting several years ahead in their career.”
How are university labs a fresh source of ideas for companies like Littelfuse?
Levi Gant said, “Just take, for example, the Silicon Carbide team that Littelfuse has at their disposal. It's a pretty limited number of engineers, and they're swamped with their day-to-day tasks of getting products to market and supporting materials for those things.”
“They don't have a lot of time to browse the research literature, go to many conferences and attend the presentations that go along with those. Whereas, with universities, that is part of their research; that is, to keep fresh and to have a broad base of knowledge, and really explore all the new things that are going on with different types of technologies.”
(Image courtesy of University of Alabama)
“So, just having exposure to multiple things automatically, makes the university labs have a fresh source of ideas and things to try along with things that other people have already found, that don't work. So, having the university to lean on as that base of intellectual knowledge has been very valuable to Littelfuse.”
Professor Lemmon added that, “The other part of having an educational mission within the university context is that knowledge is valuable for its own sake. The university exists as an educational entity, and so if there is a particular aspect of really anything like technology or another branch of engineering or science, and a faculty member decides this is something they're interested in learning about, then it automatically is part of your job description, to learn about it.”
“And so that is one of the things Lemmon loves about his job; if he wants to learn about something, it's automatically part of his job description. That's why you end up with faculty members who are the world’s leading experts in some obscure variety of animal life, or something like that. Or something that could have no commercial or economic impact whatsoever, but they're still a valuable and valued member of the faculty.”
“But then, you get others that have interest in things that do have a commercial application. And so, when you get that combination of intellectual interest, and the possibility of commercial interest, now comes a really powerful combination. And in Lemmon’s area, there is the opportunity to dig deep into this technology and to learn about what all the various players are doing, and one of the things that Levi Gant mentioned is the ability to spend time reading the literature and understanding what others are doing, so when we go to build a system or a converter, we're not starting out from a vacuum.”
“The university can begin knowing what others have tried, and the difficulties that they've encountered, and even some of the solutions that are already known. And so, they can begin from a point of already knowing where the state-of-the-art is, because it's in the professor’s job description to be familiar with the state-of-the-art as described in the literature.”
Lemmon stated that one of the things, when he was a practicing engineer in the industry, that he did not fully appreciate, was the value of the academic literature. He kind of viewed it as something that was somewhat irrelevant and impractical. But after having been a little bit closer to the academic side of things, now he came to realize that not having a knowledge of what's in the literature is a real penalty when you go to design a system. Lemmon told me that “Because you may not be fully aware of the potential pitfalls, and you may end up recreating the errors and the problems to which others have already discovered solutions. So, there are even practical constraints, as well. The access to the academic literature is a subscription-based service that is very expensive.”
“Often times, companies, and especially small companies, which of course Litteluse is not, but many companies don't consider it worthwhile to subscribe to these academic literature services, because they just wouldn't get the value out of it. And so, often times this literature is behind the paywall, and is unavailable to the practicing engineers who could benefit from it. But within the university environment, they have unfettered access to all of this literature. They try to take full advantage of that in all of their research activities.”
How do university labs benefit from the ideas of application engineers from companies like Litteluse?
Lemmon told us that this is where they get their practical compass of what matters, and what has a potential economic value. “One of the benefits of working at the university is that knowledge is valued for its own sake, but that has kind of a side effect that it doesn't help you discriminate of what knowledge is valuable from the economic standpoint.”
“One of the things that is very valuable about working with industry partners like Littelfuse is they help us narrow down to particular topic areas that are of value from an economic and market standpoint; that's valuable in a couple of ways:”
“One is that it helps ensure that his graduate students are being trained in an area in which they can then go get a job. They would consider this extremely valuable. Another benefit is it allows Lemmon to stay relevant to the evolution of this technology as it progresses. He does not want to be stuck in a cul-de-sac where he’s researching a particular aspect of the technology that's been passed over by the market, and no longer being pursued. Those are two things that he would state as a value of working with Litteluse.”
Lemmon told us that one thing that comes to mind as an example from a challenge that he is working on, that kind of touches on a couple of programs; that is, that one of the side effects of Silicon Carbide is electromagnetic interference. That's something that he is working on under a grant from the Office of Naval Research.
He has been working on that for about two and a half years, now. Under sponsorship from the Office of Naval Research, but that actually recently created an intersection between his Littelfuse work and their Navy work, because they were designing a power converter for Litteluse last year, and were having some challenges with the gate drive circuit. It was something that the PhD student who's working on this had been struggling with. They had been trying different methods to resolve it and actually, it turned out that what he was struggling with was the root-cause with some electromagnetic interference that was being created in the power converter, due to the very fast switching nature of the silicon carbide devices.
Littelfuse/Monolith Semiconductor has a Gate drive evaluation Platform, EVAL_GDEP_01, to test gate driving circuits under continuous working conditions to evaluate gate driver thermal performance and EMI immunity (Image courtesy of Littelfuse)
Since his other students had been studying that behavior, under the Navy program, they actually were able to put their heads together and identify the root cause of the problem, and help the student who was working on the Littelfuse program; they helped him resolve the issue. This brought together a disparate program; students working on two different, funded projects, but both of which were utilizing Silicon Carbide, and both of which recognized one of the challenges associated with that technology.
“This resulted in an ‘Aha’ moment for all the students involved when they added that more fundamental research that was being pursued under the Navy program.”
“They saw how it practically worked out in the context of the real-life power convertor that was being designed for Litteluse and when, and the light bulb came on and they recognized that that phenomenon that they were viewing on the bench was actually a consequence of the fundamental things that they had been studying—-they were able to very quickly come up with a solution, because of having that fundamental knowledge of what was going on down at the device level. So that was an example that you wouldn't find in a pure industry environment, because it happened due to cross-pollination of the different students on that team interacting and comparing notes on their different parts of the technology landscape.”
What about bringing the SiC gate driver closer to the power element? Also, do you, or will you do any space-related or Rad Hard work?
Lemmon commented that this is clearly something that they deal with on an ongoing basis. Management of the parasitics in circuits is one of the primary challenges and impediments right now to achieving the value proposition of this generation of semiconductors.
He said that there is still untapped potential because the packaging stands in the way of getting at the device capabilities. There is a number of things going on that they are involved in on packaging, to try to move the needle a little bit in that direction. There will be a new program from the Navy that's kicking off in June that is going to focus on advanced packaging structures and looking at module designs.
“There is an effort to try to mitigate some of the side effects that often occur in the MHz range, especially in the range of 1-30 MHz, which is what Lemmon likes to call the near-RF range. He says that ‘It's not quite up in the RF territory, but it's getting close’. What they find is that in order to really get good at designing power electronics, based on Wide Bandgap Semiconductors, they have to borrow some knowledge from the RF discipline.”
“They will have to start thinking about the system in terms of impedance matching and looking at every trace on the circuit board as an inductor and every copper pour as a capacitance to the ground plane. They have to begin to be very, very cognizant about things that historically electronics designers have been able to safely ignore.”
“And so, that's an ongoing challenge, both for Professor Lemmon and his research program, as well the power electronics industry as a whole, which has some education to do for their members, to try to impress upon them the importance of managing these seemingly tiny amounts of parasitics and the surprising amount of influence that they have on circuit behavior.”
“The Navy is working on higher voltage systems, up to 10 kV, where these challenges take on a whole new dimension at the higher voltage levels. But even industry is going to be moving in that direction in the next five years, more than likely. These are challenges are just around the bend for industry.”
Editor’s note: I really appreciate industry and university efforts to get real-world design experiences into the hands of our future engineers. This is one excellent way that this is becoming a reality in our field. The IEEE Young Professionals Program formerly “Graduates Of the Last Decade (GOLD)” as well as the IEEE student memberships are another very excellent way to get students connected to mentors right in their regional areas. Students can attend IEEE Executive meetings in their regions as well; there are certainly more mentors there as well (Students should contact local Regional IEEE chapters to get more details about this)