Aviation confidant
By Paul Brinkmann|September 2024
Herbert Schlickenmaier, president of HS Advanced Concepts LLC
If you ask engineer Herb Schlickenmaier a question about aerospace research, you will hear stories and insights informed by 50 years in the field. Case in point: A question about artificial intelligence prompts a recollection of feeding data into computers through punch cards in the 1970s, followed by an explanation of how today’s AI and machine learning models could be improved to solve more problems in aerospace. He draws such insights from 18 years at FAA, followed by 17 at NASA and six at Crown Consulting, an Arlington, Virginia, firm with a long history of providing engineering and analytical expertise to FAA and NASA. He continues to assist those agencies, and the aviation industry, as an independent consultant. I reached him at his Virginia office via telephone to learn about his latest passion: electrification of aviation. Here is our conversation, compressed and lightly edited.
—
Q: In a broad sense, what is the challenge of electrifying aircraft?
A: We can learn from the history of electric cars and electric transportation. But the challenges in aerospace are with weight, mass, volume. This is such an exciting time to be in aviation that for me, it’s the next golden age — electrifying aircraft as part of a sustainable aviation future. I’ve got a grandson, and I want him to have something to enjoy. I don’t want him living with surface transportation alone. I want aviation to be able to thrive for him and for all of his friends as they grow up. It’s just too exciting.
Q: Elaborate on the biggest specific issues that must be addressed for electrification.
A: No. 1, getting the energy storage system that’ll give acceptable range. Right now, everybody’s really looking at batteries because they’re clean, but that limits the reasonable range for the mission. Some of these aircraft are looking at a couple hundred miles, but then we’re looking at some other folks who are looking at hydrogen fuel cells that can extend range — which are promising, but technically that’s a big push. Also, things like cabling and connectors are a challenge. When we look at things like 800 to 1,000 volts for some of these megawatt-class motors, we’re talking about really big wires, like 6 centimeters. That’s almost like a structural part of the airframe — it takes up mass and volume. So the biggest challenges are energy storage systems, battery cable connectors, electric motor design and then integration into the airplane — those are the four big things that kind of keep us up at night. And underneath all of that is the supply chain, because how am I going to get these parts that are unique when the rest of the mature aviation world is already looking at that supply chain challenge? It’s tough for everybody.
Q: Are the rosy predictions of electric air taxi designs getting certified and beginning passenger flights in the next couple of years accurate at all, and why or why not?
A: Oh, yeah. Any day now, right? It’s important to ask, “Am I hearing what’s being told to the investor or what is being told to the regulatory authority?” In my experience, those two things are sometimes divergent. However, in this case, I’m talking to all the parties involved — the regulators and the companies — and they all indicate that, “Yeah, we’ve got this all under control.” I know that some of the regulatory authorities are reaching out to manufacturers, saying, “What’s your experience with this? What do you think you know?” That’s important because this is unique in history. Working on this kind of certification of an airplane is not the same thing as doing a technology risk reduction or a technology development project. There are legal questions that need to be answered, for example. There are things other than solving Navier-Stokes equations, or making sure you’ve got the phase margin in your control system or identifying what the structural integrity is. With the development of electric aircraft in the private sector, we need to be able to stand up and say, “You can put your child on this airplane and it is safe.” At some point, that’s part of this dynamic. This is not something where the [U.S.] Air Force has funding to develop some new integrated high-performance turbine engine technology like we had in the 1980s and 1990s. This is not like the development of a Mars Exploration Rover. Here, the government is trying to learn as much as we can from what’s going on in the industry.
Q: You helped NASA develop the means of ensuring that the X-57 could be deemed safe to fly, but the program ended without a flight. Do you still consider X-57 a success?
A: OK, you didn’t get weight off wheels, but the knowledge gained and published documents have been fundamental to the success of the [electric aviation] community, including what was learned from the battery fires to avoid thermal runaway and what was learned about electromagnetic interference.
In late 2016, a lithium-ion battery experienced thermal runaway during testing, resulting in fire spreading from cell to cell and the entire battery burning. A subsequent design fixed that vulnerability. — PB
The idea changed during the project lifecycle from “Hey, can we make this thing get off the ground and come back around and land safely?” to proving out components and systems. The next step was to plug the propulsion system onto the body of an airplane. And then somebody said, “Well, wait a minute. If it’s on an airplane and it’s flying at a NASA facility, there’s a whole laundry list of airworthiness criteria you’ve got to meet.” NASA has more paperwork they’ll throw at you than you can shake a stick at. NASA thought Joby [a California electric air taxi developer] was a few years away from a certificated airplane at the time. So the thought became, “Just give a grant to somebody and have them build it.”
Q: In your current role, you’ve led the NASA project to update standards for integrated megawatt-class electric powertrain systems. Which standards need to be updated? What’s the current state of that work?
A: The two big standards development organizations ASTM and SAE have four big committees, two each, that are relevant to this work. The SAE committee known as AE-10 is about high voltage, and that’s where they’re looking at all this kind of good stuff in concert with AE-40, the Electrified Propulsion Committee. In ASTM, the F44 Committee on General Aviation Aircraft is where we have a subgroup working on electric means of compliance, so things like distributed electric propulsion. I happen to be the chair of a working group building standards for distributed electric propulsion. Then there’s F39, which is the Aircraft Systems Committee of ASTM, and that’s where the electric engine standard currently exists and will be updated. FAA is involved in the process, but until the standards and regulations are updated for the emerging types of electric aircraft, manufacturers are proceeding under special conditions.
Q: What’s the role of artificial intelligence in solving some of these challenges, and how does FAA view AI in this regard?
A: Back in ’73, I was doing programming on punch cards on an IBM 026 punch card system feeding into a UNIVAC 1108 — not exactly what the kids are doing these days, but the thesis of “garbage in, garbage out” is still right on the mark. So, there is some work in large language models that could be applicable, but the challenge is that most of those large language models are kind of generic, meaning they learn from the web and not necessarily from aviation-specific material. If you could take, for instance, accident data from the NTSB, FAA regulations and materials standards — published, juried, competent — and actually train some of these LLMs, there’s an opportunity to maybe build AI systems that are competent in aviation-related matters. That could help move things forward with a minimum level of hallucination. And so there are some folks who are taking a look at that. I think that’s a neat thing. The other piece is there’s also some work looking at some digital engineering tools like model-based systems engineering, and this is something that the NASA Electrified Powertrain Flight Demonstration Project is pursuing with Georgia Tech’s Aerospace System Design Lab, where they’re looking to do a gap analysis of the regulations. In terms of FAA, we had an FAA official tell us recently that the agency is OK with you using AI in your aircraft development, in system engineering, but they do not want to see AI learning while in flight while there are people on board as passengers.
