AJ Piplica joined me on the Acquisition Talk podcast to discuss transportation at hypersonic speeds. He is the founder and CEO of Hermeus, which is developing the Quarterhorse aircraft intended to fly at five times the speed of sound. It’s based on a turbine-based combined cycle engine which uses the off-the-shelf J-85 jet engine from GE. When the J-85 accelerates to supersonic speeds, a proprietary ram-jet kicks in that uses the aircraft’s forward motion to compress air, combust it with fuel, and accelerate to even higher speeds.
Hermeus has already shown that a lesser jet engine designed to fly at Mach 0.8 at 26,000 feet can be outfitted to fly Mach 3.2 at 60,000 feet in a test facility. That’s closing in on the top speed of the famed SR-71! The first Quarterhorse prototype using the J-85 is expected to fly a couple hundred nautical miles, but the eventual transport aircraft will fly at Mach 5 for 4,000 nautical miles at an altitude of 90,000-100,000 feet.
Download the full-text transcripts.
One of the many interesting technical accomplishments at Hermeus is the pre-cooler. There’s a “valley of death” of sorts between Mach 2 and Mach 3. The speeds are beyond the limits of the turbine engine but less than what’s needed for the ramjet to fully kick in. The air entering the gas turbine engine at Mach 3 is over 800 degrees — far too hot for the materials to handle. Hermeus’ pre-cooler cools the air down by 675 degrees in one-tenth of a second so that the turbine engine can continue to provide thrust. Once the Quarterhorse accelerates past that regime the turbine engine will be bypassed and the ramjet does the rest up to Mach 5.
This capability isn’t just about accelerating civilian transportation by reducing frictions to travel and trade. It has clear military applications. AJ points to survivable airborne ISR for the first use-case as the military environment shifts from permissive to denied. Even Iran knocked out an MQ-4C Triton back in 2019. A hypersonic reusable aircraft would be hard to see coming, and even harder to knock out. All of North Korea could be imaged in 30 minutes, for example.
Since the 1970s, however, the US Air Force has focused almost entirely on stealth (low radar cross section) as the primary means of survivability. Speed, altitude, and maneuverability are other knobs that can be turned in that equation. These knobs become more important as radar catches up to stealth.
The Air Force is hedging their bets. Hermeus was awarded the largest STRATFI award yet, with $30 million coming from the Air Force including $15 million from PEO executive airlift, and a matched $30 million from private sources. In the episode, AJ talks about what his company had to do to win the contract and provides advice for other companies trying to transition into defense programs.
Contracting with DoD
I was interested to learn how the Air Force contract was structured for Hermeus’ dual-use technology. Here’s AJ:
The way the program is structured, all the design development, fabrication, everything, that’s all happening on private funding… Within the $30 million government contract, the things that we’re delivering are test reports, test data, test plans, and things related to the ground in-flight test activities that we’re doing.
That was I think a really important setup for how that contractual mechanism worked. Being able to design, develop, and fabricate outside the FAR [federal acquisition regulation] was huge. That would have had a very significant impact on our business. Even with a FAR part 12 [commercial] contract. The key deliverables that we’ve got are yet focused on the ground of flight test activities that we’re doing on the engine, on the aircraft. And then of course, as we put them together and go and try and get up to Mach five and get back.
I think if you compare the results that will come out of this program for the air force to previous hypersonic flight programs it’s pretty valuable effort. You’re probably an order of magnitude lower cost than any set of hypersonic flight test data for a new air breathing vehicle like this in the past. I think the X-51 was $350 million or so.
That contract structure makes a lot of sense. It avoids or at least defers complicated questions of intellectual property rights should government fund product development itself. And the government is gaining valuable test data at a relative bargain. AJ noted that the United States has only about 15 minutes of data on air breathing aircraft traveling above Mach 5. Ultimately, R&D is “a search, a process of discovery… R&D is not intended to buy airplanes or missiles; it buys knowledge.” This is something RAND economists understood back in a 1958 paper.
Another reason for the contract structure is to avoid certain government requirements flow downs that affect their ability to do business:
There are lots of suppliers out there that we work with on a regular basis who can manage export control information, but have never been on a government contract before. So if I’m going to go flow down a FAR clause to a machine shop that not taking up a huge amount of their time, it’s going to be difficult to effectively do business on a schedule.
I’m also guessing another reason for structuring the contract that way was because it allowed Hermeus to retain design independence. AJ explained how the company’s driving principles include: (1) vertical integration, because its hard to draw a line where the engine stops and the aircraft begins; (2) a focus on hardware; (3) quickly pushing to test; and (4) iteration.
These principles can often be difficult under a government R&D contract. When people think of a proper situation for traditional “waterfall” acquisition, they often point to nuclear submarines and hypersonic vehicles. Hermeus shows that there are always exceptions to the rule. For example, they built a ground test facility for the Quarterhorse in a month using cargo containers and completed 50 engine tests over the next four weeks.
Recommendations
AJ provides a number of important recommendations for deep tech firms looking to work with government. First, find an advocate in the government who can help you through the process. If you’re responding to RFPs without having done the early leg work, then it’s probably too late. Establishing credibility is a crucial prerequisite for turning that potential customer into a real advocate. For privately financed companies, this means demonstrating capabilities and not just brochuresmanship. Here’s AJ:
You have to continuously make progress along the way so that you’re continuously updating and breaking their mental model for what’s possible. With the first engine that we built, we had met with some DOD folks before we were even funded and were like, “Hey, we’re building this turn-based combined cycle engine. We’re going to have it done in a year.” And they were like, “We’ll see, come back with the data.” Nine months later, we did it, knocked on their door, showed the data. They’re like, “Cool. You guys do what you’re say you’re gonna do. This is refreshing.”
A second crucial task for deep tech is to solve customer problems along the way. Early revenue helps sustain companies without the loss of equity while also making it easier to raise funds when needed. As explained above, providing the Air Force hypersonic test data is a good example. Scaling goes a lot easier when companies are continuously integrating customer requirements.
Moreover, this is test data that Hermeus would have to generate anyway, so dual purposing effort is important. AJ provided a different example from SpaceX developing reuseable launch vehicles. All the tests required to successfully land the vehicle were essentially “free” after getting a customer’s payload to orbit.
A third recommendation AJ made was to get good accounting and legal teams that understands government contracting/compliance. There are a lot of implications to each FAR clause, so understanding what that means to your business is important.
A final insight was on AJ’s view on cost plus contracting. Often, these contracts are viewed as an incentive to cost maximize by venture-backed companies and come with several accounting system requirements. AJ says he’s not opposed to cost plus, but would only accept one if it either: (1) accelerates schedule; or (2) reduces risk.
Improving Acquisition
AJ made a couple important comments on how defense acquisition can be improved: (1) budget flexibility; and (2) price analysis. Since PPBE reform is in the air, I’ll start with by quoting AJ on budget flexibility:
I don’t know how many times I’ve heard, “We got plenty of O&M funding, but I wish we could squeeze this into an R&D role.” I think if you’re able to quickly get tens of millions of dollars on contract and ensure that you’re delivering, that can be quite game-changing.
AJ recommends raising the reprogramming threshold, currently at $10 million or 20% of a program element, whichever is less. New starts are only allowed for up to $10 million for the entire life of the program! In my view, consolidating budget line items into coherent portfolios is another way of getting there. But some mechanism that doesn’t take between two and five years to get funding over $1 million is crucial for the efficacy of government interaction with nontraditionals.
Pricing can often be a challenge as well, especially for deep tech companies that are in advance of established commercial sales. In those cases, government often requires a formal cost-based buildup to justify pricing. Here’s AJ again:
I think if I were king for a day, I would definitely expand the ability of contracting officers to leverage pricing justification over cost justification for things like this. Because the analogy that I just gave, if you just look at what the government paid for the X-51 program, at the end of the day the products were test data. Yes, they built a few aircraft or a few vehicles. But they were expendable… So I think you can just do that, look at those two things and say is this for $30 million a good value or not?
There are always creative ways to get to pricing without relying on accounting costs — which, by the way, can be quite misleading even with all the cost accounting standards in place. At the end of the day, price should reflect value rather than input costs.
Thanks AJ Piplica!
I’d like to thank AJ for joining me on the Acquisition Talk podcast. Follow him on Twitter @AJ_Piplica. You can find a lot more about Hermeus at their website. There’s plenty of good articles about the company, including at Forbes, Defense News, Flight Global, Janes, Popular Mechanics, and much more. They have a number of good videos on YouTube, including on building Test Site 27 in just a month and a video featuring board member Will Roper.
Leave a Reply