This Is The Holy Grail Of Rocket Science
By Y Combinator
Summary
Topics Covered
- Reusable rockets are the iPhone app store moment for space
- Solving re-entry for stage 2 rockets unlocks true reusability
- Iterative hardware development drastically cuts costs
- Software is the bridge from garage to FAA oversight
Full Transcript
The team at Stokepace is going after the holy grail of rocket science. Stoke is
building fully and rapidly reusable rockets with aircraftlike reusability.
Today, only this section of rockets can be reused. But Stoke Space is focused on
be reused. But Stoke Space is focused on the entire rocket, including a stage 2 capsule that will finally be able to survive the brutal re-entry to Earth.
This could open the door to all kinds of new opportunities in space. I think if you have something that can go up and come back to the place where you want it
to go when you want it to go, I think it is like the iPhone app store moment.
People are going to come up with like absolutely crazy ideas for how to take advantage of that.
>> So, how did two founders go from testing rocket engines in their backyard to solving for one of the hardest problems in space travel? This is the origin story of Stoke Space.
Andy and Tom invited me to check out their headquarters in Kent, Washington, just outside of Seattle. It's an area that's quickly becoming known as a hub for space tech startups.
What are you building at Stoke Space?
>> Building fully rapidly reusable rockets designed to go to space, through space, and back from space. uh really to lower cost, improve availability, and uh
improve reliability of access to space.
>> Today, there are only around 150 commercial space launches every year.
While that's a record for the industry, it's a drop in the bucket compared to where things could go if rockets become fully reusable. With only 150 potential
fully reusable. With only 150 potential transactions and most of them getting taken up by Starlink, there's just not that much availability. And so there are customers who are willing to pay quite a
bit more to get the availability. But to
make the space economy more ubiquitous, more diverse and to enable some of these new verticals and and applications yet cost is a huge barrier.
>> Describe the hardware architecture itself. Can you walk us through the
itself. Can you walk us through the designs of the Nova and the Andromeda?
>> Nova is a two-stage to orbit rocket.
First stage functions similarly to other rockets. It punches you out of the
rockets. It punches you out of the atmosphere and comes and and restarts the engines to land either downrange or back at the launch site. It has one of the highest performing rocket engines in
terms of fuel efficiency that's ever been produced. That was a big technical
been produced. That was a big technical bite for us to bite off as a small company, but I think it's one that is paying off and it's a very important component to long life and rapid
reusability. The second stage goes the
reusability. The second stage goes the rest of the way to orbit. Typically,
rockets are thrown away completely. In
the last few years, the industry has shown the ability to reuse the first stage, which has been transformative. It
allowed companies to scale from maybe a dozen or so, 10 to 20 launches per year now to that 150 number. But the second stage is still thrown away on every single mission. That's in large part
single mission. That's in large part because the stage 2 capsule is traveling at 17,000 mph as it drops out of orbit.
Because of that high speed, it eventually breaks down as it heats up to more than 2700° F during its descent.
>> So, these are multi-million dollar machines that not only have to get built, but they have to get tested and acceptance tested to prove flightworthiness, and then they go to space and they get thrown away. Here's
how their stage 2 or upper stage Andromeda capsule is able to survive re-entry. After releasing its payload,
re-entry. After releasing its payload, it'll close up and re-enter Earth's atmosphere. At that point, the custom
atmosphere. At that point, the custom heat shield takes effect. It uses cold liquid hydrogen flowing through a heat exchanger to absorb the extreme heat of
re-entry. Then 24 small thrusters kick
re-entry. Then 24 small thrusters kick in to help it slow down and land at the right angle. What rapid reusability
right angle. What rapid reusability allows you to do is to scale the flight frequency without having to scale your factories and your test facilities and all the infrastructure that comes with it.
>> It seemed like that was the thing that you thought needed to exist in the world. And if somebody else wasn't going
world. And if somebody else wasn't going to do it, you needed to do it.
>> Yeah, that's right.
>> How did you get interested in building rockets? Was this something you were
rockets? Was this something you were always interested in as a kid growing up?
>> From as far back as I can remember, you know, doing Estus rockets in the backyard with my cousins and it really just kind of took off from there. Pun um
not intended. From pretty early on in high school, it was pretty clear that I wanted to do propulsion. It was like, let's do engineering. What's the most badass thing you can do with engineering? Well, it's probably like
engineering? Well, it's probably like rocket propulsion.
>> Tom and Andy met while working as jet propulsion engineers at Blue Origin.
They left to start Stoke Space in 2019.
>> How did you guys uh ultimately decide to to leave and and start your own company together?
>> Pretty quickly, it became like, "All right, Andy and I are going to go start a company." And at the time, we had no
a company." And at the time, we had no idea what we were going to do. We were
like, "All right, let's start a company.
let's brainstorm some things that sort of like meet in the middle of our ven diagram of like desires and skills and problems that need to be solved. And
then it it was like it kind of clicked one day that like there is this problem you know people are still throwing away rockets and like we really didn't see anybody actively working super hard to like you know attack that problem with
rigger combined with the fact that we thought we had a pretty good idea on how to solve it and it was like all right that's a thing. There were more than 150 rocket companies at the time that were around and I assume you looked at all of
them. What was the thing that convinced
them. What was the thing that convinced you that you should start your own and that it would actually be better and out compete what they all were doing?
>> It was not intuitive to start a company.
It wasn't the first inclination. We
looked at a lot of those and a lot of it was, you know, a bunch of PowerPoint stuff that would that didn't seem like it had a very bright future and we were certainly both hardware people and so we
kind of understood what it would take to take some of those PowerPoint ideas into reality. Their key insight was that if
reality. Their key insight was that if you focused on reusability from day one, you could decrease cost and increase availability and reliability. But even
though they believed in the idea, it was still a massive risk. At the time when you were starting this company, you you both obviously had very comfortable jobs. You had small children. It seems
jobs. You had small children. It seems
like it's a huge decision to go and actually decide to make the leap and start a company and start from zero and nobody else in the trenches except for you two. It felt wildly irresponsible to
you two. It felt wildly irresponsible to be honest. Easily the hardest decision
be honest. Easily the hardest decision that I've ever made personally. I had a 3-month-old at home and I had a really good paying job that I was quitting.
Let's give ourselves a timebounded, you know, scenario where like in 6 months if we don't have any kind of legitimate traction, that's going to be the sort of cue to re-evaluate. And cuz we really didn't know anything about what we were
doing when it came to starting a company. And so, you know, a lot of the
company. And so, you know, a lot of the first few months was just like a was heads down engineering and b it was like what does it even mean to start a company? We started in in September
company? We started in in September 2019. So, it wasn't long before like the
2019. So, it wasn't long before like the pandemic was a real thing and then we were like really like, "Oh like was this the right move?" It was like right up against the clock on the
six-month uh thing that we finally, you know, were able to get some traction. I
think we got our first check sometime around that time frame and then we were off to the races.
>> With that first check, they started building a prototype engine and testing it in a shipping container in Tom's backyard. So, at that time, Andy's in
backyard. So, at that time, Andy's in the basement powerpointing like crazy to try to like hone the story and I was like welding this steel structure in the garage and then taking it out to the
shipping container and bolting it in.
And that was kind of like how we spent our days for the better part of like 2 months. We knew we wanted to demonstrate
months. We knew we wanted to demonstrate some hardware. We knew that with a
some hardware. We knew that with a reasonably small amount of money, we could develop a pressure-fed gaseous hydrogen liquid oxygen thruster and be able to show fire coming out the end and
say, you know, hey, early investors, we want to do this thing. Here's a teeny tiny little piece that we already did.
>> All right. So, these are the shipping containers where it all started.
>> This is where it started. So, this one was primarily for storage. This one
>> was where the actual test cell was.
>> Yeah. Walk me through this one.
>> The test stand. You can still kind of see like a imprint of it there. Oxidizer
stuff on one side, fuel stuff on the other side. Actually, the original
other side. Actually, the original workbench that I built probably the very, very first thing I ever built for Stoke is still out there. I'm pretty
sure >> we had our what we called our run tank was really a cylinder. Then we had storage outside of the wall.
>> Bunch of valves along the wall. Bunch of
valves sort of right at the test article there. Yeah, we probably built this
there. Yeah, we probably built this entire facility in 2 months, maybe. I
remember distinctly a day in January where I remember having a feeling like, okay, we're convinced now. What do we do?
>> What led you to IC?
>> We wildly underestimated the challenge in raising money. If there's one thing we totally missed on that was it. We had
no rolodex. We had no, you know, rich uncles. We had to do it all organically.
uncles. We had to do it all organically.
Right when we started to raise money, co hit. The market shut down. Nobody was
hit. The market shut down. Nobody was
entertaining anything except trying to keep their portfolios alive. So, it was a really tough environment. YC obviously
has a network second to none. They can
help us learn to, you know, speak the language and understand the fundraising process and the whole path and that's the reason to do it. YC was a big part of that early stage and I think YC also
added some momentum to the round. Right.
>> I'm curious, how did you get good at fundraising? A lot of the venture world,
fundraising? A lot of the venture world, especially at the time, was geared towards SAS. We have a a very different
towards SAS. We have a a very different story and a very different financial model than a SAS business. And so, at that early phase, my goal was just, hey,
get get introduced to a different angel who may be a little bit closer to investing in a hardware company. There
was one piece of advice that was extremely important and I carry with me all the time and I will pass on and that is get get good at hearing no.
It is tough to keep it going when all you hear is no after no after no to just keep the conviction and keep the fire alive to keep going.
>> And how much have you raised so far to build all of this?
>> We've raised about $990 million to date.
We've certainly not spent close to all of it. I think we're operating very
of it. I think we're operating very efficiently compared to some of the other groups in our industry, but it is a, you know, it's it's been a a journey
to get to this point. Over the past few years, they've hit several major milestones. In 2022, they completed
milestones. In 2022, they completed their first full engine test. In 2024,
they moved into this 168,000 ft factory.
Next up will be their first launch into orbit which is planned for later this year.
>> We have engineering space uh here as well as manufacturing. We build
everything from avionics and electronics. Every part of the rocket
electronics. Every part of the rocket gets built in here. We have engine assembly over to our left. We have
additive manufacturing to our right. We
have structures behind us. This is where it all happens. Facility is designed to build about seven vehicles per year.
How do you balance the speed of execution with the patience that's required for this deep technical work and and building something that has a long time horizon?
>> This is very tricky. I think you have to know that this is hard. You can't
analyze everything to perfection and at the end of the day rubber meets road and you have to test and so the speed at which you can iterate becomes
fundamentally important to your ability to do the hard thing as quick as possible. You do have to plan for
possible. You do have to plan for failure. So when for example if you're
failure. So when for example if you're developing a rocket engine and the engine fails, how soon can you learn the lesson from that one? Make any change that you need to change and get the next one out there? There better be one on
the factory floor basically ready to go if it's going to be fast, right? We want
the ability to make every part on the rocket ourselves. As you learn more,
rocket ourselves. As you learn more, your overall development timeline and therefore your overall development cost is tied to how quickly you can iterate.
And if you can't make parts yourself, then your iteration cycle is dependent on a out of house supplier. We'll go
learn something on a test stand, tear it down, drive it back here, pull it apart, run a change in the machine shop, drive it back out there, integrate on the test stand, and we're off and going. So, a 1
month cycle is down to a day or two.
>> Talk about the importance of software in building this company. This was
something that was surprising to me is how key it is, not just to the actual product that you're building, but to how you run your company. This company has to scale from building things in a
literal garage to flying government payloads or even humans on a vehicle that is overseen by the FAA. And somehow
you have to bridge from garage to that.
And that bridge is often very very painful for a lot of companies. In our
industry, we build a rocket and ship it and then that's it, right? It launches
and it lands in the ocean and you never see it again. But what we want to build is a vehicle that's going to go to orbit, fly around, and come back again.
And and we want to turn that thing around as fast as possible and go again and again and again. So, did it work well? How long have the parts been in
well? How long have the parts been in service? When do I need to go and do
service? When do I need to go and do preventative maintenance? When do I have
preventative maintenance? When do I have to go do unscheduled maintenance on things? All of these questions are kind
things? All of these questions are kind of staring us in the face. We're in a moment in time where software can automate a lot of that. I'm really
excited about, you know, new developments in AI and other things to make it even more seamless for a factory worker to do their job and also log
answers to those questions. And then you use software to abstract the information, slice it and dice it and send it to the different functions that need it. So, we made the decision very
need it. So, we made the decision very early on, we're going to build our own tool to to do all this. We're going to base our operations around it. We call
it Bolt Line. That's been a big part of our success.
In addition to their manufacturing and testing facilities, they're building a launch site at historic complex 14 in Cape Canaveral, Florida. The same place where John Glenn launched to become the
first American to orbit Earth back in 1962.
We're working on putting all the pieces together to get to orbit. We have a massive construction project happening in Cape Canaveral and that is going fantastically well. We have our first
fantastically well. We have our first and second stage engines which are completing their development and qualification. We have the structures
qualification. We have the structures builds that are happening right now.
We've built multiple structural qualification articles. We have a
qualification articles. We have a facility in Moses Lake where we go and test those from a structural perspective and also at cryogenic conditions.
Software and avionics is another big area where you know we have hardware in the loop testing where we have all the avionics for the vehicle turned on integrated together with the flight
computer and with the GNC software and so we're flying simulated missions all the time proving that the avionics and the software stack are robust and ready to fly.
>> What does the world look like if you're successful?
>> If we are successful we are flying this thing every day. What was the moment that you guys actually started to believe that you could pull this off and that your crazy plan would work?
>> The very beginning was convincing ourselves that this was an idea worth pursuing. I got to the level of
pursuing. I got to the level of conviction in the idea that I said we owe it to the world to try this idea.
Whether or not it succeeds, this idea is good enough. It has to be tried.
good enough. It has to be tried.
Loading video analysis...