Hi there, welcome to my talk, Experimental Aviation Risks and Reward for the Aerospace
Village DEFCON 28 Safe Mode. I decided to do this talk with a little bit lighter content
than the stuff I did last year. Since those of us involved in security and variably are
also involved with risk, I thought it would be interesting to kind of apply the risk model
to some different perspectives. Since I'm building my own aircraft, I'll talk about
some of the things that I learned, some of the decisions that you will have to undergo
if you decide to actually get involved with building your own aircraft, and hopefully
give you a new perspective on this particular segment of DII and being a maker of building
your own airplane. So, again, my name is Patrick Kiley. I'm a member of the Penetration Testing
Team at Rapid7. Previously, I've done research last year on avionic security. There's a talk
that you can look up for last year's Aviation Village, before it became this year's Aerospace
Village, where I talked about CAN bus and avionics. I've also done other research in
internet-connected transportation platforms. I have experience in hardware hacking, Internet
of Things, CAN bus quite a bit, and I'm building my own airplane. So here's kind of a quick
overview of the agenda. We're going to talk about the FAA regulations very lightly and
specifically this 51% rule. We'll talk about plans and kits, a little bit on quick builds
as well. Composite versus riveted aluminum construction, you know, just kind of the construction
materials that you can choose to build your airframe out of. A little treatise on making
changes to tested designs, talk about engine selection, avionics, and then a little bit
about the final steps after you've completed. So the FAA defines the language for what is
permitted if you're going to build your own airplane. So certified aircraft have a ridiculous
amount of regulations they have to comply with for safety to ensure, you know, consistency
and everything else. And it's one of the reasons that makes certified aircraft so expensive.
Experimental aircraft are not cheap by any means whatsoever, but they are a little bit more
advantageous. The FAA defines the rules and federal aviation regulations 21.121 section G, and then
they've got an advisory circulator that they did 20-27 where they clarify some additional stuff
because of what some of the other manufacturers are starting to do. The biggest thing is you need
to have proof, you know, you need proof that you actually, this was built by a person and not a
company. So this is basically be taking pictures of your work, keeping a blog, keeping other records
showing that you bought the components, that you put them together, something showing your progress.
The work can be started by someone else. You can actually buy somebody else's amateur built kit and
finish it yourself. Of course, you know, cabbie and emptor, you don't know what their skill set
is, so you need to have someone who knows to kind of help review it and be able to actually examine
the work up till then and see if it's sound. And then you can also have help. That's how
quick builds work. There are a lot of programs where you can actually go to a factory and do
most of the work and then you just have the instructors there to guide you so you actually
comply with that 51% rule. A repairman's certificate is a little bit different. A repairman's certificate
basically allows you to work on it after it's done. So, for example, you can work on the engine
more and wouldn't have to hire an airframe and power plant mechanic. You can demonstrate
you have the skill to do the work safely and successfully. That's what a repairman's certificate
is. It's separate from the airworthiness certificate and it's a demonstration of the
skill, not a demonstration of the work. And experimental aircraft have restrictions that
don't apply to commercial aircraft. You cannot use an experimental aircraft in a for hire manner.
But if you're just moving yourself and your family around and you're not charging for it,
you're allowed to do that. You can basically use it for pleasure all pretty much within the
limitations. When you get the airworthiness certificate, your airworthiness certificate is
going to have its own set of limitations in it and it's going to be very clearly defined
as to what you're not allowed to do with it. Beyond that, the topic is very complicated and
I would just encourage you to do the research if you want to use an experimental aircraft in a
particular manner. There's going to be someone who has actually tried and done it and the answer is
going to be out there for you on the internet. So, why would you want to choose an experimental
aircraft? It can't be just because you want to fly. If you want to fly, you should just rent.
There are many things in life that you should rent and not own. Aircraft is probably one of
them, but if you want to fly something that's a little bit more higher performance and you're an
engineering nut and like to build stuff, this might just be for you. You have to be able to tolerate
a certain amount of risk. Building something that's normally done by professionals automatically
inherits a specific amount of risk and you have to be willing to analyze that, see what it is,
accept it. It's going to be probably the most complicated thing you've ever done
because you're not just doing the airframe, you're doing the avionics, you're doing the engine.
It's a lot of different disciplines mixed in together, but you can get a higher performance
aircraft than you would be able to get for a specific dollar amount. The image that I have
here of the Long Easy, it's a very high performing aircraft. It's also very reasonably
priced. It's a two-seater tandem. One of my favorites is you'll kind of see later when you
actually see that the kit that I chose to build, it's a very solid choice for an experimental
aircraft. So how are you going to build it? There are a few different ways you can go about this.
You can buy a set of plans. I actually have a set of plans here for my kit. This is a
set of Cozy Plans. It's a very thick, bound, large bound manual that also comes with some
very large format drawings. So this is some drawings of the instrument panel using
conventional gauges. And then, you know, that's just one section or two sections of the plans
that actually cover the whole thing. The other thing that you can do, and when you buy plans,
you actually just source everything. You're going to buy all the components. For example, this one,
it's going to be all the fiberglass, all the epoxy, all the individual metal components.
You're going to buy the engine, the airframe, the avionics, and you're going to build the
airframe from scratch using just plans. It's the least expensive option. It's also the most time
consuming. The way kits work is you purchase the plans and partially completed parts, enough to
qualify for the 51% rule. They have some quick builds where a lot of the stuff is done for you,
but you still have to finish it, and it greatly decreases the amount of time that you spend on it.
But you also increase the cost. The picture I have here is actually from Sun and Fun 2017.
It's a Velocity Twin. It's a gorgeous aircraft, very high performing, beautiful inside,
canard design, which you can kind of see my preference towards those already.
And then there's another thing called builder assistance programs, where you actually go to
the manufacturer and they help you along the way, but ensure that you're actually doing
enough to qualify for 51% of the work.
And then there's different materials. So if you're going to build an aircraft,
if you want just something low and slow, or maybe a small and single seat airbag,
the Fabric Overwood is perfectly valid. The very first aircraft were made from these. You
might have some just aluminum tubes, aluminum spars, and the rest of it's going to be wood.
It also can be very lightweight. And then you cover that with a doped fabric. It's a special
type of fabric that's coated in epoxy to make it very rigid, relatively easy compared to the other
techniques, inexpensive. This was just a gorgeous one that I saw at Sun and Fun. I'm not even sure
if it's fabric covered, but it's the type of aircraft that you'll see. Probably the most
popular style out there, the one that has the most examples flying, is riveted aluminum.
The Vans RV line of aircraft are all riveted aluminum, maybe with some composite components
that you add here and there. And that's how, for the majority of aircraft history, aircraft were
built. It was aluminum that was all held together by rivets. You have to be comfortable banging
rivets, working with blind rivets, but it's very strong and light. It was only recently that
aircraft construction moved towards composite, but you'll still see a lot of riveted construction
out there for existing airframes. And now we get to my favorite composite. So this is what I chose,
fiberglass and epoxy, or carbon fiber and epoxy. You can see a beautiful carbon fiber airfoil here.
What is composite? So really, anytime you combine two materials, and when those two materials are
combined, they form a stronger material, you get a composite. Concrete is a composite material
because you have your aggregate, your sand, and your cement, and you get concrete. Sometimes,
you know, with your reinforcing structures like rebar or other structures added to it,
but it's still a composite. So fiberglass by itself is not a composite. Fiberglass,
when it's bonded with epoxy, is a composite. Carbon fiber and Kevlar, not the bulletproof
Kevlar, but an abrasion-resistant Kevlar, are also a couple of other fabric materials that
can be used in composite construction. In fact, newer aircraft like the Dreamliners,
all composite construction. 777X uses composite wings and then conventionally fabricated fuselage.
So what I chose, the Burt Rutan design, this is a little bit about those, my favorites.
He pioneered moldless composite construction. So it's a method of fabricating fiberglass parts
without a mold. You take a section of foam, and I have an example right here. So for this image,
this is a section that was actually removed from the aircraft in a building. You can see you have
your fiberglass and epoxy layer, but beyond it you have sandwiched types of foam and they're
held together with micro bubbles mixed in with epoxy to bond both the fiberglass to the first
layer of foam and then the second layer of foam together. And what you do is on this other side,
you're eventually going to get foam on this. Not for this, this is a discard piece.
And between the two, by having this space separated by the two separate skins of fiberglass,
you get much stronger construction. And you can basically take those shapes, you can
lay them out, and then you lay the fiberglass on top of it into its final design. That was what
Burt Rutan successfully did in his very easy and his long easy, and he turned that into scaled
composites. And it now is going into space with that design. He just did Voyager, the aircraft
that flew around the world without stopping, and many of his other brilliant designs. He was a big
advocate of the canard. I think some of the the work has gone away from that, but I'll kind of
go into what a canard is in the next slide. But most of his designs were canard based.
Canards are very resistant to stalls. You still have to follow weight and balance,
and you always have to ensure that it's far enough forward so that the canard stalls first.
But a canard adds from lift instead of subtracting from it, as I'll kind of show you in the next
image. This is all from Bold Method. In a canard designed aircraft, the canard always must stall
first. If the main wing stalls first, you're going to enter into a situation called a deep stall, but
it's a relatively easy thing to avoid as long as you follow weight and balance, which you're
supposed to do every time you fly anyway. You'll never get into that situation. So a traditional,
conventionally designed aircraft, you have your horizontal stabilizer that provides
a vector of force opposite that of the wing. So it's providing, in this image, 250 pounds
of negative lift that counteracts the 2250 pounds of positive lift. And then in a neutral situation,
you have 2000 pounds of weight, you know, in this particular configuration. In a canard,
the idea is that it's more efficient because both surfaces are providing lift. Both surfaces
are providing a positive amount of lift to counteract the weight of the aircraft.
Again, all this composite picture of the skin I just showed you, where you basically have a very
thin layer of fiberglass embedded with epoxy. It's actually a very friendly method to get
into. Both the aluminum and the epoxy kits, they have what are called kind of like builder's
confidence kits. You're like, I'm not going to spend $10,000 on a set of components just to
see that this is really not for me. So you can spend just a little bit and they'll give you some
epoxy, some fiberglass, and some foam and you do it. You actually go through several steps to show
that this is actually something that you're capable of. In this other image, we actually
have a canard style wing that's for a COSY that's actually up on a jig before it has the
fiberglass applied to it. So this is where I got into the selection for the plans that I
eventually chose. So Bert Rutan designed the very easy and then he does design the long easy as a
ground-up redesign of that. Long easy is a two-seat tandem aircraft. These pictures are from
Sun and Fun 2017 when I took them. And then the picture you see here is that of an air canard
that's still in partial primer. I was really interested in this one because it's the design
that I originally chose to go with. And eventually I actually licensed the set of COSY plans as well
because I just wanted to make sure everything was covered. There was some disagreement between
Nat Puffer and the guy that designed the air canard, which is a slightly roomier version of
the COSY. So just to avoid any possible issues in design, I wanted to have both set of plans
because they're reasonably inexpensive. And I could actually decide that if something didn't
give me a clear image, I figured the other set of plans would. So here's the little blurb about
making changes. One of the biggest things I learned about from just looking in the forums,
talking to people, is that any changes that you want to make can have unintended consequences.
One of the earliest things I learned is like, you know, why wouldn't you not
take the fiberglass and use carbon fiber? It's reasonably cheap now.
Well, you're really not going to save a whole lot of weight, but you are going to introduce
completely different moments of elasticity. You could actually add some vibration elements that
could cause something to shear and cause something to fail and not anticipate it.
The other big line that I heard was basically, you know, from your aircrafts actually end up
being lighter because you're not going to be able to afford any avionics to put in it for after all
the money you spent on the carbon fiber. So you'll end up with a lighter aircraft that way.
It's pretty interesting when you actually dig into it and see that, you know, maybe not carbon
fiber is the best choice for all aircraft. There are aircraft that were designed around carbon
fiber. The Burkut, which is a fast and slick version of the Long EZ, is a carbon fiber variant,
but it was designed from the ground up to have the amount of carbon fiber necessary that it has for
it. And then this is kind of a famous line from a plan. If you want to put something in your
airplane, first throw it up in the air. If it comes down, it doesn't belong in the airplane.
The designer of this was very minimalist in what he designed. He wanted this aircraft to
go without a battery. He wanted to go without a starter using just a prop start. Most people
have chosen not to do that. There are other issues and risks when you deal with hand cranking a prop.
You know, all it takes is a little slip of the foot and now all of a sudden you're, you know,
like the bad guy in Indiana Jones with a big fat spinning piece of metal right in your face. So,
no thank you. I'll pay the weight penalty and deal with the starter. And then with glass panel
cockpits, you need to have a backup battery source. You need to have a power source.
And if you're going to have a starter, you have to have a battery. But it just goes to show that,
you know, you really need to be careful when you're actually considering making changes.
Is this going to add weight? Is this going to change the wing loading? Is this going to
significantly change the performance characteristics of the aircraft? And here's really a case study in
this next example. The unfortunate accident that took John Denver's life. John Denver was
flying along easy. The NTSB actually found that the probable cause of that accident was the designer
decided to take the fuel selector valve and instead of it being in front in the area where
it's easily reachable by the pilot, he relocated the fuel valve to be closer to the fuel source.
But what that actually did is when you actually have to select it, you have to reach over your
shoulder because of the, you know, just you're kind of confined in. So you actually have to do
this maneuver to change the fuel valve. And then when he was doing that, they say he probably
unintentionally put control pressure on the rudder, causing a loss of control, causing him to
fly into Monterey Bay, taking his life. So just the thought, the designer, it's thought relocated
the fuel valve to make the fuel line shorter. By having shorter fuel lines, there could be less
risk of a fuel leak, but introduced other elements that were not necessarily reconsidered from the
fuel valve being up front. So in a long easy, the fuel tanks are behind you, the engine is behind
you. So the thought was let's put the fuel selector behind you as well. Maybe not the best idea. It's
a perfect example of the unintended consequences that you can have from actually making changes and
why you really need to take a risk-based approach to any possible changes than what the designers
actually intended on this. Because an experimental aircraft is inherently a risky activity.
That was a little section about engine selection. So when I was getting into this, I was debating
did I want to spend tens of thousands of dollars on an engine that was effectively a 50-year-old
design? And that's what they all are. All of the air-cooled engines out there are really based on
old, old technology. Pardon me. But they're designed for the purpose where they're used.
But the thing I quickly learned is that the problem you run into with automotive conversions
is, and the Subaru is a big popular one because it's kind of the same horizontally flat plane
opposed engine, you have all the cooling issues. You have to introduce some type of cooling system,
now you have a radiator, you have additional weight. You have to put in a reducing gearbox
to slow the engine speed down to the appropriate speed for a propeller.
That adds additional weight, adds additional complexity, additional points of failure.
And then you have to deal with the fuel injection system. A fuel injection system that
if you're using the automotive ECU, you may end up going to limp mode. And when you're in a car,
you can just pull over. In an aircraft, you have to find a place to land immediately if you lose
power. It can introduce an element of danger that's unexpected. And what I discovered is
they ended up not being cheaper. They ended up being heavier, more complicated, more prone to
more things that can go wrong. And because of all the changes you have to make there,
they end up not being any cheaper. So that leaves us with conventional engines. Okay,
you can do a turbine, but if you have that kind of money, you're going to be getting a super high
performance something anyway. So conventional engines, they are lightweight, they're expensive
for what you get. They are just look up IO 360 new and you'll see what I mean. Not to mention
things like an IO 540 or IO 540 turbo, you know, turbo normalized or, you know, some of the other
fancier ones. And imagine if you had an aircraft that had two of these, which you'd have to pay
for to replace them in. So there are experimental options. What that basically means, it hasn't
gone through the exact same certification process because the certification process itself is
expensive, but yet you get the same quality. So you can get an experimental version of these
engines that have the same quality for a much more reasonable price than buying something else
that would be kind of in the certificated realm. All right, avionics, one of my favorite subjects.
So when most of these designs came out, all you had choices were like, you know, your
little round steam gauges, as they call them. You have your individual airspeed indicator,
your horizontal situation indicator, you know, your compass, your turn and slip,
vertical speed, altimeter, all of your, you know, attitude indicator, all separate designs.
In a glass panel, you combine all that with a moving map, with GPS, all your controls in a
centralized point, but you're also introducing a central point of failure. So you need to have some
additional plans for that. But glass panel avionics are getting cheaper all the time. They're some of
my favorite things to mess with for those of you who've studied the previous research that I've
done. I think they're safer. I think they provide you better situational awareness. Even in, you
know, darkness, you're going to be able to see the terrain in front of you if you have a more recent
system. So you're going to be able to see that because it's going to have the terrain database
loaded into it. You're going to see that there's an antenna tower loaded and I have to refer down
to a sectional chart and manually de-conflict. The aircraft is going to tell you that there's
an issue coming up. All your approach procedures are going to be built into a database within the
avionics. But what I feel that these need are some backup gauges. Now they have electronic backup
gauges, but why not look for something that is reasonable and inexpensive and just use a couple
of must-have backup instruments as your backups next to your steam gauges. I think that's the
best way to go. All right, so a little blurb on avionics cost. And we're going to take the Garmin
G3X for example. They make certified and experimental versions of the exact same display
running the same firmware. Just it's been blessed by the FAA and received certification approval.
$99.95 for just the display. That's without any of the instruments, without your magnetometer,
without your airspeed sensors, your various electronic gyros, any of your engine instruments.
$10,000 just for the display. Whereas the exact same experimental display is $3,895.
So my opinion, best choice in avionics. What I like, the ones that I've actually displayed
them for and they're not mine. I've just done research on them. But experimental glass with a
emergency backup steam gauges is a great combination.
Okay, so now that you've done this, you've actually built your aircraft, what's the next step?
You have to get a special airworthiness certificate. So what you do is you work with
what's called a DAR and you have it inspected and you get an airworthiness certificate that tells
you you have to fly for 25 to 40 hours in a restricted area away from population. You have
fly out of a special airport that's away from everyone else. And you have to basically fly off
the number of hours to show that this aircraft is worthy to be operated on the same runways and
areas that other aircraft operate. And if you DIY this, you effectively you are a test pilot.
But there are people that you can hire for this. If you are not,
you know, this is a level of risk that you don't necessarily want to undertake. There
are people that you can hire that love to do this all the time that are very skilled,
especially maybe for your particular airframe. But effectively what you have to do is you have
to build a test plan. Same thing that test pilots would do. So the first thing you do is you fire
up the engine, you test everything. Then you taxi around in a little bit, shut everything down,
inspect everything. Then you undergo high speed taxi testing where you actually get almost up to
flight speeds and then slow back down again. And then you have your first flight. And the biggest
things I see in the plan sections are when you do this, you need to wear a parachute.
I guess that assumes you need to have taken skydiving classes, but I guess any parachutes
better than any parachute, even if you haven't been through this stuff, is better than not
wearing one if you need to get out because you've gotten into an uncontrollable situation.
But this is kind of like probably one of the most risky areas of flight. It's where a lot of the
accidents happen because something was tinkered with, there were unintended consequences, and
that's where disaster struck. So thank you. I think we're gonna have some Q&A after this,
so we'll cut over to that. And please feel free, I look forward to your questions.
Thank you for listening to my talk.