 to meet some new friends and connect with old friends and I just want to end the day by telling you some stories about our lights. It's kind of what I like to do. I've been an aviation geek for literally as long as I can remember. It probably started when I was eight or nine years old. My parents took me to an air show at Dias Air Force Base in Alley, Texas. And the Air Force Thunderbirds, that F-16 demonstration team, was the featured attraction of the day. And if you've ever seen them with Blue Angels, you know how astounding they are. They fly these amazing hyper-phonetera craft with very close proximity to each other and managed to not run at each other, which is a pretty amazing thing. And I was blown away by it, but as blown away as I was by that, the thing that really stuck with my imagination was standing nose-to-nose with this amazing machine, the SR-71 Lacker. And you just look at this plane and tell how fast it needs to go. It's got these razor sharpening edges, these curves, the engine themselves are almost as big as the fuselage. You can't quite tell that in this picture. But standing nose-to-nose with this airplane started a lifelong obsession with planes. This was pre-internet, so I went back to my elementary school and at the school library and pulled every book she could find that mentioned the SR-71 for me. And spent two or three months reading about this plane and others like it. Years later, my career has taken a decidedly non-aviation turn. I am now VP of Engineering at a company called Move Health. And we're working on software that's going to change how major procedures like me and the other placements are paid for in this country. But I'm so fascinated by the world of airplanes and everything having to do with aviation. And every once in a while, I run across an aviation story that lends a wisdom to how I practice my craft, how I build my teams. And this story today of a series of planes built by even more amazing group of engineers is one of those stories. So if you see an SR-71 in the museum, if you look at the tail sometimes, sometimes you'll see this little, not always, but sometimes. And the reason that that cute little scum is on the tail of the SR-71 is that this plane was designed and built by Lockheed Martin's Advanced Projects Division. Better known as Stump Works. Now companies love throwing the term Stump Works around. Anytime you have a project where you need rapid innovation or you need to keep something secret from the CEO, you set the Stump Works. Lockheed Martin's was the original. This is where that term comes from. And today I want to tell you about three of their most iconic planes and the stories of how they were built. And to do that, I've got to start by telling you about this plane. Clarence Kelly Johnson, because without Kelly Johnson there would not be a Stump Works. Now Kelly graduated from the University of Michigan in 1932 with an Aeronautical Engineering degree. He applied for work at Lockheed. This is when Lockheed was about a 10th anniversary company. And they didn't hire him because he didn't have any experience. So he went back to the University of Michigan. Got his master's degree in Aeronautical Engineering. And while he was there, he happened to be fortunate enough to work on this plane, the Lockheed Electron. Now at this point University of Michigan was one of the few places in the country that had wind tunnels and were actively testing aircraft in their wind tunnels. And he was fortunate enough that his professor didn't want to work at Lockheed. He decided to work on this project, testing this plane in the wind tunnel. Well that was the connection Kelly needed. And after graduating with his master's in Aeronautical Engineering, he got a job at Lockheed. Not as an Aeronautical Engineering, but as a tool designer making the 83 bucks a month. Not what he wanted to be doing, but it was his way to get in the door. Well as he was testing this plane as he was working on his graduate degree, he and his professor had different opinions about it. His professor thought it was perfectly stable and ready for production. He disagreed. He thought that it was particularly unstable and pitch. And once he started working at Lockheed, he found an opportunity in time to talk to Hall Edward, the chief engineer at Lockheed at the time about this. Hall actually considered firing him on the spot for his ordination. Because who was this young wearer who should happen to tell him that this plane that they had banked their company on? It was unstable. But he wanted to start looking at the wind tunnels. And the thing that Hall Edward quickly realized is that Kelly Johnson was right. The plane was unstable and pitch. And so for his comeuppance, he sent Kelly Johnson to go retrieve the wind tunnel wall of this plane, shut it back to the station wagon, drive to the University of Michigan. 73 generations in the wind tunnel later, this is what he came up with. I can swipe back and forth and you'll notice the difference in the tail of the planes. That was what it took to cure the instability of this plane. This plane, the Lockheed Electro, wanted to be a great success in the early days of commercial aviation in this country. And because of his work on this plane, Kelly Johnson got his promotion from tool designer to aeronautical engineer. He designed a series of planes leading up to World War II, the most famous of which he's grown in Urdo, Milwaukee, P-38 Lightning. If you've ever been to a World War II museum, you've almost certainly seen this plane. It's one of the most famous planes in World War II and it absolutely dominated the skies of Europe early in the morning. The pilots loved the flight. It was great at dog fighting, very maneuverable, had plenty of speed. And it was in the lead position until the Germans started flying this plane, the Messerschmitt and the E-262. This plane has the distinction of being the first jet aircraft ever deployed in combat. And it was way faster than anything that the Allies had. The Germans had started investing in jet propulsion way earlier than anybody else in the world and were much further in the deployment of that technology, so they were able to get this plane in combat. Well, lucky for the Americans, the British had been working on a jet engine forever. But because the UK is smaller, it doesn't have a production capability for the US, the British reached out and asked if the Americans would like to license the 1-H-1-B Godwin engine, one of the first jet engines that the Allies produced. The Air Force approached Lockheed about this project and asked them to build a one-off prototype of a plane around this engine. Lockheed at the time was busy doing this. This is the P-38 assembly line in the middle of World War II. This is what all of their facilities look like. They were trying to build these planes faster than the Germans could shoot at. And they were making plenty of money doing it. So there was a question of innovation in Lockheed's work to take on this one-off prototype, especially when the Air Force asked for it to be done in 180 days. They figured it was tremendously risky. But both Hulk, Edward, and Kelly Johnson had been wanting to get into jet propulsion. They both believed that it was the future of innovation. Beyond that, Kelly Johnson had been bugging the frats of Lockheed for a long time to let him set up an experimental aircraft division where he could put the designers and the fabricators and the mechanics all in one roof to work closely together outside of the regular bureaucracy of the Lockheed organization. And this was an opportune time for them to do that. So, Hulk, Edward convinced the frats of Lockheed to let him do that. And he gave this project to produce one plane in 180 days to Kelly Johnson. The first thing Kelly had to figure out, though, was where to work. Because like I said, all of their factories look like this in the middle of World War II. So he did what any of us would have done. We're going to service them. He had desks installed. He had phones installed. He even had an air gauge. Everything you need to make it an office, he did that. And he set it up next to a factory on Lockheed grounds. The factory that he set it up next to was a plastics production facility. And apparently, it smelled terrible. So the SB-80, as this pointed out, it doesn't make the project was a secret. And they especially didn't want any of the engineers revealing what they were working on when they answered the phone. So, Kelly Johnson gave out strict instructions not to answer the phone anything related to the SB-80 for jet aircraft. Irv Culver, a structural engineer on the project, who was known as a bit of a cut-up, took to answering the phone. It's kind of works very fair. And it wasn't long before this kind of works long. We started catching on with other engineers at Skunk Works. And the name stuck. So if you've ever called a division of your company Skunk Works, it's because Kelly Johnson set up a circus tent next to a smelled plastics factory. So the contract with the SB-80 was signed on June 24th, 1943. And this started at 180-day clock. The only concrete information they had about this plane that they were building was the dimensions of the plane. They didn't have the engine at this point. They were still waiting on it to be shoved over for the UK. So they built them off the engine and designed the plane around that. Normally when they started on the plane, they would build a mock-up on wood so that they could test part fitment and make sure they built all parts in the way that it could be repeated. Well, they would just build one of these. They didn't need to do that. So Kelly Johnson decreed that this plane that they were building would beat the mock-up itself. And the engineers were free to design parts to fit on the spot, to just stick on this plane. He also reduced the formality of the drawing approval process. So normally in Lockheed, they had a very strict process that they followed for making drawings and approving drawings. They had strict style guides that they followed. Kelly decided to do a lot. They had 180 days to put together a plane. He was going to have to test some corners to get it done. So he told his engineers that as long as your drawings convey the meaning of what needs to be built, I don't care what they look like. Just make sure the person you're giving them to can understand. That was enough. That's how they had this plane built. It worked. So by November 13, they were done 143 days after they started. They had complete aircraft. They took a part. They shipped it out to Muran Air Force Base in the middle of Mojave Desert. Why did they do that? Well, they had never built a jet aircraft in the United States at this point. They weren't exactly sure what was going to happen when they tried to take off. They wanted to make sure there was plenty of room in case things went sideways. Lucky for them, they didn't. Shortly after New Year's Day, the X-P80 took flight for the first time, and it flew like a dream. He would actually want to be the first American-built plane to fly over 500 miles an hour in double flight. And the production version of it, the Lockheed P-80 shooting star, would go into be the first jet ever deployed by the Air Force in combat. Not only that, but it would fly well into the 80s. So this plane that's got work slacked together in 143 days would go on to be in service for over 40 years. Pretty good design. About this time, World War II was wrapping up, and it was unclear as snow works just got happening to us to work on. The U.S. was in tremendous debt from the war. There was very little desire to build a plane, and there was no threat to the counter. So Kelly's Bandits expanded and went back to their places around the Lockheed Factory. Stopped building planes and weren't sure they were going to get to do it again. That didn't last for long. This picture is of Winston Churchill, FDR, and Joseph Stahlman at the Alta Conference. The Alta Conference was one of the three conferences the three Allied powers had after World War II to decide how the hurricane continent was going to be split up and go. It was actually at Alta that the decision was made to split Germany and to split Berlin. These three superpowers had united against the Axis powers during World War II. But they had such different ambitions for the European continent that as soon as the war was over, they started getting a little across with each other. And the U.S. and the Soviet Union began spending on military facilities and military but with a rapid pace trying to keep pace with each other. And we had entered the Cold War. While it wasn't just military spending that increased, espionage activity went way up as well. Around this time, something around 55% of the American population thought that they were more likely to die in thermal nuclear war than old age. And they went too far off the mark. The Russians and the U.S. had both built tremendous nuclear arsenals. Under the theory of mutually assured destruction that if we disbilled enough bombs, we'll both be too scared to ever drop them and that will keep this tension contained. In order for that to happen, each side desperately needed to know what the other was up to. And the Americans were interested above everything else in this place, the Boston Yard. Boston Yard is Russia's primary missile development facility, something that can't do area 51 in the U.S. And it was very heavily defended. The Air Force considered an overflight of the Boston Yard. But the side that was too dangerous because of how heavily it was defended. After a lot of lobbying, though, the CIA was able to get the Air Force to attempt an overflight in this plane, the Martin V-57 Canberra, which is actually a bomber. So they got in this plane. They took everything out of it. They possibly could try to make it as light as they could, put a camera at it, and flew it at about 50,000 feet, which normal operating altitude for this plane is somewhere in the high 30,000 to low 40,000 feet. So they got an extra 10,000 feet, but it limped back to base having taken any aircraft fire 12 different planes. Somehow made it back, but they decided they would never do that again. So they needed a different answer. Intelligence of the day indicated that Russian radar was blind over about 65,000 feet. And so the CIA, knowing this, decided what they needed was a plane that could fly at 70,000 feet. If they could just do that, the Russians wouldn't understand that. So they asked for this, but if they had no means of reconnaissance until they got this plane, they needed this plane fast as well. Well, Scumworks Group proposed a refactoring of an existing design, the F-104 Starfighter. This is a plane that Scumworks had built early in the Cold War. It was the first American plane capable of going wild, too. And what they thought that they could do with this plane, because it was a pretty light plane already, they decided that they could change the inch out, lengthen the wings, and reduce weight, and come up with something that would fly at 70,000 feet. Because their proposal was based on an existing plane, and along with the body part that they produced, and the PAE and the headphone and horn output that they built, both of those planes that they built into production. Their proposal went out. The plane they were building, of course, was the U-2. So the team started working on the U-2 in November of 1954. The project was so secret that the kickoff payment for the project actually went to Kelly Johnson's house, and the fall of a $1.1 million cheque addressed the C&J engineer. C.J., of course, being Clarence Johnson's initials. It took the F-104 fuselage of Mabel, the thinner. They made it out of wafer thinner only. So thin, in fact, that when a worker accidentally bumped into one of these things with a toolbox, it left a four-inch dent on the side of the plane. Normally, if you bumped into a plane with a toolbox, you might stop and paint, but you certainly wouldn't want to leave a four-inch dent on it. People at Sky Works were wondering if this plane was even going to be strong enough to fly. Eight months later, though, right on schedule, they loaded the crate of the plane up into the belly of a C124 cargo plane, and they flew out to a purpose-built airfield in the middle of the Nevada desert, right in the middle of a bunch of dried-out leg beds. Again, they weren't really sure how this plane was going to behave, and they picked an area of dried-out leg beds because there's lots of potential timelines out there. On the one side, there's lots of places you can land. If I show the two planes together, I hope so, if you don't want to see it. If I show the two planes together, you can pretty clearly see the plane on the edge, especially from the wing forward. In this picture taken by Kelly Johnson himself, is it the actual first flight of the U-2 on August 4th? Just a hair over eight months from when the first metal of the plane was cut. A month after this first flight, test pilots were breaking altitude records almost daily within Nevada desert. By the end of testing, the plane had been up to 74,500 feet, the highest any point had ever flown, and it had flown over 5,000 miles for 10 hours on a single-time gas. So it met the operational requirements to fly over Russia. Despite the ability to fly three miles higher than any other plane ever built, the U-2 is really a remarkably simple plane. Wave was everything on this plane. Every count cost the plane approximately one foot of altitude. This is a picture of the inside structure of the wing. This wing weighed about four pounds per square foot. Most aircraft wings of this state weight about 12 pounds per square foot. So it's about a third as heavy as a normal aircraft wing. Looks kind of like a sheet metal on the inside. And the problem with this wing of that U-2 earlier is it doesn't have a lot of rigidity. It's so light that one of the constant complaints from pilots is that when they would hit turbulence, the wings would flap like a seagull. And they were afraid they wouldn't break off. And they were dead. It didn't affect mission viability at all. Another interesting thing about the U-2 is it was designed with tandem bicycle style and income here. This is pretty common on gliders, but not on jet aircraft. The combined weight of this landing mechanism of the front back wheel is 200 pounds. It's the lightest landing gear that's ever been deployed on a jet aircraft. And it's easier for me just to show you how this works. I've got a video of a U-2 landing. So you can see the plane coming over. It's actually on a chase car in this video. Because the pilots in a bulky pressure suit, they actually can't see the ground when they're coming in for landing. So they have a chase car behind it that's calling out aptitudes. Two feet, one feet, six inches. They finally get on the ground. And you can see the pilots literally flying the plane down the runway, balancing it on the two wheels. And so it goes down the runway, finally bleeds off enough speed, and tips the plane over onto its wing. Then they set this crew of guys out there to start pulling on the wing and try to put a logo gear under the other side. So they've got these removable landing gear that go under both wings. And once they get stopped on the runway, they literally go out and four or five guys hang on one side of the plane, and one guy puts this logo gear in on the other side. It actually uses the same logo gear to take off. So when it takes, you can see right there taxing the logo gear in place. When it takes off, they just fall away as it takes off, and then somebody goes out and picks them up off the runway. But it saves them a lot of weight. Every three-part of the U-2 serve only one purpose. To get this payload up to 70,000 feet and fly over rush worker. This payload is this one, this particular model is in the Air and Space Museum in Washington. It's a 30 camera with a 36 inch focal length lens. At the point it was developed, it was the highest resolution camera that had ever been built. Modern spy satellites, obviously, can do much better than this, but in the 1950s, to be able to resolve an object that was two and a half feet across from 70,000 feet to pretty impressive feet. Because that's what they cared about, they hacked the rest. They could have made the wings more rigid so they didn't flap, wouldn't have cost them too much weight. But they wanted to make things the lightest possible because the thing that mattered in the U-2 was the altitude. It was how high it would fly. It's hard to land, so conventional wisdom would have been to put more landing gear in so that you didn't have the pilots going down the runway. But it mattered. They didn't need extra landing gear, what they built was fire, and it let them get to the altitude they needed to get to. And so they started to overflight to Russia. Gathered a lot of great intelligence data, but there was problem. It turns out that the intelligence they had was flawed. Russian radar was not blind, above 65,000 feet. So almost from the first overflight of Russia, they had to chase them about 20,000 feet below, now the base could do anything about it. If they couldn't fly that high, they didn't have the missiles that could shoot that high. But they knew when they were there, there were instances where it made to actually flew in formation under a U-2 trying to block its view. Because of this, they knew it was just a matter of time before the Russians figured out how to shoot it down. So they thought that they had probably 18 months to two years of operational liability for the U-2 before the Russians got this viewed out, and it would be too risky to fly the U-2 over Russia. They needed a different answer. So the CIA in there forced to put out a bit and asked to forward bits on a replacement for the U-2 almost immediately after an operation. In response to stuff where we started on the Archangel series of design studies, this isn't an early one here. By the time we got to the 11th revision, it will start looking a little bit more familiar to you. Probably think I'm about to tell you about the SR-71. I'm not. I'm about to tell you about the A-12, the predecessor of the SR-71. Now the type of watch we believe that this plane represents is almost impossible to comprehend. It's designed to fly five miles higher than the U-2 at 9,000 feet, and it's designed to fly four times faster at bottom 3.25. Now the fastest plane that has been built to this day is still the F-104 Starfire, and it can dash at top two. But the SR-71 is designed to cruise at bottom 3.25 for hours at a time. Performing at those extremes meant almost everything the team knew about traditional airplane design just didn't apply, and the CIA in their generosity gave stuff works 22 months to figure it out. So normally you would build a plane for you to go very high up the moon. You would want a lightweight material so that you didn't have a lot of weight to carry up to altitude. The problem with that is after this aluminum loses its structural integrity at about 300 degrees Fahrenheit. Flying at Mach 3.25, the early calculations indicated at this point would be about 800 degrees Fahrenheit at the nose, and 1,200 degrees Fahrenheit on the engine couplings. So if you tried to fly a plane made out of aluminum at that speed, it would literally just hold up on itself with a fresh light and aluminum can. They considered stainless steel, which would have stood at temperatures just fine. Stainless is heavy, and so if they built it out of stainless steel, they wouldn't have been able to get the altitude they needed out of this one. So Henry Combs, the primary structural engineer on the project, suggested they built it out of titanium instead. The problem is, nobody could have built anything this way out of titanium. They built the engines constantly at 1.4 out of titanium, and it was a real pain in the butt they had a hard time doing it because titanium was a phenomenal, hard to machine. But it's half the size of stainless steel, and it would have no problem standing up to the temperatures or pressures that the plane was expected to experience. So Kelly Johnson said, any material that can cut our gross weight by half is a day out of tempting, even if it will drive us nuts in the process. And he was absolutely right about it driving the nuts in the process. So they ordered to test batch of titanium to see if this is something that's even feasible for them to do. We want to show it out there and realize they had no idea how to screw it, how to weld it, how to rev it or even how to drill it. The drill heads they normally used on titanium and aluminum would just shatter when they tried to drill titanium on them. On top of that, the only US supplier titanium at this point was producing batches of wildly varying quality and didn't have the capability to produce titanium in the quantity that they needed to build the planes that they were expecting to build. So they asked for help. They called the CIA and said, we want to build this thing out of titanium. We need to help us figure out where we can get some. And so eventually, through a series of dummy companies and third parties, the CIA set a supply chain up from the leading exporter of titanium of the day. The Soviet Union. So literally the amount of the bill, the A-12, came from the very country it was being built to spy upon. And the extreme operating environment of this plane required adaptation everywhere in the plane. The early calculations also indicated that this plane, when operating at altitude and temperature, would actually stretch by about two or three inches from its ground length, just from the air heat, the friction of the air heat. So all of the systems on the plane had to cope with that friction and that stretch. They built control cables on the plane out of algae oil. And algae oil is a pretty expensive metal that's used to build high and watch strings, because you can stretch it out several times and it never loses its tensile strength. They built the engine nozzles out of this obscure nickel alloy called Hastaloy X. But this Hastaloy X can withstand the 3400 degrees Fahrenheit that the afterburners produced when they ran for an hour or two in a stretch. Off the shelf, electronics wouldn't function because of the extreme heat. Neither would oils, hydraulic fluids, greases, you name it, they had to cope with new answers for all of this stuff. Even fuel, they developed a custom fuel for this plane, so that it wouldn't be volatile at when it was up to temperature and altitude. The only problem with the fuel they developed is it had such a low flash point that you couldn't light this up. So to get it lighted, you had to uncheck the engine with triangle boring. Triangle boring is really nasty stuff that when you expose it to the atmosphere, it's spontaneously combusts with this bright green flash. That's literally the only way they made this fuel lit. One of the biggest challenges though was propulsion. You know, a plane had ever been Mach 3.25, at least no air breathing plane they'd sent off your hands this fast. So they had to figure out how to build an air breathing engine that could go over Mach 3. And Kelly Johnson put 32 year old Ben Rich in charge of this process. Then was a thermodynamic engineer that had participated in building propulsion on the F-14, so we already knew how they had built that plane and how they had gotten it to Mach 2. So, you know, Mach 2, Mach 3.25, not that big a leap. Never been done before. So they sent them off the shelf, J58 turbo fan from Kraken Whitney. This engine had been built for a fire jet that the project had subsequently been cancelled, so Kraken Whitney was really looking for a home for this engine. They wanted somebody to pay for all the research they had done on it. The problem was they required extensive modifications to even keep running at 70,000 feet because they're at 90,000 feet, so there's just no oxygen out there. The atmosphere is so thin. Their answers to that are what you see right here, the inlet cones. Now these inlet cones, when this plane gets up to speed when it's going Mach 3, actually move 26 inches back into the engine. Don't understand why you have to understand how jet engines work. A jet engine, the opening on the front is scooping as much air as possible yet. And over a series of compressors, compressing that out into a small strip, that's how it builds its speed. It's kind of like when you take a water hose and you put your finger over the end of it and the water starts running. Same theory just on a much grander scale. So in order to get this air-breathing engine to work at 90,000 feet, they needed something else to produce compression besides just the compressor. That's where these inlet cones come into place. At cruise speed, the amount of compression is in inlet cones. The amount of compression these inlet cones produce is responsible for 70% of your overall engine thrust, just the inlet cones. The afterburner is another 25%. The turbine itself is only producing 5% of the engine's thrust at altitude. And there's so much, these inlet cones produce so much compression that air coming in is negative 65 degrees Fahrenheit on the leading engine. By the time it passes the cone and gets to the combustion stage of the engine, it's at 800 degrees Fahrenheit. That's how much compression these cones are producing. But even more interesting to me is what they chose not to solve in the SR71. There's no fuel system sealant that was affected with the entire temperature range that the plane was expected to operate by. So the plane would actually just sit on the tarmac-driven fuel. You can see the puddle here at this point. That's not water. They shouldn't care. It didn't matter that the plane sat on the tarmac-driven fuel because it just they just didn't keep it fueled up on the ground. So they solved that problem. So off back to supersonic speed with fuel system sealed. No big deal. Another problem they chose not to solve is how to start the engines. I told you about the triad before anything. You have to eject to get the engines to light. You also have to get the turbines turning 4500 RPM against the not-small turbines. So they thought about putting a starter engine on the plane. Now the problem is to get those giant turbines turning 4500 RPM. They decided to put a giant starter motor on there and it would have cost them a ton of altitude. So instead of doing that, they solved it with this, the HE-330 starter cart. Now when the ground cruises the Buick, and the reason they called it the Buick is because it quite literally is two Buick VA wildcat engines coupled together. They physically coupled that to the starter shaft of the engine, cranked it to full throttle, and then injected the triad before any light off the engine. It's a subsistence combustion. The ground crew said that the hangar sounded like a stock car race when they started this plane off. But you know what? It didn't cost them any altitude. It's a pretty elegant hack. There are only two things that mattered in building the A-12 and it could go very fast. And it needed to do so very high. Biologs have higher and four times faster than the U-2. So on April 30th, 1962, a full year and 100% over budget spent works good to see what they wanted. This is a picture of the A-12's first flight, drip fuel. You couldn't start the engines without crazy chemicals and a couple of VA race car engines, but it didn't matter. They spent their time relying on the things that didn't matter, the titanium construction, the propulsion, and they hacked the rest. This plane went off 3.25 at 90,000 feet and overflew every hostile territory in the world. After building 15 A-12s for the CIA, the Air Force took over the program. And they requested that stuff works modified the plane to be a twin seater and add about 50% more cargo capacity so they could have more sensors up their altitude. That plane is the A-12's far more famous armored brother than the SR-71. It flew for 30 years and has the distinction of being the only U.S. combat aircraft to have never been shot down. Despite 3500 sores over very hostile territory and hundreds of missiles launched at it, holds about every speed and altitude record there is. For altitude, it set a record of 85,069 feet. Now, this is the official record set over a defined course. This plane almost certainly went higher than 85,000 feet. Speed, 2193.2 miles an hour, just a shade of a Mach 3.3. Now, in his book, sled driver, Brian Schuhl, a U2 SR-71 pilot, tells a story of evading missiles over Libya. And after he had dodged the missiles' rig, reconnaissance systems officer in fact, he had to remind him to back the speed off of the plane when he looked down and realized he was going just to touch up a Mach 3.5. So we know it would go faster than Mach 3.3, but this is the official record. So what does that actually mean? Well, the muzzle velocity of a 22-caliber rifle is 2,046 miles an hour. And the SR-71 goes 2,193 miles an hour, so the SR-71 at cruise speed can make the claim that it is literally faster than a speedy boat. How fast can it get places? Well, New York to London, the SR-71 could fly in 1 hour and 55 minutes. The comfort? On a good day with a strong tail line could do it in 2 hours and 52 minutes. Los Angeles to Washington, across the United States. The SR-71 can do that in 1 hour and 4 minutes. And I love this connection, it's a little bit easier to wrap your head around. And of course, we're setting up Los Angeles to Washington, they also tracked it from St. Louis to Cincinnati. The SR-71 can do that in 8 minutes and 32 seconds. If you do that in your car, it's going to take you about 5 hours and 16 minutes. It'll probably hold these records forever with the advent of unmanned drones and spy satellites, but we don't really have any meaning for a plane that goes this fast. So we're probably never going to build another one. The SR-71 was Kelly Johnson's crown of achievement. 1975, he had Lockheed's mandatory retirement age of 65. And he passed the reins on to this man, his Projet Benrench. The same Benrench that at 32 years of age designed the SR-71's propulsion system. A Ben took over stock works at his vultures time. This was close to Vietnam. US appetite for military spending was very, very low. There were no new planes being built. And Lockheed had just attempted to reenter the commercial aviation market. With this plane, the L-1011 TriStar. And it was a horrible failure. Lockheed lost somewhere in the neighborhood of $2 billion on this plane. These were $1975, so a ton of money. Benrench, after he took over, knew he had to find some significant new work to get it sold to the military pretty quick, or he was going to have to start letting his best engineers go, because they were also his most expensive. Meanwhile, the Cold War continued. And Russia would be in power in Russia for another eight years. The Soviet Union had invested 300 billion ripples in developing surface-to-air missiles like this SA-5 that were far more advanced than any attack capability that the US had that literally couldn't pierce their missile defense. So in order to maintain the mutually assured destruction that had kept us from bombing each other for all of those years, the US needed to develop something that could get past these missiles. But ideas were in pretty short supply. Until Denis O'Rolls, a 36-year-old math and radar expert in the Stanford staff, came across this paper. The method of edge waves was a physical theory of diffraction. Sounds like an enthralling read, right? Well, it was so enthralling that it actually took the Air Force ten years to get it translated from an publicly published Russia. So this paper had been written by Katerozov himself, the chief scientist at the Moscow Institute of Radiant Engineering. And it wasn't until it repulsed red in the school paper and all the way through it that he found a shocking revelation in the back. There was a formula in the back that he reasoned he could extrapolate to calculate the radar cross-section of the edge and surface of a wing and be able to come up with an accurate reading. Well, that was important because in those days, accurately determining the radar cross-section of a plane was only possible on a radar range. He would route a plane on a pole upside down for some reason. And shoot radar at it to see where it came back. So if you wanted to develop stealth technology, the only way that you could do it is to iterate and iterate and iterate and iterate. Folks like O'Rolls are good into science, could make a reasonable inference about what it gave a change to a plane might do to its visibility on radar. But it was not imperable by any means. Stealth is long and bad around its possibility, but it has always been written off as too expensive and too difficult to do effectively. But O'Rolls was convinced he had an answer to that from Seth's paper. So he convinced Kelly Johnson to urge him, can that spin range let him spend some time building software to train him just that. So O'Rolls really spent a few weeks hard at work on software to take advantage of the formulas. He walked into Ben Rich's office and handed him a sketch of this, which quickly became known around Scump Works as the hopeless diamond because they didn't think there was any hope of ever getting this thing to fly. Nonetheless, Ben Rich trusted this O'Rolls route and built up a Monca. It took out to a radar range in Palmdale, California to see how well it performed on radar to see if the math actually held up. The radar technician asked Ben Rich, are you sure that they've got them all on pole? Can you stick your head out and look? And so Ben does this. About the time he sticks his head out the window, a crow leans on the plane. And the radar operator says, never mind, I've got it. Ben didn't have the heart to tell him that he was having a crow on the airplane. And it was at this point that they knew that they were on to something big. About this time, there's a dark side of the holding design competition for stealth on an aircraft. Lockheed and Martha won the first phase and were each given $1.5 million to refine their designs. And to build 38 footballs to be tested at the Air Force's most sensitive radar range, and why it seems to be next to none. That's what you see in this picture is Lockheed's 38 football. Again, more radar range problems though. The plane design was so good that the only thing they were seeing on radar tests was the pole. Now, until this point, the Air Force thought that this pole was invisible. They've never seen it on radar before. But this plane was reflecting so little in the way of radio information that they clearly saw the pole on radar. So in order to get an accurate reading on the plane, they also went to work and built them a better pole too. You can see that pole there. The pole itself cost around $500,000, but it was no longer visible on radar. I came up with a really interesting way of determining a cross-section of radar-afflicted planes. They knew that they could calculate the expected reflectivity of a sphere. So they took steel ball bearings and started gluing them on the front of the plane. They started with... They started with 3 inches a little bit bigger than us. This is a 2-inch ball bearing. And they went smaller and smaller and smaller. They knew that as long as they were seeing the expected cross-section of that size of sphere on the radar that they were not seeing the plane. Well, they got all the way down to a 1 eighth inch ball bearing. I don't know if you can see that or not. But it's tiny and smaller than they may be. I think you can still see the ball bearing. They still weren't seeing the plane, even that small. So while he'd obviously been on the design competition handling, the next step was to build an actual plane in a group that had... Once it had the things that the model didn't have like inches and intakes and landing gear and a pilot's helmet and a shield, they would still be stuck in. The Air Force 1 and 2 prototype planes in 14 months. And it's not works very well. And sure enough, 14 months later, like plot work, they had a plane ready to fly. Now this plane is literally a bucket of spare parts off the surfless shelf. The flight control computer came out of the F-16. The navigation came out of the B-52. The seat that also came out of the F-16 heads up display from the F-18. The engines from the T2B Buckeye trainer and on and on. Literally the only thing you can do about this plane is the exterior skin. The biggest thing they had us all obviously was the aerodynamics. If this looks like it ought to fly to you, it's because you've seen enough pictures of the F-117 over the course of your life to have conditioned your brain that this is a plausible representation of an aircraft. Note that at this point you've never seen a thing like this that could fly. It was actually unstable in all three axes of flight. It was unstable in pitch, in yaw, and in roll. Up until this point, the only plane that had ever been deployed that was unstable in any axis of flight was the F-16. And it was only pitch unstable. But that's one of the reasons they used the flight control computer out of the F-16, because they already had code to deal with the pitch instability, and only had to code for two more axes. And the way it worked is the computer would constantly calculate what it needed to do to keep this plane stable on flight, and it would sum that with what input the pilot was giving in, and that's what it would send out to the control services. And by doing that, they essentially hacked the laws of physics and made this plane stable. It wasn't so stable at first. It's the first couple of test flights that did the nickname the wall in Goblin, because it took a while to get the software just right. But they finally got it perfectly. True form, they got it to fly, because most of the test flights were at night. This is actually the only picture of the half blue in the air that I've ever seen, like maybe the only one there is. Now they got it in the air, though, they came to see it live up to its promise of stuff in this. And so they took it out to the Nevada desert and flew it into one of these, the target acquisition radar from a hot missile battery, the most sensitive radar equipment that the U.S. had at the time. The plane flew right overhead of this radar dish. Never picked it up. The missiles never pointed at the plane. Didn't even see it. Less than five years later, the first of that 117 detachment was operational out of Tahoe Bautestrange report, which is a snap in the middle of Area 51. Now this would have been sometime in the late 80s, so it was a pretty good chance that any reports of UFOs in that area around 1989 or so were actually this plane. Pilots were actually just about to go flying because it looked so strange, but once they got it off the ground, they found it was actually a joy to fly. It was very stable in the air and it had great controllability. The American public found out about this plane on the first night of the desert storm. Until the 22 of 117s flew into Baghdad. The air force had actually privately expected about a 30% loss rate on this plane. They expected 30% of these to be shot down because of how well-defended Baghdad was. And it was a single plane at night, more for the rest of the desert storm. The stealth technology lived up to the promise. They couldn't detect it. This whole point, like I mentioned, is one big hack. They needed to be invisible on radar. They got very close to that by essentially not caring about aerodynamics at all and hacking their way around the laws of physics and aerodynamics and the inherent instability of the design. The reason this plane is made up of all flat surfaces isn't because you have to do that and for it to be stealthy. It's because the technology of the day wasn't sophisticated enough to calculate the radar reflectivity of the curved surface. So they just built it out of flat surfaces. So how do they do it? All of these amazing planes, each of which was groundbreaking in some significant way and several more planes we haven't even talked about these were just the three headlines. The story ends in the same place as the game. With Kelly Johnson and his strategy team have at its peak 23 designers and 105 fabricators that created the PA around a mocked-up engine in 143 days. Complaint them subsequently used for 40 years. Not much about Kelly's philosophy of how snow works built planes changed over the years, even when the rains passed in the rich. He was a proponent of prototyping and learning and tried to find a picture of half blue on the tarmac. Next to the F-117 stealth fire but there aren't any. And I realized it's because they actually managed to crash both of the half blue prototypes before the first F-117 was ever built. It was that much of a throwaway prototype. You like to iterate. You can see the H-12 on the right here and the SR-71 on the left. The H-12 could go a little faster a little higher than the SR-71 but it turns out it didn't really peak tier. It revised it to a two-seater double payload capacity. Sacrificed a little bit as to this but getting it more closely to the mission just like we do with software once you actually start using the software you refactor it to fit the ways that your users are actually losing it because sometimes you don't get it right on time. Kelly also had some general rules about how to run an organization like he does. If you Google Kelly's rules you can find a list of all of them but I'll just share a couple that are particularly relevant to what we do. The first is to use a small number of good people. Kelly's rules actually say the number of people having any connection with the project must be restricted in an almost vicious manner. Use a small number of good people 10 to 25 percent compared to a so-called normal project. So the H-12 arguably the most sophisticated plane that had ever been built when it was built at its peak at 75 design engineers worked on to give you some contrast when Boeing built the 777 at its peak there were 10,000 design engineers working on the 777 that's with the assistance of computer-aided design. Kelly and his team were still doing drawings by hand. Kelly hired smart people in his organization and he trusted them to do good work in the business and he trusted them to bring their expertise to the table. A very simple drawing and drawing release system of great flexibility for making changes must be provided. We mentioned earlier the lightweight drawing system that Kelly put in a place for the PA. It became one of his rules and he kept his process to the minimum necessary to give the team the required context. Teams did lightweight drawings not copious documentation required elsewhere in life. The drawings only needed to convey their meaning to the people who were going to use them. He kept the process as light as practically possible also giving everybody the information they needed to do their jobs so they didn't spend a lot of time feeding the process. They spent a lot of time getting stuff done. I actually had a great tweet about this. Team Pathology is either hanging on the processes suited to a smaller team or early adopting processes suited to a larger team. About 10 years ago a friend of mine started a software consultancy and the first thing we did is probably the first thing you would do if you started a software consultancy we went out and spent a thousand bucks on a license for JIRA and we spent the better part of two days getting it installed on VPS so we could track our work. There's two of us. We had one client. We had absolutely no need for that level of process. You can guess we were not knowing concern for very long. You need enough process so that everyone has the context they need but not so much that people turn off their brands and just want to do what the cart says to do. You need people to bring their expertise to the table. You need them to think about what they're being asked to do and to do it in the right way. I think that's right for the software. I think that's right for long-term sustainability but to do that you have to give them the context and the freedom to make those decisions. This is what Kelly Johnson got so right. There's no way he can deliver this level of innovation off his own. The processes he put in place allow all of his staff to bring their expertise to the table as well and the collected output is the fabled, the whole is greater than the sum of the parts. He set the right priorities and that helped his team make the right decisions on how to make compromises. The most important rule Kelly had was that there should be but one object to get a good airplane built on time. You want me to get a good airplane? Deliver the value that the customer needed and hit the key specs and compromise wherever it was necessary to hit those key specs. He was a prerogantist. Every decision was rounded with the most value and the shortest amount of time for his customer while working the best out of his people. Because of the freedom and trust he gave his teams and because of how clearly he laid out goals for his projects they were both able to deliver some of the most amazing airplanes ever built. You two landed on a terrible landing here. Pilots used to say that it was the easiest point of the world to fly between 60,000 feet and 6 inches. But the team decided it was worth fixing the landing here and sacrificing the altitude it would have taken to add even one more bogey. Those were the problems. It was the right compromise for that design. The SR-71 is the fastest plane ever built and it couldn't stop starting because Star Wars would have added a lot of weight. It sat on a tarmac drinking the fuel because it just didn't matter. The team spent all their time figuring out how to build a plane out of titanium so they could hit Mach 3 at 90,000 feet and just didn't worry all that much about the rest. The other one said the team piloted almost every law aerodynamic design so they could have a plane that was invisible on radar. It bucks conditional wisdom in almost every way possible and it does it because Ben Rich trusted Dennis over Ulcer enough to believe that this design could be radar invisible and it was worth the time for them to impact the way around everything else they had to do to build a plane that looked like this. Kelly's and later Ben's teams had unprecedented input into what they were built. They had incredible freedom and trust for the bosses. If you don't have that great work you should push for it. You should push the not be just a simple line worker writing software and pulling the next car off of the staff. You should have input into what your team is built you should have input into what it's going to take for your team to build an exceptional product because you have the expertise to have that input. If you don't have that input you should either work to change your employer or you should change your employers. And if you're a fever you need to find ways to give your team that freedom you need to find ways to push decisions as far down into the stack as you possibly can so that everybody on the team is bringing their expertise to their because if you can spread out the decisions if you can get everybody to contribute you're going to build something better than you ever could if it's just you or you and a couple other people making decisions. You need to make sure that everybody on your team knows that the two or three most important things in your project so that they can make the right decisions they can pick the right places to take on technical debt. Take the time to build and refine a process that works for your team and gives them context and if you do that there's no time when you'll be able to build something. Thank you.