Any time a plane flies, silently assisting with the craft’s lift and trim is the huge group of people that made sure it was safe. Engineers, obviously, are essential, but there needs to be somebody directing proceedings at the intersection between the work of engineers and the pilots who’ll eventually be in control of the machine.
Enter Aaron How, Royal Australian Air Force test pilot.
“My role is quite specific,” he says. “I’m not a typical frontline pilot now that I’m into this area as a test pilot.”
Despite the apparent specificity – test the airplane – How’s role is incredibly diverse, because problem solving doesn’t happen only when the complete new aircraft is tested in flight. Instead, every change, update and part installation needs to go through rigorous examination. It follows that there is no ‘typical’ day.
“[My duties] can be from flying a brand-new aircraft for the first time, which is quite rare,” How says of his day-to-day tasks, “all the way down to a very small modification – like just changing the smallest thing, maybe a switch on [the] control sitting in a different place.”
So how does one get to the renowned, even exalted, role of test pilot?
How relates his background fairly matter-of-factly. He started flight training on PC-21 aircraft – a Swiss-built single-engine turboprop – at Sale, Victoria.
“I’m not a typical frontline pilot now that I’m into this area as a test pilot.”
“[To become] a useful pilot of the air force, doing operational missions, that takes about four years, four years of flying, plus or minus probably a year depending on what aircraft you go to,” he says. “And from there, I did a three year [duty] tour, which you have to do before you become a test pilot, because you have to have some operational background and know how the aircraft gets used in an operational sense, not just in the training sense.”
Then he spent a year at test pilot school. The RAAF sends candidate test pilots either to the Empire Test Pilots’ School, in England; or the US Air Force Test Pilot School at Edwards Air Force Base, in California; or the US Naval Test Pilot School at Naval Air Station Patuxent River, in Maryland.
Now, How says, “I do sort of a conversion of how we can apply that test pilot knowledge we learned at school to the Australian situation”.
“There’s no university requirements for my job,” he says. “But it’s very helpful to have background knowledge in engineering, so I did a Bachelor of Technology at ADFA. That’s helped me a lot as a test pilot.”
Fast facts: RAAF aircraft
- The RAAF has over 250 aircraft, which fall into four main categories: combat, mobility (including VIP transport, refuelling, search and survivor assistance, medical evacuations), intelligence and surveillance, and aviation training.
- The air force is currently in the process of replacing their Hornet craft with 72 new F-35 Lightning planes, which are being built overseas by Lockheed Martin and shipped to Australia.
- By the mid-2020s, the defence force plans to have unmanned aircraft stationed at Edinburgh, SA, which will be flown from a ground station and used for surveillance.
The test-pilot caper isn’t just about giving a green light when a new update is deemed safe; the hidden part is making sure that the update is user-friendly. If a high-tech device is incorporated into an aircraft but its meaning and use are too difficult to comprehend, the new device is essentially useless.
“All the techniques we use are supposed to be simple to understand,” says How. “If you make them so complex that only an engineer can understand them, then things can get dangerous.”
Ultimately, the philosophy behind making a plane flyable is that it should be simple enough to understand for somebody with only a high school education.
“It’s actually quite a hard process to make something that’s normally quite complex – which a fighter jet [and] any aircraft we use in the air force [is] – and make it simple to understand,” How says.
This means that How needs to use a combination of skills: understanding the engineering of the plane, knowing how the plane should fly, figuring out whether other people would find it intuitive, and then translating that knowledge to multiple people from different backgrounds.
“All the techniques we use are supposed to be simple to understand. If you make them so complex that only an engineer can understand them, then things can get dangerous.”
“You can be the smartest engineer or smartest pilot out there, but if you can’t communicate your ideas to people, and not just communicating but communicating [in a] clear and easy way, then all of the skills you have inside [are] essentially useless,” he says.
After all, “none of what we do is individual, everything of what we do is in a team environment.
“We’re never really alone. We kind of work as a team with an engineer and a pilot to problem-solve any situations we come up with.”
This means he must communicate with makers and users – either through talking to them during and after a flight test, or through a report – while also constantly keeping up with new research.
“We work hand in hand with a flight-test engineer,” he says. “Those flight-test engineers have done that same school that I did, the test pilot school, but instead of flying the plane, they’re flying next to us or behind us, or [they’re] even in the telemetry room on the ground.
“We’re never really alone. We kind of work as a team with an engineer and a pilot to problem solve any situations we come up with.”
“They’re reading all the same publications we’re reading, and they’re helping whilst I’m flying to problem solve that bit of kit, or how to fly the aircraft, or the problems I’m facing with the aircraft.”
All very well: but is the job really as cool as we all imagine it to be? How certainly doesn’t go nuts trying to paint it that way – but the answer (hell yes) leaks through.
“Once whatever they’re designing for the aircraft comes to me, I assume that all of that work has been done,” he explains. “I look into the documentation of how it’s going to deal with the situation that we’re going to fly that piece of equipment on, and then it’s up to me to check that. So I validate what the engineers tell me is true.
“What’s so difficult about designing a piece of equipment for an aircraft [is that] the conditions [it] has to go through is so extreme. You know, it has to go from temperature extremes of probably minus 50 or 60 degrees, up to maybe plus 70 or 80 degrees. And if you’re going supersonic on a fighter jet, potentially even higher than that – you know, over 100 degrees.
“You’re going from pressures of one atmosphere down at ground level, all the way up to almost a complete vacuum, depending on how high you are, at 60–70,000 feet. You get some different environmental conditions – rain, hail; you might have bird strikes. You have to make sure nothing tears apart from the plane.
“And the g-forces as well for fighter jets. You’re up to eight, nine, g, and down to maybe minus three or minus four, g.”
Like most others, How’s work was affected by COVID-19. Usually, he would travel to various air force bases around Australia to do his job, but he was stuck in Brisbane in 2020.
But being confined in one place didn’t mean his work was stalled. His in-depth knowledge of flight and engineering help him troubleshoot an important problem on the ground.
Flight simulators allow new trainees to get a feel for flying while remaining safely on the ground, but How explains that sometimes new pilots get up into the air and anticipate that the plane will behave as it did in simulation.
In reality, there may be many other factors, such as changing conditions, that affect how the plane will handle in the air, so the simulations need to be more reflective of what true flight is actually like.
“My job was to try to align that simulator to the aircraft, because we found deficiencies where the simulator would act in one way, whereas the aircraft will do [something] different.
“And one of those situations, for example, is when you stall an aircraft, you start to get turbulent flow over the wings. In that case, you’ll get some outcomes of the aircraft that are very difficult to predict.
“By [just] using the equations that the program is using, they won’t get you the exact solutions that happen in the aircraft.”
“The equations that the program is using, they won’t get you the exact solutions that happen in the aircraft.”
This comes down to programming. The instructions that are fed into the simulation’s governing program are not going to account for the plethora of changes and randomness that would be experienced when a plane stalls. Those “imperfections” need to be written into the simulation code.
Of course, programmers are not engineers, so How had to tell them exactly how the simulation needed to behave in order to better reflect true flight. That requires a deep knowledge of the engineering, and the ability to then achieve an outcome without having to rely on explaining the engineering to the programmers.
“I just try to tell them the outcome of what their programming is going to do to the students, you know, and what effect we want to give to the students,” he says. “We’re trying to align the aircraft as closely as possible.
“Sometimes you have to go back to the fundamentals of what we’re trying to fix here – because I get very wrapped up in trying to fix all the very specific cases. Whereas, the outcome we want might be a generic outcome.”
“The best pilots fly more than the others; that’s why they’re the best,” wrote Tom Wolfe in The Right Stuff. Next time you step onto a plane, think about Aaron How and his dedicated band of flight-test collaborators. Your plane wasn’t passed “safe to fly” by accident.