Experimental Variations

That’s not a runway, but when gravity asserts its dominance, a straight, long, dirt road is about as good as it gets. Knowing you can safely land somewhere during your flight testing becomes part of your risk tolerance. (Photo: R.E. Butcher)

I have often told the story about landing our RV-3 on a dirt road on its second flight when the engine quit after my aftermarket fuel valve handle slipped on the shaft, trapping me between tanks. The root cause of this was that I had failed to Loctite a set screw that was used to pick up the “flat” on the D-shaped fuel valve shaft—the set screw being there so that you put it in from either direction, making the handle a “universal fit” (it could go on one way or 180 degrees the other). In an attempt to make something universal, the handle set a little trap for a builder—but it was still the builder’s (my) responsibility to make sure that compatibility was assured. This is, in fact, true of the entire Experimental/Amateur-Built category of aircraft—the builder is responsible for making sure that they have checked and assured compatibility of the pieces that they choose to put in their aircraft.

This is probably a new concept for those coming from the certified aircraft world. I was recently at the CubCrafters facility in Yakima, Washington and getting a tour of their factory. In the final assembly area, they showed me a Carbon Cub that was in final prep for delivery to a Fish and Wildlife department that was going to use it to control coyote populations. To make sure that the gunner (in the back seat) didn’t accidentally shoot a hole in the airplane’s prop, they had painted the back four inches of the prop tip white to make it visible. But since this was a certified airplane, with a certified engine and a certified prop, and that paint was not part of the design specification for the prop, they had to have the prop manufacturer do certification paperwork to allow that paint. That’s how certification works—configurations are carefully examined and controlled.

Something simple, like forgetting to use Loctite on a set screw can have profound consequences.

In our Experimental/Amateur-Built world, you can pretty much do whatever you want, so long as you don’t put anyone but yourself at risk. That gives us remarkable latitude to explore and experiment—but it also adds a tremendous responsibility to do the background work, research, and (sometimes) the engineering to make sure that when you bolt a specific prop on a specific engine, neither the prop nor the engine is going to become a grenade.

The roots of homebuilding aircraft go back a long way—well back into the earliest days of aviation, to be honest—but the real modern homebuilding movement began in the 1950s with a few simple designs built of simple materials and (usually) “barn find” engines and propellers. Frequently, people built airplanes around the powerplants they could find, and “if it bolted together, it fit.” Modern builders might look in horror at the drilled crankshaft flanges of the GPU versions of the mighty little O-290 from Lycoming. Designed to bolt onto a generator in a ground-based power cart, it was never intended to swing a propeller with all the attendant gyroscopic forces. Yet many of them flew, and many of the aircraft on which they flew did aerobatics. I know of one very experienced NASA test pilot who watched his prop spin away from his homebuilt Pitts Special (then looked for a place immediately below him to land—the glide on a Pitts is pretty poor).

Engine and prop compatibility is a big deal in the certified world, and prop manufacturers like Hartzell, McCauley, and Sensenich use engine test stands, flying aircraft, and heavily strain-gauged props in test programs to determine what will work together and last until TBO. Torsional vibration is not something easily seen or heard—you need instruments to tell when it is happening—or you just wait until parts start flying off (not a recommended way of testing, by the way … ). Very minor variations in an engine/prop system can have dramatically different results in testing—changing from a magneto to electronic ignition, for instance, can change the slope of the force pulse that goes into the prop because ignition of the fuel/air mixture in the cylinder is quicker. Similarly, different compression pistons cause the power pulses to be shaped differently—and the effect on the prop hubs and tips can be consequential.

Most pilots have a healthy fear of fire, loss of control, or structural failure. Let me add a subset of “structural failure,” and that is prop failure. Failure of a prop blade often leads to separation of the engine from the airframe, which creates a tremendous control problem as the center of gravity shifts drastically aft in an instant. I’d add prop failure to the top of my worry list if I were you. But even if you don’t fail a propeller, untested configurations which give different vibration signatures can cause failures in ignition systems, cracking of exhaust, bearing failures … any number of things can be lurking in an untried or untested configuration that might bring you to grief. Folks setting up airplanes for racing might take chances with higher-horsepower modifications, but they are exposing themselves to danger for less time than the person trying to fly an airplane for 2,000 hours. Good risk management says—stay in the middle of the envelope and stay away from the edges—dragons can lurk out there.

Propeller design and testing is serious business. These photos show how a prop is instrumented when doing vibration analysis on specific engine/prop configurations. It is not a trivial task. (Photos: Paul Dye)

It is the responsibility of the builder to make responsible decisions; decisions based on data, not just TLAR (That Looks About Right) “engineering.” Now, yes—we are experimenters—so we experiment. Few folks have supercomputers and NASTRAN software in their garage shop, so complete vibrational analytics before flight are kind of out of the question. Most of us are going to end up taking a few risks if we want to operate outside of the existing knowledge base. But research can help quantify those risks. Finding out what engines a particular propeller is approved for by the manufacturer is a good place to start. If you have an engine (or engine configuration) that is not approved, a call or email to the propeller’s engineering department might help you to understand just how far outside the database you’re going to be operating. Most engineers like to talk about their work and what they know—and what they don’t know. Yes, corporate lawyers will probably inhibit a completely free and open discussion—but you can still learn something by reading between the lines.

The thing about risk is that you have to openly acknowledge and accept some risk if you are going to color outside the lines. At least you have to acknowledge it to yourself, unless you want to simply be a daredevil. Accepting that risk means that if you operate outside the approved envelope and components or equipment start to fail, you don’t blast the company on the internet—you accept that experimentation has risks, you took them, and you came out on the short end of the stick. Hopefully, you learned something that will help your future experiments—and eventually, the entire community will benefit from the knowledge. Sometimes, it helps the manufacturer’s engineering department by providing them data that they couldn’t afford (or get permission) to gather themselves.

In aviation, the idea of “moving fast and breaking things” (a popular modern concept) is not always the best way of going about designing a new flying machine—at least if you are putting human beings aboard during that testing. Unmanned flying machines? Sure—knock yourself out! But once you have humans (especially yourself) onboard, caution is the watchword. Real flight testing is—as I have said many times—an incredibly boring activity … if you do it right. Make sure that you have backups for critical functions, make sure nothing is going to catch fire—and make sure that you aren’t inviting a loss of control or structural failure. Keep yourself in a position for a safe emergency landing until you are sure of the reliability of your craft.

And finally—recognize that minor changes or variations in what is normally accepted (or recommended by a manufacturer’s engineers) put you into the realm of experimenting. Keep your eyes wide open, your mind working, understand that you are accepting risk—and plan accordingly.

The post Experimental Variations appeared first on AvBrief.com.