Practical Men

American innovation at the dawn of flight

Kevin Cameron's Top Dead Center banner
Technical Editor Kevin Cameron shares his wealth of motorcycle knowledge, experiences, insights, history, and much more.Cycle World

Because the Wright brothers achieved powered, controlled flight in 1903, the accomplishments of their competitors tend to be forgotten. But it remains notable that the four-stroke aero engine devised by engineers Stephen Balzer and Charles Manly for Prof. Langley’s “Aerodrome” gave four times the power of the Wrights’ engine—from a very similar weight.

Langley, with backing from both the US Army and the Smithsonian Institution, believed he could scale up the successful performance of a steam-powered model (which with its 13-foot wingspan had made flights of 1/2 to 3/4 mile) to the full-scale machine, which twice flopped ignominiously into the Potomac River from a catapult mounted atop a houseboat.

The Wright brothers believed their own long series of experiments, which included learning to fly full-scale man-carrying gliders, and they were rewarded with success.

Langley, with money to spend, sub-contracted his engine, having decided gasoline engines were the coming thing. The Wright brothers, focusing on their “flyer,” regarded the engine as a detail. When no suitable engine could be found, they designed and hand-built something that would serve, giving about 12 hp.

Balzer, born in Hungary, was, like the Wrights, largely self-taught. Langley chose him as engine contractor in 1898 because he had in 1894 built the first gasoline auto to run in New York City. It was powered by an expertly machined three-cylinder rotary gasoline engine. "Rotary" in this case means that the crankshaft remained stationary while the crankcase/cylinder assembly whirled around it. Tens of thousands of such engines—most with nine cylinders—powered combat aircraft in WWI. They offered advantages that Langley needed:

  1. Radial cylinder construction is light because all cylinders share a single short crankcase and single-throw crankshaft.
  2. A rotary's crank sits still while crankcase and cylinders revolve, making the extra weight of a flywheel unnecessary.
  3. The whirling of the cylinders had provided adequate engine cooling for the 2-1/2 to 3-hp auto engine.

Balzer’s design contained important innovations. One was the use of a cam ring, concentric with the crankshaft. Lobes on its OD operated the valves of all three cylinders. Balzer’s cam ring can be found in practically every one of the hundreds of thousands of aircraft radial engines ever built, propelling such aircraft as the Boeing B-17 and Lockheed "Constellation."

Because a flat tappet could not work with a cam ring (lobes would strike its edge rather than its flat working surface), Balzer devised roller tappets with guides to keep them aligned to the cam ring. Such roller tappets are to be found today in every Harley-Davidson Big Twin and Sportster engine, and their use by Harley dates back at least 87 years. In addition, they are widely used in hot-rodding and by Detroit automakers eager to cut low-rpm valve-train friction.

Langley wanted six cylinders to smooth out the delivery of power, but Balzer knew that only radials with odd numbers of cylinders give evenly spaced firing, so five cylinders were chosen.

How do you connect five pistons to a single crankpin? Balzer found a way; he gave the inner end of each connecting rod an arc-shaped shoe that bore against the crankpin and was held there by L-shaped retaining rings at each end of the crankpin.

When Balzer tested the completed five-cylinder (378.7ci) he found himself up against scale effects that he didn't understand. Instead of being adequately cooled by its own rotation as the small three-cylinder had been, just five minutes' operation made it hot enough to act as its own ignition source. This is the dreaded "squared-cubed law" in action; the bigger engine generated heat in proportion to its volume (which is roughly the cube of dimension) but gained cooling surface only in proportion to the square of dimension.

Here’s an example: Start with a cube 1 inch on a side. Its volume is 1 cubic inch (1 x 1 x 1) and its surface area is 6 square inches (six sides, each of 1 square inch). The ratio of the two is 6 square inches of surface for 1 cubic inch of volume.

If we now double the dimension to 2 inches, volume becomes 8 cubic inches (2 x 2 x 2), but surface area becomes (2 x 2) x 6 = 24 square inches. Now the ratio of the two has fallen to only 3 square inches of surface for every cubic inch of volume—only half what it was in the first example.

An engine cylinder has a more complex shape than a cube, but the same general idea applies, so Balzer’s large five-cylinder engine overheated. In air-cooled engines this shortfall of cooling surface area is made good by providing cooling fins. Balzer considered them, but he also had a second scale problem: intake valve action.

Early four-cycle gasoline engines were usually built with a cam-operated exhaust valve and an intake valve held shut by a light spring, such that the piston’s suction stroke would “suck” the valve open, allowing fuel-air mixture from the carburetor to be drawn into the cylinder. Early motorcycle engines began with such “automatic intake” systems derived from the widely sold French de Dion engines.

Automatic intake worked on Balzer’s small engine, but when it was scaled up (the five-cylinder was 33 inches in diameter), inertia force from rotation more strongly held the intake valve on its seat (its stem pointed away from the crankshaft). No matter what Balzer did to improve cooling, no matter what springs he applied, this rotation effect reduced valve lift so much that this engine of over 6 liters displacement made only 8.5 hp. It was an impasse.

Langley, whose backers expected to see powered flight, in April 1901 decided he must switch engineers. He and chosen replacement Charles Manly decided after some testing to seek ideas abroad. Meetings took place with Hiram Maxim (who made millions from his invention of the machine gun) and the French Count de Dion, who had by 1900 sold 20,000 motor tricycles and countless engines in various sizes worldwide. Just as the DKW RT125 motorcycle was the model for postwar two-strokes from BSA, Harley, MV, Honda, and Yamaha, so these de Dion engines were copied worldwide and later improved by Indian, Curtiss, and a hundred other makers.

Both Maxim and de Dion disliked having the whole engine spin, and de Dion suggested building an inline-four cylinder. But placing cylinders inline required more weight in crank and crankcase.

On his return to the US, Manly, realizing that the whirling of Balzer’s engine was limiting the opening of its intake valves, converted the large engine from rotary to static operation. With its valves now opening as originally intended, its power doubled.

Balzer had made a single-cylinder test engine and demonstrated that it ran very well indeed when cooled by wrapping its cylinder and head in damp rags. Manly now obtained similar encouraging results from the five-cylinder in its new, non-rotating form.

It must be noted here that later workers found other solutions to the intake problem, the best of which was the mechanical cam operation employed by engines today. Thousands of Allied and Central Powers aircraft of World War I (1914–1918) were powered by rotary engines, but as they reached 200 hp the windage loss from the whirling cylinders became excessive, forcing later designs to adopt Manly’s static form.

Encouraged by the good performance of the single with “wet-rag cooling,” Manly set about the construction of larger, water-jacketed cylinders of 5-inch bore, increasing displacement of the five-cylinder radial to 491ci (8 liters). In place of the improvised Prony brake (friction dynamometer) inherited from Balzer, two large water brakes were built. Manly hired several expert machinists to produce the necessary parts. In 1900 it was possible to hire such people for $2.75 to $3 a day.

Once a first round of leaks of Manly’s new brazed-on water jackets had been corrected, and light, thin steel pistons of de Dion pattern made, the engine made a 10-minute run at 600 rpm, giving 25 hp. Progress! In the course of constructing the water jackets, Manly subjected himself to prolonged brazing heat and flux fumes that are believed to have damaged his eyesight.

As so often happens in engine development, reaching a new power level reveals new problems. The pressure of the bigger new pistons on the con-rods’ little brass “shoes,” bearing on the crankpin, caused seizure and a comprehensive engine wreck. Con-rod brasses were broken and cylinders distorted.

Manly thought again, devising a more durable big-end construction. Instead of each connecting-rod having its own arc-shaped shoe, bearing through its own small area against the crankpin, one master connecting rod was given a big-end bearing that encircled the crankpin, and each of the remaining four link rods was given an arc-shaped shoe bearing against the outside diameter of the master rod's big end. As before, the shoes of the link rods were held against the big-end OD by flanges.

Balance is a minor issue in slow-turning engines, but Manly’s more durable master rod system now allowed higher revs, and shaking forces rise as the square of rpm. This engine was to power a fragile, kite-like airplane, so its vibration had to be reduced. Manly now placed balance weights on the engine’s external flywheels. He had previously lightened the valve mechanism to avoid “smash-up” of the valve train—the “punch rods” (pushrods) by 30 percent and the valves by 17 percent—and also reduced the suddenness of the cam-ring’s lift profile. Changes such as these would become routine tools of engine development in the years to come, but Manly’s work was original and pioneering.

By the spring of 1902, the engine was giving 51 hp at 950 rpm and proved its durability by making three 10-hour runs.

In October and December 1903 the two attempts to launch Langley’s “Aerodrome” failed.

In 1914, as part of the endless legal wrangling with the Wright brothers, Glenn Curtiss rebuilt Langley’s "Aerodrome," equipped it with floats, and demonstrated on June 2 that it could take off from Keuka Lake, near his Hammondsport facilities. It was powered by the Manly-Balzer five-cylinder radial engine.

Imaginative intelligent persistence is central to R&D.