Revisiting Rotaries
Peter Garrison

When I wrote to an old friend, who had built a beautiful Falco himself, that
my homebuilt was at last flying, he replied, "I at the same time envy you
the airplane and think there-but-for-the-grace." I knew exactly what he
meant. Building an airplane is endless trouble, and once that ends,
maintaining it is endless trouble too. But it's sweet trouble. Time spent
taking things apart and putting them together, machining, drilling, shaping,
laminating, is, at least for some people, time well spent. People who don't
understand that wonder how you can stand to spend years alone in a garage or
hangar puttering with pieces of metal and fiberglass. But then they don't
understand how a yogi can spend so much time sitting crosslegged in the
snow, either.

The population of amateur builders has changed over the past couple of
decades. The majority used to scrounge materials and build from plans
--sometimes, as in the case of the once popular Wittman Tailwind, very
sketchy plans indeed. Today they order hardware and materials over the
Internet from huge suppliers and assemble airplanes from largely
prefabricated kits.
 
Like any old goat who sees the rest of the flock taking a different route
than he, I grouchily disapprove. I fancy the design-it-yourself,
build-it-from-scratch types are the loftier sort. But when I arrange amateur
builders in a hierarchy that -- quite accidentally -- places my category
near the top, I still see others higher up. For however bold may be the
enterprise of designing a new airframe and building it from scratch, it
ranks below that of another class of homebuilders: the ones who build their
own engines and test them in the air.

Two factors motivate amateur builders to seek alternative engines. One is
cost. Certified engines, especially ones in the 200-and-up horsepower class,
are prohibitively expensive for ordinary folk, as are their parts and
maintenance. The other is a sense, often quite intuitive and unfocused, that
modern technology must have something better to offer than engines whose
basic plan was laid down more than half a century ago.

The natural place to look for an alternative engine is under the hood of a
car in a wrecking yard. Auto engines are cheap and readily available, and
they come in all shapes and sizes. They are known to be reliable, though
that reliability assumes a duty cycle -- 20 percent of power most of the
time rather than 70 percent -- quite different from the aeronautical one. 
Auto engines are much lighter today than they were when their blocks and
heads were made of cast iron, and despite the added weight and complexity of
liquid cooling, one can easily convince oneself that it offers possibilities
of drag reduction and superior temperature control.

A basic problem for aviation applications of auto engines is that they are
without exception designed to deliver their peak power and torque at nearly
twice the rpm at which conventional propellers want to run. The typical rev
limit of around 2,700 rpm was chosen by the ancients to allow propellers of
reasonable diameter to maintain subsonic tip speeds without reduction
gearing. A direct-drive engine needs a large displacement -- six to nine
liters -- to achieve its characteristic power-to-weight ratio of 1.5 pounds
per horsepower at low rpm. Auto engines yield the same power with less than
half the displacement, but at twice the crankshaft speed.

Gearing an engine down sounds like a simple matter, but it turns out not to
be. The engine delivers its power in a series of impulses as cylinders fire
one after another. The propeller reacts like a big spring, its blades
flexing in response to the power pulses. The flywheels and torque converters
that are used in cars to smooth out this turbulent relationship would add
unacceptable weight to an airplane, and so the reduction gears (or chains or
belts) must instead be more robust and carefully manufactured than you would
at first suppose.

Nevertheless, people do put auto engines in planes and they do work.
Aluminum-block V8s and V6s have been popular, as have liquid-cooled Subaru
flat fours. In my opinion, however, the most promising engine for aviation
use is the Mazda rotary.

There are fundamental reasons to prefer the rotary, of which its famous
smoothness is the least important. Compactness is more important; the fact
that it prefers to run on auto gasoline helps; but most important of all is
its simplicity.

Rotaries consist of a three-sided rotor spinning inside an oval, slightly
eight-shaped case. The three edges of the rotor slide along the inner
surface of the case, forming three separate combustion chambers whose
volumes increase and decrease in turn, like those of conventional cylinders.
In order to ensure complete combustion throughout the wide, flat combustion
chamber, two plugs per cylinder are standard. Breathing is similar to that
of a two-stroke engine, through openings in the casing that are covered and
uncovered by the moving rotor. There are no valves, no connecting rods, no
camshafts, no crankshaft.

What makes the Wankel an appealing engine for aviation is that long list of
parts that it doesn't have. The very few moving parts that it does have,
furthermore, are extremely unlikely to break. The only thing in the core
engine that can fail is one or another of the apex seals at the corners of
the rotors, which are analogous to piston rings; but even if these break or
stick the engine continues to run. The rotary engine combines, at least
potentially, turbine-like reliability with the low cost of a simple
mass-produced engine that does not require any high-temperature alloys.

Mazda engines come in two- and three-rotor configurations of 1.3 and
2.0-liter displacement, with outputs in the 200- and 300-hp range
respectively. (In auto racing, where they have been very successful, they
achieve more than twice those power levels.) They have been produced by the
millions, and cores are readily available from discarded cars. They are
inexpensive to buy and to overhaul.

Naturally, it's not that simple. Most of the automotive accessories -- fuel
and ignition system, manifolds, exhaust headers, radiators and so on -- are
unsuitable for aviation use. Then there is the matter of the reduction
gearing or PSRU (Propeller Speed Reduction Unit). These are available from
several sources, but are suitable for use only with wood or fixed-pitch
propellers, not with conventional oil-controlled aluminum constant-speeds.

Estimates of the cost of rotary installations for homebuilt aircraft are all
over the map. George Graham of Bradenton, Florida, reports spending a mere
$2,000 on his engine, but doubts that he could repeat the accomplishment
today. His side-by-side two-seat canard does better than 150 knots at 7.5
gph; it has logged more than 320 hours in the air, on top of an initial 40
on the ground. Ed Andersen of Matthews, North Carolina, who has logged more
than 180 hours in a rotary-powered RV-6A, estimates that powerplant costs
might run $5,500 to $10,000, depending on the builder's choices of accessory
types and vendors. Some improvisation may be needed. "We're still not quite
to the buy it and hook it up stage," he comments, but adds that trying
different approaches is what true experimenting is all about. "Once you
standardize it, the experimenting is over."

Oregonian Perry Mick dispensed with the reduction unit altogether,
installing a direct-drive ducted fan turning at 5,700 rpm on the back of a
Long-EZ. He gets 137 kts at 8 gph -- not great, but then again the entry to
the fan is largely blocked by the airframe.

Another extremely active promoter of the rotary -- the spider, so to speak,
at the center of the web -- is Paul Lamar, who maintains a frenetically busy
Internet newsgroup at http://home.earthlink.net/~rotaryeng/.

Given all this amateur activity, and if the rotary is so great, it's natural
to wonder why there are no commercial rotaries for aviation. To be sure, new
aircraft engines are few and far between. Toyota certified an aviation
version of the Lexus V8 a few years ago, but has made no attempt as yet to
commercialize it. Honda is reported to be developing an aviation piston
engine (and a turbine, too). Continental developed a four-cylinder diesel
under a NASA contract, but then shelved it. France's Renault has also
introduced a four-cylinder aviation turbo diesel which has a takeoff rating
of 230 hp. It weighs about 100 pounds more than a comparable Continental or
Lycoming, consumes about one to two fewer gallons per hour. It reportedly
costs somewhat more than comparable gasoline engines.

I find it hard to understand why these big companies are interested in
getting into the aircraft engine business at all, given the small size of
the market, the great difficulty of getting the few existing airframe
manufacturers to try anything new, and the rich potential for generating
product liability suits. They must hire very unreliable consultants. It is
perhaps revealing that these tentative moves all come from huge companies
with virtually unlimited funds. Mazda, which belongs to Ford, is the only
company possessing both the capital necessary to involve itself in
developing an aviation rotary engine and the technological competence to do
so; but, if it ever considered the possibility, it evidently decided to
steer clear.

It would be a mistake, however, to apply Darwinian principles to the
aircraft engine business and to conclude that if an aviation rotary does not
exist, it must be that one does not deserve to. The same argument could have
been made, a million years ago, about Man. But, now that I think about it...

A twenty-year-old NASA study found a 340-hp "stratified charge omnivorous
rotary engine" superior to both a diesel and a small turbine as a future
engine for general aviation. Nevertheless, despite participation, at one
time or another, of Curtiss-Wright, Lycoming, John Deere, and General Motors
in rotary development, the job of making the rotary into a practical
aviation engine has now fallen into the hands of a bunch of homebuilders.

I hope that the combined efforts and ingenuity of these amateurs will
settle, once and for all, the question of the rotary's suitability for
airplanes. I suspect that they will find in its favor. In the meantime they
remain the noblest and most daring of homebuilders, hairy-chested heroes for
whom the evidence that their latest idea was not a good one may well come in
the form of an engine failure just after takeoff. They continue undaunted,
and their numbers increase. The world owes them a debt of gratitude.