Do-it-yourself airplane design
In my quest for a better back country airplane, I wanted more power than my Rotax 582-powered Avid had available, and I wanted more cargo capacity to carry the fun stuff. This started the building/designing saga described in the November ’96 and January ’97 issues. The design was centered around a good, affordable engine with at least 100 hp, and so the research began.
Horizontal Equals Reliable
The horizontal opposed engine has been used by the aircraft industry for more than 60 years and is still being manufactured today by Continental and Lycoming. What is it about this design that makes it cling to the propeller-driven aircraft industry like a starfish on a round rock? The answer is that this design runs smoother with less harmonic vibration than other four-cylinder designs.
The crankshaft is shorter on a horizontally opposed cylinder layout than the crankshaft of an inline engine of the same displacement. The shorter crank means less rocking couple effect, and these factors total to the one feature necessary to win an aircraft engine popularity contest: reliability!
If I didn’t mind parting with the bucks, the Continental 0-200 would be the choice; it is the standard of the industry in the 100-hp range. This well known, no-longer-manufactured engine is a horizontal-opposed, four-cylinder, OHV pushrod powerplant that turns out 100 hp or more, depending on how it is tuned. I just can’t handle spending in excess of $5000 for parts to zero-time this engine when you can do the same to some auto conversions for less than $1000.
In the world of auto conversions for airplanes, there are just two recently manufactured auto engines with the horizontal opposed cylinder layout: the VW and the Subaru series. In the 100-hp category, the Subaru EA-81 is the engine of choice over the VW because of its strength.
The EA-81 and the Rotax 912 are very similar. The biggest difference is in the rod bearings, and also how they are cooled. The Legacy EJ-22 has since replaced the EA-81 in Subaru’s current autos and is becoming a popular conversion for 160- to 180-hp aircraft. Its drawback for smaller aircraft is that ready to fly, the EJ-22, plus a propeller reduction drive, weighs somewhere over 300 pounds, depending on how it is configured.
The EA-81 engine was introduced by Subaru in 1980 and stayed in production as the auto engine through 1984. It is still being manufactured in Japan and sold as an industrial powerplant. You can buy a used Subaru car and turn it into an engine donor, or you can zip down to your favorite junk yard for the engine only. Look for the EA-81 letters cast onto the block, as the 1600cc engine was still being sold on some models up through 1982. The stock engine you are looking for will be a used 178lcc (108.68-cubic inch), and can deliver 72 hp at 4800 rpm and 92 foot/pounds of torque at 2400 rpm. It has a compression ratio of 8.7:1, and the overhead valves are pushrod operated.
In 1985, Subaru introduced the EA-82—basically the same block as the EA-81 except the valves were actuated by overhead camshafts and timing belts rather than pushrods. The EA-82, in production until 1989, is not as popular with kit builders because it is wider, the timing belts are short life, and you can get as much power out of a properly tuned EA-81. Either of these engines can be found with the turbocharger option.
Importers who sell used engines from Japan are another Subaru source. For reasons more political than practical, the Japanese government requires car owners to replace the entire engine at about 30,000 miles. These engines are sold in the United States for about $500, depending on the engine, and they make very good rebuilding cores.
If you were to transfer a stock EA-81 from car to plane in place of the 582 Rotax, you would be replacing 65 hp with 72 hp with about 100 pounds of added weight, so the increase in weight easily eats up the extra 7 hp. You can run the Subaru as a direct-drive engine, which is the good news. The bad news is that it will only deliver about 50 hp at 3500 rpm. Higher rpm is not practical as the propeller diameter must be quite small to avoid the problem of supersonic propeller tips, which loses efficiency in addition to creating a terrible racket.
Specially designed propellers have been developed for Formula 1 racers that can operated in the 4000 rpm range, but these would not work in STOL situations. A turbocharger must be used with the direct-drive application to get any notable power increase over stock below 4000 rpm.
The normally aspirated EA-81 must be able to run at 5500 rpm to develop 100 hp, and this requires a prop speed-reduction unit (PSRU) to slow the propeller speed into the lower 2000-rpm range. As the PSRU will increase power package weight, it will be necessary to pare every pound to have a good powerplant.
Here lies a problem for the engine builder: There are basically three ways to get close to 100 hp from an EA-81 Subaru engine.
To make the U.S. car engine an outstanding performer, you need to rework the engine’s induction system so it will develop something like 100hp. High-rpm power is obtained by increasing the engine’s ability to breathe, which requires cleaning up the heads with porting. Better breathing needs longer-duration valve opening, which results from changing the camshaft timing. Carburation must be matched with the greater engine demands.
The turbo-equipped EA-81 can be purchased from importers for about the same price as normally aspirated engines and will give you 94 hp at-4800 rpm. But you need the electronic brain box to run the engine, and it was left behind in Japan. Even from the local junk yard, the turbo presents additional building problems such as getting the engine and turbo under the cowl and then keeping it cool.
Also, there are reliability factors associated with engine management. You need to be a person who loves to look at gauges if you want to keep the turbo engine out of trouble. Another problem is high fuel consumption.
The idea of buying the used car and moving the whole assembly into the plane is done often, but it seems that a person’s affinity for gadgets on the engine is inversely proportional to the resulting number of forced landings. Simplicity is a virtue.
The HO model (Subaru’s Asian engine) is another way to go. This engine is used in trucks in Japan and comes with dual carburetors and a different cam and cylinder heads. It is imported with the rest of the 30,000-mile Japanese engines and is hard to find but can be bought from used engine importers.
The HO intake valves are in the center of the heads with the exhaust valves on the outer position, which is opposite from the stock cylinder heads. This feature improves high-rpm breathing. The HO will save you the trouble of changing the cam shaft, as required with the stock engine, but will have more questionable reliability due to the stress of truck duty in Japan.
Because the HO was never marketed in the U.S., dealers will have trouble supplying parts—or even knowing what you are talking about. Due to their scarcity, most professional rebuilders avoid them as they cannot promise volume delivery or parts on a regular basis.
As usual, there is no free lunch. The idea that you can get a complete 30,000-mile engine for $400-$500 is mind boggling for the cheapskate who has been wondering where he is going to get $8000-$9000 for a Rotax 912. The trouble with the stock engine is it comes with so may EGR tubes, pipes and sensors that it looks like a candy bar being devoured by half a dozen night crawlers. This stuff has to be eliminated.
The stock alternator is heavier than necessary, and the starter cannot be used without an adapter as it mounts on the engine’s bell housing, which must be left behind with a belted PSRU. An exception is the Ross Aero planetary PSRU, which bolts to the stock bell housing.
To get a PSRU for the engine, you might choose AMAX, which sells a unit with starter for about $2000. The RFI PSRU sells in the $1500 range with no starter. And to rebuild the engine to original specs will cost $500-$1000, depending on who does the work and what you might spend for parts.
The stock carburetor setup will not fit under the cowl of the average airplane, so you must come up with a carburetor or injector system, plus a manifold. The engine mounts must be made or bought, as well as an adapter to fit the engine to the frame. I spoke to one machinist in California who had his plane in the air for $2800 by manufacturing all his own parts, including the PSRU. If you are like me and are gifted with a hammer and chisel but not blessed with machinist skills, you will probably be looking at $4000-$5000 by the time you get rid of the ugly pounds and are in the air.
My instinct was to build the engine myself, as I love engines and like a challenge, but the determining factor was that I had yet to find a pilot who was doing his own engine who did not go through a long tedious process of getting the right parts and then making them work harmoniously.
If I already had the plane perfected, I would be ready to tackle an engine-build project, but since I had a hat full of unknowns with the controls and other systems, I didn’t want to worry about the engine also. But Catch 22 was that I could not build the cowl and finish the frame until the engine decision, which would determine the thrust line.
Hope appeared on the horizon when I was visiting the Avid Aircraft factory one day and spotted a static display Stratus Subaru engine that had been to Sun ’n Fun and Oshkosh a couple of times. My original interest in the Subaru was spawned in 1991 when Reiner Hoffman installed the EA-81 in his Avid Flyer with spectacular results.
I knew that he had gone on to research and develop a conversion kit, and I personally knew several Stratus owners with more than 500 hours of trouble-free flying. I called Stratus to see if he would part with the powerplant, but he said it was for display only and he would want to overhaul it before he could guarantee it. I talked him into letting me use it to build the plane. He agreed to overhaul it at a later date, and I had an engine that I could bolt on and design around.
Getting It Together
I brought home the prototype Avid Subaru mount, along with the engine, and could hardly wait to get the engine on the frame so I could begin to build and fit the cowl and get a preview of the finished product. I completed the cowl, slipped it on and stood back to admire. What’s wrong with this picture?
It took some scrutinizing, but it seems that the engine was pointing a little upward rather than the customary downward. I called Dean Wilson, the original Avid designer—and he advised, for my plane, the engine should have 1.5-2° of down thrust, and the engine should be offset to the side about 3° to counter P- factor and torque.
He explained that if the engine had no down thrust, it would want to change pitch radically with every throttle change, and if the engine is not offset for P-factor, it will change heading with throttle variations.
Wilson also gave me a useful tip: If you take a board 4 feet long and taper it from %-inch at the wide end to 0 at the other end, the resulting angle will be 1°. Armed with this, I leveled the plane and set a level on top of my 1° stick on top of the PSRU and found I had about 3° of up-thrust rather than any down. Stacks of washers under the engine mount bolts got me the necessary down-thrust. (Avid Aircraft has corrected this problem on the mount they offer for sale.) All of these factors must be considered if you are making your own engine mount.
With the thrust line nailed down, I could finish fitting the cowl. I had a radiator shop make a larger version of the Avid Mark IV cooling system (which I had designed), and I mounted it up front in the cowl. I owned an IVO three-blade prop, and Ivo-prop traded me for an IVO Magnum ground-adjustable prop that I installed after trimming and fitting a spinner.
Some time was lost completing the plane because I kept sitting in it having delirium tremens as I mentally flew off into the limits of my imagination with the engine going pocketci, pocketa.