One of the things that bothered me most about building (or owning) an aircraft was the staggering cost of the engine and engine parts. I bought my 193 HP Jeep, with the entire body and extras and every was already put together, for $20,000. The cost of a factory new 180 HP aircraft engine is about $24,000 and it does not even come with spark plugs (8x$17), an alternator, control cables, wiring, intake ducts, air filter, oil filter, oil, oil cooler or starter, AND you have to install it your self. A sodium exhaust valve commonly found in trucks costs $12. The same valve in an aircraft, made of the same material but of a slightly different shape, costs over $200. And these engines really are not that reliable. They are constantly at the limits of their structural tolerance. That is why there are so many recalls and structural failures with certified engines. Check out my page on Certified engine Bloops, blunders, and Recalls to see examples. There really has to be a better way. One rule, no amount of savings is worth compromised safety.
So, while on deployment to Japan, I began reading everything I could about engines and especially auto conversions for aircraft. I read several books: Contact!.. a collection of people who have installed auto engines in air planes, Fuel Injection.. a book about the theory and application of fuel injection, and a book on Turbocharging. I frequently went to the websites of several companies that build or promote the use of Subaru, Chevy, and assorted diesel engines. None of them even came close to measuring up to a certified engine, they were usually too heavy, big or (especially with the diesels) the engines just were not available and the companies claims were doubtful. I decided that there was no way that I would use one of those engine in my aircraft. One exception stood out, the Mazda 13B Rotary engine. The rotary (Wankel) engine is very unique. Of the thousands of car designs on the road today all except the Mazda RX-7 (and a handful or early low production Wankels) use piston engines. The rotary has had a hard time of making a big impact on auto industry. It has its strengths and weaknesses. I am thankfully amazed that Mazda is still producing these remarkable engines. Luckily, the shortcomings of the rotary are minimised in aircraft, while its advantages really shine.
1) The rotary's primary strength is its excellent power to weight ratio. That is, for a given weight the rotary will make more power than a piston engine, or make the same amount of power at a reduced percentage of design workload and/or it will weigh less. In a car reduced weight is good, in an air plane it is great. Climb speed, take off distance, landing distance, and even how much fuel and cargo that can be carried are greatly affected by engine weight. A 180 HP normally aspirated Mazda 13B weighs about 40lb less than a 180 HP Lycoming (bare O-360). Add a PSRU and radiators and normally aspirated 13B rotary installation will weigh about the same as a comparable certified engine. But add a turbo charger, another rotor, or both and the rotary engines will weigh solidly less than certified engines of comparable HP. If you add a P-port to the stock rotor housings the normally aspirated HP jumps up to 250 HP at 7500 RPM. The rotary then becomes comparable to 100 pound heavier six cylinder air cooled air craft engines.
2) The stock 13B components (rotors, e-shaft etc..) are extremely stout and can withstand abuses well beyond their normally aspirated power outputs. Track racers routinely get 200+ h.p. PER ROTOR (i.e. 400 HP for a 13B or 600 HP for a 20B) using STOCK rotors and housings (turbo or superchargers used to boost the power). Drag racers sometimes push that number up above 300 HP per rotor, again with modified stock components. Those stock components are not likely to ever fail in aircraft use, which would usually be less than 150 HP per rotor. Ultimate durability of those components is very high, and replacement and rebuild is exceptionally cheap and easy.
3) The rotary runs very smoothly. This is due in part to its famed rotary motion instead of the back and fourth motion of pistons. But the more important cause of the rotary's smoothness is the LACK OF TORQUE REVERSAL. A 4 cylinder piston engine has a portion of its cycle where it is between power pulses and the prop is actually pushing the engine – an area of negative torque. This happens twice each revolution and is largely responsible for some of the prop/engine RPM limitations seen with certified engines. The rotary on the other hand has no such torque reversal, the torque is always positive (two rotors or larger). This is much gentler on the prop (and PSRU if installed). Another important cause of the rotary's smoothness are the smaller but more frequent power pulses. A 4 cyl aircraft engine has 2 power pulses per revolution of the prop. A 2 rotor rotary with a 2.8:1 gear drive, will have 5.6 power pulses per rotation of the prop. Although this will make for a higher pitched engine sound, it minimises the amplitude of torsional vibration that is associated with each power stroke. That in turn reduces the stresses on everything from the engine mount to alternator and avionics.
4) Cost. Lets face it, the big draw of alternate engines is cost. Unfortunately, the cost of creating a unique installation sometimes offsets the lower price of the engine block. The rotary is not different from other auto conversions in this regard. Proponents of aircraft engines often mention that for the $10-$12k it took to convert an auto engine, you could install a half-time Lycoming and been flying much sooner. Or they point out that the firewall forward kits for auto engines usually cost as much as a new Lycoming. True on both counts, but the savings on an auto conversion comes primarily from lower operating and maintenance costs. That half time Lycoming will very likely need a rebuild sooner rather than later, and Lyc rebuilds cost $15k or more. A rotary can be rebuilt for less than $1000, uses Mogas (or Avgas), and is designed to be used with inexpensive ancillary components such as alternators, starters, injectors, fuel pumps, fuel filters, turbochargers and so on. All of those components come in versions with extensive use in auto applications with corresponding reliability – often greater reliability than what is seen in expensive aviation components (alternators and starters in particular). Follow the link on the left to see my firewall forward costs.
5) Easy and inexpensive fuel injection. Fuel injection offers several advantages over carburation. Automotive and purpose built for aviation EFI systems are reliable and inexpensive. EFI allow for more precise control of mixture to individual rotors (or cylinders). It automates altitude compensation and can provide complex mixture mapping and timing advance settings. More importantly, it virtually eliminates the risk of carb icing which takes down more than a few planes each year. The other advantage to automotive style EFI is high pressure/high flow fuel systems that eliminate the risk of vapour lock.. another major problem that claims victims regularly. Note that aircraft fuel injection systems are not high pressure high flow and are thus still susceptible to vapour lock and hot start issues.
1) The apex seals need lubrication. This would be best accomplished by a cleanly burning oil separate from crankcase oil. Car owners would never tolerate having to add a separate oil that was not readily available and made the car smoke a little. Airplane owners however, are not so finicky. They are used to having to tinker and baby their old aircraft engines. Adding a second oil to a reservoir (or the oil to the fuel itself) is not a big detractor.
2) Due to the unusual shape of it's combustion chamber (higher surface to volume ratio), the rotary is about 10% worse on BSFC (fuel burn) than a piston engine. In the automotive market, this can hurt a little. But compared to carburated aircraft engines designed in the 1940's, the fuel injected rotary does pretty good. Further, because of its lack of exhaust valves, the rotary can be excessively leaned in cruse flight without damage to the engine and thus improve its fuel efficiency. Water cooling can significantly further reduce cooling drag (only if great care is given to system design), bringing the MPG of the rotary above that of its piston aircraft rivals. More importantly, by using Mogas the cost per flight hour and/or mile traveled will invariably favour the rotary over a piston aircraft engine. In practice, rotary powered RV's have fuel consumption that is about the same as those powered by certified engines.
3) A gear drive, or prop speed reduction unit (PSRU) is needed. This adds weight and cost as well as a possible failure point. There are several bolt on PSRU's on the market. All are of high quality and none have had an in-flight failure or other major issue that I am aware of.
Other Auto Conversion Considerations
1) Unique installation issues. Anytime someone creates a "one-off" firewall forward design, issues are bound to arise which may present a significant risk of in-flight failure. Such issues engine mounts, alternator mounts, and the cooling, electrical and fuel systems. Although significant, these issues are minimised with close vigilance. I estimate that somewhere around 500 hours of flying time is enough to wring out most of those issues and thereafter the auto conversion will not be an increased risk of systems issues. I plan to make a separate page someday detailing some of the issues I had in my first 200 hours of flight time, so keep checking back if it is not listed to the left.
2) Resale. There is no doubt that resale on most auto conversions will be significantly less than the same plane with a certified engine. There is little chance the cost savings during the build period will compensate for the lower resale price. So if resale is likely in the first few years after the project is finished, then a certified engine is clearly the best choice. However, as flying hours accumulate the cost savings of the auto conversion will continue to accumulate. At the same time the conversion will "prove itself" and the perceived reliability will increase. At some point, the continued cost saving and other benefit of a proven rotary installation may eventually add to the value. Nonetheless, I intend to keep my plane indefinitely so resale is not an issue for me.
3) Insurance. Insurance will be more difficult for an auto conversion, as many companies will not offer it. There is an initial flight period (of varying lengths) where insurance will be particularly difficult. When a policy is issued, however, they do not seem to be significantly more expensive than those for similar aircraft with certified engines. I have been self insured for my first 200 hrs of flight but will probably obtain liability insurance in the near future, but I do not feel that hull insurance is a good investment for most experimental aircraft. 10-15 years of paying the insurance will usually mean that your insurance payments will total up to the value of the aircraft. And the most you can be reimbursed is for the stated value, but if you want your repairman certificate back, you still have to build it again yourself anyway.
4) Additional construction time. Why do we build experimentals instead of just buying certified aircraft? Essentially because we are willing to trade time and effort in construction in exchange for better value, versatility, and the experience of building. The argument is exactly the same for an auto conversion. I estimate that my auto conversion accounted for 1 additional year in the build process, which was 5 years total for my quick build RV-6. By spending the additional construction time I saved money and gained a more powerful engine that is cheaper to operate and maintain. I also know my engine and its systems better than most RV builders know theirs.
5) Be Unique. I really enjoy showing up at a fly in, and despite the fact that my unpainted and unfinished RV-6 would be one of the ugliest RV's of the lot, it would still often get the most attention because of the engine. I guess you might say that making my own engine installation was a lot more challenging and rewarding for me than spending an equal amount of time sanding the paint for that show-stopping finish. Personal preference I guess.
About 4-5% of accidents are caused by structural failure of the engine. The vast majority of aircraft accidents are caused by pilot error, fuel mismanagement, weather, airframe structural failure, engine accessory failure, and faulty maintenance. But still, wouldn't it be nice to essentially eliminate that 4-5%? When a rotary is turning at, say 7500 RPM, that is perfectly OK. That straight e-shaft has none of the long crank throws (journals) that are such the weak link in certified engines. At 7500 engine RPM, the rotors are only turning at 2500 RPM - those components are not going to fail anywhere in the aircraft operation realm. And oh, those ARE the only moving components. There are no valves, cam shafts, lifters, lifter springs, pistons, heads, or head gaskets that all fail occasionally in certified aircraft engines. The remaining trick is to install all the other systems in a manner that is going the make the engine live up to its potential in terms of reliability
As long as Mazda continues to make a rotary (and the new RENISIS engine promises more aircraft potential than even it's predecessors), it will gradually become a predominant engine in aviation. If you are building an aircraft, and care to save many tens of thousands over the life of the air-plane, and would like something that is potentially much more reliable.
JMHO - David Leonard