See if you agree with my conclusion…
We used to call them homebuilts, which a big majority have always been, and that activity is one of the few areas in private aviation that is still growing. The goal has always been to encourage pilots to build their own airplane for the purpose of self-education and recreation. Regulations are kept to a minimum to keep costs down and in return the builder is supposed to do over 50-percent of the work on the airplane. The activity is right there with apple pie and motherhood but that doesn’t keep naysayers from having their “aha” moments.
From time to time those “aha” folks will come up with the startling (to them) discovery that the safety record in what the NTSB now calls E-AB (experimental – amateur built) aircraft is not as good as it is in store-bought certified airplanes. Depending on whether fleet size or flight hours is used as a measure, the E-AB fatal accident rate has been shown to be from two to seven times higher than for certified airplanes.
A look at the accident history of these airplanes can put this into context. In a 2012 study the NTSB compared E-AB airplanes with certified airplanes used for the same purpose. In both groups that basically meant piston singles though they did also include gyrocopters which would be almost exclusively E-AB.
The NTSB found 131,700 certified piston singles and 21,300 E-AB airplanes that fit. (I have seen E-AB fleet estimates as high as 33,000.) For some reason the ratio of accidents to airplanes is often used when talking about E-AB airplanes where the number is usually in relation to 100,000 flight hours for certified airplanes.
In raw numbers, there were 70 E-AB fatal accidents in 2009 and 31 in 2013. That covers the highs and lows and in perusing the 2015 and 2016 preliminary accident reports I counted 52 for 2015 and 43 for 2016. The big difference between the high and low annual numbers is something that has never happened in similar certified airplanes, where the rate stays pretty steady.
Those are significant numbers when compared to an estimated fleet of from 21,300 to 33,000 airplanes. If you split the difference on fleet size and annual accidents, that would be one fatal accident for about every 600 E-AB airplanes. Using numbers I found in various reports I found that number to be about one fatal for every 1,000 certified singles. In its 2012 report the NTSB found a greater disparity than that.
As an aside, last year there was one fatal accident for every 8,136 registered motor vehicles in the U. S. There is no comparable number for U.S. airline flying because in almost all recent years there have been no fatal accidents.
It is assumed that certified airplanes fly more hours each than E-AB airplanes so the fatal accident rate per 100,000 hours would show a greater difference. Anything based on flying hours is suspect because nobody really knows how many hours are flown and the FAA has resisted calls to get an accurate count by simply requiring a post card after each annual inspection that shows how many hours the aircraft flew that year.
The higher incidence of accidents in E-AB aircraft is just as logical as the fact that the fatal accident rate in private (general) aviation is almost infinitely higher than it is in airline flying. When more freedom is granted by reducing regulations and eliminating stifling procedures then the risk goes up. In airline flying, risk is managed by making all pilots and carriers toe the line on procedures and regulations where in private and especially E-AB flying, risk management is left solely to the pilot. For E-AB that freedom extends to the building and testing of the airplane though the FAA does have requirements for the flight testing of newly-built E-AB airplanes.
E-AB activity is also well-supported by the EAA. The association offers excellent information and guidelines for builders, from construction through the required 25-hour (40 if an uncertified engine is used) test flight period, and individual EAA members and local chapters also offer support and encouragement to builders and pilots as they build and enjoy their airplanes. If you want to get a feel for this, look at the website for Chapter 1600 in Delaware, Ohio.
One of the first friends I lost in an airplane accident was flying an airplane he had built himself, from plans. It was small, a single-seater, and the only time I saw it was after the accident. There was not much left and what was there looked pretty flimsy to me.
The accident happened in the traffic pattern and it was likely a low-speed loss of control and spin-in. According to an NTSB study, loss of control is a factor in 43.9-percent of E-AB accidents v. 29.7-percent for certified airplanes.
That difference there is not as much as I expected it to be. The low speed handling and stall characteristics of a Cessna 172, for example, are carefully addressed in the certification requirements where in an E-AB those characteristic are left pretty much to the designer and builder. Maybe this tells us is that even without regulation, fairly careful attention is paid to low speed handling qualities in E-AB airplanes.
Flying qualities likely have a lot to do with the fact that a disproportionate number of E-AB accidents happen in the first hours the airplane is flown after it is completed, or in the first hours flown by a subsequent buyer. A pilot moving among certified airplanes won’t find many differences in basic handling qualities where there is likely a much wider variation among E-AB airplanes.
I don’t have a lot of hours in E-AB airplanes but I have flown a number of different ones and have seen a wide variation in flying qualities, from terrible to excellent.
I’ll tell you about the excellent ones first.
When I flew the Swearingen SX-300 my first thought when I looked at the diminutive wing from the cockpit was about Ed Swearingen. I first met Ed when we were in our 20s. I was in the Army at the time and an L-23 (Twin Bonanza) showed up that had been heavily streamlined (probably with body putty) to get some extra speed.
Ed had something to do with that and with a lot of other airplanes as we both made our way in the aviation business. So when I looked out at the tiny wing, I thought it was okay because I trusted Ed. The airplane was delightful to fly though it sure was loud inside.
The Questair Venture was the same deal. I trusted Jim Griswold, who was key in the development of the Piper Malibu and who designed the Venture. When we finished the photo mission in that airplane, a low cloud deck had moved in and I had to fly a low-weather ILS in the airplane on my first flight in it. Jim’s pilot who was with me was not instrument rated. I wondered about the legality of that but decided it was one of those “don’t ask, don’t tell” situations. It was nice to fly on instruments.
I flew the Glasair, too, and found the handling qualities there to be good.
Those airplanes are small with big engines. There are no secrets in airplanes and you could look at any of them and tell that they would glide like a brick. I tried power off approaches and landings in the SX-300 and the Glasair and they reminded me a little of a helicopter autorotation. You had to arrive at the flare point with enough energy and there, in an instant, you had to make the peanut butter and jelly come out even.
That’s what E-AB is all about. If you want a little airplane with a big engine you can have (build) it and the stalling speed can be higher than the 61 knot limit for certified airplanes. That there will be more risk in case of an engine failure is obvious.
In the case of the E-AB airplane with the worst handling qualities, the Prescott Pusher, the problems there were pretty obvious too. It was a chunky, short-coupled airplane with a t-tail, a long wheelbase and a 180 horsepower engine. That suggested a sluggish takeoff with a lot of pull required to rotate and then an immediate reduction in back pressure. That’s a recipe for a pilot-induced pitch oscillation, which I had seen before but never that bad. It was twitchy as all get-out in pitch.
I offer those examples just to illustrate the potential for a wide variety of handling qualities when airplanes are designed outside the certification rules.
Pilots who have accidents in E-AB airplanes tend to have more flying experience than those wrecking certified airplanes. They are also older. There is not enough difference, though, to set the two groups of pilots apart.
Powerplant failures lead to almost twice as many accidents in E-AB airplanes and these also often come on early flights after build or purchase. Because power failures are often followed by a low-speed loss of control, these two accident causes do go together. I think the fact that less than half the E-AB powerplant failure accidents occur in airplanes that use a certified engine helps explain the number of power failures.
A frequent cause of power failures, especially early in the life of an E-AB airplane, is the lack of a proper amount of fuel and air being fed to the engine. This is a design matter, well covered by the regulations for a certified airplane, but if not properly addressed in the design and build of an E-AB airplane it can lead to an early power failure. The NTSB did note that the number of E-AB engine failure accidents could be substantially reduced by requiring that the builder document a functional test of the fuel system. This is actually in the EAA-recommended procedures before a first flight.
There are less weather-related accidents in E-AB airplanes than certified airplanes even though a lot of them are well-suited for cross country flying, where weather accidents usually occur. Properly equipped, they can be flown IFR, too, and just now, at 9:25 a.m. I counted five E-AB airplanes in the IFR system.
I attended my first Experimental Aircraft Association fly-in before the event moved from Rockford, Illinois, to Oshkosh. I didn’t know a lot about the homebuilder activity at the time and, as I walked the grounds and looked at the airplanes, my main thought was that a lot of them did indeed look like they had been built in a basement. They were, however, the pride and joy of the individuals that built them.
At the time, most of us thought of homebuilts as Pober Pixies and Breezy Pushers, both simple airplanes that were of greatest value to those who just wanted to get into the air, literally in the case of the Breezy which you fly on, not in. I did it once, it was a hoot, and I never went back for more.
How things do change. As time ran more build-kits were developed and fewer airplanes were built from plans, though that activity still goes on.
Van’s Aircraft currently shows eight different models of the popular RV series in both two- and four-place versions. When I had my airplane based at Hagerstown, Maryland, I had hangar neighbors who had built and were flying RVs. The workmanship in the airplanes was beautiful, better than any factory-built metal airplane, and the owners used the airplanes just like you would use a Piper, Cessna or Cirrus. The airplanes are fast, efficient and comfortable.
Where a lot of people equate amateur-built with inexpensive, that isn’t necessarily so. An example is the Lancair Evolution, a four-place pressurized airplane powered by your choice of three different P&W PT6A turboprop engines. It is an airplane you would use like your King Air except if you wanted to see how the King Air was doing you would need a rear view mirror in your Evolution.
You would also need at least half a King Air checkbook. Used Evolutions run well north of a million bucks and I doubt if you could build a new one for less than a million and a half.
The Evolution was developed by Lancair and the kits are sold by The Evolution Aircraft Company which also operates an extensive program to help buyers complete their airplanes in a manner that satisfies the FAA requirement for builder participation. It is definitely not a garage or basement project and professional help is a necessity.
The company is careful to make sure the airplane is built in compliance with those FAA E-AB rules and indeed builder attendance is required for a good deal of time during the process though what the builder does is steered away from the critical stuff.
I was interested in a comment on the Evolution website about approved ice protection equipment. The point was made that there is no type certificate for the airplane so no required supplemental type certificate can be obtained for icing flight approval. The equipment is available, it can be installed, and it will offer the same protection as it would if approved so I suppose that is just a bit of trivia.
There are likely ways for a person to have one of the many kit-built airplanes built by someone else and flown through a loophole in the regulations. There are also kits that look almost like completed airplanes when they reach the builder.
It is also perfectly legal to sell a completed E-AB airplane and there are a lot of them for sale in and on Trade-A-Plane. It is interesting to me that the NTSB found that new owners have many of the same problems that were experienced on early flights by the builders themselves so if you purchase a used one, approach it like it just hatched.
It is difficult to get an idea about the safety record of different E-AB airplanes because the FAA registration rules allow the pilot to register the airplane under whatever name he wishes. Most builders use their own name with a reference to the type, like John Smith RV, but nothing specific is required.
Just looking at the list of accidents, no conclusion can be drawn other than that the pressurized airplanes the Lancair IVP and Propjet and the Evolution might be there more often than things like RVs. This same situation exists in certified airplanes where P210s are wrecked more often (per 100,000 hours) than T210s and Malibu/Mirages more often than Saratogas.
In its study, the NTSB did note that in-flight structural failures don’t often occur in E-AB airplanes.
They do occur, though, and one E-AB accident is worthy of note here.
Steve Wittman was an icon in the homebuilding and air racing business for many years. He built some outstanding racing airplanes and did well flying them in a lot of races. Steve was also one of the founders and a force in the Experimental Aircraft Association. That little airport where pilots gather in great numbers every summer is, in fact, Wittman Regional Airport.
Steve lived in Oshkosh and Ocala, Florida, where it is said the winters are more pleasant. He built a one-of-a-kind airplane to fly between the two places and named it the O & O for Ocala and Oshkosh. It was a two seater, maximum takeoff weight of 1,850 pounds, powered by a Continental O-470 rated at 205 hp. No doubt that it was a good cross-country airplane.
Steve and his wife were lost when flying that airplane from Ocala to Oshkosh in April, 1995. He was 91 at the time.
The NTSB probable case and findings were as follows: Aileron-wing flutter induced by separation at the trailing edge of an unbonded portion of wing fabric at an aileron wing station. The debonding of the wing fabric was a result of improper installation.
The airframe had apparently failed as they cruised along in good weather.
Mistakes have been made when designing, building and maintaining all types of airplanes including all classes of private (general) aviation airplanes, airliners and military airplanes. A lot of those mistakes have led to accidents, as this one did, but that has never stopped aviation from flying on.
I have never owned an experimental airplane nor have I flown them extensively but I feel a bond with the folks who do.
The accident record in private flying when I soloed in 1951 was about like the accident record in E-AB flying today. When I started, we were free spirits, a band of brothers, who loved to fly. It was almost holy and we weren’t looking for anyone to save us. We knew that what we were doing was a lot riskier than knitting and we didn’t care. That is all true of many if not most E-AB pilots today and I think that in both cases the accident record is simply the result of the interface between the romance of flying and the pilots who practice the art.
So if you ask me what is wrong with experimental pilots, I say nothing. Just leave them alone and let them enjoy flying.
- From the archives: how valuable are check rides? - July 30, 2019
- From the archives: the 1968 Reading Show - July 2, 2019
- From the archives: Richard Collins goes behind the scenes at Center - June 4, 2019
I own both a certified airplane and an E-AB. When getting into the E-AB, risk wise, it feels a little bit like getting on my motorcycle after driving the car. I have always suspected that the feeling is pretty close to reality. But something would be lost if I stopped riding the motorcycle or flying the E-AB (oh hell, it’s a Flybaby, not some generic acronym). As long as one understands and accepts the risk, it’s ok. I don’t much believe in exposing others to that risk unless they also well understand it.
I think that there is one major flaw in your argument. That is the assumption that E-AB owners fly just as much as owners of certified airplanes. I will say from experience that is not true. I fly 4 to 5 times as much in my E-AB than I could ever afford to fly in my piper. Why? cost. I spend 20 an hour to fly E-AB and 85 to fly my piper. So guess what, I fly more in E-AB. So I think when comparing the two regardless of fleet size the total hours flown by the E-AB fleet and those from Certificated are close. Then when looking at the rate per 100,000 hours the accident rate is close.
It appears that your experience is unusual. I was offering the most favorable comparison I could find. If you look at the fatal accident rate per 100,000 flight hours, the NTSB shows E-AB at 5.27 compared with 1.56 for like certified airplanes. AOPA shows E-AB at 6.35.
Richard, from your article and my experiences in the EAA world, it’s clear there are many explanations for crashes of E-AB aircraft. Each is unique, and as you have written many times, most crashes result from chains of decisions and events leading up to the bad end. I have never built an airplane, and there are some homebuilts I would not sit in, much less fly. But the ones I have flown in were meticulously built and tested, making flight in them a non-anxious delight. Two RV-12s I’ve flown in were masterfully designed, marvelous machines, little airliners. Like you, I say of aircraft homebuilders, leave ’em be. It would be so easy for the gov’t to screw this up, a potential tragedy, because homebuilt aircraft may be the salvation of personal flying in America.
There are several points in the article I question. First, your state up front that “two to seven times” more EAB accidents than certified and then later you estimate accidents at 1:600 for EAB and 1:1000 for certified. That is 67% and not 200% to 700%. Next you assume fewer hours per year for EAB vs certified. In my admittedly small contact list, the opposite is true. Later in the article you meantion engine failure. However the NTSB data shows “pilot error” and not “equipment failure” as the predominant cause.
This article would be more accurate in its concerns if you looked at EAB pilots vs certified aircraft pilots. I do believe there is a psychology at play with a number of EAB accidents.
I think one problem with comparing statistics between E-AB and type certified airplanes is that there is a first-flight test period with every E-AB but only one truly first-flight test period with each model of type certified airplane. If you eliminate the accidents during the first-flight test period for E-AB you will find the comparison will change significantly and be much more comparable. Conversely you could include the accident statistics from the first-flight period of type certified aircraft but since the number of models introduced is so small compared to every E-AB aircraft built, and by definition a first-flight test period is required for each E-AB and not for subsequent type certified aircraft of the same model the statistics will still be skewed.
Well said! And as someone who has flown fixed and rotary wing and ridden many motorcycles, I think you make an apt point.
E-AB testing is conducted by the builders. Some are very experienced and some are barely legal, let alone proficient. I met builders who are master craftsmen but on the piloting side they lack basic skills and experience. I found nothing wrong with that, but try to politely suggest anything that will sharpen piloting skills on a certified AC before E-AB testing.
A Glasair owner in a next door hangar introduced himself as private pilot and Engineer (structural) in west central Fla. He kept bragging about his speed mods to his plane and never missed the opportunity to tell me that the “factory” people knew nothing about speed mods. He also revealed to me that he had a “small” problem with oil temperature. When he showed me his “mods” it was clear at first glance that in his quest for speed he buried the oil cooler under the cowling where no air circulation takes place. On top of that the air intakes of the cowling were reduced to half size and the bottom rear part was closed off for “aerodynamic benefit”. His theoretical airspeed gain numbers were minimal at best. I asked, why deviate from the original plans. First excuse, he wanted to be the fastest Glasair owner. Second: -I am an Engineer my time is very valuable, need to make the 120 mile trip to KFMY fast.
I agreed to fly with him after he made some changes I suggested, the oil temp was very high even with power reduction in the short flight and his piloting skills extremely poor.
He revealed to me that he had several accidents in the past alone and with a rusty CFI who refused to fly with him again. I declared that I am the second CFI that will not fly in that modified plane again.
Had similar experience in a Wittman W-10 with its owner builder. Great craftsman but weak pilot,he knew Steve Wittman very well and visited him very often for advice. He lost his medical for years and when he received a temporary waver, I agreed to fly with him so he can become proficient again. His taxi coordination was poor so I suggested ground runs at various speeds, tail raise and deceleration. As he was getting better at it, one day shortly after partial throttle acceleration on the runway the tail was erratic and my rudder inputs where almost useless to help him. I used all my focus to keep the plane under control and avoided getting off the 150 ft width of the runway only to find us on the opposite side. The owners hand was frozen on the throttle, eventually I pulled it back and managed to stay on the center line. It was rudder mechanism structural failure bellow the floorboard. The next day the owner came to me with the broken part in his hands apologizing for the failure. He said that many W-10 owners re engineered the part as SW did not worry about its integrity.
That is the closest I came to an accident in my four decades in aviation but I learned a valuable lesson. Career test pilots go to school like aeronautical engineers do, then they work together during testing. The FAA has the final word for Certification to protect the Public in general. We all benefit from that.
Even gifted people like Steve Wittman can overlook a small but fatal detail. It is all about Human Factors or the Human Element. Ego is a Human flaw but some of us are blessed to be able to control it.
Fly safe !!!!
Chris Wow. Now that’s a story. I’m not a pilot just have 2 son’s learning. I’m glad you’re alive to tell this!
GREAT article Richard. I graduated from RPI with degrees in Aeronautical Engineering/Science in 1955. Uncle Sam paid to make me a pilot in the class of 57H back in 1955-57…almost ‘washed out’, but graduated to receive my wings… flew for about 60 years in Piper Cubs, T-28s,T-birds, and B-47s in the Air Force, F-86Hs in the Mass. Air National Guard, and a Cessna 182 that my wife and I owned…. flew to AirVenture some ten years plus. I am now a ‘Ground Pounder’, but mentor with youngsters and ‘oldsters’ about many Aviation topics… VERY rewarding experiences that continues EVERY day!
Wow. Now that’s a story. I’m not a pilot just have 2 son’s learning. I’m glad you’re alive to tell this!
Whoops, I clicked the wrong reply button! Sorry Joel!
Good article Richard, kudos. The comments demonstrate the passion EAB builders have for the self build freedom. Keep poking ’em. At the least some frank observations my stimulate useful introspection. Maybe some useful learning.
I agree with the conclusion. And from my PoV, too, good luck to them, and let them continue in peace.
I’m in the other group. I want to be able to focus on flying rather than the airframe, and I’m grateful to the all designers, engineers, and test pilots at Piper who came up a PA-28-161 that makes me look great on my good days, and keeps me alive on my…um…less-than-good days. The spirits of all those people, the people before who they copied and learned from, and even the little-loved FAA regulators, are there helping in the cockpit every time I fly.
From the article David > “So if you ask me what is wrong with experimental pilots, I say nothing. Just leave them alone and let them enjoy flying.”
In the back of my mind I’m thinking it should have been written… ‘let them enjoy dying from flying?’ The old pilots thing. That would be me if I were to take it up myself.
I’m not a pilot but I can agree with your sentiments here. With a son active in flight school, I would never let him touch an experimental aircraft at this point…. Just a thought, (but he would jump at the chance tho). I see you in a lot of forums, it’s always good to read your take on things. I dunno, but there are so many cases of much older pilots dying doing what they love to do in the ntsb archives & kathryns report. (isn’t that what relatives of dead pilots always say?) Maybe it’s a case of ‘God finally catches up with them’ or something like that.
Just some random thoughts. Thanks for your comments. They are always informative!
As a fairly recent RV-8 purchaser and 1/2 owner I can attest to the challenges of moving into the E-AB class of airplane. Both the required Tail Wheel check-out (C-170B, and then on to the RV), and the performance increase were enough to get my serious attention, both before, and after I was approved to solo. But the decision to do the E-AB was a well researched lesson in Risk Management, and I decided to do it.
During delivery the previous owner cautioned me that it would take about 50 hours to get reasonably comfortable (?!!!), and that was independently seconded by another RV-7 pilot. At a chance encounter airport lunch meeting, I shared a picture of the new plane with a local and well respected FAA Flight Examiner; after the “Nice plane….” dialog took place he looked me right in the eye and said “Nothing matters more than you getting properly trained”. I now fully concur all of their opinions.
I am so glad to be among the E-AB group of pilots. The experience is much the same, but at the same time very different from Part 23 Certified airplanes. Just as previously mentioned, like motorcycles, they are not for everyone. But for me they are more fun to fly, cheaper to own and operate, and the performance increase is, well, just that!
I do believe that the safety record of E-AB does need some improvement. I’m kinda’ new to this group (and maybe it’s already being done), but I bet that just identifying, and then formally addressing the top two or three safety problem areas through EAA, AOPA and the mainstream aviation media (which I think does not do its part in representing the E-AB segment of the flying community) would be a very good thing.
Richard;
Insightful analysis, as usual. I wonder if something can be learned from looking at the accidents per flight (each take-off and landing cycle). The airlines do a terrific job of staying safe, but I would wager that GA has a significantly higher flight cycle per hour rate, which would make the accident rate per flight cycle look slightly better for us amateurs compared to the pros.
From my observations, there are builders and flyers. Having assisted on a number of E-AB aircraft, I’m an ATP, A&P, and a CFII SMEL. I have observed that builders, sometimes, lose focus when they feel their project is complete. Beautiful builds don’t necessarily mean ” good handling characteristics”. Example, helped on a scale P-47 that was an outstanding performer, 160 mph cruise w/ an O-235. It would overwhelm a lesser pilot on landing who was not capable in an aircraft like a Pitts. Back to the discussion, builders want to prove themselves, and their work. They are, often, to push limits to prove their hard work is validated.
There are just the same two groups when it comes to model aircraft, there are builders and there are pilots, but just a few of us are both excellent pilots and excellent model builders.
Especially if the model is a one-off, possibly made after a photograph or some very basic drawings, say from Jane’s All the World’s Aircraft.
And just because the finish of the model is superb it does not mean that it handles well in the air, nor that the RC equipment is optimally set up.
So not so surprisingly many do crash during the first handful of flights, no matter if the pilot is a seasoned pro or not.
And homebuilt aircraft is much the same, it is the first hundred flights that is the real challenge, after that it is weather and health-issues, pilot and engines!
I’ve given thousands of hours as a flight instructor and I still cringe when I have to perform a flight review or a check out to pilots I didn’t train. Airmanship and basic flight skills are grossly lacking in a lot of pilots. I’ve also flown a number of experimental airplanes and their flight characteristics require a higher level of experience and skills. This is a bad combination. Please don’t let affordability get in the way of safety.
I always like reading the archived Richard Collins stuff. I have learned so much from it.
The Prescott Pusher in the photo is flown out of 3CK by a retired professional pilot. I have never had the honor in flying in it but the owner modified the kit design to add the two aft rakes to provide additional slow speed stability to the back end. I am told that screams right along to make KOSH in about 40 min.
I am sure Richard would be pleased to know it is one of two that are still flying but there is an ambitious builder working on a jet powered one as we speak.
I am the builder and owner of the Prescott Pusher pictured in the article. I agree that the airplane as designed is a lousy performing and handling airplane; this dawned on me during the building process, but I was not ready to throw away some 50,000 dollars paid for the kit and other stuff. So I decided to at least try to figure out what was needed to make this a well performing and handling airplane. All together it took me 11 years before the airplane flew for the first time. I have well over a hundred modifications, aerodynamic, structural, 100 more HP, CS prop, stability, pitch control gearing, stall warning system, procedures and more. It has now been flying for almost 22 years, and while the handling is different during TO and landing, in most other phases of flight, it is perfectly normal and a great performing airplane.