Homebuilt Safety Survey

For buyers seeking alternatives to spam cans, the record is mixed at best.

Homebuilt airplanes have found fans among pilots seeking the speed, utility and rakish good looks that, until recently, the average spam can couldnt provide. (Okay, so the Lancair Columbia still isnt a spam can and certainly isnt average.)

With most new aircraft production limited to mere refinements of old designs, sporty new styles have an undeniable appeal. Many buyers-notably those with more money than time-are turning to completed homebuilts instead of new or used certified aircraft when putting their offers on the table.

But homebuilt airplanes are licensed as experimental by the FAA for a reason. The only scrutiny the designs face is from the marketplace. A design that builders like will succeed, but one that maims or kills wont find many buyers.

One of a Kind
Theres no stall testing required, no climb rate at gross weight requirements, no prohibitions against spins or loops. Compounding the problem is that every homebuilt airplane is a one-of-a-kind flying machine. Small vagaries in construction technique, minor mistakes, even a light or heavy touch sanding a wing surface might combine to make the airplane handle differently than any other.

The recent well-publicized crash involving singer John Denvers recently purchased Long-EZ and another accident involving New York Times Magazine writer James Gleick have focused a spotlight on the murky world of the purchasers of completed homebuilts.

A recent issue of Trade-A-Plane lists 78 completed experimental aircraft for sale-almost as many listings as for A-36 Bonanzas. They ranged from Rotax- or Volkswagen-powered biplanes to a fire-breathing Glasair powered by a 750-HP turboprop.

Many of them hadnt flown off the 25 hours of flight testing the FAA requires; some were even waiting for the buyer to specify paint and interior. Clearly the buyers of some of those airplanes will be test pilots in the truest sense of the word, if not within the FAAs designated test window, then in the coming months and years when the aircraft faces flight situations impossible to test during the short, mandatory break-in period.

Others will be flying proven designs that have been flown several hundred hours by their builders. The question buyers need to ask: How safe are the airplane designs and the amateur builders construction?

Many Designs
There are more than 600 companies offering plans or kits for amateur-built aircraft, according to the Experimental Aircraft Association. The FAAs stats recently showed registrations for 19,643 amateur-built airplanes and builders are actively working on about 40,000 to 50,000 more, according to some estimates.

The popularity of homebuilts has increased dramatically in recent years. Twenty years ago, the typical airplane builder was a blue-collar trade worker who was good with tools. These guys had more time than money and focused on finding cheap ways to get into the air. With the advent of nearly complete kits using composites, the image of the homebuilt community changed dramatically, says EAAs Tom Poberenzy.

Aircraft designers went from selling plans to producing fast-build kits, where most of the materials are supplied and 49 percent of the work is already done. Designs evolved from simple tube-and-fabric Cub look-alikes to sheet metal sport planes and composite rockets.

The menu for builders has certainly expanded, Poberenzy says. But the homebuilt movement was not designed to be an alternative to certification.

A caveat: Many completed aircraft are advertised as pro-built. The FAA allows individuals to build their own airplanes for their own education and recreation, but frowns on those who build for profit.

Check the aircrafts licensing carefully. If the builder didnt satisfy the FAAs desire for an individuals educational effort, the aircraft may be licensed in the exhibition category, which means it can only fly to, from and in air shows. More than one buyer has been burned by this loophole.

The Accident Picture
The sales success of the Vans RV series aircraft and the relatively high-profile of the Glasair, the Varieze, the Long-EZ and the Lancair have opened the world of homebuilding to people who had an interest in new airplanes but didnt have the time to build their own. When production of new aircraft virtually evaporated in the late 1980s and early 1990s, fast-build kits took on added significance for pilots who wanted something new.

Yet the growth has come with a price. In the last 10 years, nearly 1900 amateur-built aircraft have been involved in accidents. Many were merely runway deviations or minor metal bending, but 590 involved fatalities, according to National Transportation Safety Board records.

The accident rate has held remarkably steady. Or so it would seem, given the limitations of the NTSB and FAA accident tracking data. The true measure of safety-if there is such a thing-is accident rate, or the number of crashes per hours flown.

Unfortunately, although we have a good idea of the number of crashes each year, the number of hours flown is but a wild guess. And comparing accident rates of homebuilts to production airplanes is impossible because most homebuilts dont fit the operations parameters the FAA uses to estimate usage for the rest of the general aviation fleet.

They tend to be operated out of small uncontrolled airports, are generally not flown as much as production aircraft and many are of the low-and-slow variety.

Further complicating the comparison is that many builders spend their time and money on construction, so when it comes time to launch, their flying skills have eroded significantly. For example, the builder of a Lancair 360 crashed in 1995 on the first test flight. The airplane lost power on takeoff and stalled, killing the pilot. He had not flown in the previous 10 years except for five hours in a low-performance airplane just before the test flight.

Still, a look at the fleetwide total accident picture is useful. According to NTSB figures, each year, 1.3 percent of amateur-built aircraft can be expected to be involved in accidents of some kind, with a third of them involving fatalities. In the GA fleet as a whole, there are on the order of 2000 accidents a year, which works out to about 1 percent of the total fleet. On average, about 20 percent of these involve fatalities.

In an effort to measure safety, we created an index to measure the probability of an accident by looking at the number of accidents in the past five years divided by the number registered. We used five year total here, not yearly totals. While this is an imperfect method of assessing the likelihood of crashing, it does allow some comparisons to be made between aircraft of the same general flight characteristics.

It also illustrates an important point: The aircraft models we studied are all popular designs and perhaps prove the concept of market-driven safety. With a single exception, all were less likely to be involved in accidents than the amateur-built fleet as a whole.

Our survey revealed that the total amateur-built fleet showed an index of 5.4 accidents overall and 1.6 fatal accidents. By comparison, during the same period, the GA fleet as a whole had an index of 5.2 accidents and 1.0 fatal. Even with its considerable flaws, this comparison suggests that the overall accident incidence for homebuilts may be similar to that for GA as a whole, although the likelihood of an accident being fatal appears to be higher.

On the other hand, if your hunch is that homebuilts, on average, fly half as much as spam cans, then the accident rate is double that for certified airplanes. Since no one in the industry can prove or disprove that theory, its as good as any. Take your pick. Heres a look at some specific models:

Avid
The Avid designs are noteworthy for their low fatality rates. Of the two fatal accidents in the past five years, one was a student pilot taking off illegally with a passenger from a plateau airport and stalling into the valley. The majority of Avid accidents have been due to loss of power-many times unexplained -and minor damage during the subsequent forced landing. Several accidents have been blamed on carb ice in aircraft where no carb heat mechanism had been installed.

The Avids showed no incidences of VFR-into-IMC and only two cases of fuel exhaustion, suggesting that these airplanes are mostly used for fair-weather local pleasure flights. The fact that the Avid flyer is relatively slow probably figures in its low fatality rate.

Kitfox
Although the popular Kitfox has a similar accident rate, the character of the accidents is much different. Many accidents are blamed on loss of control, including ground loops, noseovers and inadvertent stalls. In fact, only 19 of the 51 accidents are attributed to something other than stalls or loss of control.

Five of the accidents were caused by flaws in construction and six were the result of power loss for unknown reasons. Interestingly, a high percentage of Kitfoxes are operated out of private strips or even unimproved fields. Many of the loss of control accidents occurred when the pilot was trying to dodge a tree, fence or something else you wouldnt find on a runway.

Challenger
Challenger designs are plagued by a high accident incidence and a high fatality rate. Essentially beefed-up ultralights, Challengers have been subject to both mechanical failures and operations by inexperienced pilots.

A quarter of the accidents involving Challenger aircraft in the past five years have come at the hands of student pilots, unlicensed pilots, and people making their first few flights. Add in three crashes right after the pilots added optional doors and another when floats were added, and its clear that this low-cost aircraft is seeing its share of rookie mistakes.

However, the Challenger has also suffered six equipment-induced crashes, primarily involving engine trouble. The high fatality rate is possibly explained by the fact that occupants really have no cabin to protect them during a collision.

Glasair
Glasairs, a model which all but wrote the book on composite aircraft performance, have proven to be fairly safe, despite their reputation for blistering speed and a handful of high-profile mishaps.

The types accident record is better than average for the fleet and even exceeds that of the low-and-slow Avid Flyer and Kitfox. More than a third of the accidents reported were due to fuel exhaustion or fuel starvation, but five of the eight crashes were the result of improper maintenance of the fuel system. Of the nine fatal accidents, three are listed as cause unknown and two were VFR-into-IMC. Three others were tied to improper maintenance. Only one accident was directly attributed to the type of construction-a crash in which static charge built up on the fiberglass airframe and shorted out the radios, causing disorientation in IMC.

Lancair
Lancairs have compiled an accident record slightly better than average, but the number of accidents involving fatalities is relatively high. Thats not surprising given the fact that Lancairs tend to be cross-country machines rather than fly-around-the-patch sport airplanes. 0Two of the fatal accidents were VFR-into-IMC, one occurred when the pilot had a heart attack and one when the pilot attempted to take off with the airplane 340 pounds overweight. Those can scarcely be blamed on the airplanes design or manufacture, although there have been complaints about inadequate tail surfaces in early kits.

But more troubling is that a third of the accidents were caused by mechanical failure, according to NSTB reports. In fact, an airplane registered to Neico Aviation, designer of the Lancair and manufacturer of the kits, lost its propeller on the way home from Oshkosh in 1996, killing a company employee.

Although this didnt appear to be due to a design flaw, it could be either just bad luck or somewhat symptomatic of the spotty quality control under which homebuilts are constructed.

RV
The popular RV series is in a class by itself. Its among the most numerous of all homebuilts with good market success and a generally good safety record. It has a low total accident rate and its fatality rate is substantially better than the fleet as a whole. Having said that, however, the RV series has had a bad recent couple of years, suffering a spate of 16 crashes in 1998 and 1999.

The accident record shows no trends that would indicate a design weakness and this might be difficult to correlate in any case, since there are many model variations of the RV series. Twenty-six incidents were the result of loss of control, including noseovers, stalls, hard landings and other pilot blunders, many of which may be due to the types early tailwheel design.

Sixteen involved loss of power, including suspected carb ice and seven were the result of fuel exhaustion. The record shows 20 fatal accidents in the past five years. Of them, two resulted from low-level aerobatics, one was an apparent suicide, one involved an improper modification and one came after wing failure in what the installed g-meter recorded as a 9.2 G maneuver. As with the Lancair, Vans suffered one fatal crash of a factory-built model, an RV-8 which appeared to break up. That accident is still under investigation by the NTSB.

Long EZ/Varieze
Burt Rutans Long EZ and Varieze designs stack up well against any of the homebuilts. They racked up the lowest total accident rate of the models we studied. Together, they recorded only four accidents attributed to stalls or loss of control, a testament to the designs ease of handling and the advertised stall-proof capability of its canard control surfaces.

Two of the accidents involved inexperienced pilots. One was conducting high-speed taxi tests the other was a departure stall when the pilot had one hour in type.

The remainder primarily involved engine failures, cables breaking, bolts coming loose and other reflections of poor maintenance or construction. Because of this models popularity, there were also a number of accidents after modifications. Some were flights testing the mod, others were improper mods that failed.

Safetys Not All
Of course, theres much more to buying a used homebuilt than analyzing accident statistics. But the record shows that inexperienced pilots and poor maintenance is to blame for most of the accidents that plague homebuilt aircraft and not always the designs themselves.

Thats not to say homebuilts in general are free of design faults that may give an inexperienced pilot less margin when he finds himself in a tight spot, however. Because the flight characteristics may be far different from anything you or your flight instructor have ever flown, initial test flights can be high hazard operations for some aircraft.

Teaching builders how to fly their new toys led the EAA to create its Flight Testing Techniques for Homebuilders program. Pilots considering a buying a completed homebuilt can also take advantage of the program. Rather than teach you how to fly a particular model, it demonstrates how test pilots approach flying an aircraft with unknown flight characteristics, slowly expanding the envelope from one flight to the next so that squirrelly handling doesnt come as a fatal surprise.

Assessing the mechanical purchase of a homebuilt is a bit less certain. A good mechanic can examine a potential purchase and be fairly certain whether its been well-built and well-maintained, but homebuilts may have hidden flaws a mechanic wont find on a pre-buy inspection.

The FAAs Airworthiness Alerts, a report issued monthly, routinely cites mechanical flaws found on homebuilt aircraft. Some are traced to builder technique, others are the result of poor design, unclear instructions or cheap parts.

The two biggest problems cited in the last two years have been poor workmanship and improper modifications. Faulty welds top the list of workmanship issues, followed by incorrect assembly.

But the reports also raise the troubling issue of parts that are improperly specd for the application-some of which were supplied by the kit maker. The landing gear actuation switch on a Lancair, the throttle cable on a Kitfox and the tailwheel leaf springs on an Avid have specifically been blamed for accidents on those aircraft in the past two years.

Those types of flaws may be detected by a mechanic and they may be found by someone who has built the model being considered. But they may also remain hidden until the airframe hours pile up.

More insidious are maintenance issues that may result from having the builder/pilot/seller authorized to perform annual inspections and other maintenance. While builders are handy with tools and know the airframe, not all are mechanics, nor are they inclined to spend unnecessary time, effort and money maintaining the airplane. Unlike a hired mechanic, who makes money from finding things wrong, builders have only the pride of ownership and desire for flawless operation for motivation. Some individuals may be perfectionists because theyre the ones strapping themselves in for takeoff. Others, however, may operate under the close enough philosophy. Which type of person built the airplane you may be considering is something for you and your mechanic to judge.

Conclusion
Picking a completed homebuilt airplane to buy will be a time-consuming process, but worthwhile for someone who doesnt like the looks, performance or utility of any of the certified aircraft out there.

In any given issue of Trade-A-Plane, there will be 70 to 80 listings of completed aircraft, ranging from $10,000 toys to $400,000 masterpieces. With the exception of the Challengers, any of the popular kit planes listed here appear to have adequate if not blemish-free safety records, in our view.

Theres no compelling evidence to suggest widespread or serious design or kit flaws that make any of these airframes deadly. But to put it bluntly, some may require a degree of pilot skill that you dont have.

Once you decide what types interest you, ask the EAA for help in finding a model near you owned by someone who might take you flying in it. But remember, an RV-6A built by one person may be a drastically different machine than the same airplane built by someone else. Unless you know what youre doing, stay away from kit planes in which the builder has departed drastically from the original plans.

Popular kits, each with hundreds of completed aircraft flying, are relatively safe bets in the risky game of experimental flight. Just dont expect them to fly, look or work like a certified cousin you may have flown. Check them out well. After all, this is one time the FAA isnt looking over your shoulder.


Also With This Article
Click here to view “A New Safety Goal.”
Click here to view “The Insurance Angle.”


-by Ken Ibold
Ken Ibold is editor of Aviation Safety magazine. Contact EAA at P.O. Box 3086, Oshkosh, WI 54903-3086, 920-426-4800