May 27, 2018
During the initial climb, when the airplane was between 500–800 feet AGL, the engine backfired several times before losing power. The airplane entered a 180-degree turn back toward the airport; however, during the second half of the turn, the bank angle increased, and the airspeed slowed. The airplane entered an aerodynamic stall and spin and subsequently impacted terrain.
Post-accident examination of the engine revealed that the left magneto was loose on its mounting pad and free to rotate by hand. This would have resulted in erratic engine-to-magneto timing for that magneto and would likely account for the engine backfiring and the partial loss of engine power.
Looks like the pilot tried to execute the Impossible Turn resulting in a fatal stall spin into the ground, when the photos reveal a perfectly flat pasture in the landing area.
FIRST FLY THE PLANE.
We know (or can make an educated guess) how many accidents occur as a result of an attempted “Impossible Turn”, but we don’t know how often they’re successfully pulled off. I actually witnessed one…working in my hangar (near the departure end of the runway) when the engine of a C152 on takeoff started running rough. I could see it all through its turn, and it managed to set back down on the runway. No injuries, no damage that I could see, other than it had run off into the mud.
Of course, in this case, it wasn’t a total engine failure. But this one…and similar cases…don’t end up in NTSB reports, so we don’t have any idea how often they’re successful.
My airplane has the glide ratio of a manhole cover (open-cockpit, wire-braced). I once tried a simulated impossible turn (at altitude, of course). With a slight delay to simulate surprise, I lost 500 feet before completing the turn. Pretty much convinced me to NOT try it for real.
One factor to consider is that even if you complete the 180-degree turn, you aren’t lined up on the runway. You need to fly either a question-mark or light-bulb-shaped path if you intend to land back on the runway you took off from. Also, for much of the maneuver, you don’t actually have eyes on the runway…it’s behind you, and you don’t see it again for some very critical seconds. Even if you have a low-wing and bubble canopy, no one is used to judging their approach path looking back over their shoulders.
Glider pilots are trained to complete a turn back and land from a 200′ simulated rope break.
I had an engine failure towing a glider flown by my wife. I rocked the wings left, right, and pulled the release. Looking over my shoulder, my wife was already half way through the turn back. After setting up to land 45 degrees left of heading, I noticed that my glove had snagged the mixture lever (Cessna 188 AgWagon), advanced the mixture and completed a normal landing.
After landing, I found the rope at the departure end of the runway. We had released it simultaneously.
Dan, you’ve got the key word right there: “Trained.” Glider pilots are trained to do this, most power pilots aren’t. My student days are well in the dim past, but I know they taught me to, like Barry said, land straight ahead with only a bit variance to either side.
In the 70s, the BBC had an excellent TV series about British Royal Flight Corps pilots learning how to fight an air war in 1914. The second episode was titled, “Never Turn Back.” And, of course, in the episode, someone tries. The series is available in full on YouTube. Ironically, the young actor who plays the main character in the series later played Squadron Leader Rex in “Piece of Cake”.
I’m a just a thousand-hour pilot, with most of that time in a low-powered wood homebuilt, so I’m not suited to pronounce yea or nay regarding the impossible turn. Sure, it’s possible. But it takes practice, training, a compatible airplane, and, ultimately, luck. It’s not *always* possible; the problem is, one has only a second or two to decide if this particular instance is one of them.
While I don’t have the piloting chops to participate in this debate, I do have a bit of experience looking at the statistics. I once did an in-depth study of EAB engine-failure accidents in a ten year period. There were 476 reported accidents involving engine failure in the 2011-2020 time period. The pilots maintained control of the aircraft in most of them, but in 74 (about 15%) they lost control of the aircraft in the process.
What surprised me was the survival rates. I recorded what the pilots that maintained control of their aircraft eventually hit. Roughly 40% found a nice pasture or field, about the same number encountered trees. About 18% landed back on the airport grounds (though not on the runway), 12% on road/ditches, about 5% lost the lottery and force landed among buildings. (note that some cases saw more than one type of terrain involved, such as fences in those open fields).
Let’s look at crashing among building, which has the highest rate of fatalities involved. Yet 75% of those accidents did NOT produce a fatality. Again, these are the cases where the pilot retained control of the plane through impact.
Overall survival rate for pilots who retained control during an engine failure? About 89%. Them’s pretty good odds.
Overall survival rate in cases where the pilot stalled, either trying to reach a better landing zone or doing the impossible turn? Less than forty percent.
Two issues we can learn from this (again):
1: you have two ignition sources. One of them can make the engine misfire or stop the engine. Turn the bad one off and the engine will work just fine. This is best done using toggle switches. It is a bit more difficult with a key switch.
2: How many engines have quit because the mixture was accidentally moved to engine cut off? The mixture range should not have a cut off!
It should only go from AFR ~11 at full rich and AFR ~16 where the engine is still running but with reduced power.
The only reason there is a cut off with the traditional mixture is that the “P” lead can fail and in this case the mixture would be the only way to stop the engine.
Lesson: Don’t use anything that has “P” leads.
The original Fly Baby had toggle switches, and you’re right, I do prefer them. I *have* inadvertently turned a key switch all the way to the off position in flight on a Cessna. However, the reflex reaction was to turn it BACK immediately, which resulted in a nice little BANG and the engine (and the pilot’s heart) running again. So I think most pilots’ reactions would be the same; to turn the key back.
I don’t think the mixture cut-off issue is a major one. I went through the ~5,200 accidents in my EAB database and found only two cases where an accident occurred due to the mixture being pulled to cutoff.
My Fly Baby has a Stromberg without the idle cutoff…it is funny how some pilots are shocked and appalled to find me flipping the mags off to kill the engine. But it always starts up again without a problem, even if the shutdown is just long enough to fill the fuel tank.
I did have a mag stay live after maintenance, which meant I had to turn off the fuel to kill the engine. Took about 30 seconds in my airplane.
While I certainly respect the work you’ve gone on electronic ignition, I do have to point out that the single leading cause of failures in auto engine conversions has been a loss of power to the engine controller or ignition system. For instance, out of 40 engine failures of Subaru-powered aircraft, ten were due to the loss of power to the electronic ignition. *None of them were due to a fault in the ignition system itself*, but the builders failed to provide the power required Builders have to be careful in the design and construction of the support systems.
In contrast, Lycoming-powered homebuilts in the same time period suffered 12 cases where the magneto was at fault…but the installed base is significantly higher. There are about 30 Lycoming-powered homebuilts for every Subaru-powered one.
OK, I forgot that people still turn their engines off by going to mixture cut off, very bad idea, makes starting much harder! Have not done that in 40 some years.
“I don’t think the mixture cut-off issue is a major one.”
That might be true but my policy is to eliminate very single failure point I can because with my luck…..
Magnetos: In the 1992 book “The Magneto Ignition System,” the author, John Schwaner, has this to say about magneto failures:
“Since 1985 the National Transportation Safety Board has cited magnetos as a cause or factor in 92 accidents involving 22 fatalities and 21 serious injuries. This is surprising in that only one magneto is required to operate the engine.”
“Magnetos are reliable and provide a redundant safety factor, but they have the potential of causing engine failure if they are not inspected and serviced. When the magneto bearings fail on one obsolete magneto model, the rotor seizes causing the gear teeth inside the engine to break. The broken gear uncouples both magnetos causing instant engine shutdown. Worn impulse-coupling rivets can cause engine failure when the impulse coupling engages in flight. The adage “…if it’s not broke don’t fix it.” translates to “…operate it until it fails and hope it doesn’t take the engine (and us) when it does.”
“Low voltage and hard starting can be caused by shorted secondary turns within the magneto coil.”
“Any time the magneto is “fired” without the secondary circuit attached to spark plugs or ground, the full secondary voltage stresses the winding insulation. Eventually the insulation breaks down and the windings short.”
“Shorting of a secondary winding may create an open circuit as the winding burns.”
In writing of impulse coupling failure, the author states” The flyweight moves outward until the heal(sic) of the flyweight strikes the stop pin. At first, the heal(sic) just grazes the stop pin causing a telltale mark on the flyweight. Eventually, the heal(sic) strikes the stop pin with enough force that it pivots the flyweight outward where it engages the stop pin. When this happens during engine cruise, there is enough force that the flyweight disintegrates into the accessory gear train. Pieces of the impulse coupling then lodge into the engine gearing, breaking off gear teeth and decoupling both magnetos from the engine. The engine quits.”
Other failure mechanisms have to do with arcing of the distributor block, oil or moisture in the magneto, especially on super-charged engines, heat, failures of the capacitor, weak armature magnets, incorrect timing of the “E” gap, and incorrect impulse coupling lag-angle when faster or slower starters are installed which can cause kick-back .
And I don’t think magnetos got any better.
Auto engines and their ignitions are not designed or suitable for safe flight, that was also established long time ago.
KS
Part of the results of my analysis of homebuilt aircraft accidents has been a number of articles and online postings relative to the safety records of auto-engine conversions in homebuilt aircraft.
Yes, the record isn’t good. I presented a summary of the results as part of an EAA webinar I did ~30 months ago, “The Real Culprit.” The PDF of my charts is available for download at:
http://www.wanttaja.com/eaa23.pdf
Sure, the safety record of auto engines doesn’t match that of dedicated aircraft engines.
The problem is…it’s the only option for many. Ideally, we’d pour our money into making our aircraft as safe as possible. In reality, though, many homebuilder are faced with the decision to either use a less-safe engine or not fly at all.
Take me, for example. I’ve got a Continental C85 in the front of my bird, an engine that went out of production over fifty-five years ago.
Ideally, I’d replace it with a Rotax 912…I think that’d be a KILLER engine, on a Fly Baby.
The problem is, I’d be installing a $35,000 engine on a homebuilt worth, maybe, a third of that. The economics just aren’t there.
The cheap surplus engines of WWII and the immediate post-war boom have been gone for 30 years or more.
The VW got popular as an aircraft engine not because it is incredibly safe, but because it could be strapped to the front of an airplane like a Volksplane for only a few thousand dollars.
It’s somewhat similar, for me, as far as the accessories of the engine are concerned. My Eisemann magnetos have been out of production longer than the engine they’re mounted on. Yet…replacing them with new mags would cost me $2,000-$3,000. I *already* had to install a transponder and an ADS-B Out on my $10,000 airplane. Not really enthusiastic about replacing the Eisemanns with something from the latter half of the 20th century.
I’m not quite as poor as I pretend to be, but I’m every bit as cheap as I sound. The biggest problem in homebuilding now is the lack of cheap engines, preferably with those with some reliability. I’ve been hoping more Rotax 912s end up on the used market, but haven’t seen too many. So bring on the auto engines!
As far as magneto reliability itself, I have a database of over 5,200 homebuilt accident from 1998 through 2024. I read the narrative (not just the Probable Cause) for each, and assign what *I* think is the main cause of the accident. I occasionally differ from the NTSB Probable Cause.
For example, of those 5,200 accidents, 1,883 involved homebuilts carrying Lycoming engines. In 121 cases, the accident was due to a mechanical failure of the engine. Twenty-four of cases involved the ignition systems, split evenly between conventional magnetos and electronic ignition systems.
So that’s 12 accidents out of 1,883 that were caused by magneto problems. That’s pretty good odds. Sure, it’d be GREAT to reduce them further, with a more-modern ignition system.
In my case, though, A) I’m cheap, and B) It’s a single-seat airplane (only my own hide being risked), and C) I don’t carry hull insurance, only liability. If my airplane gets destroyed (ideally with me stepping unharmed from the wreckage), no longer having to pay hangar rent will compensate me for the value of the airplane in about two and a half years.
Ron, look at it this way:
If you install a $35,000 Rotax in your $10,000 Flybaby you can have $50,000 worth of fun with it and be $5000 ahead!
KS
Great idea, Klaus! Now, can you call my wife and talk her into it? She has some silly notions about a kitchen remodel…. 🙂
As per your excellent presentation, The Impossible Turn back to Departure actually requires about 360° of turning, plus a distance factor to accomplish. That’s why it’s recommended that one land within 45° of straight ahead, unless substantial altitude has been gained at time of engine failure.
Again the Prime Directive: FIRST FLY THE PLANE
Maybe. A 210 degree turn will get you pointed at the runway, that’s all you need to survive.
That, and sufficient airspeed, and a plane capable of making it around, and the wind being favorable, and…..
All true. “Lose not thy airspeed…” The point I was making is that it’s not necessary to be lined up with the runway to survive as often there are cleared areas both sides. It might not be pretty but if you can use some braking before going into the weeds you will probably be able to walk away or at least crawl out of what’s left of the plane.
First rule in performing the “Impossible Turn” is to unload the wing
Absolutely correct.
Emergencies can turn out to be survivable only if we Practice Practice Practice and Practice them on every airplane we are rated to fly not just type.
Every airplane flies a bid differently and in order to accept that we have to acquire experience first not just a logbook sign off.
The impossible turn is determined by the pilot before take-off in the months and years of realistic training as well as during the planning phase before flight and finally during Pre Take- Off briefing.
How many of us do that routinely ?
I had the privilege to practice a lot just because I am a CFI and I love to teach Emergencies to pilots. Some pilots had fears even with the word emergency but I changed their mindset by breaking everything down to small easy to understand parts vital to survival.
Avoid turning back when in doubt ( 99%), fly the airplane first per check list and call the memory items loudly even when solo.
Panic is the enemy, practice eliminates panic when well prepared and mentally ready before airborne.
Make emergencies become a second nature, you will enjoy every flight to the fullest.
” We are as safe as the next flight, not the last one. ”
” Safety is no accident”
I used to own and RV-6A. We practiced this return at altitude a number of time before doing the actual return to the field. The number was 1000 feet AGL. If we did not meet that altitude and the engine quit, we were landing straight ahead. If those altitudes listed are accurate, he was never going to make it.
I recommend everyone reading this to go fly at altitude and figure out your minimum return altitude and memorize it!