Home

Wing Fatigue-- A Layman's Explanation

In reading the discussions on the T-34 Association Internet bulletin board about wing separations, mainly in response to the T-34 mishap occurring in Texas on November 19, 2003, it appears many people do not understand the concept of metal fatigue and how it relates to aircraft wings.
 

By Dave Marshall

 

I was ignorant once too, both about wing fatigue and what I could do to an aircraft before "destroying" it due to pulling G's. It took a tragic mishap and loss of life to educate myself and the aircraft community I was flying in at the time about wing fatigue.

 

Wings, being made of metal, will flex went loaded up with G's. They even flex in 1 G level flight. Next time you are on a commercial jet watch the wings as the aircraft approaches liftoff and you can see how much they flex up near the wingtips. How much a wing flexes depends on how much G is applied at the time. Sooner or later, depending on how much and how often the wing is flexed, the wing will fail. Want a demonstration? Bend a paper clip back and forth. If you bend it 90 degrees each time it will fail quickly. Bend it 10 degrees and it will still fail but you have to bend it many more times. What else did you notice? I am willing to bet the paper clip failed not at the extreme but at some intermediate point where it wasn't flexed all the way. Was there any visual indication the paper clip was getting ready to fail? Probably not, all of a sudden the last couple of flexes seemed easier, then there it was in two pieces. Get the point? Metal, when flexed, will eventually fail if flexed repeatedly. Every time you apply G in your T-34 you are flexing the wing and spar and bringing it that much closer to failure. The T-34 wing and spar are not indestructible. They can and will fail. What determines when it will fail is how the aircraft is operated prior to the failure.

 

Many years ago in another life I was a military pilot. My aircraft community lost an aircraft when the wing separated and the aircraft crashed. The aircraft had used up all the calculated fatigue life available and fatigue life expenditure (FLE) had exceeded well over 100 percent. Since I had information of a privileged nature concerning this mishap I am not at liberty to discuss the aircraft type, location, date, or any other mishap details but I can summarize some of the lessons learned.

 

-- The aircraft was being operated within the published flight envelope. It was not exceeding airspeed or G limitations prior to or during the wing failure. We knew it was operating within the G limitations because it carried counting accelerometers that would register a "hit" whenever 4, 5, 6, or even 7 G's were reached. We weren't getting high G hits outside the published.

 

-- Fatigue life was monitored using the accelerometers. Readings were pulled monthly and fed to the engineers who calculated remaining fatigue life. When the fatigue life expended (FLE) reached certain levels G restrictions were imposed on the aircraft. When the 100% threshold was met the aircraft was grounded.

 

-- FLE on the mishap aircraft was improperly calculated for many reasons. All data was still available so FLE was re-calculated and it was discovered the aircraft had FLE well over 100%. The wing had been "bent" so many times that it failed just like a paper clip fails when bent.

 

-- The aircraft was actually in 1 G, wings level flight when the wing failed. You don't have to be under high G loading to cause the wing to fail at some given point in time. The damage was already done. We knew the aircraft was in level flight because another squadron aircraft in the same flight witnessed the breakup.

 

-- Sadly, up to this mishap, we the operators had a collective mentality that the aircraft could never be broken. Following this mishap we began to operate much more intelligently regarding when and how we pulled G's.

 

FLE vs. G's pulled is not a straight-line curve, it is more exponential. At 6 G a much higher percentage of fatigue life will be expended than at 4 G. FLE is also gross weight dependent. Most, if not all, military aircraft have an absolute G limit which is most likely at a significantly lower gross weight than takeoff weight. As gross weight increases the amount of G one can pull is reduced. So if you are operating your aircraft in a high G environment vice straight and level you are consuming a far higher percentage of available wing life each flight and pushing that wing closer to failure.

 

Every time you hit a certain G level the same amount of fatigue life is expended regardless of how long you are pulling the G. A five second 4 G pull uses as much fatigue life as a split second pull. It is possible to instantaneously pull a very high G load within the normal airspeed flight envelope without the body noticing it. Even if pulling less G and you flew through wake turbulence or other turbulence you could instantaneously hit a higher G load. Your body wouldn't feel it but that higher load would consume more FLE nonetheless. What's important in wing fatigue and calculating FLE is the fact you hit the level, not how long you were at that level. Bend your paper clip and leave it bent-- it won't fail. Bend it repeatedly and it will fail.

 

Flight hours flown is irrelevant regarding FLE. What does matter is how many G's were pulled and how many times the G's were applied. This would account for why a single owner/operator aircraft probably being operated in a more benign environment would last longer than an aircraft being leased/flown in an air combat or out of control flight environment that repeatedly subjects it to high stresses. The squadron with the mishap described above took delivery of a brand new aircraft from the factory. Operating within the normal flight envelope the wing was used up in less than three years of normal operations and had less than 2,000 hours on it when restrictions were imposed. The aircraft was operated in a training environment and had a greater number of higher G hits than comparable aircraft in the fleet even though it had fewer flight hours.

 

What does this mean to the average T-34 operator? Hopefully a great deal. Do you know how your aircraft was operated in the past? Do you know how many G's were pulled and how many applications of those G's were made? Do you have any idea how much fatigue life has been expended over the years? Do you know the condition of your spars? Is there any corrosion in the spars? How do you intend to operate your aircraft?

 

You should be asking yourself the above questions in regarding what fix to apply to your aircraft. Personally I would opt for a fix relative to the type of flying I would be doing. The military flying club I am associated with operates two T-34B aircraft. The flying club program prohibits aerobatics, air combat, and out of control flight training so we opted for the least expensive AMOC we could find. But if I were operating in a higher G arena like flight demonstration and air combat I would personally opt for a beefier fix, such as new spars or a doubler plate. But that is a personal choice.

 

Since the original wing separation mishap the FAA has imposed many restrictions on the aircraft. Many operators probably felt those restrictions were too harsh and too restrictive. The latest mishap proves one very big point-- the FAA's positions have been validated. Hopefully this will get everybody's attention out there. There has been much discussion about whether the Texas Air Aces aircraft was being operated within the FAA's restrictions concerning the wing. The investigation will tell the story, let it run its course and we can all read about. But one thing should be painfully obvious-- the FAA has been proven right in a very bad way. Let's not do it again.

 

--


About the author. Dave Marshall is a former Navy carrier pilot and graduate of the Navy's aviation safety officer school. He is also a licensed pilot and former manager of a military flying club operating, among other aircraft types, two T-34B aircraft.

 

Updated December 13, 2004. Click here to send us feedback on our website.