Preliminary results of the ERAU PA-28 accident

Don't the Lance and Seneca have a similar wing attach structure to the Arrow?

Sure, but you misunderstand my motivation. IOW, I have zero quarrels with my wings or how they're attached, it's the FAA buffoonery and ensuing peanut gallery panic I'm trying to skirt here. I used to believe having a really proliferate airplane was a good thing for my ownership economy. Now I'm potentially finding out this thinking was flawed. I realize that thanks to the monkeys around me, we can never have nice things.

To the point, Lances are virtually absent from the training environment, and PA-23s are old enough and low enough in training fleet volume these days they aren't likely to spring up a wing spar AD at this point on account of a derelict puppy mill dream-seller. I suppose the Seneca I is the only one I'd be walking into a trap on that list, since it's fairly proliferate in the training environment. so I'll concede you that one and scratch it off the list.
 
Last edited:
There is still the view there's a higher level of uncertainty working with composites, so stuff is overdesigned.

My business owns both steel and composite structures (pressure vessels). The Federal regulators require the carbon fiber composites to be re-tested and re-certified every 5 years, half the interval for steel. And we have to scrap the composite structures at 15 years no matter what. The steel units have been used for multiple decades if they are maintained properly.

Composites have little ductility. When they fail they snap or shatter. Because of that our composite pressure vessels have to be completely without flaws on the exterior surface. Any damage, even so much as a scratch, is cause for immediately scrapping it as there is no reliable repair method with carbon fiber, and the flaw is a potential stress concentrator.

Exactly. If you look at the spar on these planes, you can picture a discussion going sort of like this: “ok so we built a sizable spar out of carbon fiber, what do you think?” “You know, composites fail in weird ways and are hard to really inspect...” “alright, what about if we made it 4x beefier?” “Seriously? Ok yeah, that should do it!”
 
and therein lies the core of the issue, was it a freak incident where someone seriously botching a landing or someone pulling a stunt, or is this kind of thing "common" at ERAU. I mean, if you want to tool around in the air, find a freaking Aerobat, Citabria, Great Lakes, something other than an ARROW!

It appears several posters have assumed that systemic and ongoing mishandling during training and/or maintenance oversight must have been causal factors in this accident. I think it more likely the accident airplane experienced a single event that initiated the fatigue damage, and the overstress failure occurred before a maintenance check could discover it.

One would expect certain things to have already occurred if the ERAU Arrow fleet was abused with "impunity" (from another post) and it was "common".

ERAU airplanes would have had to remain in service without any specific spar connection inspections after repeated significantly abnormal landings or turbulence encounters for a spar failure to occur.

It is almost certain ERAU instructors are required to report flight encounters or landings which could have damaged the airplane. The CFI should be aware enough to discern the difference between a hard landing and one which buckles the firewall or damages the MLG. Flight in moderate to severe turbulence would not go unreported.

This also would not have been the first ERAU Piper Arrow to shed a wing in flight. Similarly, post sale inspections by subsequent owners would have disclosed some instances of similar spar damage (which would probably have resulted in FAA involvement) or some number of these former ERAU Arrows would have had a spar failure.

However, Piper didn't build them with the caveat they had to be treated gently. They had to assume the Arrow would become a staple of the trainer fleet, and that the average aircraft would experience a certain percentage of their lives operated by students in a wide variety of weather conditions (like summertime Florida convective turbulence) and varying degrees of landing skill.

I think it unlikely that the incident is indicative of willful indifference by ERAU.

What a sad incident. There was nothing the student and examiner could have done to save themselves. May they rest peacefully.
 
Last edited:
Another interesting note. They stated that the right wing spar showed a similar degree of deterioration. To me this means that whatever abnormal loads were being applied they had been occurring more or less equally.

The report didn't say that at all. It said the right wing center section lower spar cap had cracks at the same location (the two bolt holes on the end of the center section lower spar cap stub), but they were just 0.047 in depth.

The left wing lower center section spar cap had cracked through 80% of the material.

The right wing also exhibited fatigue cracks in the lower spar cap at the same hole location extending up to 0.047-inch deep. .

Note that only the left center section lower spar cap had fatigue damage. When it failed, the remaining components of the spar connection joint (upper cap and doublers) failed from overstress.
 
Last edited:
Stress corrosion cracking? It shouldn’t have failed so early in its life but it did. Tensile stresses. Humid environment. Cracks at fuselage-wing joint on both sides.

Wonder what they’ve been using to wash the planes?

From the NTSB preliminary report:

None of the surfaces exhibited visible evidence of corrosion or other preexisting damage.
 
FTFY

I don’t have a dog in this fight, just a bad feeling that is isn’t going to end well for Piper owners.
Not long after I bought the cherokee, I had the infamous SB 1006 done, which involved removing the fuel tanks. Made it very easy to check the bolts and such. About 10 yrs later I had a fire on the ground (thanks to the shop that screwed up the brake system and no longer exists). This time the wings were removed, a few skins replaced on the wings, and I assume/hope the shop (Beegles) examined the spar, bolts and everything around that area.

Can a dye penetrant test be done without taking the wings off?
 
I'm going to say no. The upper and lower caps of the wing main spar have doubler plates at the joint with the center section, and several rows of bolts connecting the doubler to the wing and center section spar caps.

The bolt head, nuts, and washers conceal the perimeter area of the bolt holes, precluding their inspection.
 
There are cover plates over the bolts on the bottom of the fuselage and access to the center section once the rear seats and plywood cover are removed. As noted previously the cracks would have been visible to diligent inspection.
oh krap, removing the back seat without damaging the entire side panels....there goes the new interior....
 
Last edited:
I feel better about the oddball Grumman spar.
I will admit at least 2 times I have flown after this incident, I have grabbed a wing during preflight and shook it a little.

View attachment 62063
I always wiggle the wings on the cherokee. Even more fun when it's Young Eagle day, and I explain I want to make sure the wings are still glued on.
 
The black is fine oxide the forms at high localized temperature from fretting. The fretting is from the initial crack propagation being subjected to repeated cyclical stresses over some period of time.

From the NTSB preliminary report:(no corrosion)

Well somebody is wrong...
 
Sure, but you misunderstand my motivation. IOW, I have zero quarrels with my wings or how they're attached, it's the FAA buffoonery and ensuing peanut gallery panic I'm trying to skirt here.
My point was that there's nothing to stop the buffoonery and panic from extending from the PA-28R to its progeny, including PA-32R and PA-34.
 
I wiggle the wings on the Mooney during preflight, but I doubt it would have helped those guys.
 
Well somebody is wrong...
I believe the NTSB report stated visible corrosion. I think your point on stress-corrosion may be closer. Since there is no fire damage to the spar the lab guys will be able count and follow the failure marks right to the crack origin. And determine the number of cycles it took to failure. You can see those faint "beach" marks on the spar and they tend to move away from the bolt holes. It can be a pit .001" deep or a minute stress riser from installing those bolts that could start the failure process. Not to mention the failure is through both bolt holes (steel bolt/aluminum spar). Will be interesting how the spar load cycles vs flight hours will work out.
 
I wiggle the wings on the Mooney during preflight, but I doubt it would have helped those guys
I give it a firm wiggle. Actually learned that from my brother in law who flew sailplanes.. since they routinely took the wings off it was a good way for them to check if anything rattled or seemed loose, etc. I imagine though that on the accident plane it probably won't have done much. NOW, had they peeked under the plane, I wonder if maybe there would have been some odd or concerning looking evidence under the plane. I can also imagine the look of otter shock on the pilot if you wiggled a wing and it buckled and came off. Doesn't the Mooney have a solid one piece spar? I wish all planes did

Didn't @denverpilot say a buddy of his scrapped a flight once because the wings weren't even when they did the preflight? I think he said it turned out the thing had a hard landing and the plane never flew again

engine may fall off with the wing
I remember the first time I flew in a 402 and we hit some moderate bumps, it was very odd to see the propeller and spinner moving independently of the wing and nacelle (cowl?) in the bumps. Also weird to see the same thing on a 757, you have this (comparatively) massive engine out there just dancing away on the wings. Metal (when not fatigued!) is incredible.


*I must say though.. in recent years as I spend more and more time flying Cirrus I am appreciating more the strength of composites. I might even dare say I feel safer on a composite plane now!:stirpot:
 
Lol, you do realize the extruded I beam construction is stronger in bending moment than a hollow cylinder right? Torsional rigidity goes to the tube, but this quality is usually attained in conventional wings by the addition of a much smaller secondary spar, where the control surfaces and drag devices are usually attached to. Which is basically a sunk cost in all spam cans since they all have flaps and ailerons regardless.

Grumman put a tube in lieu of a properly constructed I beam as a cost cutting measure. The geometry of your spar is in fact not stronger than the geometry of failed wing in question. But let's not let facts get in the way of the placebo we all need to rationalize our need to break ground, myself included. The more you know....

Yeah but there’s more to spar strength than just geometry. What material / quality is the spar made of? How thick is the material? How large is the spar itself?

I know you’re a PA28 fan because you own one but the fact is, there are 22 in-flight breakups listed on NTSB JUST for the PA28 alone. Not a single in-flight breakup is listed for not only the AA5 but the AA1 series as well. Even the ones that crashed from over G showed signs of wrinkled wing skin but no spar failure. Sorry, but I’ll take the strength of Grumman over a Piper any day of the week.
 
Late to the party, but it doesn't surprise me too much. Those PA28Rs were used for power off 180's a ton, and probably got dropped in a ton as well. The commercial course only had maybe 10 hours in the Arrow, but it was all commercial maneuvers and whatnot. I never flew them (as while I went to ERAU, I saw the value in 61 training). Wouldn't surprise me to see this be ERAU trying to be cheap, as it seems like everything they do is about lining their pockets with cash.
 
I have a soft spot for Pipers but I also prefer composites (like the Cirrus)... so I won't take sides. But, I've never had a composite component let me down in life.. (aviation or otherwise) but I've had many steel or metal parts fail or disappoint me. For anyone doubting the strength of composites check this out. That small little foil, which already has a sizable bend built into it, withstood 4,600 kg of force.. or a little over 10,000 lbs!! I'll happily fly on a Grumman wing any day.. same with any composite plane

*note: those guys walking around so close during that test are buffoons.. if that piece had let go and released 10,000 lbs of force in an instant there would have been shrapnel flying every which way
 
...*I must say though.. in recent years as I spend more and more time flying Cirrus I am appreciating more the strength of composites. I might even dare say I feel safer on a composite plane now!:stirpot:

That would be irrational.

Composites and metal behave differently, including how they fail. One isn't necessarily better or superior to the other.

All airplane design is a compromise between competing objectives. Strength, weight, corrosion resistance, fatigue life, ability to inspect/repair and so forth. There's lots of glassed in metal fittings in a composite airframe. ;)
 
I believe the NTSB report stated visible corrosion. I think your point on stress-corrosion may be closer. Since there is no fire damage to the spar the lab guys will be able count and follow the failure marks right to the crack origin. And determine the number of cycles it took to failure. You can see those faint "beach" marks on the spar and they tend to move away from the bolt holes. It can be a pit .001" deep or a minute stress riser from installing those bolts that could start the failure process. Not to mention the failure is through both bolt holes (steel bolt/aluminum spar). Will be interesting how the spar load cycles vs flight hours will work out.

My read of the NTSB preliminary report suggests it has ruled out stress corrosion cracking, and put it down to fatigue. On closer inspection as the investigation progresses they may change that view, but what has been published so far seems pretty clear there's no evidence of a chemical environment (such as chlorides) to promote SCC.

That there's a stress concentration at the bolt holes should not surprise. The question is the selection of materials and configuration of the attachment fundamentally inadequate, did this airplane exceed the airframe cycle limits, or is there something unique about the way this wing/spar was fabricated and bolted together that is different from all the others (e.g. an important step in the manufacturing that was overlooked on this one assembly).
 
I'm sure they didn't recognize fretting as corrosion....but, it is.
Agree that many folks haven’t been exposed to some of the more technical concepts that are part of metallurgy. In particular some of the failures accelerated by a corrosive environment appear to be normal fatigue failure until they get they fancy cameras out. Cracks early in life and without deformation are clues. Not conclusive at all but they are clues.
 
I have a soft spot for Pipers but I also prefer composites (like the Cirrus)... so I won't take sides. But, I've never had a composite component let me down in life.. (aviation or otherwise) but I've had many steel or metal parts fail or disappoint me. For anyone doubting the strength of composites check this out. That small little foil, which already has a sizable bend built into it, withstood 4,600 kg of force.. or a little over 10,000 lbs!! I'll happily fly on a Grumman wing any day.. same with any composite plane

*note: those guys walking around so close during that test are buffoons.. if that piece had let go and released 10,000 lbs of force in an instant there would have been shrapnel flying every which way
While it's certainly true composites can be very strong, that test doesn't take into account the strain rate at which force is applied. Most materials can withstand a truly massive load when it's applied slowly, but when all of that force comes in a sharp whack, they're much more brittle and prone to yielding.

Composites are not great at absorbing shock loads, vs metals which generally are.
 
My read of the NTSB preliminary report suggests it has ruled out stress corrosion cracking, and put it down to fatigue.
I think we're getting a few terms crossed. Corrosion fatigue and stress corrosion are not the same thing. Corr fatigue requires the wet environment or as you suggested with interaction with chlorides. Stress corr is more similar to cyclic fatigue and usually can only be seen at the molecular level. The NTSB report said they sent those items out for further study. I read it as the spar did not corrode apart. The difference between stress corr crack and cyclic fatigue crack is stress crack starts at a micro-pit which usually starts from dissimilar metals in bore holes.
 
Last edited:
2007 Arrow "cross country training" accident in Texas where the wing came off. I would never buy an aircraft which was a "trainer". This is not normal usage when there is a hot shot instructor in the mix.
Also, the graphics depict a "T" Tail, it was a straight tail.
 
As long as you don't have a soft spot in your composite, you're good!

Believe it or not, in the early days of composite parts in the USAF, the standard field NDI to check for delaminations was tapping with a nickel and listening for funky sounds.

Cheers
 
the standard field NDI to check for delaminations was tapping with a nickel and listening for funky sounds.
And still practiced today on helicopter honeycomb structures down to using a coin if you can find the correct tapping hammer.
 
I think we're getting a few terms crossed. Corrosion fatigue and stress corrosion are not the same thing. Corr fatigue requires the wet environment or as you suggested with interaction with chlorides. Stress corr is more similar to cyclic fatigue and usually can only be seen at the molecular level. The NTSB report said they sent those items out for further study. I read it as the spar did not corrode apart. The difference between stress corr crack and cyclic fatigue crack is stress crack starts at a micro-pit which usually starts from dissimilar metals in bore holes.

I know they are not same thing. My point was there is, at least at this point in the investigation, no evidence of stress corrosion cracking, or that the environment needed for that ever existed in this instance.

However, the text from the preliminary NTSB report is specific to visual evidence of corrosion, so I'll accept SCC as a contributing factor can't yet be ruled out either.

"...Preliminary examination of the left wing main spar revealed that more than 80% of the lower spar cap and portions of the forward and aft spar web doublers exhibited fracture features consistent with metal fatigue...The fatigue features originated at or near the outboard forward wing spar attachment bolt hole..."

"...None of the surfaces exhibited visible evidence of corrosion or other preexisting damage..."


 
Believe it or not, in the early days of composite parts in the USAF, the standard field NDI to check for delaminations was tapping with a nickel and listening for funky sounds.

Cheers

I don't think that method as a test has changed much.
 
I give it a firm wiggle. Actually learned that from my brother in law who flew sailplanes.. since they routinely took the wings off it was a good way for them to check if anything rattled or seemed loose, etc. I imagine though that on the accident plane it probably won't have done much. NOW, had they peeked under the plane, I wonder if maybe there would have been some odd or concerning looking evidence under the plane. I can also imagine the look of otter shock on the pilot if you wiggled a wing and it buckled and came off. Doesn't the Mooney have a solid one piece spar? I wish all planes did

Didn't @denverpilot say a buddy of his scrapped a flight once because the wings weren't even when they did the preflight? I think he said it turned out the thing had a hard landing and the plane never flew again


I remember the first time I flew in a 402 and we hit some moderate bumps, it was very odd to see the propeller and spinner moving independently of the wing and nacelle (cowl?) in the bumps. Also weird to see the same thing on a 757, you have this (comparatively) massive engine out there just dancing away on the wings. Metal (when not fatigued!) is incredible.


*I must say though.. in recent years as I spend more and more time flying Cirrus I am appreciating more the strength of composites. I might even dare say I feel safer on a composite plane now!:stirpot:

To add some levity to an otherwise horrible incident...
In your description i’m picturing the first pilot wiggling the wingtip, only to have the plane go “ker-thump” onto the ramp, and the other pilot looking at the first and saying “what’d you do?” like Chris Farley at the gas station in Tommy Boy...
 
While it's certainly true composites can be very strong, that test doesn't take into account the strain rate at which force is applied. Most materials can withstand a truly massive load when it's applied slowly, but when all of that force comes in a sharp whack, they're much more brittle and prone to yielding.

Composites are not great at absorbing shock loads, vs metals which generally are.
This.

While it's true that the strength-to-weight ratio of aerospace composites is superior to aluminum and most steel alloys, the Achilles Heel of composites is that they have low (really low) elongation, generally around 1 or 2 percent. Metals typically have elongation in the 10 to 20 percent range. Elongation is a measure of brittleness; low elongation is brittle, high elongation is ductile. Materials with low elongation when stressed beyond their yield strength, particularly when the stress is very high and locally concentrated like an impact, fail spectacularly (shards and splinters everywhere), while materials with high elongation will typically deform by bending and/or denting rather than breaking.

The debate over materials choice for different applications isn't simple.
 
This.

While it's true that the strength-to-weight ratio of aerospace composites is superior to aluminum and most steel alloys, the Achilles Heel of composites is that they have low (really low) elongation, generally around 1 or 2 percent. Metals typically have elongation in the 10 to 20 percent range. Elongation is a measure of brittleness; low elongation is brittle, high elongation is ductile. Materials with low elongation when stressed beyond their yield strength, particularly when the stress is very high and locally concentrated like an impact, fail spectacularly (shards and splinters everywhere), while materials with high elongation will typically deform by bending and/or denting rather than breaking.

Yes.

Exhibit A, Robert Kubica at the 2007 Canadian GP. Data show a peak momentary impact of 75g, but he lived to tell the tale.

accident-kubica-2007-canada.jpg
 
Yes.

Exhibit A, Robert Kubica at the 2007 Canadian GP. Data show a peak momentary impact of 75g, but he lived to tell the tale.

View attachment 62256

That’s an amazing photo, could have easily been a photo of his death.

Energy dissipation is saving his life there. All the parts getting launched is keeping his body from being destroyed and the coroner writing “blunt force trauma” as his cause of death.
 
That’s an amazing photo, could have easily been a photo of his death.

Energy dissipation is saving his life there. All the parts getting launched is keeping his body from being destroyed and the coroner writing “blunt force trauma” as his cause of death.

Indeed...he was a lucky man to survive that.
 
Back
Top