Preliminary results of the ERAU PA-28 accident

[Fearless Tower]

On another note, my Hats Off to the NTSB on this. Never have I seen a preliminary report, so extensive and with in a week of the accident. Good work...

What is interesting about this prelim is that while there have been alot of rumors being circulated around the internet that the wing had sever bolts missing and the rest were sheared off, the NTSB prelim mentions nothing about missing bolts. As detailed as that prelim is, I'd think if they had found missing bolts, that would be noted.

 

[Fearless Tower]

If I was ERAU I would have G-Meters in the airplanes that could not be reset except by the A&P at inspection. I realize that doesn't tell the whole story. I could certainly envision a bunch of college kids who are into flying treating a rental like a rental and trying a roll, loop or who knows what. Easy enough to mess up pull a crap ton of Gs and walk away that time but leaving the next guy to pay the bill.

I don't know how prevalent it is, but supposedly, a great deal of the ERAU fleet has some kind of flight data recorder system that maintenance downloads frequently and identifies whenever an aircraft has been overstressed.

 

[James331]

If I was ERAU I would have G-Meters in the airplanes that could not be reset except by the A&P at inspection. I realize that doesn't tell the whole story. I could certainly envision a bunch of college kids who are into flying treating a rental like a rental and trying a roll, loop or who knows what. Easy enough to mess up pull a crap ton of Gs and walk away that time but leaving the next guy to pay the bill.

Or one of those "you're fired" lights like some of the cheap cargo operators have that go on and can't be turned off if you exceed a limit

 

[dell30rb]

What is interesting about this prelim is that while there have been alot of rumors being circulated around the internet that the wing had sever bolts missing and the rest were sheared off, the NTSB prelim mentions nothing about missing bolts. As detailed as that prelim is, I'd think if they had found missing bolts, that would be noted.

If you look at the picture of the spar, it looks like it failed right where the bolts attach, in the middle of the bolt holes. At a minimum I think you would be missing those two bolts.

 

[drotto]

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.

Does this similar degradation influence the type of " abnormal abuse" that the plane may have been put through? Are landings or stressful flight maneuvers more likely to cause such equal and excessive wear?

Sent from my SM-G950U using Tapatalk

 

[wsuffa]

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.

Does this similar degradation influence the type of " abnormal abuse" that the plane may have been put through? Are landings or stressful flight maneuvers more likely to cause such equal and excessive wear?

Higher stresses on a Piper (low wing) than on a Cessna (high wing) because of the configuration and placement of the landing gear. Each landing on a low wing plane puts somewhat more stress (and leveraged stress) on the spar where it connects with the fuselage (and where this cracked).

 

[Zeldman]

"Good god man, it's a yoke not a shake weight" lol

 

[jallen0]

Does anyone thinks this effects the chinese order they just placed?

 

[PaulS]

The report reads evidence of fatigue cracks and cracks caused by overstress. This does not necessarily mean the fracture was caused by a pilot overstressing the wing. What I think they are saying here is that there is evidence of fatigue cracks, which would be a long term issue and evidence of overstress cracks, which probably occurred when the wing finally failed after being weakened by the fatigue cracks. The fatigue cracks could be caused by repeated hard landings overstressing the wing, but looking at that photo I think there may be inclusions in the aluminum which could act as stress risers and cause the fatigue cracks. The inspectors rightly have sent this to a lab for analysis which will hopefully provide more answers. To lay this at the feet of abuse of the airplane or ER maintenance is definitely premature and most likely incorrect conclusions.

I suspect cheap Chinese aluminum, but that is just an educated guess and worth what you paid for it.

 

[drotto]

The report reads evidence of fatigue cracks and cracks caused by overstress. This does not necessarily mean the fracture was caused by a pilot overstressing the wing. What I think they are saying here is that there is evidence of fatigue cracks, which would be a long term issue and evidence of overstress cracks, which probably occurred when the wing finally failed after being weakened by the fatigue cracks. The fatigue cracks could be caused by repeated hard landings overstressing the wing, but looking at that photo I think there may be inclusions in the aluminum which could act as stress risers and cause the fatigue cracks. The inspectors rightly have sent this to a lab for analysis which will hopefully provide more answers. To lay this at the feet of abuse of the airplane or ER maintenance is definitely premature and most likely incorrect conclusions.

I suspect cheap Chinese aluminum, but that is just an educated guess and worth what you paid for it.

But what is important here for the big picture is this a design flaw in the PA28? Was there specific prolong abnormal stresses to this plane, if so what and how? Was this just a case of a bad batch or metal? Or just plain bad luck? Sometimes, an individual mechanical item, or structural item will fail without any clear cause or indication.

I doubt there is a design flaw in the PA28 just based on the shear number produced, and very very low rate of this type of failure. I suspect the issue will be something specific to this plane or how this plane was used, but it will likely trigger an AD for all PA28's.

 

[Fearless Tower]

Does anyone thinks this effects the chinese order they just placed?

Nah. That would require Chinese capability to make a better wing. That ain't happening.

 

[Fearless Tower]

To lay this at the feet of abuse of the airplane or ER maintenance is definitely premature and most likely incorrect conclusions.

But that's the internet way!

 

[Velocity173]

Reminds me of the T-34 issue. Not a high rate of failure but wings did fold under G load and ADs were issued. Don’t think you can point to one over load or any particularly hard landing. It comes from prolonged abuse.

 

[James331]

Would the AD apply to the restart new pipers or all PA28?

 

[PaulS]

Would the AD apply to the restart new pipers or all PA28?

Man, if I were flying one of these airplanes, I would want to know that there is no evidence of any cracks pretty quickly, never mind waiting for the FAA. There is just not enough information.

 

[drotto]

Would the AD apply to the restart new pipers or all PA28?

Were there any design changes made between the old PA28 and the new PA28, if nothing significant then I suspect the AD may be all. I would not be surprised if they put in a time qualifier however, like all PA28's with more than say 5000 hours total time. My gut would say that the AD would require visual inspection and checking the bolts at the spar junction, and an installation of a access port if the area is not easily visualized. It would seem onerous, but there could be an extreme AD requiring bolt/wing removal at again around 5000 hr TT for visual inspection. I have heard that removing the wings is not that hard, but an AD that could cost an owner 1 k to 2 k would be a massive hit on the model.

Are all models a similar setup? Would this affect PA24's, PA32's, PA34's, etc. Or is the wing design different enough or "beefier" that it would not require the AD? Does this even just affect Arrow's? Is the wing design significantly different from the fixed gear version, or does the fix gear result in different or less stress on the wing?

 

[James331]

Were there any design changes made between the old PA28 and the new PA28, if nothing significant then I suspect the AD may be all. I would not be surprised if they put in a time qualifier however, like all PA28's with more than say 5000 hours total time. My gut would say that the AD would require visual inspection and checking the bolts at the spar junction, and an installation of a access port if the area is not easily visualized. It would seem onerous, but there could be an extreme AD requiring bolt/wing removal at again around 5000 hr TT for visual inspection. I have heard that removing the wings is not that hard, but an AD that could cost an owner 1 k to 2 k would be a massive hit on the model.

Are all models a similar setup? Would this affect PA24's, PA32's, PA34's, etc. Or is the wing design different enough or "beefier" that it would not require the AD? Does this even just affect Arrow's? Is the wing design significantly different from the fixed gear version, or does the fix gear result in different or less stress on the wing?

Id guess the same design, but I wonder if the quality of the materials is different with the new one, maybe some chinese materiel or something?

 

[X3 Skier]

Having been involved in the analysis of several structural failures in airframes and engines, the fracture surface examination and fracture mechanics analysis will eiliminate much of the mystery here. The crack initiation site, the density of the striations, the overload portions of the fracture face and the final critical crack length will lay out the total history of the part.

As a typical case, the crack initiates from some defect, such as a machining mark, inclusion in the metal, forging defect, or other problem. It then propagates in fatigue with occasional tearing from overloads until the remaining metal cannot sustain normal or abnormal load and fails.

Use in a rough training environment will accelerate the propagation of an initial crack either in fatigue or over stress. The real answer of “why” is hidden now but should be visible without much difficulty.

Cheers

 

[Bell206]

X3 is spot on. While fatigue and over-stress result in similar failure modes, they have different initial causes. The picture of the spar shows that difference. In general, the darkened part of the crack area is the fatigue area and the silver/shiney part is the over-stressed area.

 

[JGoodish]

Not sure where the predictions of ADs are coming from. One failure, without an identified root cause? Seems to me that any airframe could be compromised by improper operation, but there should be much less of a fleet risk for an otherwise stout airframe not certificated for aerobatics.

What am I missing?

 

[Shepherd]

Yikes.

 

[Capt. Geoffrey Thorpe]

Both pre-existing fatigue cracks AND overstress fractures

Once the fatigue cracks had progressed far enough (over the course of hundreds if not thousands of hours), then what is left will be overstressed by normal loads. The overstress failure happen because of the fatigue cracks.

A hand-full of high load events (effing up a roll for example) will not cause fatigue.

 

[Fearless Tower]

Not sure where the predictions of ADs are coming from. One failure, without an identified root cause? Seems to me that any airframe could be compromised by improper operation, but there should be much less of a fleet risk for an otherwise stout airframe not certificated for aerobatics.

What am I missing?

Call it cynicism, but many of us have seen how the FAA reacts to stuff like this with other catastrophic failures.

The reason for concern is that this appears to be a problem potentially related to defects in manufacturing combined with fatigue rather than improper operation. I’m not saying that IS what was going on, but a very real possibility based on what we know so far. The truth will come out when they start pulling other comparable Arrows apart to see if this is a bigger problem.

 

[Fearless Tower]

Does this similar degradation influence the type of " abnormal abuse" that the plane may have been put through? Are landings or stressful flight maneuvers more likely to cause such equal and excessive wear?

Arrows tend to take more abuse in this area than other PA28s largely because of two reasons:

1) when you pull the power out, they tend to sink more than other PA28s

2) they are largely flown by low time pilots training for Commercial or CFI and those pilots are more likely to make hard landings in the airplane because of #1.

 

[GRG55]

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.

You may be correct.

But then how many Cherokees are out there, and how many of their wings gave fallen off? Compared to, say, the number of Bonanza V-tails that have come apart in the air.

I can see an AD coming out of this, but perhaps not cataclysmic.

 

[weilke]

I always thought it would be interesting to have a lab, maybe someone like UL, etc., disassemble an aging plane (but still technically within its lifespan) and check out its overall condition, and even stress one to failure and see how it stacks up against textbook numbers

God no ! Cessna did that with some 400 series aircraft and came up with the SIDS.*


I don't believe the usage history of this arrow was representative for other Cherokees. 10 years of commerical candidates practicing their spiral dives and chandelles gave plenty of opportunity for inadvertent high G events.




* You are of course right. As these planes age beyond their expected lifespan, tearing some apart to look for failure points is the right thing to do. It's just that any kind of mandatory $20,000 inspection is going to wipe out most of the budget aircraft on the market. Unless more pa28 start to shed wings I am willing to chalk this on up to a random event.

 

[hindsight2020]

Once the fatigue cracks had progressed far enough (over the course of hundreds if not thousands of hours), then what is left will be overstressed by normal loads. The overstress failure happen because of the fatigue cracks.

A hand-full of high load events (effing up a roll for example) will not cause fatigue.

Indeed, and we're saying the same thing. It's noted that the final failure is an overstress failure due to the fact the crack has propagated to the degree it compromises the tensile strength of the component beyond that of normal delta-sigma experienced in normal operation. Ultimate failures are always "overstress" in nature, but that's academic at this point.

The point I was making regarding prior overload events, is that they do reset the delta-K because there is plastic deformation embedded in the cycle count of the crack due to said overloads. It introduces higher and different tensile preloads (plastic deformation) and at that point the crack will not exhibit the crack propagation model it was following before and up to the intermediate overload event. Multiply these overloads by dozens over years of ham-handed and lying pilots, and that fatigue life estimate is as good as garbage.

The good news is that these discontinuities are easy as day to see in the material as recovered, and we fully expect the materials lab to be able to recount these events based on the striation density and plastic deformation events present et al. Though I suppose it would be face saving for ERAU flight ops that they find craptastic chinese metal as the culprit for an early failure sample, I'm not convinced this doesn't have to do with a mx directorate completely out to lunch to just how abusive the ERAU flight line is to these spam cans.

You may be correct.

But then how many Cherokees are out there, and how many of their wings gave fallen off? Compared to, say, the number of Bonanza V-tails that have come apart in the air.

I can see an AD coming out of this, but perhaps not cataclysmic.

From your mouth to God's ears. Let's hope cooler heads prevail. I can categorically tell you if they slap me (Hershey Arrow mind you) with a repetitive wing pulling or dye penetrant AD, I'm done. The airplane is not virtuous enough for me to undergo that kind of recurring expense. Never mind it is actually damaging to the spar structure mind you, just like the Comanche torque tube horn inspection, completely counter productive behavior. I do hear the Beechcraft owners have to deal with the lower spar web carry through repetitive inspection, with the doubler kit installation (if needed) becoming quite pricy. Any Bo/Debbie owners care to chime in with the cost and invasiveness of the recurring? I consider 500 hours at the edge of becoming onerous to be honest.

At any rate, if it's anything resembling wing pulling I'll just sell it and move into my upgrade airplane sooner than I had planned. I'm not messing with that expensive kabuki for a 130kt sub-1000fpm climb 3-seater. At that point the added airframe costs of 182RGs, Comanches and C33As become a wash for a lot more capes gained. It'll definitively kill the cherokee market at the bottom, so again let's hope they show restraint in how wide they throw the diaper-wearing penalty net on us unsuspecting private non-training use owners.

 

[kyleb]

A hand-full of high load events (effing up a roll for example) will not cause fatigue.

But if you damage something (start a crack) by really effing up a roll, fatigue will finish the job for you much faster than it would have otherwise.

 

[Fearless Tower]

I don't believe the usage history of this arrow was representative for other Cherokees. 10 years of commerical candidates practicing their spiral dives and chandelles gave plenty of opportunity for inadvertent high G events.

Maybe, maybe not.

A large portion of the Arrow fleet is used just like this one: primarily rentals/flight schools. They are largely high time/high cycle airplanes.

I don’t think the issue is going to affect all PA28s, but I would not be surprised if they find this is a larger problem with the PA28Rs

 

[paflyer]

Not sure where the predictions of ADs are coming from. One failure, without an identified root cause? Seems to me that any airframe could be compromised by improper operation, but there should be much less of a fleet risk for an otherwise stout airframe not certificated for aerobatics.

What am I missing?

The knowledge that the FAA has been known to act stupidly, to paraphrase a remark made by the 2009 Nobel Peace Prize winner, especially when it costs them nothing to "comply."

 

[Tantalum]

been known to act stupidly, to paraphrase a remark made by the 2009 Nobel Peace Prize winner

Maybe the FAA just needs to "recalibrate" their words

 

[zaitcev]

I suspect cheap Chinese aluminum, but that is just an educated guess and worth what you paid for it.

These things happen. SpaceX strut that held a COPV is one. Just last week Kel-Tec announced a recall for SUB-2000 barrels made in 2010. Both companies blame a supplier for substituting a bad quality metal (well it was steel, not aluminum).

 

[PaulS]

These things happen. SpaceX strut that held a COPV is one. Just last week Kel-Tec announced a recall for SUB-2000 barrels made in 2010. Both companies blame a supplier for substituting a bad quality metal (well it was steel, not aluminum).

You would be amazed at the contempt some have for specifications, my experience isn't in aviation parts, but human nature is human nature. No one knows if that is the issue here, but it could be.

 

[mtuomi]

My question would be, is that a point where you'd expect the structure to be overstressed? If you pull 7G's, you'll rip the wings off at around 40% span length (as been proven by multiple accident reports, Piper number is 163% of design limit load when the wing breaks off).
I'd imagine there are points with higher stress elsewhere in the wing structure, especially when abusing the plane?

 

[hindsight2020]

The knowledge that the FAA has been known to act stupidly, to paraphrase a remark made by the 2009 Nobel Peace Prize winner, especially when it costs them nothing to "comply."

Indeed, textbook case of moral hazard.

 

[hindsight2020]

I'd imagine there are points with higher stress elsewhere in the wing structure, especially when abusing the plane?

Not with as as many stress concentration points as the attachment fittings to the carry-through structure, especially when you combine it with impact loads caused by landing. Any time you introduce holes and fittings, here comes the K-factors to mess up your day. Bolt holes are low hanging fruit here, but that's true of any structure.

We'll see how far the overreaction goes. ERAU can afford the inspection AMOC, non-revenue owners maybe not so much. Good ol gubmint, picking winners and losers since 1776.....

 

[mtuomi]

Not with as as many stress concentration points as the attachment fittings to the carry-through structure, especially when you combine it with impact loads caused by landing. Any time you introduce holes and fittings, here comes the K-factors to mess up your day. Bolt holes are low hanging fruit here, but that's true of any structure.

We'll see how far the overreaction goes. ERAU can afford the inspection AMOC, non-revenue owners maybe not so much. Good ol gubmint, picking winners and losers since 1776.....

Makes sense. Still weird that this hasn't been an issue before - I honestly hope they will find a smoking gun in the past of this airframe. Else, I smell a VERY intensive and expensive Emergency AD in 3...2...1....

 

[WannFly]

I am going to stop by Walmart and get some diapers. Not for the fear of wing separating, but for the obnoxious AD that’s coming out anyway now

 

[kyleb]

Makes sense. Still weird that this hasn't been an issue before -

I expect they will find or infer something bad happened to that airplane during usage, maintenance, or construction. Arrows have been around long enough that if there was a design flaw that led to premature wing removal, it would have come to light before now.

 

[denverpilot]

I hate to say this but the BEST thing that can happen for PA28 owners is that the FAA and NTSB find a culture of “normalization of deviance” is happening at the school and the school has questionable instructing or maintenance practices.

That takes a lot of pressure off of the AD process to get momentum and feet under it.

Sadly though, that’s not what I really want to hear out of such a large and well known school because it’ll come back around and instead of biting aircraft owners in the butt, it’ll bite CFIs in the butt with some unforeseen new regulatory silliness on the training side of things.

No matter how you slice it, I don’t see a fatal accident with a wing departing the aircraft with a Designated Examiner on board as not coming back to bite SOMEONE in the butt.

It’s more a matter of who and how many.

Stuff is going to roll downhill on this one, just a question of how big will the rolling rock be and whom it’s aimed at before it’s released to crash down the hill...