Minnesota Crash

[bradp]

1 hour ago, eman1200 said:

I’m just curious what’s “brilliant” about it? He states all the facts that we already know, then says it’s a LOC incident. Uh, ya, we know that. The only thing “brilliant” about that video is he says mooneys are great planes.

And minor point he gets the wing construction confused a bit - which is a little odd since he previously had an E.  Wonder if I take for granted we Mooney pilots know how our planes are put together.  I think there is value to every pilot parking his or her butt on a creeper during annual at least once.  

[Shadrach]

2 hours ago, eman1200 said:

I’m just curious what’s “brilliant” about it? He states all the facts that we already know, then says it’s a LOC incident. Uh, ya, we know that. The only thing “brilliant” about that video is he says mooneys are great planes.

Yes, well Juan’s channel isn’t for Mooney owners by Mooney owners. If this were a Bonanza or a Navion crash there’d be a good chance you wouldn’t have all the details that you have. In such a case his prelim might be educational. He’s playing to a much broader GA audience. He makes mistakes sometimes but is typically humble and takes ownership of them. 
I subscribe to his channel because he gives a short, fact driven, synopsis of most GA accidents. I simply don’t have the time to look into every significant  GA accident.

i’m terribly sorry to have subjected you to free media that didn’t meet your standards...

Edited August 10, 2021 by Shadrach

[Shadrach]

1 hour ago, bradp said:

And minor point he gets the wing construction confused a bit - which is a little odd since he previously had an E.  Wonder if I take for granted we Mooney pilots know how our planes are put together.  I think there is value to every pilot parking his or her butt on a creeper during annual at least once.  

If you read his pinned comment, his description of where the wing broke was based on the sheet metal joint, not the spar. The wing skin overlap at that station is double riveted for the first 2/3 of the wing. I suspect that’s what prevented the wing from separating from the airframe. It could’ve been worded better for sure. Dude is an airline pilot and an A&P. When he does these while on the road they are sometimes a little rough around the edges. 
 

Edited August 10, 2021 by Shadrach

[Davidv]

3 hours ago, eman1200 said:

I’m just curious what’s “brilliant” about it? He states all the facts that we already know, then says it’s a LOC incident. Uh, ya, we know that. The only thing “brilliant” about that video is he says mooneys are great planes.

After having watched the video I can only imagine his "brilliant" comment must have been sarcastic.  Juan seems like a nice enough guy, but he essentially summarizes what anyone could have read on Kathryn's report.  The truth is that we have no idea what caused this crash (which may have been wing spar failure in flight), so any analysis (even brilliant ones) are just speculation after regurgitating the facts that the general public knows already.

[MooneyMitch]

It would be interesting to know if NTSB will interface with Mooney International.   Obviously, all is speculation at this point.

Probably goes without saying, but thought I’d say it anyway. 
 

Perhaps, I should have said it would be interesting to know the extent of the NTSB interface with Mooney, given the uniqueness of the speculative spar failure issue. 

[Davidv]

Just now, MooneyMitch said:

It would be interesting to know if NTSB will interface with Mooney International.   Obviously, all is speculation at this point.

Of course, one of the first things they do in this kind of incident. 

[eman1200]

Yes, well Juan’s channel isn’t for Mooney owners by Mooney owners. If this were a Bonanza or a Navion crash there’d be a good chance you wouldn’t have all the details that you have. In such a case his prelim might be educational. He’s playing to a much broader GA audience. He makes mistakes sometimes but is typically humble and takes ownership of them. 
I subscribe to his channel because he gives a short, fact driven, synopsis of most GA accidents. I simply don’t have the time to look into every significant  GA accident.
i’m terribly sorry to have subjected you to free media that didn’t meet your standards...

We’ll clearly my standards are low given that I read and replied to your comment. All I’m saying is he simply read all the ALREADY WIDELY KNOWN FACTS off an index card. Absolutely nothing “brilliant” about that. Has nothing to do with mooney knowledgeable people, they were all facts people already knew.

And yes, I watch his channel and appreciate what he does.

[aviatoreb]

With eyewitness accounts, and the video, seems likely that yes indeed that wing did fold up in flight.  I had been dubious, and also thinking it was a video artifact possibly.  And it also doubtful that it seemed as if the wing folded at the very last second within the last few feet of the ground.  Maybe not.

Ok, so for the wing to fold up like that it must have torn a huge huge hole in the skins?  Would all the fuel just fall right out almost immediately?  Would the folding cause a spark?  I am almost surprised it didn't catch fire from that event, while still in the air.  Maybe this is why the fire was not more intense since by the time it hit the ground maybe a good fraction of the fuel had already departed the wings.

[DXB]

15 hours ago, lithium366 said:

I am almost positive this scenario is what happened. He described the most plausible scenario 2 days after crash when he could read enough and did not say anything new. He could tell this to investigators at the scene and this would be in the news but looks like he just walked away just to post the comment 2 days later on the youtube? Just wanted to say that so far the only real evidence for us is the crappy video and chances of this comment on the youtube of being a red herring are very high. This is the beauty of the Internet: I can create a new account and comment that an airplane was flapping it’s wings and everyone will talk about this

I'm not sure the youtube "eyewitness" post is the most plausible scenario exactly (pilot exits the clouds at 800 AGL descending at 20,000fpm, somehow manages to exert enough elevator authority to snap the spar and fold the wings up while starting to level off).  But I certainly agree the Youtube comments section is a rather poor tool for accident investigation.  I also stand corrected on an inflight Mooney spar failure having never previously been documented.  https://aviation-safety.net/wikibase/36780  Perhaps the actual events here might have looked more like this.

Edited August 10, 2021 by DXB

[EricJ]

I don't think it's possible for a pilot to induce the g-loads necessary to make an airworthy Mooney spar fail in a short period of time.   A failing elevator/stabilizer, however, depending on what it does when it fails, could potentially twist and cause an abrupt pitch-up before separating from the aircraft.

I mentioned previously (I think) that that would be similar to the trim tab failure that caused the Galloping Ghost crash at Reno.   The tab failed in such a way that it caused an abrupt pitch up, which g-loc'ed the pilot and resulted in an uncontrolled crash.

I could see a failing stabilizer causing an abrupt pitch up on any airplane, not excluding Mooneys.   What caused the stab to fail is unknown as well.   Flutter past Vne?  Existing unrepaired damage?   Bad repair?  Since the pieces have all been recovered I think there's a pretty good chance we'll know eventually.

Edited August 10, 2021 by EricJ

[Scott Dennstaedt, PhD]

I've been reading on some other forums that there are some pilots that think this accident is related to convective turbulence or a microburst. Here's a more protracted discussion in my blog that documents the weather at the time of the accident that demonstrates otherwise.  

https://www.avwxtraining.com/post/weather-analysis-for-recent-mooney-accident-near-minneapolis

Enjoy.

[jlunseth]

42 minutes ago, DXB said:

I'm not sure the youtube "eyewitness" post is the most plausible scenario exactly (pilot exits the clouds at 800 AGL descending at 20,000fpm, somehow manages to exert enough elevator authority to snap the spar and fold the wings up while starting to level off).  But I certainly agree the Youtube comments section is a rather poor tool for accident investigation.  I also stand corrected on an inflight Mooney spar failure having never previously been documented.  https://aviation-safety.net/wikibase/36780  Perhaps the actual events here might have looked more like this.

Dev, I did some searching and found a couple of others. They are not easy to find because they have been so few in number and the Mooney has been around long before the internet, therefore records are not necessarily in a searchable database, and I may not have found all.

One was a one-line report of a spar failure in an M18C.

"A72 Mooney M-18C 31 Mar 64 1 D Roanoke Rapids, N.C. USA. Wing failure (spar). Ref: NTSB file 2-1036"

https://apps.dtic.mil/sti/pdfs/ADA137254.pdf

The other was a J. https://aviation-safety.net/wikibase/44066

There may have been another J in California, I could not find enough of a record to verify.

The Rocket and J were wing failures, but both were different from the current accident in that the failure of the horizontal stabilizer reportedly caused the wing to fail in the negative direction, that is, upward if the aircraft had been right side up at the time. In the Rocket incident the factory concluded that the failure of the stabilizer caused the aircraft to tumble up and over, hence the negative wing failure. As we know, structural integrity in the upward direction is about half that in the downward direction, or at least that is how the FAA rates normal category aircraft. 

The issue of concern to us as Mooney pilots is that there have been at least four incidents now, of failure of the horizontal stabilizer and elevator, which likely is a flutter issue, not a load issue. This video has been around forever and is worth watching for anyone who needs a lesson in the effects of flutter. Turn the sound off, the music is aggravating.

The lesson is that we cannot exceed Vne. Not that any of us have been going around doing that just for fun, but rather, if put in a potential accident scenario we must obviously protect Vne at whatever cost.

Also, I was the one who said 800 based on having looked at the weather off and on during the day, earlier than the accident. Scott Dennstaedt gave us a RAOB that put the ceiling at 1200 and there were a few others who provided more current weather. Not that an extra 400 feet would make that much difference to your point about the force required to fail the wing by pulling up.

From the sparse accident history it does appear that failure of the wing can be caused by the failure of the stabilizer. It appears to me it would be fair to say, though, that this is the first time that has occurred in the positive direction rather than the negative (excluding the Mite accident in 72 - different wing). 

[Shadrach]

1 hour ago, DXB said:

I'm not sure the youtube "eyewitness" post is the most plausible scenario exactly (pilot exits the clouds at 800 AGL descending at 20,000fpm, somehow manages to exert enough elevator authority to snap the spar and fold the wings up while starting to level off).  But I certainly agree the Youtube comments section is a rather poor tool for accident investigation.  I also stand corrected on an inflight Mooney spar failure having never previously been documented.  https://aviation-safety.net/wikibase/36780  Perhaps the actual events here might have looked more like this.

I don’t think so. That aircraft completely lost its horizontal stabilizer at high-speed. This cause the fuselage to rotate  around the center of lift tail up and nose low loading the wing in the opposite direction (negative g). The wings pulled away from the aircraft’s fuselage, not toward it.  Given the Minnesota aircraft wings folded in almost the exact opposite direction and in the direction the wing with stress to be strongest, I still feel like an immediate pull was the likely culprit. I allow for the possibility that a horizontal stab failure while  inverted could cause a positive G overload. However, given how close the separated parts were to the scene of the crash, I think the odds are that the airplane came apart below the deck.

Edited August 10, 2021 by Shadrach

[DXB]

5 minutes ago, Shadrach said:

I don’t think so. That aircraft completely lost its horizontal stabilizer at high-speed. This cause the fuselage to rotate  around the center of lift tail up and nose low loading the wing in the opposite direction (negative g). The wings pulled away from the aircraft’s fuselage, not toward it.  Given the Minnesota aircraft wings folded in almost the exact opposite direction and in the direction the wing with stress to be strongest, I still feel like an immediate pole was the likely culprit. I allow for the possibility that a horizontal stab failure while  inverted could cause a positive G overload. However, given how close the separated parts were to the scene of the crash, I think the odds are that the airplane came apart below the deck.

Agree the stab most likely came off below the deck.  I’m less certain the final wing position in the last video frame was the initial one whenever the wing failed.  If the eye witness is real and reliable, then it happened as you say.  However the sequence of events in the K accident, which broke the spar in the weaker direction, seems perhaps more plausible.

[Shadrach]

Just now, DXB said:

Agree the stab most likely came off below the deck.  I’m less certain the final wing position in the last video frame was the initial one whenever the wing failed.  If the eye witness is real and reliable, then it happened as you say.  However the sequence of events in the K accident, which broke the spar in the weaker direction, seems perhaps more plausible.

Perhaps. Indeed it seems unbelievable to me as well. I keep reminding myself that 30,000 ft./min. is 340mph straight down. There are very few wings GA or otherwise, that would withstand a rapid, max effort pull from such a speed. Modern military fighters, yes... just about everything else, no...

The descent rate appeared to decrease to 18,000fpm seconds before impact. I theorize the the aircraft descent rate decreased after (and as a result) of the spar failure.

[Hank]

Descent rate accelerated to 30,000 fpm then decreased to "only" 18,000 fpm = 205 mph DOWN, likely due to the Big Pull. If memory serves, the consecutive ADSB values are 18,000; 30,000; 18,000; impact. So that's when he pulled and broke the plane.

Acceleration would be huge to reduce speed by 12,000 fpm between ADSB blips every few seconds, to say nothing of the momentum of the airplane. Remember, this is not the TAS of the plane, just the downward portion of the velocity vector. 

[David Lloyd]

On 8/9/2021 at 12:42 PM, Shadrach said:

I've plotted the descent and climb rates leading up to the crash.  He clearly was struggling to maintain pitch control if not roll as well.  Regardless of when the wings folded, this was clearly a loss of control followed by a severe overstress of the airframe.

The first plot is a larger picture of where the problem likely started or exacerbated. The second is a more granular plot of the final moments.  It must have been absolutely terrifying for all involved. thankfully the end was immediate and painless...RIP

Pitch control problem seems to become evident at 18:38:34

More granular look at final descent.

I think the first picture is the whole story.  At 18:38:34 something happens and the plane goes from a stable approach at a reasonable speed  to a short series of increasing pitch, turns, altitude and speed excursions ending 72 seconds later when the airplane descends from the clouds and pulls the yoke hard enough to break the airplane. The next to the last ADSB hit

showed 147 knots ground speed, but factoring in the 30,000 FPM descent would result in an airspeed considerably above VNE.

So, what happened at 18:38:34?  I have had several attitude indicator failures over the years as well as about 10 vacuum pump failures.  Two of the AI failures acted the same as with a vacuum failure:  over two or three minutes, the AI would oscillate slowly up, down, right, left with increasing amounts and rates as the gyro wound down.  Except the last one.  A couple seconds from a normal indication to a 45 degree bank and 20 degree pitch, oscillating back and forth, increasing the amount and rate, before suddenly parking itself in a 90 degree, nose high indication.  Start to finish about 10 seconds.  Thank God I wasn't in cloud.  Could his AI have failed?  Down, up, left, right, up, down, left, up, down!  Did the autopilot initially follow?  A sudden failure in a bumpy cloud might be overwhelming.

All of us would like to think something beyond our control happened here.  The elevator broke off causing the plane to go out of control.  Or the wing broke in half due to some hidden damage.  Or a thunderstorm.  Microburst.  No, we will wait two years for the NTSB report and find a mundane answer.  

[A64Pilot]

I don’t believe the rates of descent are possible. You just can’t go from level flight at what assumption is less than gear speed as his gear were down to over 300 kts in 2,000 of altitude?

Remember the gear are down so their drag at those airspeeds are quite substantial. I doubt the aircraft could achieve those speeds from 10,000 ft.

I accept you guys analysis on reported speeds, just believe the data is flawed

‘I also don’t believe the horizontal failed due to flutter, but from a tremendous overload, remember to pull the nose up at high G, the Horizontal has to develop a large downward force, I believe the wing and tail failed at about the same time, neither from flutter, failure modes are different so the investigation will be able to tell.

I believe that because in other accidents where the tail has failed, the wing actually fails downward as after the tail failed there is a huge neg G load on the wings, but in this case it seems apparent that the wings failed from positive G.

Edited August 10, 2021 by A64Pilot

[EricJ]

1 hour ago, jlunseth said:

From the sparse accident history it does appear that failure of the wing can be caused by the failure of the stabilizer. It appears to me it would be fair to say, though, that this is the first time that has occurred in the positive direction rather than the negative (excluding the Mite accident in 72 - different wing). 

The failure mode of the stabilizer may be different between Mites, short/mid bodies, and long bodies, or even just different from incident to incident.

I think it's becoming evident that something led to the stab failure, which caused the severe pitch-up which resulted in the spar failure.    The stab failure was likely flutter or other effects from exceeding Vne.

This is definitely educational for us, or me at least.   I'm learning a few things, like the spar is unlikely to fail from a pilot-induced action, the tail (stab/elevator) is the likely failure in an event exceeding Vne, and the stab failure can induce pitch changes severe enough to fail the spar.   

This also makes the recent Canadian F-model incident more remarkable, as it appeared that that airplane exceeded Vne several times during its incident.   Whoever gets that airplane needs to look carefully at the tail.    So that's another learning point:   inspection after a Vne event will want to look carefully at the tail/stab/elevator.

 

[A64Pilot]

This was a friend of mine and my mentor as a civilian test pilot, long story but Rockwell recreated the failure in the wind tunnel and from the onset to complete disintegration  was less than 1 sec.

https://aviation-safety.net/wikibase/2731

The aircraft had been dove to VD several times in the past, what was different in this test flight was that Ralph had the trim tab driven to full down so that he didn’t have to hold so much control pressure, Rockwell in order to make manufacturing easier only had one trim tab on the tail, this along with the control doublet which was part of the test set up the flutter.

The other person onboard was the Flutter Engineer who had been told if your so sure it’s safe, then you get on board for the flight, which he did.

‘I don’t know about then, but today you don’t take observers along on a high risk test flight like a dive to VD.

On edit this is Ralph’s Book, he literally wrote the book on GA flights testing, he got his Doctorate  in Aeronautical engineering from Aachen University in Germany and I believe was taught by one of the Horton brothers, but not sure about that.

https://www.thriftbooks.com/a/ralph-d-kimberlin/1235913/

Edited August 10, 2021 by A64Pilot

[Shadrach]

7 minutes ago, A64Pilot said:

I don’t believe the rates of descent are possible. You just can’t go from level flight at what assumption is less than gear speed as his gear were down to over 300 kts in 2,000 of altitude?

Remember the gear are down so their drag at those airspeeds are quite substantial. I doubt the aircraft could achieve those speeds from 10,000 ft.

‘I also don’t believe the horizontal failed due to flutter, but from a tremendous overload, remember to pull the nose up at high G, the Horizontal has to develop a large downward force, I believe the wing and tail failed at about the same time, neither from flutter, failure modes are different so the investigation will be able to tell.

I believe that because in other accidents where the tail has failed, the wing actually fails downward as after the tail failed there is a huge neg G load on the wings, but in this case it seems apparent that the wings failed from positive G.

I would not assume that the gear were down. In fact I think he was behind the airplane before the point where he would’ve normally dropped the gear. An object will accelerate just from gravity at about 22mph per sec.  I too have been trying to work out the acceleration numbers, but I wouldn’t know where to begin. It’s above my mathematical pay grade. Seems impossible to accelerate such speeds so rapidly. However the only thing I have to go on are the a ADSb returns. My guess is that a significant amount of his vertical speed is getting lost in the averaging of the data points. What looks like incredible acceleration to us is as much a change in vector as it is acceleration.

[Shadrach]

9 minutes ago, EricJ said:

The failure mode of the stabilizer may be different between Mites, short/mid bodies, and long bodies, or even just different from incident to incident.

I think it's becoming evident that something led to the stab failure, which caused the severe pitch-up which resulted in the spar failure.    The stab failure was likely flutter or other effects from exceeding Vne.

This is definitely educational for us, or me at least.   I'm learning a few things, like the spar is unlikely to fail from a pilot-induced action, the tail (stab/elevator) is the likely failure in an event exceeding Vne, and the stab failure can induce pitch changes severe enough to fail the spar.   

This also makes the recent Canadian F-model incident more remarkable, as it appeared that that airplane exceeded Vne several times during its incident.   Whoever gets that airplane needs to look carefully at the tail.    So that's another learning point:   inspection after a Vne event will want to look carefully at the tail/stab/elevator.

 

The other thing that stands out is the one piece tail. The size of the the hardware used to affix the empanage to the fuselage is a running joke with many mechanics. However looking at this accident as well as the Canadian Mooney with the much better outcome, it does not appear to be a weak spot at all.

[A64Pilot]

I thought the gear were down in the pictures? I’m looking at them on an Ipad so I don’t see things as well as if I had a big monitor.

‘Even gear up, I’, not so sure a Mooney can get over 300 kts in less than 2,000 ft, that’s only a few seconds to double your airspeed. But yes if he could get to over 300 kts, it’s likely to be shedding parts due to flutter.

Edited August 10, 2021 by A64Pilot

[A64Pilot]

1 minute ago, Shadrach said:

The other thing that stands out is the one piece tail. The size of the the hardware used to affix the empanage to the fuselage is a running joke with many mechanics. However looking at this accident as well as the Canadian Mooney with the much better outcome, it does not appear to be a weak spot at all.

Even a .25 bolt in shear between three pieces has its shear load over two points, and the force to shear a .25” bolt between two tight fitting lugs is huge.

‘If it had even been a close thing, it woud have been child’s play to upsize the bolts, but they weren’t because they didn’t need to be.

I assume they are AN bolts not even NAS? NAS are stronger of course

[jlunseth]

I don't think a rapid pull up is necessary. I think of the phenomenon this way. In normal flight the wings generate upward lift and the horizontal stabilizer generates counterbalancing lift that keeps the nose up or down to cause level flight. If the aircraft suddenly enters a steep dive, it is actually doing the beginnings of an outside loop. During the initial phase of that loop the lift from the wings is positive, that is, upward along the vertical axis of the aircraft if the aircraft were in normal flight. The angle of attack begins to change and at some point, who knows exactly where, the lift generated by the wing becomes negative, that is, in a downward direction were the aircraft in a normal horizontal position. I am certainly no aerodynamics engineer, but in my mind if the aircraft already has some movement in a horizontal direction and it starts to go vertical, the angle of attack may change to the top of the wing (negative angle of attack) before the aircraft reaches vertical. When, in this transition, the horizontal stabilizer fails, the counterbalance to the wing's lift is suddenly removed, the wing either does an immediate upward movement if the wing lift is still positive, or an immediate downward movement if the lift has gone negative. When it does so, the angle of attack of the - now high velocity - apparent wind also changes abruptly, it increases. That movement abruptly increases the lift being generated by the wing, and if that lift exceeds the load capacity of the wing, the wing fails. It does so either in an upward or a downward direction depending where in that incipient outside loop the aircraft is, that is, whether the wing is generating positive or negative G's at the moment of stab failure. That explains why, in some accidents, the departure of the stabilizer has caused a negative failure of the wings, and in this most recent one, a positive failure.

However, neither failure mode necessarily requires a rapid pull-up by the pilot, it simply requires the departure of the counterbalancing stabilizer from the airframe. Could a pull-up start the sequence? Sure. Is it necessary? No. All that is needed is an abrupt change in the angle of attack caused by the loss of the counterbalancing stabilizer.