Another Flapperon Failure

27 posts in this topic

Posted (edited)

A Facebook post detailed a flapperon failure on a Mark IV. It concerned failure of one lollypop link in the flapperon controls, which seems to have killed all flapperon control. The poster had very poor english, but posted the below pictures. The plane was in Brazil, the failure was quite recent. The pilot was injured but survived. I posted the pics below, and a sketch of where it lies in the controls. Those rod ends look like Aurora MM-3 and MW-3 that take an AN3 bolt. Aircraft Spruce sells them and the Aurora catalog is at http://www.aurorabearing.com/pdf/aurora-bearing-610-catalog.pdf, see page 12. Even though the rod ends are trapped and can't rotate off, my experience shows that a set nut is required to hold the threads from vibrating and failing. The threaded rod failure shown appears to be because the bearing doesn't rotote freely, and puts the threads under bending - a real no-no.

I'm going to take mine off today and inspect.

broken link.jpg

broken link 2.jpg

broken link 3.jpg

Edited by nlappos

Share this post


Link to post
Share on other sites

Posted

Nick can you document with pictures? I’m assuming I knwo where this is and what broke, but having an issue conceptualizing what your fix for this was.

1 person likes this

Share this post


Link to post
Share on other sites

Posted

In the middle picture, the black part (heim joint fitting) hanging down from the bellcrank connects to the tube going forward under the seat that is just under you left thigh when you are sitting in the seat.  The male heim joint fitting that is connected to the mixer broke.  JImChuk

Share this post


Link to post
Share on other sites

Posted

I find it hard to believe that those rod ends could ever fail under normal use. JMO  

Share this post


Link to post
Share on other sites

Posted (edited)

Had to have been the bolt yea? All the heim joints look intact. Also just a shower thought, was that finding the outcome of an accident investigation? Or just an observation of the post crash from the owner and company? I guess I’m asking because that bolt could have snapped during the crash, and the cause of accident be something much more standard. Not to be Monday morning quarterbacking, just asking questions from the eyes of an old army aviation safety officer.

Edited by Matthewtanner

Share this post


Link to post
Share on other sites

Posted

Nick can you document with pictures? I’m assuming I knwo where this is and what broke, but having an issue conceptualizing what your fix for this was.

I wrote the guy twice to try and figure out what I was looking at, he provided the circled sketch. I figure today I'll just pick it all apart (thanks 1avidflyer!) to look at them all.

culater, I also find it hard to imagine how that can crack without abuse.

Share this post


Link to post
Share on other sites

Posted (edited)

The broken parts are  male and female heim joints screwed to each other.  (see picture) The male one is broken off.  I also would wonder how that could have broken off in flight.  Just a wild guess would be maybe if there was flutter in the flaperons, but would think maybe something else would have broke before the heim joint.  JImChuk

PS  twice now I've copied and pasted pictures of male and female heim joints, and they show up in the message at first, but not once it gets posted to the forum

 

 

Edited by 1avidflyer

Share this post


Link to post
Share on other sites

Posted

I don't think Paul Harvey has spoke just yet. I'd say there is more to the rest of the story that hasn't been told. I've seen these wear out, rust up, and freeze, but never seen one broke like that in normal flight loads.

Share this post


Link to post
Share on other sites

Posted

I have seen 3 of these spherical rod ends fail in the past 10 yrs.  one was on a cessna 150 the other on a Cessna 172. Both failures were on the push pull rod that go from the rudder pedals to the steering collar in the nose wheel. When one rod end fail, the nose hear turns sideways and digs in flipping the cessna 150 and in the case of the 172, running it into a ditch. In both case both planes were totaled. The 3rd failure was a threaded rod end on a PA-31 Navjaho nose gear up lock cylinder. IN this case the rod end had a few tiny rust pits right next to the jam nut. So small rust pit= stress riser and the jam nut put the threads under tension and eventually it failed.

ON the Cessna 150 and Cessna 172 no indication of previous damage or rust, however both failed right where the nut puts the threads under tension.

If a person could look at the rod end that failed, on the Avid, look to see if it was a brittle fracture(frosted appearance on the fracture surface) or if it was ductile overload which would leave tell tale signs of 45 degree shear lip. 

At any rate, Id inspect rod ends on flying aircraft and clean with MEK , NO wire brushes, and inspect as closely as possible with a 10X glass. Any sign of rust or imperfection, dispose of the rod end and get a new one. ONe other possible thing to do is make a flat plate with holes on each end that lays over the rod ends as a fail safe prevent. THe flat plate would have to be slotted a little bit.

Other items to look for are make sure the threads are rolled  and NOT cut threads. Make sure the rod end is free to rotate and not binding. Also look and see what happens to the geometry of the rod end if the flaperons are extended more than 23 degrees.....it might be possible to cause binding by exceeding the flap down setting.

Just my ideas.....

2 people like this

Share this post


Link to post
Share on other sites

Posted

Good AN fittings do have rolled threads, but they are not cheap. The cheaper one probably have cut threads. As much as 10 bucks difference in price apiece. The nose gear links take a beating most of the time so they are subject to high stress. Even though I've never seen one break. The aileron bearing should not be under that much stress, installed correctly under normal flight loads. I stand by the fact that something else is going on in this situation. Good rod ends do not break in normal use. In a kitfox or avid, I can't see how one could put enough pressure on the controls to snap a rod end or even fracture it. Unless it was a cheap knockoff, and they do exist. I am running across a few fittings made in Mexico that will not seal because the threads are not centered. Forged parts are a big thing in comm. aircraft and sometimes we use a few of the same parts. And john doe homebuilder out there is completely unaware. My story and sticking to it! But I have been wrong before, ask my wife.

Share this post


Link to post
Share on other sites

Posted

According to the drawings I have looked at that heim joint assembly is in upside down to boot. My guess is the same as Nlappos. Binding between the heim and the f1 frame created a side load in the root of the threads concentrated at the jam nut and thru a number of cycles eventually led to failure

1 person likes this

Share this post


Link to post
Share on other sites

Posted

According to the drawings I have looked at that heim joint assembly is in upside down to boot. My guess is the same as Nlappos. Binding between the heim and the f1 frame created a side load in the root of the threads concentrated at the jam nut and thru a number of cycles eventually led to failure

that would be my guess as well.  I know when I was building my kitfox I had to make a small shim about 1/16 inch thick from 4130 tubing basically making a high misalignment heim joint. Otherwise with full stick the body of the heim bearing would hit the mixer and bind the joint. I believe it would have broke in a few hrs it left like that.

Share this post


Link to post
Share on other sites

Posted

I have sent an email to the owner now and an invite to this forum. I also think that the incorrect assembled heim joint could be the cause. If it binds during full deflection, it could result in a fatigue break. 

1 person likes this

Share this post


Link to post
Share on other sites

Posted

Good catch saskavid on seeing the parts are in the wrong position.  I dug through my pictures of my MK IV rebuild, and did find the one picture I'll post here that shows the parts.  If you enlarge the picture, you can see that mine are in the right order, but interesting to note as well, it looks like his joints were screwed together as much as they could be.  I found a picture of that bellcrank also, and mine have a good amount of thread showing between the two joints.   Wonder what that did to his stick position?  I suppose if the front bellcrank that hinges at the seat truss is adjusted all the other way, maybe that would center the joystick???JImChuk

Photo0445.jpg

Photo1011.jpg

Share this post


Link to post
Share on other sites

Posted

I browsed the load specs for various versions of Heims rod ends from a vendor. Even the cheapest imported version has such high failure specs. in pounds I can only see a slight chance that it was the part at fault. If it failed in the jam nut location it's more likely the failure was caused by over torquing the jam nut causing stretching of the thread in that area and weakening it. A link to the vendor page I browsed here: https://www.fkrodends.com/products/rod-ends/

Remember that bridge that collapsed in Florida last year? A You Tuber I follow did a couple of videos on this and the mass consensus was Post tension failure. That's a possible explanation of how a rod end could fail if the jam nut was over torqued. A link to Ave's video here with a warning for adult language: https://www.youtube.com/watch?v=KtiTm2dKLgU

As one person commented rolled threads are stronger than cut threads; I'm going to add generally this is true. You can screw up rolled threads during manufacturing too! I've made hundreds of thousands of threaded parts in my life time both rolled and cut. A properly cut or rolled thread will each perform to the  task it was designed for. It was also mentioned a rolled thread costs more. I can't understand this. Maybe if they don't have a roll threader head for their cnc and need to send the part out or run the threading operation on a separate machine would this increase the cost. Roll threading is faster and more efficient in the long run but many of the younger cnc guys have never been exposed to this technology. And it was mentioned the threads were not centered. Very hard to produce off centered threads on a lathe and almost impossible to produce them off centered with a roll threader head. But that's the male end I'm talking about. The female end is another story. Locating a cast or forged part in a fixture so the hole for the thread is centered every time is impossible given no two forgings or castings are the same. Who ever mentioned rolled threads did not mention if the threads were internal, external or both. There are rolled taps to produce rolled internal threads also. The tolerances needed for the hole for a rolled internal thread are something most younger machinists don't understand and you end up with a high cost factor due to operator error with screwed up parts and broken taps.

That's my two cents worth; if we look at this problem from enough perspectives were going to get a whole dollars worth sooner or later.

1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

My guess is that failure was caused by bending at the nut, where threads are absolutely vulnerable to bending. The lollypop bearings are supposed to relieve any bending, making the joint purely tension-compression, but if the bearing is trapped, the bending can move down the threads and cause cracking.

 

Edited by nlappos

Share this post


Link to post
Share on other sites

Posted

My guess is that failure was caused by bending at the nut, where threads are absolutely vulnerable to bending. The lollypop bearings are supposed to relieve any bending, making the joint purely tension-compression, but if the bearing is trapped, the bending can move down the threads and cause cracking.

 

Maybe it was bending, but those fittings are designed to withstand a side load within reason. See the first link I included in my comment above to the vendor and angle deviation information is listed under the part specs. That number will give you a starting point if side load was a factor or not. If the side load can deviate beyond the manufacturer specification I'd say it's either a design flaw or failure by the builder to build the linkage according to print. That's if it failure of the part occurred in flight and was not a result of the crash or lever abuse. All bets are off until we can rule out everything else as the cause of the crash. I did not read any official crash investigation yet.

Share this post


Link to post
Share on other sites

Posted

Most hardened parts don't bend much at all before braking.

Share this post


Link to post
Share on other sites

Posted

IF you Google "spherical rod end failure" you will get a few hits and if you look at them, you will se most fail right where the jam hut is... Any thread under tension is a bad deal and in this case it is a  single load path or what I like to call a "single point of failure" which could result in loss of the aircraft.  I never liked the threaded rod ends on the end of the lift struts on Model 4 kitfox. They are not very big and a few have failed with the plane going into a uncontrollable spiral.... IF a kitfox has a GW of 1500 # and it has those little rod ends, then look at a  Cessna 120 or 140. It has a gross weight of 1650 and he HUGH rod ends on the end of the lift struts....3/4" diameter or larger threaded rod ends...

PUt the threads under tension, install a jam nut to increase the tension, add in rust pits and more than likely you will exceed the damage tolerance of the part and it will fail.

IN the case of the flaperon rod end, it was probably binding somehow and eventually failed.

1 person likes this

Share this post


Link to post
Share on other sites

Posted

from another thread about thread failure:

October 19, 2007 - October 31, 2007

 



      
      I will say that AVIDs do have what seem to me to be a stronger wing attach fitting
      by eliminating the rod end. 
      
      Kitfox used rod ends  I think to compensate for variances in build tolerances but
      I don't think it anything to worry about.  I would check with John McBean
      to find out the exact procedure of what you are asking.
      
      
      You only need one  thread to fail ,so no matter how far it screwed in the rod end
      the one that will fail will be out side of it .   Does anyone know the actual
      shear strength of the threaded part of the wing strut attach point ?   I would
      bet each of the 4 attach points are 4000 to 6000 pound tensile and we fly
      at 1200  or so lbs  divided by  4 = 300  each ?  MAke sense ?  If I  am correct
      that means at 10 gs you are pulling 3000 pound force and I guarantee you that
      you won't see 10gs.    I am done talking outta my ass now.   
      
      
      >  I would think a connection that important, and under that much stress would
      have all the bolts in sheer, not tension. If those 8 or ten threads fail, your
      cooked. 
      > 
      > 
      
      
      --------
      Rotax Dealer, Ontario Canada
      Flying Videos and Kitfox Info
      http://www.cfisher.com/
      
      
      Read this topic online here:
      
      http://forums.matronics.com/viewtopic.php?p=140724#140724
      
      
________________________________________________________________________________
From:"fox5flyer" <fox5flyer(at)idealwifi.net>
Subject:Attachment fitting at top of lift struts.
Date:Oct 19, 2007

One thing that needs to be remembered is that AN fittings have "rolled" threads, not cut threads. This makes them much stronger than the normal non-AN threaded fitting and much less susceptible to stress cracking at the threads. Personally, if there was anything to worry about I believe it would be the fittings at the horizontal stab where there is a history of breakage in the IV. For what it's worth. Deke S5, NE Michigan ________________________________________________________________________________

Subject:Re: Attachment fitting at top of lift struts.
From:"dave" <dave(at)cfisher.com>
Date:Oct 19, 2007

> I believe it > would be the fittings at the horizontal stab where there is a history of > breakage in the IV. > I think that perhaps the breakages was due to rough ground handling using the horiz.stab braces for handle rather than using the Fuselage handle. Would this be the likely culprit ? He have rolled thread on the strut atttach point? SO what is the tensile of them and the rod ends ? -------- Rotax Dealer, Ontario Canada Flying Videos and Kitfox Info http://www.cfisher.com/ Read this topic online here: http://forums.matronics.com/viewtopic.php?p=140728#140728 ________________________________________________________________________________

From:"Ken Harrison" <kenharrison(at)comporium.net>
Subject:Re: Attachment fitting at top of lift struts.
Date:Oct 19, 2007

OK, that makes sense. Except I wonder if one of the lurking engineers on the list could calculate the in-flight load on one of those fittings. Because I think you have to account for the fact that the strut is not attached to the wing at 90 degrees. The in-flight vertical load per fitting is 300 pounds, but the strut is at about 60 degrees from vertical (or more, I didn't actually measure it). Just for example, if the strut were at 45 degrees to the vertical, and had a 300 pound vertical load from the wing, the tension felt by the strut would be 600 pounds, and 600 pounds of compression felt by the inboard wing spar. (Now I'm showing my ignorance.) Is that how it would work? The same as the vertical and horizontal components of lift felt by a wing in turning flight? Anyway, I think I'm just worrying for no reason. It sounds like the connection is plenty strong. I'll go back to worrying about ground-looping. Thanks for the info. -----Original Message----- From: owner-kitfox-list-server(at)matronics.com [mailto:owner-kitfox-list-server(at)matronics.com] On Behalf Of dave Sent: Friday, October 19, 2007 5:51 AM Subject: Kitfox-List: Re: Attachment fitting at top of lift struts. Ken, I think you will find that there to my knowledge has ever been a inflight breakup of a Kitfox from the wing strut attach fitting. I think the testing showed in excess of +14G and still no failure. But I stand to be corrected. I will say that AVIDs do have what seem to me to be a stronger wing attach fitting by eliminating the rod end. Kitfox used rod ends I think to compensate for variances in build tolerances but I don't think it anything to worry about. I would check with John McBean to find out the exact procedure of what you are asking. You only need one thread to fail ,so no matter how far it screwed in the rod end the one that will fail will be out side of it . Does anyone know the actual shear strength of the threaded part of the wing strut attach point ? I would bet each of the 4 attach points are 4000 to 6000 pound tensile and we fly at 1200 or so lbs divided by 4 = 300 each ? MAke sense ? If I am correct that means at 10 gs you are pulling 3000 pound force and I guarantee you that you won't see 10gs. I am done talking outta my ass now. > I would think a connection that important, and under that much stress would have all the bolts in sheer, not tension. If those 8 or ten threads fail, your cooked. > > -------- Rotax Dealer, Ontario Canada Flying Videos and Kitfox Info http://www.cfisher.com/ Read this topic online here: http://forums.matronics.com/viewtopic.php?p=140724#140724

Share this post


Link to post
Share on other sites

Posted

Kitfox_Strut_Fairings_027_177 (1).jpg

download.jpg

Share this post


Link to post
Share on other sites

Posted (edited)

There 3250lbs each and they are nothing special nor are they rolled threads Just your standard Heim bearing.

the Heim bearing fails on the outer edge the part that is around the bearing itself not the threads and remember its all just held together with a 1/4 inch bolt ha but I have never heard of one failing yet.

 

Edited by TJay

Share this post


Link to post
Share on other sites

Posted

I don't think its panic time just quite yet. 2 or 3 cases in a thousand aircraft with a few thousand hours under its belt isn't a disaster. Of course, it probably was to the two that it happened to. Situations like this can destroy an aircraft's reputation in a second. It is happening now with the QCU challenger aircraft. 4000 aircraft and 7 or 8 had wings fold in flight. Turns out that most ended up being owner/ builder error. But the wing fold issue stuck like glue. And earned it a damaging reputation. In most cases, its very hard to determine the what and why of how it happened. Builder error, bad design, defective part. incorrect assembly, we might never know but rod ends are used in almost every aircraft without issue or concern. Inspect them, of course, maybe even modify the setup if its a major worry. but word spreads fast in this internet world and what was once a good solid aircraft is now a flying death trap because of 2 or 3 cases. Tube failure due to corrosion has been around since there were tube aircraft and will continue. That's why we have conditional inspections every year, but a lot of them are pencil inspections which hurts us all. This type of group is great for info, but for the lurker who is thinking about buying an Avid/Fox, it can be very negative. I suggest we tread lightly till we know more, if we ever do.

1 person likes this

Share this post


Link to post
Share on other sites

Posted

I don't think its panic time just quite yet. 2 or 3 cases in a thousand aircraft with a few thousand hours under its belt isn't a disaster. Of course, it probably was to the two that it happened to. Situations like this can destroy an aircraft's reputation in a second. It is happening now with the QCU challenger aircraft. 4000 aircraft and 7 or 8 had wings fold in flight. Turns out that most ended up being owner/ builder error. But the wing fold issue stuck like glue. And earned it a damaging reputation. In most cases, its very hard to determine the what and why of how it happened. Builder error, bad design, defective part. incorrect assembly, we might never know but rod ends are used in almost every aircraft without issue or concern. Inspect them, of course, maybe even modify the setup if its a major worry. but word spreads fast in this internet world and what was once a good solid aircraft is now a flying death trap because of 2 or 3 cases. Tube failure due to corrosion has been around since there were tube aircraft and will continue. That's why we have conditional inspections every year, but a lot of them are pencil inspections which hurts us all. This type of group is great for info, but for the lurker who is thinking about buying an Avid/Fox, it can be very negative. I suggest we tread lightly till we know more, if we ever do.

Well said. Jabiru has the same problem

Share this post


Link to post
Share on other sites

Posted

Its not the lift strut rod end I am concerned about. It is the welded in stud that the FEMALE rod end attachs to. This is on some of the early model 4 kitfoxes. I see the factory change and started using a MALE rod end that threads into the end of the lift strut. The earlier style is the one that worries me. It has CUT threads placed under tension....not good.

THis silver strut has a stud welded into the strut end and a FEMALE rod end threads onto it....The threads are CUT threads...The aviation industry learned all about cut threads on Piper lift strut forks when Univair made replace forks with cut threads. Even the rolled piper threads failed after so many cycles and hours so finally Piper came out with a lift strut that had a really large diamer HD strut fork that does not have a life.....

The WHite lift strut in the photo below is a later model Kitfox strut. Note the end of the strut is drilled and threaded to accept a MALE rod end....

This is better but I think I will design some kind of secondary load path for th strut attach point.... Probably a flat strap that picks up on the lift strut wing fitting and then is riveted to the lift strut tube with 4 or 5 steel high shear cherry max rivets. 

WHen new and tested even the cut threads will hold a lot, but they are WAY more suspetable to fatique crack(the cut thread already has a built in stress riser), add to that flight cycles, people pushing on the struts and you are asking for trouble....

The way Avid and early kit fox struts were built is the way to to go. Put the attach bolt in shear. The CAA accident aircraft C-FOLD  (theres a good N number!!!) shows a threaded strut end fitting that failed and they think it failed on take off. WHo know the history.. was it in a previous accident? How old is it? Was it rusted? Anyway according to he accident report the aircraft lifted off and a loud crack was heard followed by an uncommanded roll to the left...... it was the left rear lift strut attack threaded part that failed.... The plane crashed and was BER...

 

dft_avsafety_pdf_023415.pdf

download.jpg

134.jpg

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now