Titanium end on pushrod breaks inflight

[Matt Colan]

Went flying today with the new airplane. I had been battling some motor run gremlins that I think I had solved after the first flight. Second flight while flying level, the airplane suddenly threw itself into the ground. During the autopsy, I saw the titanium rod snapped at the end of the turnbuckle for the elevator. I have never seen this happen before to anyone. I guess the better question is, has this ever happened? The idea of using these ends makes me wonder if it’s worth using them again or not.

The caveat to all of this is I was also in the middle of a complete runaway motor run…which turns out was because the motor had come loose. Could the harmonic and vibration from the motor being loose and running away have caused this?

[frank williams]

WOW .... thats terrible ... Very sorry to see that .... you say they were titanium .... what brand ..... turnbuckle? ... with a hex area in the middle?

[Paul Walker]

The short answer is yes.

It is also dependent on the geometry around the end. Can you include a picture of that?

[Paul Walker]

Before you suggest that you have a pic, I see the aft end of it. I am looking for the front side of that joint.
 Who made it?
Were the threads rolled or cut?

[Matt Colan]

Started to cut it away and see what I had

Paul, here’s the pushrod end. I got it from Okie Air. Where he got it made from, I’m not sure

[Dave Hull]

That's really awful, Matt. Very sorry to see this happen.

The break is right at the maximum moment in the Ti part, and I suspect on any small parts run you'll find that the threads are cut, not rolled. If you have a good magnifying glass even, check to see if there is a root radius on the threads. Not proof, since the cutter should have had a tip radius, but if it is sharp, then that's going to be the problem.

[john e. holliday]

That is sad to see.  I myself have not used titaniam because it is a brittle material.  It looks like the tap ends in a sharp surface instead of a taper.   Might check that on the next one or is the thread going into yhe pushrod?  Looking agaian it looks like solid rod.  Better luck next time.

[Brett Buck]

Went flying today with the new airplane. I had been battling some motor run gremlins that I think I had solved after the first flight. Second flight while flying level, the airplane suddenly threw itself into the ground. During the autopsy, I saw the titanium rod snapped at the end of the turnbuckle for the elevator. I have never seen this happen before to anyone. I guess the better question is, has this ever happened? The idea of using these ends makes me wonder if it’s worth using them again or not.

The caveat to all of this is I was also in the middle of a complete runaway motor run…which turns out was because the motor had come loose. Could the harmonic and vibration from the motor being loose and running away have caused this?

    As noted, yes, and I spent most of the last two weeks examining a similar problem (thankfully, *before* it happened). You already have the world's foremost expert on titanium on the task, so I have nothing to add.

    I have used either medium steel all-thread, RC pushrod ends, or the stainless inserts from Central hobbies, partly because of potential fatigue issues with other materials.

    I am very sorry this happened. It looks fixable to me, but before you make any drastic decisions, give it a day or two.

     Brett

     

[Mike Griffin]

Started to cut it away and see what I had

Paul, here’s the pushrod end. I got it from Okie Air. Where he got it made from, I’m not sure

Matt,

Okie Air took all of Tom Morris's inventory as far as parts for control line systems.  I do not know if this was one of those or not and if that helps any after the fact.  Sorry about the crash.

Mike

[Ty Marcucci]

Yes, and I posted a photo of it on here a few years back.. The unit failed as it had been bent due to a previous crash, but I used it again, at the cost of one airplane. New or never again.

[Ken Culbertson]

I suspect that the part was defective.  These parts are made on a small screw machine which is really just a tiny turret lathe.  I ran them when I was going to school.  The threads would have been cut by a die and judging by how close they are to the hex I can see what happened.  Titanium is difficult to machine and the cutting tools need to be changed often. If the die gets too close it will snap the threads or worse, twist them.  The design is also bad.  The threads should start following a small taper.

Now I have to worry about mine.  Sorry Matt, that plane was beautiful and a real pleasure to watch fly.

Ken

AND, I have no idea where the rest of the thread is.  It just disappeared last night.  There is another post after this one so maybe it just decided I did not need to see it and everybody else can.  Happens a lot lately.   

Well whatever it was, it came from my end!  Thanks Dennis.  If you ever want to know how long a post can be, even Ted couldn't have reach the end!                                                                                                                                                                                              

[Air Ministry .]

Its Why theyre called HARMonics ! . 

And is why japanese parts fall off pre unit Bonnevilles inside a week ! .

Straight Eight Cranks are subject to them . And Other Things . Like SINGLE CYLINDER ENGINES .
This is why the Alfa used TWO four cylinder cranks . And people use counterballancers .



And WHY Ducati got a dual rod single going quick . One had a counterballance .


and suzuki copied it .



theres a ' PERIOD ' of max smoothness & one of max roughness , generally . Overlapping things can multiplicate the amplitude & things .

[Dennis Adamisin]

Matt: not much useful aside from the general observation - that STINKS!

Ken C: you might want to take a look at and edit out some blank space in your post #10...

[Paul Walker]

Matt,
I typed a long response on my phone, and it disappeared before being included.
Bottom line, possibly the wrong material, bad geometry, and not rolled threads.

I am at the NWR currently as the ED. Busy at the moment. I can elaborate later when this is over.

[Dan McEntee]

Went flying today with the new airplane. I had been battling some motor run gremlins that I think I had solved after the first flight. Second flight while flying level, the airplane suddenly threw itself into the ground. During the autopsy, I saw the titanium rod snapped at the end of the turnbuckle for the elevator. I have never seen this happen before to anyone. I guess the better question is, has this ever happened? The idea of using these ends makes me wonder if it’s worth using them again or not.

The caveat to all of this is I was also in the middle of a complete runaway motor run…which turns out was because the motor had come loose. Could the harmonic and vibration from the motor being loose and running away have caused this?

    I think this could have been a contributing factor. If the part was defective, it may have never failed unless it was exposed to specific conditions. The hammer and tong effect from a PA.75 loose in it's mounts could affect a lot of things. Since you were in level flight, no excessive load on the pushrods at that time. I think it's a very real possibility. Since you will be rebuilding it, it's a perfect time to examine as much of the entire air frame as you can. I've seen FP.40s shake the nose off a model before! Rolling the threads on such a part is preferred also, but the tooling required for that isn't too common. I have a friend that has the tool for rolling threads to restore./make motorcycle spokes and such, but I don't recall if it can be set up for left hand threads or not. Sometimes I think Titanium is best left alone for what very little weight savings they provide on such a small part.
   Type at you later,
  Dan McEntee

[Ken Culbertson]

I saw this happen and I saw the expression on everybody's face (including mine I assume) when Matt opened the rear hatch and saw the ball link just dangling there with the pushrod threads attached to nothing.  Everyone was saying to themselves "I use those, am I next?"  That prompted me to take a look at mine and I was a bit shocked what I saw.  All of mine are defective and it is easy for me to see what happened.  IMHO it was most likely the vibration that caused a defective part to fail.  You have to understand how these parts are made to see what happened. I can think of many ways to set up a screw machine to make one of these but only two to cut the threads.  They are either cut by a lathe or a die.  Even if they are molded which one of the Okie set appears to be, they have to pass through a die. The Okie one that failed appear to be using a die.  Why they would choose Titanium for this is beyond me.  If you look at the bellcrank pushrod on my Trifecta you will see what went wrong.  Look at the threads at the very end where it transitions to the pushrod.  They are twisted.  Simply put, the die was either dull or not releasing cleanly and the shaft was twisted.  Steel basically says "so what" and moves on when this happens, titanium cracks.  Little tiny internal cracks.  I am not going to change mine out, electric does not generate the vibration that triggered Matt's failure, but I doubt that I will ever use these connectors on a PA ship again.

There are steel alternatives and one that I particularly like.  The turnbuckle at one end of the pushrod.  It requires a threaded pushrod at one end but that is not difficult using a blind nut, an endcap and some Kevlar thread.  Another thing it lets you do, especially with logarithmic horns, is have all of the ratio and alignment adjustments on the flap horn and eliminate that pesky elevator hatch.

Weight is not an issue at all with Titanium vs Steel.  The Okie set weighs 5 grams, the turnbuckles are 4.5 grams if you count something on the other end.

[Lauri Malila]

Sorry for your loss, Matt.
A couple of things come to mind. Why the threaded part has to be so long? With your building skills, I'd think that something like 1/8" of length adjustability is enough. Long and thin part will flex more, and eventually crack earlier. Also, the bigger diameter chunk in the middle of rod is unnecessarily long. Just useless mass thus inertia that increases the risk of bending and fatigue.
All corners should be well radiused.
Did the part break from middle of thread or from a corner?
For small cottage production I see nothing wrong with threads cut with die, but you need a very good quality and sharp die that cuts without inducing too much stress into the part.
Also, in this case I'd prefer something else than imperial thread (is that 4-40?), M3 Ould be better, or even a fine pitch M3. L

[John Carrodus]

Oh crap oh dear.
I have just repaired and rebuilt a model given to me in pieces. During the rebuild it was obvious the pushrod had been vibrating like a tuning fork which in turn led to failure in the connections and crash. There was evidence also that the tail assembly had developed stress cracks where it joined the fuse. I have since been informed the engine had been modified and needed an unbalanced prop to -wait for it- to stop it vibrating!! When it crashed it had a balance prop - I believe. Them harmonics thingies can really bugger stuff up and even bring down bridges.

[Matt Colan]

Sorry for your loss, Matt.
A couple of things come to mind. Why the threaded part has to be so long? With your building skills, I'd think that something like 1/8" of length adjustability is enough. Long and thin part will flex more, and eventually crack earlier. Also, the bigger diameter chunk in the middle of rod is unnecessarily long. Just useless mass thus inertia that increases the risk of bending and fatigue.
All corners should be well radiused.
Did the part break from middle of thread or from a corner?
For small cottage production I see nothing wrong with threads cut with die, but you need a very good quality and sharp die that cuts without inducing too much stress into the part.
Also, in this case I'd prefer something else than imperial thread (is that 4-40?), M3 Ould be better, or even a fine pitch M3. L

Hi Lauri, it broke on the threads, just past the middle, hex portion of the rod. I’ve thought about making the adjustable portion much shorter in the future, long enough for the ball links to thread in with adjustment on each side. The threads are 4-40. What makes an M3 thread superior to a 4-40 out of curiosity?

Dennis Nunes was kind enough to send me an article he wrote on how he sets up his controls. He used a cut off 4-40 bolt and a nut JB welded to the carbon rod. Something I might do as well on the replacement pushrod. Another concern I have is if this part failed, what would the others look like and how fast are they going to fail in the future? Hmmmm….

[Lauri Malila]

I have no exact figures here now, but I will try to find/measure something.
But 4-40 has slightly smaller o.d. and much coarser pitch (1/40”, 0,635mm). That leaves clearly less cross area than with M3 with 3mm diameter and 0,5mm pitch.
So, with a finer thread you win both in bending stifness and in stressing less the material during manufacturing.
But it’s a trade-off; 4-40 holding power is superior in aluminium and plastic due to the coarser pitch, so with M3 I’d make the inside threads a little longer to compensate.
I never understood why people use ball links in stab pushrod. There is no sideways movement. L

[Dick Pacini]

Matt, is that your new NATS plane for this year?  Do you have a backup plane or can that one be repaired successfully?

[Matt Colan]

Matt, is that your new NATS plane for this year?  Do you have a backup plane or can that one be repaired successfully?

Ole reliable will be pressed into service again. A 6th and 5th place at the last two NATS and never coming in lower than 2nd at a local contest still shows how good it flies

[Crist Rigotti]

Ole reliable will be pressed into service again. A 6th and 5th place at the last two NATS and never coming in lower than 2nd at a local contest still shows how good it flies

You mean how good you fly!

[Jim Svitko]

This is not the first time I have heard of such a failure.  One other time, years ago, I saw that same fitting break on another model, at nearly the same location.  In another instance, it was 4-40 all thread rod that failed in a similar manner.

I am not the expert on such things but I noticed one common factor regarding these failures: there was a rather large amount of unsupported thread.  When under vibration, maybe that unsupported portion does act like a tuning fork.  It reminds me of a diving board that is fixed at both ends, but free in the middle.  The fixed ends look to be under high stresses.

After seeing these failures, I tried another method.  I have small clevises machined for me.  I leave just enough exposed thread to make adjustments.  So far, no issues of any kind.  The only problem with a clevis is that the minimum adjustment amount is one half turn.  The thread pitch and throw affect how much change that half turn makes.

[Phil Spillman]

Hi Matt, So sorry for your loss. Hope it never happens again.

Phil Spillman

[pat king]

There are lots of grades (alloys) of Titanium. Some are absolutely NG for fasteners (read threaded parts) or parts subjected to cyclic loading. For our applications I recommend staying away from Titanium parts unless you are a metallurgist and know the grade used for the part.

Pat

[Air Ministry .]

Yea , a total bumah .

Blankety Cars - ' wrecks ' the impact can total / overstress parts . A frontal on a old front engine job can
wreck the crank , tailshaft gearbox and bend the diff . Suspension bits can be a no no .Overstressed .

ANYWAY . I think the LENGTH - Overhang , mightve had a bit to do with it . Cantilever ( its called )
if the pushrod got Excited ( technical term ! ) UP & DOWN . i.e. tangental / across the thread ( Rather'n ALONG ) .

Ones Best having say 4 threads only ' clear ' as its the minmum diameter in the contraption .
Some think ONE adjuster / ball joint is enough . Its harder to break good quality 1/8 in. Wire , and bearing areas good on a 1/4 in. long hard brass bush fitted to the horn .
If theres no acid left init .  Triple Check - rinse cetra .  lube with pressure through ! ,  Maybe a larger plastic tube held on it & a long steady blow ora syringe . 
( Test a few in a vice too see the bends NOT cracked , on THAT wire . )

Anyway , sorry to see any decent plane & or engine splattered .