What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

7A elevator spar crack

This is a very helpful thread, inspection prior to my next flight..

I may be stateing the obvious but it seems to me that if one of the bearings is screwed out relative the the hinge line (ie a miss-aligned hinge as per the discussion), then this would result in a force on the elevator or rudder spar that in trail would be normal to the spar, but would vary in direction according to the control surface deflection. So moving the surface from full up to full down would correspond to one cycle of such a force. The least stiff part will 'give', this is probably the spar at exactly the point the cracking has occurred. It would be a large misalignment to be felt as binding.

I would expect cracking on the opposite side of the bearing as well if the bearings were misaligned only in the sense of one being screwed out beyond the hinge line.

Is there any evidence of the .063" doubler cracking?

What I am most concerned with would be the rudder bearing alignment. The lower bearing is tied to the fuselage and Tailwheel assembly and the lower part of the VS spar can be pulled out of alignment when mounted on the fuse. Also, any pressure on the VS forward spar when attached will also pull this alignment out. So it is mandatory to carefully realign these bearings after VS final attachment.
 
Last edited:
In my sample size of (1), I found a 100% failure to properly align the empenage hinges. I don't know if it was built that way or moved into that condition, but I certainly found a significant problem upon disassembly.

All good now though. I'll check it in a few hundred hours, but I suspect that I'll find each bolt will still slip easily through each hinge point.

My vote is that "we" are creating our own problems.
 
Checked my -6 elevator today, and thought I found a crack! Until I wiped my finger over the area and the hair that was stuck in the LPS residue came free. No cracks!
 
Not much news today

I checked the doubler with dye penetrant and it was not cracked.
Someone asked about it.

If anyone was ever concerned about epoxy resin bonding to the lead counterweight you can relax! Unbelievable bond. Tip had to be destroyed to get it off.

Elevator is apart now and I will start the re-assembly tomorrow.
I plan to put it back together per the plans.

Thanks for all the comments.

Mark
 
Stuff deleted....

Like most pilots, during preflight, I grab the elevator and give it a wag and a shake, then move on to the other one. Today, since I was doing a very THOROUGH job, I notice a little bit of a (tiny) clunk when I moved the left side. I moved to the right, and got the same feel. I then grabbed both of them - and they moved about 1 degree relative to each other !!

This is EXACTLY what happened to me back in June. Scared the poop out of me. I was doing my walk around and gave the elevator my usual good shake and got a "thud" in return.

I have replaced the bolt and nut, but I think I'll take the advice I saw elsewhere in this thread and put another bolt through.

Now my walkaround includes reaching around the rudder, gripping both elevators, and then trying to move them in opposite directions.
 
I may be stateing the obvious but it seems to me that if one of the bearings is screwed out relative the the hinge line (snip)...It would be a large misalignment to be felt as binding.

All true. However, Mark is sure his bearings were not misaligned, and reports no binding or odd feel. Right now I'm inclined to discount misalignment as a cause.

This may address why it happened there, through no apparent failure of the builder or maintenance, but not what developed a loading condition sufficient to fatigue this area. Warm days on the gulf coastal areas do present a lot of bumps in the warm months (many of them) at relatively low altitudes, but what other reasons could this bird be seeing more loading cycles than another? And, is it only on the lower edge and not showing much loading on the upper edge because this is the typical load direction for keeping the nose up and it is not precipitated by some overwhelming aerodynamic cyclic load?

Bill, perhaps the cycles are vibratory, not aerodynamic. Explanation as we go along.

But first, here's a sketch with what I think is the bending mode when cycled:

116pfrp.jpg


In this illustration the tip of the elevator has been forced upward in relation to the fixed stabilizer pivot. The resulting load is as shown. The lower rivet is in tension. The spar material below the rivet is unsupported.
 
Loose jamb nuts

Yesterday I check my -9A and found the HS jamb nuts tight and no issues. BUT... all 3 of the VS/rudder jamb nuts were loose! I was stunned because the middle hinge nut was close to 2 turns loose!

The only mechanism I can picture is application of rudder loads causing the rudder spar to twist just enough to break the face to face friction of the nut to the spar. Then with induced vibration in the airframe, the nut continues to work it's way loose.

While I look at these nuts during inspections, I've never put a wrench to them since first flight. Not good. Locktite & an index mark goes on next!

Also, I wiggled the elevators against each other and found them to move separately by 1-2 deg! Re tightened the pushrod tube attachment bolt and all is as it should be now. I have to agree, I think a second bolt through these horns is a good idea.
 
Dan, your sketch makes sense. Also looks like that's what would happen if the stick was locked and the elevator loaded upwards. Or, an up-slam against the stop. That's why I still like an external lock between the stab and the counterweight. Reduces the arm and supports the outboard area of the elevator.
 
...That's why I still like an external lock between the stab and the counterweight. Reduces the arm and supports the outboard area of the elevator...

Has Van signed off on this? Because Cessna doesn't. Throwing a big gust load at that little counterweight structure may do a lot of harm.
 
Loose lock nuts on these bearings is very common. There is a bit of a settling in period that initially causes the nuts to loosen. As Dan's picture points out there is a load on this bearing. Back and forth, back and forth, with the "nut" supplying the main contact area. Imagine a slight gap between either the nut-plate and the backing plate, or the backing plate and the spar. It will take a bit of time, with pressure from the locking nut, to settle these parts together. Perhaps that is why some of the nuts never come loose and others do. For this reason I do not like witness marks, or lock tite, on these parts as it indicates that the parts were tight at the time but does not tell you if they are still under the proper torque. It does not take much flexing, and compression of the spar area for this nut to get loose. The next time you tighten that bolt it will stay tight much longer, the next time you snug it up it will likely never move as the parts have settled into place.

Bottom line is that these nuts need to be checked on a regular basis but the good news is that after time they will settle in.
 
Last edited:
Maybe a stupid question, but why isn't it standard practice to put a star lock washer under the jam nuts on the rod end bearings to keep these from coming loose?
 
Lock Washer

This is a very interesting discussion! Reading the last couple posts here regarding the lock washer idea made me think about the Flap rod-end-bearing install (just because I'm there now)... It calls for loc-tite, flat wash, lock washer then jam-nut. Just food for thought. I wonder if that would be a good /bad idea. I'll let you engineer types fret that out. You know more than I. I just follow directions closely ;)
 
This is a very interesting discussion! Reading the last couple posts here regarding the lock washer idea made me think about the Flap rod-end-bearing install (just because I'm there now)... It calls for loc-tite, flat wash, lock washer then jam-nut. Just food for thought. I wonder if that would be a good /bad idea. I'll let you engineer types fret that out. You know more than I. I just follow directions closely ;)

If using two non self locking jam nuts...getting the correct torque on each nut is difficult.

Here is a good two nut article...

http://www.boltscience.com/pages/twonuts.htm

Here is some good NASA Fastner Design information. See page 7 for the two nut paragraph...

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19900009424_1990009424.pdf
 
Maybe a stupid question, but why isn't it standard practice to put a star lock washer under the jam nuts on the rod end bearings to keep these from coming loose?

I always put a star washer and a thin AN960L under the rod end nuts. Why? Because I've found mine loose.
 
Thanks Dan

Dan,
Thanks for your theory, drawing and explaination.
It makes perfect sense to me.
So we need to reduce the vibration or spread out the load.


Mark



All true. However, Mark is sure his bearings were not misaligned, and reports no binding or odd feel. Right now I'm inclined to discount misalignment as a cause.



Bill, perhaps the cycles are vibratory, not aerodynamic. Explanation as we go along.

But first, here's a sketch with what I think is the bending mode when cycled:

116pfrp.jpg


In this illustration the tip of the elevator has been forced upward in relation to the fixed stabilizer pivot. The resulting load is as shown. The lower rivet is in tension. The spar material below the rivet is unsupported.
 
another thought

When ours was pulled apart, we noticed that the hinge bearing rod end (on the damaged attach point) was adjusted well out of the nut plate to the point where it is pretty much flush with the back side of the nutplate. I would think having this lever longer would not be helping the situation.



Dan,
Thanks for your theory, drawing and explaination.
It makes perfect sense to me.
So we need to reduce the vibration or spread out the load.


Mark
 
causing stress concentrations leading to cracks. Also more susceptible to corrosion. A better choice would be Loc Tite or a self locking jam nut on top of these nuts.

This engineer adds a vote for LocTite (thread lock compound).

Also, earlier this week (not waiting for an SB, AD, etc, etc.), I inspected my empennage and found loose elev jam nuts and differential play between control surfaces. I added added the second fastener with spacer (AN3), then inspected, lubed and replaced the main AN3 fastener connecting horn to pushrod. No more differential play, and my elevator assy is much more able to withstand asymmetrical loading of spiraling slipstream, and taxiing on rough surfaces, etc.
 
Last edited:
Maybe a stupid question, but why isn't it standard practice to put a star lock washer under the jam nuts on the rod end bearings to keep these from coming loose?

As Bob A. points out, the sharp edges of the star washer would dig into the spar, creating a crack initiation risk.

Another consideration is that for this application to function correctly, the clamping force must be distributed evenly around the hole in the spar where the rod end inserts. The teeth of the star washer might create an uneven clamping force distribution, which would be magnified when up or down forces are applied by the control surface. And that would be a fatigue liability.

As I've asserted before, this is a sub-optimal application of male-shank rod ends, though one with a long and pretty much satisfactory service history. In order for it to function as desired, care must be used to stay within the envelope of established practice. That is, the rod end extension (number of exposed threads) must be minimized (or at least kept to what the plans suggest), and the nuts must kept tight.

That said, and with all due respect to the previous cautions about doing too much "beefing up" and just creating problems elsewhere, I suspect that overall strength and fatigue resistance of the structure would be enhanced with a modestly-sized doubler on the forward side of the spar that would help distribute stresses more evenly into the thinner parts of the surrounding structure.

Actually, and I know I'm not alone in this, what I think would be best would be to ditch the threaded rod ends altogether and go to a machined part with a sizable footing attached to the spar at multiple points so that it does not rely so much on clamping forces. But that cuts close to the kind of idle speculation that we're trying to avoid.

Regardless, I have full confidence that the engineering crew at Vans is taking this issue very seriously and proceeding with all haste to understand it in all its complexity and to address it. I've known Van for twenty-odd years, and I've met a lot of his engineers, and they are not the kind of people to take this stuff lightly.

Thanks, Bob K.
 
All true. However, Mark is sure his bearings were not misaligned, and reports no binding or odd feel. Right now I'm inclined to discount misalignment as a cause.



Bill, perhaps the cycles are vibratory, not aerodynamic. Explanation as we go along.

But first, here's a sketch with what I think is the bending mode when cycled:

116pfrp.jpg


In this illustration the tip of the elevator has been forced upward in relation to the fixed stabilizer pivot. The resulting load is as shown. The lower rivet is in tension. The spar material below the rivet is unsupported.

Dan I really like and appreciate your sketches. They're always very helpful for all of us to visualize otherwise complex concepts. In this particular case, though, I think the load path may be a little more complicated.

What is not shown in this sketch is what the actual load path "should" be. The jam nut should be transferring all of the rotational torque load into the spar itself, and then into the skin throug all the spar/skin rivets. In reality, some portion of load will always be picked up by the nutplate into the doubler and only then into the spar - really an indeterminate load path. The idea, I think, is for most of the load to be transferred directly to the spar by sandiching it between the nutplate/doubler on the aft side, and the jam nut on the forward side.

Someone earlier mentioned the idea of a star lockwasher. Another poster thought, in my opinion correctly, that a lockwasher directly on the spar would dig in and possibly exacerbate a cracking issue. I find that I agree with both of these posters. What strikes me as a rather good idea, though, is what Bob suggested: using a washer against the forward side spar, a lockwasher on top of that, and finally the jam nut. Other than adding a little weight and possibly decreasing the flutter frequency, a washer and lockwasher would spread the load over a slightly greater area into the spar.

Of course with literally thousands of aircraft flying per the original design, due diligence and inspections (definitely a torque stripe witness mark) are the easiest, cheapest, and likely best way to prevent problems with the control surfaces.
 
...Actually, and I know I'm not alone in this, what I think would be best would be to ditch the threaded rod ends altogether and go to a machined part with a sizable footing attached to the spar at multiple points so that it does not rely so much on clamping forces. But that cuts close to the kind of idle speculation that we're trying to avoid...

No, you are most certainly not alone. I'm toying with doing a machined ball bearing hinge fitting as a replacement for the rod ends. I am not a fan of the stock Vans setup, but I fully understand why it was done.

And in the end, while the stock setup is far from an example of engineering elegance, it is obviously "good enough" if assembled and maintained properly.
 
All true. However, Mark is sure his bearings were not misaligned, and reports no binding or odd feel. Right now I'm inclined to discount misalignment as a cause.



Bill, perhaps the cycles are vibratory, not aerodynamic. Explanation as we go along.

But first, here's a sketch with what I think is the bending mode when cycled:

116pfrp.jpg


In this illustration the tip of the elevator has been forced upward in relation to the fixed stabilizer pivot. The resulting load is as shown. The lower rivet is in tension. The spar material below the rivet is unsupported.

Dan, again, a picture is worth a thousand words, thanks! Since the stiffener is not cracked, your view seems the most likely loading case, and it also indicates that the stiffener is NOT more stiff than the spar with channeled and skins attached. If it was more stiff (channel shape) the one lonely rivet would not be carrying so much load. I am assuming it is to scale. I was loosely referring to a (yet unidentified) cyclic load overlaid on a static load, thus it failed only on one side where it is biased in favor of your sketch. (Imagine a sine wave where the "zero load" line is not centered between the peaks) It remains unknown why this bird has the failure when so many others don't and have more hours on them. Reference - 3800TT on N477RV flown by Mike Seager and many others, including 6's. This point, from a design perspective, would not indicate a fundamental design deficiency. But - it ain't over until it's over.


There has been a lot of discussion about the jamb nut, but it was tight and seems to have no relation to this failure mode.
 
All true. However, Mark is sure his bearings were not misaligned, and reports no binding or odd feel. Right now I'm inclined to discount misalignment as a cause.



Bill, perhaps the cycles are vibratory, not aerodynamic. Explanation as we go along.

But first, here's a sketch with what I think is the bending mode when cycled:

116pfrp.jpg


In this illustration the tip of the elevator has been forced upward in relation to the fixed stabilizer pivot. The resulting load is as shown. The lower rivet is in tension. The spar material below the rivet is unsupported.

Great picture Dan, but I would suggest another theory.

If you regard the hinge pint as fixed, and put a large over travel on the elevator so that the stop gets hit hard -- the elevator spar will become twisted since the stop is at the fuselage and the tip is unrestrained.

A twist of the spar would put the stress at the circle as drawn.

Perhaps these two cracks come from the control surface banging very hard to it's stops, and perhaps only a few times, rather than an ongoing event?
 
I can count on zero fingers the number of lock washers that I've seen on (structure on) big airplanes.

The reason is that they don't work, and can actually cause clamp load to decrease over time, by digging in. if you think about the mechanics of a lock washer, it's only different from a regular washer when the nut is REALLY loose already.


My first impression is that the rodends are fine. IF (big if) there is an issue, it's with the fact that the rodend is really just bearing on the doubler and the jamnut.

Lockheed/Boeing/Gulfstream would (if they were to use the rod end) machine the spar from bar and leave a thick spot where this joint is. They'd probably use a real nut on the back side, for much higher bearing area. I think on the RV, there might be reason to have a doubler on both sides of the spar, especially one that doubles the Web and caps.... IF there's a fundamental issue, and I'm not sure there is.

The roller bearing bracket idea has merit as well. This would let you make a bracket with say 4 3/16 hikes so that you can have 4 AN3'S metal lock nuts, and generate some nice clamp load over a nice large area.

Dan I really like and appreciate your sketches. They're always very helpful for all of us to visualize otherwise complex concepts. In this particular case, though, I think the load path may be a little more complicated.

What is not shown in this sketch is what the actual load path "should" be. The jam nut should be transferring all of the rotational torque load into the spar itself, and then into the skin throug all the spar/skin rivets. In reality, some portion of load will always be picked up by the nutplate into the doubler and only then into the spar - really an indeterminate load path. The idea, I think, is for most of the load to be transferred directly to the spar by sandiching it between the nutplate/doubler on the aft side, and the jam nut on the forward side.

Someone earlier mentioned the idea of a star lockwasher. Another poster thought, in my opinion correctly, that a lockwasher directly on the spar would dig in and possibly exacerbate a cracking issue. I find that I agree with both of these posters. What strikes me as a rather good idea, though, is what Bob suggested: using a washer against the forward side spar, a lockwasher on top of that, and finally the jam nut. Other than adding a little weight and possibly decreasing the flutter frequency, a washer and lockwasher would spread the load over a slightly greater area into the spar.

Of course with literally thousands of aircraft flying per the original design, due diligence and inspections (definitely a torque stripe witness mark) are the easiest, cheapest, and likely best way to prevent problems with the control surfaces.
 
Last edited:
I can count on zero fingers the number of lock washers that I've seen on (structure on) big airplanes.

The reason is that they don't work, and can actually cause clamp load to decrease over time, by digging in. if you think about the mechanics of a lock washer, it's only different from a regular washer when the nut is REALLY loose already.

Don't tell your Lycoming engine or any GA type-certified aircraft manufacturer that lockwashers don't work. They're all over.

I also work in the big aircraft world. You're right, few lockwashers. If our RVs all had two or more turbine engines I agree that we wouldn't be having a lockwasher conversation.

Another thing I see few of in big aircraft is AD material rivets. Only to attach nutplates. Certainly you wouldn't suggest we get rid of those too?

As for "really loose", we've heard multiple reports just within this thread that the jam nuts on several aircraft were just that: really loose. Would lockwashers hurt? I don't see how.
 
Last edited:
Don't tell your Lycoming engine or any GA type-certified aircraft manufacturer that lockwashers don't work. They're all over.

I also work in the big aircraft world. You're right, few lockwashers. If our RVs all had two or more turbine engines I agree that we wouldn't be having a lockwasher conversation.

Another thing I see few of in big aircraft is AD material rivets. Only to attach nutplates. Certainly you wouldn't suggest we get rid of those too?

As for "really loose", we've heard multiple reports just within this thread that the jam nuts on several aircraft were just that: really loose. Would lockwashers hurt? I don't see how.

Might not hurt, I just don't think they help. They do show up on engines, especially on accessories, less so on major parts. Might be the engineers designing for maintainers. Most Mx guys will properly torque rod bolts to spec, but not necessarily vacuum pump nuts.

As far as rivets, the lower wing skin is like 1/2" 7075-T6 at the centerline of a c-130, it takes a lot of 1/8" AD rivets to splice that. Taper-loks would be great on RVs, but you'd spend more on those $250 reamers than you'd spend on all your other tools :)
 
Elevator pushrod bolt

I don't currently have a dog in this fight as I'm a long way from flying, but.....

several here have reported loose nuts on the elevator pushrod bolt and mentioned adding the second bolt. While the second bolt may be a good idea, it doesn't address the loose Jesus nut. If the pushrod bolt comes out, your second bolt isn't going to help.

There have been discussions in the past regarding the nylon lock nuts. My testing indicates these nuts only provide 1 to 2 inch pounds of friction. Should there be any conversation about improved captivation of these nuts?
 
Should there be any conversation about improved captivation of these nuts?

What about a Nord-Lock washer between the nut and spar? Nord-Locks seems to be plenty good enough of a lockwasher for the prop bolts on ground adjustable props.
 
What is not shown in this sketch is what the actual load path "should" be. The jam nut should be transferring all of the rotational torque load into the spar itself, and then into the skin throug all the spar/skin rivets.

Sure, the locknut clamps the nutplate, doubler, and spar together. I can draw a locknut into the sketch, but it won't change the end result. The load path still passes through the circled spar section.

I am assuming it is to scale.

Not precisely. No spar handy at my computer yesterday, so the sketch was scaled from photos and preview drawings. Somebody PM key dimensions (spar height, doubler height, rivet locations) and I'll tidy it up.

I was loosely referring to a (yet unidentified) cyclic load overlaid on a static load, thus it failed only on one side where it is biased in favor of your sketch. (Imagine a sine wave where the "zero load" line is not centered between the peaks)

You're considering stress ratio, as am I.

There has been a lot of discussion about the jamb nut, but it was tight and seems to have no relation to this failure mode.

Agree.
 
Castle nut on Pitts

My pitts has a castle nut on the drilled bolt that connects the elevator horns to the push tube. There are 2 of them and a aluminum block spacer that sets the 2 elevator horns at the exact width for the push tub end bearing to fit between the horns with a washer on both sides of the bearing. The access is a clear inspection plate so you can visually inspect the bolts as well. I try to use tamper paint for all the nuts on everything I have now. I started using it when the U-bolts on my Citabria were loose. I have never seen the tamper paint cracked on any of my applications and it also forces you to go around your aircraft and check every accessible nut and mark it.

I don't currently have a dog in this fight as I'm a long way from flying, but.....

several here have reported loose nuts on the elevator pushrod bolt and mentioned adding the second bolt. While the second bolt may be a good idea, it doesn't address the loose Jesus nut. If the pushrod bolt comes out, your second bolt isn't going to help.

There have been discussions in the past regarding the nylon lock nuts. My testing indicates these nuts only provide 1 to 2 inch pounds of friction. Should there be any conversation about improved captivation of these nuts?
 
Since apparently the cracked spars in question didn't have a bearing mis-alignment problem (still my first guess), it becomes clear (to me at least) that the spar is not strong enough to carry the load and should be either beefed up at the hinge point or replaced altogether with a heaver gauge.

I don't see the "nut" being the problem, even though when loose the lower rivet gets loaded in tension, that's only because the spar is flexing, which again indicates it's too weak. Sorry Van's!
 
My apologies if this has been said already in this long thread...

Regarding misalignment of the bearings, be aware that they are not necessarily aligned just because all of the bolts freely slip in while the surfaces are in the neutral position, for example.

It only means that each rod end is matched with its corresponding attachment bracket. All bolts can freely go in, yet huge forces can be generated when the elevator (or other surface with greater than two bearings) is moved. This might be difficult to detect in moving the surfaces, as it may be a very subtle, and smooth, force feedback.

One way to check for this without complete dis-assembly is to remove one bolt at a time, and verify that the bolt goes in freely at both extremes of surface deflection, as well as at the neutral position.

The bearings need to be in a straight line.
 
14 pages of thread for a issue that so far only seems to be affecting one aircraft out of a fleet of thousands? I think the op is great if it makes us check for cracks, alignment of the bearings, and jamb nut tightness, but is the part of the thread trying to engineer solutions to what may not even be a design problem helpful? I'm not trying to stifle discussion, I just hope new builders don't take from this discussion that they should think about mods for this area.
 
Last edited:
Mine are doing well

It?s time for my condition inspection, this year the tail group got a very close look, nothing loose or moving, no loose lock nuts, no cracked spars, elevators or rudder, no cracks of any kind anywhere.

What I did note is that my rod end bearings are all adjusted such that the elevators and the rudder are all as close to the stabilizers as possible, barley more than enough room for the lock nut between the spar and the end of the threads on the rod end bearing, about one thread left showing, this should make these assemblies as strong as they possibly can be, most ideal condition. There is also no binding of any kind, elevators and rudder move very freely.

Early 80s RV-4 with 2708 hours.
 
Since apparently the cracked spars in question didn't have a bearing mis-alignment problem (still my first guess), it becomes clear (to me at least) that the spar is not strong enough to carry the load and should be either beefed up at the hinge point or replaced altogether with a heaver gauge.

I don't see the "nut" being the problem, even though when loose the lower rivet gets loaded in tension, that's only because the spar is flexing, which again indicates it's too weak. Sorry Van's!

Be careful with a statement like this. In Mr. Burn's case, there is very clear evidence that the assembly was exposed to a significant overload at some point, as evidenced by the crease on the spar at the end of the doubler.

Simply beefing up components is not necessarily a good way to handle failure of aircraft parts when the cause of the failure is unknown.
 
Be careful with a statement like this. In Mr. Burn's case, there is very clear evidence that the assembly was exposed to a significant overload at some point, as evidenced by the crease on the spar at the end of the doubler.

Simply beefing up components is not necessarily a good way to handle failure of aircraft parts when the cause of the failure is unknown.

Keep in mind that we are talking about a failure of a Primary Structural Element. Be it a wing spar or a elevator spar, cracks, deformation of any sort is a MAJOR failure. The fact that there are 2 aircraft with very similar cracks is definitely cause for speculation and concern. We do know the cause of the failure...the deflection of the spar common to the rod-end. What caused that deflection is unknown and may remain unknown. Whether it be assemble error, excessive aerodynamic force or under engineering of unanticipated normal loads...you can be sure, if I find mine cracked I am not going to put a new stock spar back in it's place. The stock spar material is not sufficiently handling the structural load. It is NOT uncommon for a manufacturer to "beef" up parts of the aircraft that time has determined that the original design was insufficient.
 
Last edited:
Food for thought....

With regard to all of the "not strong enough" and "needs to be beefed up" comments....

Keep in mind that pretty much this exact same design was used on all RV's.

There is now over 8400 known to have flown. Thousands of those have thinner skins than an RV-7 elevator (.016 vs .020)

If there was a general issue with the design, it would have become evident many years ago. At the very least, on RV-8's. They use the exact same (prepunched) elevator.

I said it once, but I guess many don't believe... this type of a crack does not occur from one (or even several) overload/bend events localized at the crack. It is many (probably many thousands, possibly even many ten thousands) load cycles. The damage from a single overload in a location like this would be a visible crease or bend. Not an original flat surface with a (fatigue) crack.

Deductive reasoning says that something has been causing a localized load cycle on this (maybe these two) airplane, that does not occur on thousands of others that have much more flight hours on them.

The challenge is discovering what. Once we know that, a solution is right behind.

My opinion is that we haven't yet heard all the details. I am not saying anyone is withholding info... It is often something that is missed, and not considered significant because of lacking experience in evaluating this type of problem.

It is entirely possible that it is a combination of factors, that on these two airplanes, they just happened to have them all, but the majority of the fleet does not.

I still am of the opinion that it is some type of alignment issue. The photo evidence of the lateral edge of the doubler working on the face of the spar web could be an indicator that there is a lateral alignment issue at the hinge point.

If I was the investigator in charge would want to see photos of all three hinge points showing a thread pulled through them, and dead center (or nearly so) in the hinge bracket holes, and the same done with the hinge points on the elevator (if there is misalignment, it could be on either one, or both); and a video showing the damaged elevators, two rod end hinge points slipping into the stab. hinge brackets without any prying or unusual force required. ( impossible to do now, since the elevator has been disassembled)
 
Last edited:
Since apparently the cracked spars in question didn't have a bearing mis-alignment problem (still my first guess), it becomes clear (to me at least) that the spar is not strong enough to carry the load and should be either beefed up at the hinge point or replaced altogether with a heaver gauge...

Clearly there is something here that calls for investigation. But to say that the structure is "not strong enough" is too likely an oversimplification that could cause more problems than it solves. Throwing more metal at a crack is likely to just move the cracks to somewhere worse.

In my experience, when something cracks it is far more constructive to say that there is something wrong with the distribution of stiffness in the structure. As much as we might wish otherwise, stresses follow the path of greatest stiffness, not the path of greatest strength. And often (though probably not in this case), the solution is to make part of the structure lighter and therefore less stiff.

Thanks, Bob K.
 
My apologies if this has been said already in this long thread...

Regarding misalignment of the bearings, be aware that they are not necessarily aligned just because all of the bolts freely slip in while the surfaces are in the neutral position, for example.

It only means that each rod end is matched with its corresponding attachment bracket. All bolts can freely go in, yet huge forces can be generated when the elevator (or other surface with greater than two bearings) is moved. This might be difficult to detect in moving the surfaces, as it may be a very subtle, and smooth, force feedback.

One way to check for this without complete dis-assembly is to remove one bolt at a time, and verify that the bolt goes in freely at both extremes of surface deflection, as well as at the neutral position.

The bearings need to be in a straight line.

Exactly

The only way they can be validated to be in alignment is with reference to a straight line (taught thread, laser, etc.), and checked individually.

Actually, if one can be checked with a high level of accuracy (the horizontal stab is the easiest), then the other can be considered aligned if all three bolts insert easily without any pushing or prodding on the elevator.
 
I removed the doubler from the rod end attach point of the spar tonight.
I was shocked at what I saw!

Look at the crease line caused by the edge of the doubler on right side of photo. It's really worse than the photo shows. Sighting down the spar, it's still very straight.

20130925_211431.jpg


20130925_212006.jpg




Thoughts?

Mark

As a follow up to my previous post, and after looking at you photos in more detail, I see what could potentially be fretting at the hard contact points between the spar and doubler. Each of the rivet holes has the appearance of fretting that can be caused by movement. The thing you refer to as a crease line, could be caused by the associated edge of the doubler moving against the spar web very slightly.
Was there any indication of smoking/loose rivets?
The repeated loads that would be applied by the constant hinging action of the elevator, with a misaligned hinge point could cause this.
 
If there has only been one crack found does this seem a with overall strength or distribution of stiffness in the structure? I would bet a problem with assembly or maintenance. The loose rudder jamb nuts is a design deficiency, one that a Van's employee told me needs to be inspected every preflight.
This thread has got a lot of people looking at their empennage, has only one crack been found?
 
If there has only been one crack found does this seem a with overall strength or distribution of stiffness in the structure? I would bet a problem with assembly or maintenance. The loose rudder jamb nuts is a design deficiency, one that a Van's employee told me needs to be inspected every preflight.
This thread has got a lot of people looking at their empennage, has only one crack been found?

I believe the count on this thread is two crescent shaped cracks found, and lots of loose jam nuts.

However at least one of the cracked spars had tight jam nuts.

If this isn't correct, could someone update?

-----------------------------------------------------------


OK. How about a more radical theory?

Perhaps a loose jam nut is actually a safety enhancer?

The jam nut just stiffens the mounting of the hinge rod end in the self-locking nutplate attached to the elevator spar. The rod end can't actually back out due to the flanges on the horizontal spar.

If the original construction has a small misalignment built in, then a loose jam nut would allow the rod end to "wiggle" slightly in it's nutplate mount. This could actually reduce the cyclic loading on the doubler as the elevator is moved.

Perhaps these loose jam nuts are compensating for a building misalignment? Especially if it's only on one of the three hinge points...
 
Last edited:
Scott (RVbuilder2002),
There was evidence of loose rivets. But only the two that hold the nutplate.
See the second photo in post #1.
See the black ring around the rivet? The top rivet looked almost the same.
They were not loose to the point that you could feel or see anything. Just the evidence of the black ring.
I don't believe what you see around the rivet holes is fretting.
I believe it's just marks that are a result the rivet being set.

Mark
 
I believe the count on this thread is two crescent shaped cracks found, and lots of loose jam nuts.

However at least one of the cracked spars had tight jam nuts.

If this isn't correct, could someone update?

Gil,
To my knowledge this is correct.

Mark
 
Clearly there is something here that calls for investigation. But to say that the structure is "not strong enough" is too likely an oversimplification that could cause more problems than it solves. Throwing more metal at a crack is likely to just move the cracks to somewhere worse..

What can be worse than having a cracked spar?
 
Exactly

The only way they can be validated to be in alignment is with reference to a straight line (taught thread, laser, etc.), and checked individually.

Actually, if one can be checked with a high level of accuracy (the horizontal stab is the easiest), then the other can be considered aligned if all three bolts insert easily without any pushing or prodding on the elevator.

Correct me if I'm wrong but it would appear to me that drilling the center bearing holes is the critical step. Misdrilling these holes may cause a vertical misalignment of the attach points that cannot be adjusted out and will cause the spar to flex with each control cycle.
 
Last edited:
Back
Top