What's new
Van's Air Force

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

RV-8 Elevator Failure

Ron RV8

Well Known Member
I am posting this event on behalf of a friend who does not wish to enter
the conversation directly. It is however a potentially catastrophic
safety issue which should be brought to the attention of the RV
community.

The aircraft, an RV-8 was not built by the current owner? Because he is
uncertain of the extent of possible hidden damage he is replacing the
entire empennage.

ISSUE:

* Torsional failure of an unbalanced elevator through flutter or metal
fatigue (undetermined).

RECOMMENDATIONS:

* If you did not personally balance your elevators (or trust the person
who did), it is suggested to check the balance.
* When doing so, it is suggested to disconnect the elevator pushrod to
check the balance of each elevator independently.

DETAILS:

* Aircraft - RV-8
* Total Time - 580hr
* Conditions - Moderate Turbulance
* Duration of incident flight - 32min
* Airspeed well under Vne due to turbulence.

THE INCIDENT:

My friend was returning from a fly-in under turbulent conditions. He
noticed the autopilot release following which the aircraft started to
climb. He reset the trim, reset the autopilot and continued to home
base, about a 20 minute flight. He did not notice any changes to the way the
aircraft handled.

While putting the aircraft away he noticed some cracks in the right
elevator which he did not notice during the pre-flight inspection. On
further investigation he discovered that the elevator had lost torsional
rigidity and was well on it?s way to complete failure.

He found the right elevator was overbalanced to the front by almost 9
oz. It had the old style weight installed which looks like a rudder weight.

When the video was made, the left elevator did not appear to be cracked and already had the trim tab and motor removed and installed on the new one. As a result, the left elevator was not checked for an accurate balance but appeared to the owner to be slightly underweighted when removed from the aircraft.

It is suggested that it may be prudent to advise owners that have not
built their aircraft of this potential failure mode. Since the state of
balance of the other elevator can mask an imbalance situation it is
suggested that the elevators be disconnected from each other to check
the balance.

Scale shows 8.7 oz leading edge heavy



Pens point to cracks









VIDEO: https://www.youtube.com/watch?v=ec-eWnCr7cQ
 
First, thanks for posting.
How do you know the imbalance caused this?
I hope they quit beating it up and zip the skin off to see what the structure looks like underneath. There may be construction errors or a material defect.
 
The video shows a hint of what may be a damaged (cracked/kinked) trailing edge in the area where the torsional flexing is occurring.
If the trailing edge of the elevator is compromised, it would have a negative impact on torsional rigidity.

Is he sure that it didn't get damaged in some way while at the fly-in?

If nothing similar to a flutter event was ever noticed (If it happened he likely would notice), I think it is unlikely this was related to improper balance of the elevators.

BTW, you wrote "failure of an unbalanced elevator". Did you mean it had no weights on it, or did you mean not properly balanced?
 
Last edited:
seems like a significant finding. I doubt there is any test data for a 9 oz nose heavy elevator in turbulent conditions. thanks for posting.
 
The cause of the failure is unknown

There is no way to know if there was damage at the fly-in. I will ask the owner, (he lives in another city), to look at the crack at the trailing edge to see if there is obvious damage that might have led to the crack.

The elevator had a balance weight on it that was ~9 oz too heavy. It would seem that it might provide torsional flexing in rough air and is suggested as a possible factor. It is likely that some degree of metal fatigue may have occurred before the subject flight. We don''t know...

I think that if it had fluttered it might be gone, possibly along with part of the tail. The concern is that without rigidity it would seem more prone to flutter and he considers himself lucky to have landed in one piece..

The elevator is being preserved intact in case someone with more knowledge in evaluating structural failures wishes to have a look at it...
 
Those weights are about 820 grams each. Each side takes two on each elevator on the 7. 9 oz is 255 grams, so 16% off two of the weights. If the 8 is similar then, there is no way the weights are missing on the left side.

Is there a way to share the other "blanked" photos?
 
The crack in the trailing edge is a PROBLEM. Since it was not built by the pilot, I would be interested in knowing what the construction looks like inside.

Having built RV-6 elevators over 20-years ago and RV-8 elevators last year, the design is similar and share a lot of the same parts. This is the first report that I have seen of a problem. There may have been other problems but over 20-years of success says a lot about the design.

I can say that the new weights that are used on the RV-8 elevators will balance both elevators. IF like you say to balance them individually, the excess weight off the right elevator transferred to the left elevator with the trim servo will balance the two out as individual assemblies. That is what I did with the RV-8 elevators recently built and balanced.
 
Somebody sit on it?

From the the way it twisted so easily in the video makes me wonder it someone didn't sit on the outboard edge of the elevator while it was tied down. This comes to mind because that much damage from flutter would likely be felt through the stick.
 
Lots of things running through my mind, the first one is that the autopilot disconnected and it started climbing, how aggressive was the climb and is it possible that the aircraft was significantly out of trim when the turbulence was entered. When on long trips in my 7 with the autopilot on I will turn it off about every 30 minutes just to make sure the EFIS is not lying to me when it seeks trim input. Cracks along the end of the internal stiffener is also a little curious, we have seen this in rudders. Flutter, not an expert at all but what I have read does not bring it to to top of my list. Balancing the Elevators, I don't remember what the plans say so if someone can provide that info to this thread it would be helpful.
 
When you say the scale shows the leading edge is 8.7 oz heavy, how is the elevator supported? Do you have it supported at the hinge bearings? How was that done? So it is free to pivot, and then you put a scale under the counterweight, as shown in the picture?

It is hard for me to imagine there is enough room in the countereweight area to get enough extra lead to overbalance that much. The proper counterweight pretty much fills up the available space.
 
When you say the scale shows the leading edge is 8.7 oz heavy, how is the elevator supported? Do you have it supported at the hinge bearings? How was that done? So it is free to pivot, and then you put a scale under the counterweight, as shown in the picture?

It is hard for me to imagine there is enough room in the countereweight area to get enough extra lead to overbalance that much. The proper counterweight pretty much fills up the available space.

Yes, lead would have to be added, there is room in the tip. And, it is at max radius so would be the best location.
 
From the video, I am trying to figure out what failed to allow it to rack so much. Did the connection to the spar tear so the ribs are not connected to the spar? Maybe with a mirror or camera, a weak spot could be found. I checked my elevators and they have zero rack. I can only imagine the force it took to damage it that bad. From flutter videos I have seen, there is plenty there to do the damage. I HOPE that if someone put that much force on the elevator at the show that they would let someone know.

I would love to see pictures with the skin off of the spar and ribs.
 
I think, based on a friend's experience with aileron flutter, that if there was flutter there would be absolutely no doubt about it. One doesn't forget that.
 
I think, based on a friend's experience with aileron flutter, that if there was flutter there would be absolutely no doubt about it. One doesn't forget that.

I experienced aileron flutter on a Moni Motorglider that ripped up wing skins and actually broke the rear wing spar in-two at the fuselage.
I promise you, when you experience flutter, there is no question that it happened. Witnesses on the ground estimated the wing tip was moving at least 2' vertically.
I was VERY fortunate to get the aircraft slowed down enough to stop the flutter and got it safely on the ground.
 
Last edited:
Comment from owner

"There was no damage prior to the flight. I was stopped for ten minutes
total at Unity where I never left the airplane. The elevator trailing
edge was flawless up until this point. The auto pilot does have an auto
trim so it wasn't an out of trim issue either when the plane pitched up.
The autopilot and the autotrim has worked spot on up until this incident
meaning that when the autopilot is disengaged it is in perfect trim. It
certainly is an unexplained cause of destruction to the elevator that I
can not explain and it just happened now at this point in the airplane
history. According to plans the elevators are supposed to be
individually balanced and these ones clearly were not. In my opinion this
is builders error and the smoking gun for this incident and something that
should be not taken for granted. I to agree that the design is good but
certainly rattled at how fast good can go bad."

Posted by Ron Townson who is not the owner...
 
Last edited:
What were the Max/Min G-Meter readings? Any EFIS data recorded? Autopilot servo torque limit? Reported to NTSB? (or Canadian equivalent)

Thanks,
Paige
 
Last edited:


When you say the scale shows the leading edge is 8.7 oz heavy, how is the elevator supported? Do you have it supported at the hinge bearings? How was that done? So it is free to pivot, and then you put a scale under the counterweight, as shown in the picture?

There's the winning question. How about it Ron...was the above photo taken with the elevator pivoting on its hinge bearings?
 
The pictures and text would indicate that we are looking at the under side of the elevator, generally out of typical, easy visual inspection.

The two cracks look like classic prop pulse induced fatigue cracking, certainly not from a single or low cycle damage. (Classic meaning it was not uncommon on .016" elevator and rudder skins of 6's and 4's.)

The kink seems to have been caused by a single overload, such as would happen if someone leaned on the one end of the elevator while it was locked at the other end via control stick or other means. Or, outboard locks holding the elevators, but someone hauling forward or back on the stick. It is very hard to believe this kink happened in flight. If it had, whatever started it would have likely not stopped there...

The two failure modes may be unrelated.

It might be good to check the integrity of the seam between the top and bottom skins, where they curl together. This seam is pop-riveted, but it would seem to be an important contributor to torsional rigidity.

Regardless, we're glad you caught this!
 
Are the rod end bearing hinge lock nuts tight? any play here that could have excited flutter?
 
Are the rod end bearing hinge lock nuts tight? any play here that could have excited flutter?

As already mentioned previously in this thread, if this damage had been caused by flutter, the pilot flying the airplane would have been well aware of it occurring (and I assume it would have been mentioned).
 
The pictures and text would indicate that we are looking at the under side of the elevator, generally out of typical, easy visual inspection.

The two cracks look like classic prop pulse induced fatigue cracking, certainly not from a single or low cycle damage. (Classic meaning it was not uncommon on .016" elevator and rudder skins of 6's and 4's.)

The kink seems to have been caused by a single overload, such as would happen if someone leaned on the one end of the elevator while it was locked at the other end via control stick or other means. Or, outboard locks holding the elevators, but someone hauling forward or back on the stick. It is very hard to believe this kink happened in flight. If it had, whatever started it would have likely not stopped there...

The two failure modes may be unrelated.

It might be good to check the integrity of the seam between the top and bottom skins, where they curl together. This seam is pop-riveted, but it would seem to be an important contributor to torsional rigidity.

Regardless, we're glad you caught this!

I agree, but since I have not physically inspected it, it is still just one more guess.

What I do feel totally confident in saying is that this having been caused by a counter balance weight being 8.7 oz too heavy is highly unlikely.
I am not saying this to judge the person who owns the airplane.... but I do hope it doesn't start panic within the RV community.

Even at 8.7 oz over balanced, the vertical acceleration/gusts the airplane would have had to experience during the mentioned turbulence, to cause damage to the elevator, would likely have caused failure of something else way before the elevator would have.
 
I agree, but since I have not physically inspected it, it is still just one more guess.

What I do feel totally confident in saying is that this having been caused by a counter balance weight being 8.7 oz too heavy is highly unlikely.
I am not saying this to judge the person who owns the airplane.... but I do hope it doesn't start panic within the RV community.

Even at 8.7 oz over balanced, the vertical acceleration/gusts the airplane would have had to experience during the mentioned turbulence, to cause damage to the elevator, would likely have caused failure of something else way before the elevator would have.

I am glad you said this - +1. Something to be understood for sure.
 
No one has claimed an extra half pound of mass balance is The Cause. The question asked was how the half pound was measured, i.e. is it really over-balanced by that much?

Nobody has enough information, yet.
 
No one has claimed an extra half pound of mass balance is The Cause. The question asked was how the half pound was measured, i.e. is it really over-balanced by that much?

Nobody has enough information, yet.

Unless I interpreted it wrong, in his first post the OP was implying that the owner thought the damage was caused by the over balance of the elevator.
(note he wrote "unbalanced elevator", but I assume he meant not properly balanced)

I am posting this event on behalf of a friend who does not wish to enter
the conversation directly. It is however a potentially catastrophic
safety issue which should be brought to the attention of the RV
community.

The aircraft, an RV-8 was not built by the current owner? Because he is
uncertain of the extent of possible hidden damage he is replacing the
entire empennage.

ISSUE:

* Torsional failure of an unbalanced elevator through flutter or metal
fatigue (undetermined).

RECOMMENDATIONS:

* If you did not personally balance your elevators (or trust the person
who did), it is suggested to check the balance.
* When doing so, it is suggested to disconnect the elevator pushrod to
check the balance of each elevator independently.

DETAILS:

* Aircraft - RV-8
* Total Time - 580hr
* Conditions - Moderate Turbulance
* Duration of incident flight - 32min
* Airspeed well under Vne due to turbulence.

THE INCIDENT:

My friend was returning from a fly-in under turbulent conditions. He
noticed the autopilot release following which the aircraft started to
climb. He reset the trim, reset the autopilot and continued to home
base, about a 20 minute flight. He did not notice any changes to the way the
aircraft handled.

While putting the aircraft away he noticed some cracks in the right
elevator which he did not notice during the pre-flight inspection. On
further investigation he discovered that the elevator had lost torsional
rigidity and was well on it?s way to complete failure.

He found the right elevator was overbalanced to the front by almost 9
oz. It had the old style weight installed which looks like a rudder weight.

It is suggested that it may be prudent to advise owners that have not
built their aircraft of this potential failure mode. Since the state of
balance of the other elevator can mask an imbalance situation it is
suggested that the elevators be disconnected from each other to check
the balance.


Since the only problem discovered/mentioned was the unbalance of one elevator, to me it was being implied that this was the cause......
 
As the owner and pilot of this aircraft I will answer some questions. I do know that the elevator was in good condition prior to flight. The turbulance experienced that day would have been uncomfortable with someone not used to flying but it wasn't more than I or my ship could handle. The airplane itself never experienced more than a couple g on the occurance flight and the speed was considerably less than cruise speed to penetrate the turbulence. I did feel something not right when the autopilot struggled to maintain pitch on the aircraft, and once it was disengaged the plane flew home just fine, however nothing would have made me even consider flutter or buzz at that time with the speed lower than typical cruise. The ONLY facts I do know is that I left with a straight undamaged and uncracked elevator, and returned with a badly damaged and fatigued elevator. Somewhere along the flight the elevator received lots of damage and i am entirely certain it never happened on the ground. All jam nuts and hinge points were still tight after the flight. The 8.7 ounces of extra weight on the elevator may have or may have not been a contributing factor but with the extra weight this contradicts Vans plans for having balanced elevators. The weight was taken with the elevator balancing on its hinge points. I will try to answer any questions either private or here but this is really all I know. I will try to post another video of the elevator before it was removed from the aircraft and let fall from neutral position once the bolt was removed connecting both elevators. Thank you.

https://youtu.be/y2C9qtbquUA
 
Last edited:
Weighing method

Thanks to Jarvis (the owner) for entering the discussion as it is much better to have first hand answers to the questions this has raised. My apologies to the group for not responding more promptly to some questions but I currently have a house full of company and was hoping that Jarvis would join the fray...

The question has been raised wrt how the elevator was supported while weighing the degree of imbalance. The objective was to get the event out to the RV family quickly in order to potentially save a life so we did not build a jig to accurately measure the exact imbalance. We opted instead to use "fingers" on the bottom of the hinges under the pivots. This definitely introduced the potential for error so the person holding the elevator purposely rolled his fingers back and forth in order to observe the degree of error that might be introduced. It was found that the maximum "finger" error was in the order of a few tenths of an ounce, insignificant when compared to the total imbalance in excess of a half pound.



Please be aware that this is not intended as a criticism of Van's design, and it is suspected that building error is probably a contributor. We do however want to make people aware that failure of this nature is possible and that a check of elevators constructed by others may be prudent.

It would be nice if someone with the skills necessary to evaluate structural failures (perhaps Vans?) would do a post-mortem on the elevator and advise all of us what caused the event... If in fact it can be traced to hidden damage it would be valuable to know that. To the best of my knowledge the elevator is still intact and available for analysis...
 
Since the only problem discovered/mentioned was the unbalance of one elevator, to me it was being implied that this was the cause......

Fair point.

I did feel something not right when the autopilot struggled to maintain pitch on the aircraft...

Which makes sense, given the elevator's demonstrated flexibility on the bench.

The ONLY facts I do know is that I left with a straight undamaged and uncracked elevator, and returned with a badly damaged and fatigued elevator. Somewhere along the flight the elevator received lots of damage and i am entirely certain it never happened on the ground.

Jarvis, if you're absolutely sure the trailing edge was undamaged prior to boarding the aircraft, the cause must be something else. Note my use of "prior to boarding", as we should note the possibility of damage during ground operations. Such damage could be (for example only) a fair size rock kicked up by a main tire, or running over an object sticking up...a stake, or light, or a Yak 52 boarding ladder (don't laugh, there was a time when a Yak ladder was the most common FOD item seen at 08A). Any such possibility at the points of departure and landing?
 
I will try to reiterate... With ABSOLUTE CERTAINTY there was no damage prior to flight. No one touched my airplane as I was there the entire time. I remember doing a walk around before leaving there and before leaving home base .5 hour prior and this was so evident that there is no possibility that this damage would not have been noticed. Also no dents or scratches that indicate foreign debris contacting elevators, HS or Wings which are more in line for FD than the elevators. And I never mowed over any runway lights either...:eek:
 
Glad you are safe

Good thing your elevator wasn't zippered up. That folded trailing edge may have been just the margin you needed to bring it home safely. Otherwise, we would be wondering why that elevator was located so far away from the main crash site.

I will try to reiterate... With ABSOLUTE CERTAINTY there was no damage prior to flight. No one touched my airplane as I was there the entire time. I remember doing a walk around before leaving there and before leaving home base .5 hour prior and this was so evident that there is no possibility that this damage would not have been noticed. Also no dents or scratches that indicate foreign debris contacting elevators, HS or Wings which are more in line for FD than the elevators. And I never mowed over any runway lights either...:eek:
 
If the root cause of this was the cyclical nature of moderate turbulence causing vertical oscillations opposing the excess weight of the counterbalance, that force would likely have had to been translated to the control pushrod and since it was on autopilot, the torque variable might have captured this.

What AP do you have and does it have the ability to trend the torque being produced by the servo? Mine logs vertical acceleration and torque among lots of other things.

I would love to see the inside of that elevator just to make sure there was not something else going on in there to contribute to this event.

This is the first time I have ever seen anything even close to this reported on a RV. Have seen a few cracks in the skin aft of the stiffener but nothing like this...

Thanks very much for sharing and like others, the questions are in an effort to find root cause not to question your understanding of the condition of the elevator prior to flight.
 
Last edited:
Perhaps the owner could send the elevator to Van's for inspection. And perhaps Van's could comment about their findings here on VAF.

Dave
 
I will try to reiterate... With ABSOLUTE CERTAINTY there was no damage prior to flight that was visible to the operator.

There - fixed that for you. There seems to be some question about the internals, but no question about visibility of damage externally. You could easily have had internal damage to the rib structure that was not identifiable without manually applying torsion to the elevator - and lets be honest here, how many of us would do that on a pre-flight?
 
Lets be nice guys, the objective here is to figure out what happened and perhaps avoid someone having an accident...
 
I will try to reiterate... With ABSOLUTE CERTAINTY there was no damage prior to flight.

And I (among many) absolutely believe you. Just working down the list of possibilities, a process of elimination. There are other useful questions, but we're not going further unless you wish it.
 
A couple things to note:
The location of the crack is typical of many I have seen, through the last rivet of the stiffener. Although debated for years, the current theory, and most palpable, is improperly bent trailing edge causing preloading of the skin.
You can quickly determine if this preload is there by laying a straight edge from the spar to the trailing edge. The skin should be perfectly straight all the way to the trailing edge radius. Any deviation could mean a preload is present.
While this type of crack was more prominent in the .016 skins, .020 skins are not immune.
I would be curious to see if this particular elevators trailing edge was adequately, and properly bent. I would also be curious to know if the instructions where followed in regard to stiffener length and distance to the trailing edge and if the builder secured the stiffeners per plans with a blob of RTV.
 
I would also be curious to know if the instructions where followed in regard to stiffener length and distance to the trailing edge and if the builder secured the stiffeners per plans with a blob of RTV.

I only have access to my 6 plans. If indeed these cracks are through the last rivet, and the call out for the distance is the same as the 6, which is 3/4" from the trailing edge, the rivets shown are nowhere close to that,
judging from the straight edge laid next to the largest crack. Most straight edges are 1" wide.
If this is indeed what I am seeing....
Please verify.
 
I only have access to my 6 plans. If indeed these cracks are through the last rivet, and the call out for the distance is the same as the 6, which is 3/4" from the trailing edge, the rivets shown are nowhere close to that,
judging from the straight edge laid next to the largest crack. Most straight edges are 1" wide.
If this is indeed what I am seeing....
Please verify.

You can see the markings on the ruler. Looks like the crack is 1" from the edge.
 
You can see the markings on the ruler. Looks like the crack is 1" from the edge.

Jumped to a bigger screen. I see that now. Thanks Bob. It appears the crack develops right through the aft rivet location.
Do the 8 plans call out 1" spacing from the edge or 3/4" ?

The more things we can elliminate as a potential cause, the closer we can be to an answer.

I too have never seen a failure like this, even in elevators with several cracks and stop drilled. Some continued to fly for a long time with no problem.
 

Yes, I have several. Already explained I could not see on the small screen, that is why I asked if others saw what I did.

Still, if anybody knows what the trailing edge to first rivet spacing is, that might add a clue, but I doubt it is the smoking gun. However, you add that to another build error, like not tying the stiffeners together at the trailing edge, and/or building a preload into the skin by an improper trailing edge bend, you could have a much weaker assembly, prone to cracking.
I am NOT saying any of this is the case with this particular airplane, or if it is, or contributed to, the cause of the failure. That is way above my pay grade.

Sticking with the theme of the post, folks should know how to inspect their elevators for proper construction if they did not build them.
 
I've read through this a couple times now but still not clear on a couple of things.

How long into the flight did you notice the autopilot was having the pitch problem? And I'm assuming you noticed it after encountering some turbulence, yes?

Thanks for sharing.
 
I've read through this a couple times now but still not clear on a couple of things.

How long into the flight did you notice the autopilot was having the pitch problem? And I'm assuming you noticed it after encountering some turbulence, yes?

Thanks for sharing.

Somewhere around halfway through the flight or 15 ish minutes in. Yes the entire flight was bumpy.
 
If the root cause of this was the cyclical nature of moderate turbulence causing vertical oscillations opposing the excess weight of the counterbalance, that force would likely have had to been translated to the control pushrod and since it was on autopilot, the torque variable might have captured this.

What AP do you have and does it have the ability to trend the torque being produced by the servo? Mine logs vertical acceleration and torque among lots of other things.

I would love to see the inside of that elevator just to make sure there was not something else going on in there to contribute to this event.

This is the first time I have ever seen anything even close to this reported on a RV. Have seen a few cracks in the skin aft of the stiffener but nothing like this...

Thanks very much for sharing and like others, the questions are in an effort to find root cause not to question your understanding of the condition of the elevator prior to flight.

The autopilot is a trutrak controlled by a Vision 385. I am not aware that it logs any torque data but it does have a very strong clutch that takes a bit of grunt to overpower.

I too have seen these mentioned stress cracks. I have one on the other elevator that I have been monitoring for over a year and it has not gotten any larger over the course of that time. It will not have a chance to now as it is getting changed out too. I haven't heard of these cracks ever appearing all at once and usually develops over a course of time. The effected elevator never had any noticeable cracking anywhere prior to incident.

I cannot get pictures of the elevator insides at the moment as I am out of province at work.
 
A couple things to note:
The location of the crack is typical of many I have seen, through the last rivet of the stiffener. Although debated for years, the current theory, and most palpable, is improperly bent trailing edge causing preloading of the skin.
You can quickly determine if this preload is there by laying a straight edge from the spar to the trailing edge. The skin should be perfectly straight all the way to the trailing edge radius. Any deviation could mean a preload is present.
While this type of crack was more prominent in the .016 skins, .020 skins are not immune.
I would be curious to see if this particular elevators trailing edge was adequately, and properly bent. I would also be curious to know if the instructions where followed in regard to stiffener length and distance to the trailing edge and if the builder secured the stiffeners per plans with a blob of RTV.

The elevator shows no evidence of having to much preload as the skin is flat up to the radius of the trailing edge. Have not opened up elevator yet so stiffener length is unknown. The RTV Silicone I would say was not installed in this elevator and I am not sure it was in plans back in early 2000.
 
Back
Top