Interesting video on rudder bracket failure

[Scott Todd]

Peter, Since you agree with Kerry that the forces are pushing the horn down, could you draw a simple Free Body Diagram to show it? Anyone?

If its pushing down when mounted on top, then its also pulling down thru the bold head when mounted on bottom. The forces should be similar.

I can see a twisting moment imparted by the original push rod but I think the distance from the top to the bottom of the horn is negligible compared to the forces.

Scott

[gammaxy]

Scott, I agree. I've tried my hand at a free body diagram and agree that top is worse, but not so much that bottom mounted seems obviously immune given the wide ranges of forces seen on landing.

For my analysis, the force is generated when steering the tailwheel. Steering left applies a force that tends to pull the bracket off. It's a 3D problem, so it also tends to pull off from the left side where the pushrod is attached. The shop head of the rivets are against the thinner metal where they can work the holes open over time.

It seems to make ~5 degrees difference (top vs bottom) if using the rod end bearings many people use. I calculate about a 30% increase in downward force mounting on top instead of the bottom.

But, when you compare this to the plans design without rod end bearings which pivots more flush, the angle is only about ~2 degree difference versus the worst case. This results in a ~15% increase in downward force when mounted on top.

[Rynoth]

Regarding the original video, it's really weird to me that the bucked-end of those rivets are all clean. I would have thought it likely that they would have taken some material with them when they separated. I'd be really curious to see a pic/video looking up into the bottom of the rudder with the parts separated. It almost seems like they never actually engaged the rib. Perhaps the rivet-holes in the rib were too large?

[peter anson]

Peter, Since you agree with Kerry that the forces are pushing the horn down, could you draw a simple Free Body Diagram to show it? Anyone?

If its pushing down when mounted on top, then its also pulling down thru the bold head when mounted on bottom. The forces should be similar.

I can see a twisting moment imparted by the original push rod but I think the distance from the top to the bottom of the horn is negligible compared to the forces.

Scott

I'll go along with Gammaxy's calculation. I took drawing number SNX-L03 and drew in the effective link direction with (A) the standard Sonex link (B) a rod ended link to the bottom of the rudder horn and (C) a rod ended link to the top of the rudder horn. According to my calculations, compared to the standard steering link, the rod end link to the bottom of the rudder horn reduces the vertical component of the forces by 20% and the rod ended link to the top of the rudder horn increases the vertical component by 15% which might be a fluke but happens to agree with Gammaxy.

I guess Kerry would argue that 15% is a significant increase but the fact is that rivets are intended primarily as shear fasteners and in this case they are having to carry a tension load. I'm almost a little surprised that we haven't seen a few more instances of this happening. In an earlier post someone suggested that the holes might have been drilled a bit oversize. Could maybe avoid the possibility of this occurring by installing rivets from the top or fitting a few threaded fasteners.

Peter

[Rynoth]

Regarding the downward pressure on the drive horn, and having no idea how to do a free-body diagram, what strikes me is the angle of the connecting rod between the drive horn and the tailwheel. It might be less about whether the connecting rod was mounted on the top or bottom of the drive horn, and more about the angle of the rod where it makes its vertical drop.

I whipped up a crude (and exaggerated) visual of the difference.



At least in my Waiex, which is setup stock and doesn't use rod end bearings (it's basically a plate at each and with a welded connecting rod in between), it looks more like the second example. The descent of the connecting rod begins and ends very near the attach points.

In the top figure, I tried to illustrate the impact of not only mounting on the top, but also the rod-end bearings being used that effectively extend the horizontal lengths of the rod, resulting in a much steeper angle when the connecting rod descends to the tailwheel. Image is exaggerated for illustration of the difference.

It seems to me that the top image would have a much larger affect of vertical forces on the drive horn, further amplified by the lever action of the longer horizontal sections introduced by the use of rod end bearings.



And, just to take this concept to its extreme, here is an example of a bottom-mounted setup that is probably far worse in terms of vertical pressure on the drive horn than either of the examples above.

[gammaxy]

Rynoth, I treat the rod as a "two-force member". This means that forces (no moments) are applied at exactly two locations on the rod. The rod end bearings seem to satisfy this as long as they are installed so they don't bind at either end (if they did, you'd get a moment). The plans-built rod without end bearings is less obvious, but I believe it is typically installed loose enough to rotate without applying a significant moment.

https://eng.libretexts.org/Bookshelves/Mechanical_Engineering/Mechanics_Map_(Moore_et_al.)/05%3A_Engineering_Structures/5.02%3A_Two-Force_Members

The key takeaway is the shape of a two-force member doesn't matter for the purpose of calculating forces at either end. Obviously, it has structural implications for the rod, but I think it can be ignored for this analysis as long as both ends are free to rotate.

[Rynoth]

Interesting Chris, thank you for that explanation. I can see how the issue could be isolated to the mounting location(s) using that analysis, provided there is no binding at the attach points.

[mccool]

Notwithstanding that a spin when in that situation would ruin your day, if I had to lose 1 of my primary controls, the rudder would be my 1st choice.

[dbdevkc]

According to my plans (Onex 017) the horn does NOT go inside the rib but on the bottom, just like in the video.

Interesting ... I don't doubt what you are saying but I wonder why Sonex LLC would change from the way it is done on the Legacy aircraft. As I understood from Jeremy the OneX is just a Sonex that was downsized.

Can someone confirm how it's done on the new B model?

Curious ...

I'm curious about that as well. It would seem to me that with the forces on the rudder horn, the shop head of rivets would be more likely to pull through the thinner rib material. Except for a few odd occurrences done for spacing/fitment reasons, I was under the impression that the shop head should be against the thicker material.