RV-12 Fatal Crash - Control Failure 6/6 Auburn, WA

[Hengelo]

@Dan Thomas it’s barely 9am here but I think I need a stiff drink after reading that.

 

[Spring Ford]

the control geometry IMO doesn’t support that conclusion

Interesting idea. I am struggling with it a bit, maybe you can assist me?

Perhaps you could illustrate a control geometry where a control rod in service as a push-pull operating rod (has free spherical bearings at each and therefore unable to transmit a turning moment), can be varied in length in such a way as to NOT affect the position of either the operating lever (eg Joystick) or the end device (eg Aileron).

I don't think I can find a way to make it work.

 

[DaleB]

Interesting idea. I am struggling with it a bit, maybe you can assist me?

Perhaps you could illustrate a control geometry where a control rod in service as a push-pull operating rod (has free spherical bearings at each and therefore unable to transmit a turning moment), can be varied in length in such a way as to NOT affect the position of either the operating lever (eg Joystick) or the end device (eg Aileron).

I don't think I can find a way to make it work.

I don't think what happened here, as it happened, would have varied the relative angles of the two sticks so that it would have been obvious to a routine look. If you were specifically checking for parallelism you might see a difference, but how often does one do that? You don't... unless you suspect there's a problem with the control linkage, in which case you're probably already pulling up the floor pan.

If you have right hand threads on both ends the rod can be turned back and forth until you either hit the end of the thread or fall off without any change in the center to center length.

And that, yeah. I don't think that was the case in this instance; I think (based on the preliminary report) the inboard end simply became unthreaded from the rod. Good reason to make the control rod long enough to engage well over half the threads. I'm building a plane now that will use control rods, so you can bet I'll be keeping that in mind (along with making darn sure there are properly torqued and marked jam nuts).

 

[Capt. Geoffrey Thorpe]

Perhaps you could illustrate a control geometry where a control rod in service as a push-pull operating rod (has free spherical bearings at each and therefore unable to transmit a turning moment), can be varied in length in such a way as to NOT affect the position of either the operating lever (eg Joystick) or the end device (eg Aileron).

If both rod end bearings have right hand threads, the push pull tube can be screwed back and forth until you either hit the end of the threads or the tube falls off the end without changing the center to center length.

edit: Rewind (again) - got mixed up on the as installed vs as built configurations

As built, it would be easier for the threaded end to unscrew completely from the rod without turning the rod on the other end - and, yes, that would have tweeked the stick position.

 

[455 Bravo Uniform]

If that’s about 1” of thread and the top of the stick is 3x longer at the hinge point vs the hinge to lower attach point, that’s about a 3” difference off of parallel with the other stick at the top of the stick?

As the left rod was getting “longer”, the left aileron would be going up making the left wing appear heavier, no?

 

[Capt. Geoffrey Thorpe]

As the left rod was getting “longer”, the left aileron would be going up making the left wing appear heavier, no?

No.
All you do is move the "neutral" position of the stick. To make a wing heavier you have to do something aerodynamic (ignoring any use of springs for trim).

 

[455 Bravo Uniform]

In a typical configuration one end of the rod will have reverse threads. The jam nut will usually have a mark on it to show that it’s reverse threaded. The rod end threads look a bit obviously weird, if one is used to decades of seeing “normal” threads. Not sure how an RV-12 is done.

I’m going to guess that the outboard end is the one that is reverse threaded in this build. Otherwise that normally threaded female would not have threaded into the inboard end, and that normally threaded locknut would not have gone into the rod end (also normally threaded) that was installed vertically on the bracket.

A bit of a legal conundrum whether to make it even more clear in the instructions or in the kit parts (different thread sizes) to make sure there is even less chance of this happening again.

I’ve learned a bit in this discussion. I’m scared of myself, I don’t know what I don’t know.

 

[tsts4]

Interesting idea. I am struggling with it a bit, maybe you can assist me?

Perhaps you could illustrate a control geometry where a control rod in service as a push-pull operating rod (has free spherical bearings at each and therefore unable to transmit a turning moment), can be varied in length in such a way as to NOT affect the position of either the operating lever (eg Joystick) or the end device (eg Aileron).

I don't think I can find a way to make it work.

My point is that as the stud rotates and backs out and that end of the push rod drops relative to the assembly the angular relation to the other end which is rotating around another pushrod bearing isn’t sufficient to induce a noticeable stick displacement relative to the other stick. As the builder and owner of a plane with side by sticks I can tell you first hand that visually noticing anything beyond a significant displacement would be a miracle. What you might notice is slop in the stick movement which in a pushrod/torque tube system should be nonexistent.

 

[Dan Thomas]

Please do.

No plans to do that. Too many here already don't believe things could be as bad as all that. Moreover, my experience is mostly Cessna and Citabria, limited Piper, a bit with Cirrus, Lake and Maule, none with Mooneys. Some with homebuilts, including my own. When it comes to electrical systems though, they're all similar, and the laws of electricity are the same in any airplane. And the maintenance shortcomings will be there across all of them.

 

[2-Bit Speed]

No plans to do that.

 

[Dan Thomas]

No.
All you do is move the "neutral" position of the stick. To make a wing heavier you have to do something aerodynamic (ignoring any use of springs for trim).

I have encountered POAers who think that rigging one aileron down a little will correct wing-heaviness.

But it can't. One aileron down will get pushed upward by the airflow, which will then move the system to push the other aileron down until the pressures on them are balanced. The control wheels will be off-level then, and the wing will still be heavy.

Service manuals give instructions for correcting wing-heaviness. Cessna uses eccentric bushings on the aft wing spar to change the incidence of the wing. Ragwing airplanes had two struts per wing, and the aft ones are adjustable to change the washout a little to make the airplane fly level. Piper uses a ground-adjustable trim tab on the left aileron, and they also say this:


If I recall right, the Cirrus has a centering detent in its aileron system, and that detent is moved by a trim motor. The Cessna Corvalis uses an electric trim tab on the right aileron.

 

[Daleandee]

I have encountered POAers who think that rigging one aileron down a little will correct wing-heaviness.

But it can't. One aileron down will get pushed upward by the airflow, which will then move the system to push the other aileron down until the pressures on them are balanced. The control wheels will be off-level then, and the wing will still be heavy.

Service manuals give instructions for correcting wing-heaviness. Cessna uses eccentric bushings on the aft wing spar to change the incidence of the wing. Ragwing airplanes had two struts per wing, and the aft ones are adjustable to change the washout a little to make the airplane fly level. Piper uses a ground-adjustable trim tab on the left aileron, and they also say this:
View attachment 130799

If I recall right, the Cirrus has a centering detent in aileron system, and that detent is moved by a trim motor. The Cessna Corvalis uses an electric trim tab on the right aileron.

I went through this with a couple of guys working on a Challenger Sport Plane. They were complaining that the plane was hard to get out of a left turn and asked if I wanted to fly it to see what the problem was. Nope! I tried to explain that the ailerons were simply going to react to the airflow and that they were simply changing the stick position but they wouldn't listen.

The problem did get fixed when I checked the wing washout and found one side had 2º and the other side had zero ...