What's new
Van's Air Force

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

serious control system issue

jwilbur

Well Known Member
Doing my first annual. When checking the elevator control in the tunnel I found the jamb nut was loose about 6 turns. Upon further investigation I found that the F-1065 cross bar was scraping the safety wire which in turn had worked the jamb nut loose. This happens only when the stick is pulled all the way back and while all the back it moves left/right -- exactly what you do when landing.

I know for a fact this was not scraping when I finished the build. I believe the push tube just twisted enough to bring the safety wire straight up top. So this had to be fixed. My plan was to first remove the bearing from the assembly, loosen the other end and roll it to put the safety wire on the bottom so it could never interfere again. Then I was going to adjust the segments as necessary to prevent any lingering interference. .... Well, as soon as I started removing the nut/bolt holding the rod-end bearing, the safety wire fell out. The safety wire loop on the bearing side had broken and the safety wire was being held only by being compressed in the assembly. In other words, it was not doing anything at all.

This part in the RV10 is on-purpose designed to have the rod-end stick out far more than half the threads in order to prevent interference with the tube itself. The safety wire is meant to prevent the tube from ever turning and working loose. I know I built this to plans and had it checked by another builder at the time.

Based on this experience I'm not happy with this design. In my case the tube twisted enough to put the safety wire on the top where it was then the cause of the interference. Further, I believe the constant working of that safety wire back and forth during landing (especially when forward CG - like flying solo for the first 40 hours) caused the safety wire to work harden and break where it went around the rod-end. I'm not 100% sure what I'm going to do to prevent this from happening again, but I'm considering safety wire on both sides in such a way to prevent the push-rod from rotating in either direction.

I'm not special so I'm assuming I'm not the only one where this has (or is) happening. I suggest everyone have a careful look at this next time you have the tunnel exposed.

In my case I caught it while doing a control system check with everything opened up for my first annual. When I had the stick full aft and moved side to side I could hear something scraping. And that lead to this find.


From the plans. Page 39-10
elevator_push.jpg


What I found after disconnecting the rod-end bearing from the control assembly. When I first looked here I saw the jamb nut had worked loose. I tightened it by hand and kept looking at what was going on. When removing the bolt from the rod-end the safety wire revealed that it had broken.
20180812_171353.jpg
 
Not sure what the other end looks like since I don't have my plans yet but if it has no other constraints to prevent it, I would install extra nuts on the other rod end bearing so that the push tube could not physically rotate enough to unscrew that end. I would also consider putting a sleeve on that end between the bearing head and the nut to prevent it from unscrewing the other end as well. A low profile sleeve should clear F-1065 if the nut does.

It is hard to tell from the photo, but it also looks like the safety wire was twisted a bit tight. AC 43.13 states 6-8 twist per inch are acceptable.

Glad you caught this!
 
Last edited:
Thanks for sharing this with the forum. I am on this page of the build right now and will be very attentive to how this is done. If you have any further insights on possible cause, please let us know.
 
a loose jam and broken safety wire. I don't like it. recommend to verify the rod could not become free by turning the rod both directions until it hits the stops at each end. this is goofy use of safety wire. safety wire should be straight and not able to wind up around around tie rod end threads. add loctite to the jam nuts.
 
Sure seems like an odd use of safety wire to me too. The movement you are trying to prevent is perpendicular to the twisting moment on the rod end, so the longitudinal change on the rod end requires more turns in order to make that twisted safety wire too tight to prevent further movement, on a bearing that is already more than half way out of the tube - - by design..... sure seems unusual to me. I just haven?t seen this before, but apparently it usually works if installed correctly.
 
Jam

If the jam nuts are tight I don't see how it could move in the first place I have a RV6 and the plans just call for jam nuts why would the RV10 need safety wire and why would the 6 not ? Asking to learn.
Bob
 
If the jam nuts are tight I don't see how it could move in the first place I have a RV6 and the plans just call for jam nuts why would the RV10 need safety wire and why would the 6 not ? Asking to learn.
Bob

Generally speaking, if the control rod rotates in service, one rod end unscrews, the other screws in. Most control rods have the rod ends screwed in more than 50%, so if tbe rod rotates in service the end going in will run out of threads (stopping further rotation) before the other end comes out. But in this particular application the rod ends could not be screwed in 50% of their length (due to clearance issues) so there was the possibility that if the rod rotated in service, one rod end could come loose. The safety wire was supposed to keep this from happening.
 
Generally speaking, if the control rod rotates in service, one rod end unscrews, the other screws in. Most control rods have the rod ends screwed in more than 50%, so if tbe rod rotates in service the end going in will run out of threads (stopping further rotation) before the other end comes out. But in this particular application the rod ends could not be screwed in 50% of their length (due to clearance issues) so there was the possibility that if the rod rotated in service, one rod end could come loose. The safety wire was supposed to keep this from happening.

Exactly!

This type of issue needs to be analyzed with a bit of logical reasoning.......

RV-10’s have been in service for 16+ years. From the first prototype through many hundreds of amateur built examples (almost 900 and still climbing), and this is the first instance of this type of problem that I am aware of.
That is not a reason to be casual about the discovery, but I think it is a good indicator that design is sound and if executed properly, will produce the desired result.

In the condition inspection forums/seminars I give, I recommend that the inspection of the control system be systematic. Starting at the stick grip and working through to the opposit end on component at a time.
I.E., ball in rod end free to move but free of play. Jamb nut on rod end properly torqued. Proper engagement of rod end to push/pull tube end. Push/pull tube end properly riveted to push/pull tube, etc.

This is a good idea for every condition inspection. Including the very first one that you do (and have to make an entry in the aircraft maint. Records as having been done) before the inspector will issue an airworthiness certificate.

Do not lose sight of the purpose of safety wire. It is not to keep things properly assembled (for the majority of things safety wired on aircraft, they will stay tight just fine on there own if torqued properly in the first place). It is to add an extra level of safety in case things don’t stay the way the were intended to.
 
Last edited:
This is most definitely NOT the first time that this has happened on a customer built RV-10. Forgive the tone if that came off wrong, as I'm not being argumentative. This isn't the only forum, and I'm not sure which forum it was reported on, but I know its happened to at least 2 other builders. One of them I think it actually unscrewed all the way out. One of them is a good friend of mine.

I checked on mine immediately after he reported this and found that mine looked good, but I was a bit paranoid after hearing what happened to him.

So far, I have done nothing additional to mine, as it's built per-plans and working fine. BUT, if someone comes up with an improved design or some sort of anti-rotation method, I would surely be interested in it. Also, I would think that in this application, loctite would be appropriate, as well as the jamb nut.
 
Actually, I just got a note back from a friend who said that Bob Kauffman crashed his RV-10 nose first from that exact problem...that the rod end unscrewed. (I think that was on landing).

This is second hand info, so I can't make any guarantee to it's accuracy, and my memory is fuzzy on that one. But, if this is true, then this has indeed actually been a factor in at least one incident.
 
Actually, I just got a note back from a friend who said that Bob Kauffman crashed his RV-10 nose first from that exact problem...that the rod end unscrewed. (I think that was on landing).

This is second hand info, so I can't make any guarantee to it's accuracy, and my memory is fuzzy on that one. But, if this is true, then this has indeed actually been a factor in at least one incident.

Tim, I heard (3rd hand) a different version: The nut securing the bolt thru the rod end bearing was removed to allow the alignment jig to be placed on the bolt end, during control alignment. Afterwards the nut was forgotten, and eventually the unsecured, horizontal bolt worked out of the bearing. Same general area but it was not the rod end that came loose.
 
When I rigged my control rods, I put a drop of green wicking Loc-Tite on the threads of the bearing thread next to the threaded insert, then ran up and tightened up the jam nut and put another drop in front of it. I did the per plans safety wire also. You know the Loc-Tite is doing its thing if you need to adjust down the road, but it does not make it overly difficult to turn. I don't think it would back out on its own if the jam nut came loose.
 
I replaced the jam nut with an AN364 low profile elastic stop nut, plus the stop wire after reading something somewhere on the same issue.
Martin
 
I?ve seen a couple of cases where the safety wire was broken, but never a jam but being loose. Both jam nuts would have to come loose for the rod to turn, which would indicate that they weren?t tightened correctly in the first place, imho. If you?re not comfortable with the stock setup get a rod end for the back of that rod that is the same as the front so there is no way for it to come apart, which is a scary scenario.
 
I got in there and fixed it today. Followed the plans (as I did before). The difference this time is I made sure the safety wire was routed thoroughly along the side (actually kind of under), of the tube and not so tightly wound as I had done originally.

Regarding the discussion that has been happening. In my case, the tube did not rotate. The rear jamb nut was so tight I couldn't even loosen it today.

As I evaluate this, I had two errors:

1... The jamb nut was not tight enough. This time I got it as tight as I could in the tight space and used thread locker as others have suggested.

2... The bigger problem in my case was the safety wire. It actually caused the problem (likely as a result of me doing something wrong). Somehow it ended up working itself over the top of the nut. It either broke first and ended up there or it ended up there and the interference caused it to brake. No idea which. This time I routed it thoroughly down the side - almost below the push tube. I still am not happy about it but the majority of builders have not had this problem. It could be that I just wound it too tight and it broke from vibrations. As I stated originally, I didn't know it broke until I loosened the nut. The broken ends were still in place around the bolt/bearing. For all I know, it broke when I installed it. ... I stress, however, that there was NO interference or scraping of any kind when I got my FAA inspection. This developed over the course of the past year.

So another comment about the issue. This is not just about preventing one of the ends from coming out (as important as that is). The tube can NOT be allowed to rotate. If it were to rotate forward, even if there were enough threads on the far side to keep it all together, it would severely hinder up elevator performance. The push tube would become your new up-elevator stop.

The whole point of this (apparently controversial) design is to keep the push tube away from that cross bar (see pictures in first post). Any improvement in the design has to not only keep the two ends from separating, but has to keep the push tube well away from that cross-bar. ... Note in the plans how it states even to position the jamb nut so as to have increased clearance. If designed to plans, full aft stick puts the cross bar right over the jamb nut. Rotate the nut off the flat and the bar can touch it.

Is there any reason the push-tube must be the diameter that it is? Something like the size of the cross-brace would eliminate the interference or any chance of it, I think. Not sure what the engineering reasons are for the push tube we're using, so I'll be leaving it alone. But just wondering. Maybe someone else would have an idea why the tube is what it is.
 
Tim, I heard (3rd hand) a different version: The nut securing the bolt thru the rod end bearing was removed to allow the alignment jig to be placed on the bolt end, during control alignment. Afterwards the nut was forgotten, and eventually the unsecured, horizontal bolt worked out of the bearing. Same general area but it was not the rod end that came loose.

This was the cause that I heard as well.
 
This is most definitely NOT the first time that this has happened on a customer built RV-10. Forgive the tone if that came off wrong, as I'm not being argumentative. This isn't the only forum, and I'm not sure which forum it was reported on, but I know its happened to at least 2 other builders. One of them I think it actually unscrewed all the way out. One of them is a good friend of mine.

I checked on mine immediately after he reported this and found that mine looked good, but I was a bit paranoid after hearing what happened to him.

So far, I have done nothing additional to mine, as it's built per-plans and working fine. BUT, if someone comes up with an improved design or some sort of anti-rotation method, I would surely be interested in it. Also, I would think that in this application, loctite would be appropriate, as well as the jamb nut.

If the screwed all the way out instance is the one that you thought resulted in the landing accident at north Las Vegas, that was not the cause of that accident ( see other post in this thread).
So maybe that leaves one other? (Note: I said I hadn't heard of any...... I didn't say i knew for a fact that it hadn't)
Do you hav any specific details? ( such as a thread name on the RV-10 list to search for)
 
.. Well, as soon as I started removing the nut/bolt holding the rod-end bearing, the safety wire fell out. The safety wire loop on the bearing side had broken and the safety wire was being held only by being compressed in the assembly.
From the plans. Page 39-10
elevator_push.jpg

I believe the root cause here may be a subtle but important deviation from the instructions. You state the safety wire was being held by being compressed in the [rod end] assembly. But the plans say that the wire should not interfere in any way with the rod end, e.g., it should go around the part of the rod end in contact with the clevis, and not be ?compressed in the assembly?. Properly assembled, the wire ROTATES a small amount, with respect to the clevis/rod end, with fore-aft stick movement. But compressed by the assembly, as yours was, will subject it to an up-down flexing with fore-aft stick movement, leading to work hardening, fatigue, failure.
 
I did what Jessie said, same length rod bearing on the aft end of the push rod. I think there was a discussion about it in the forums at that time.

Lenny

If you?re not comfortable with the stock setup get a rod end for the back of that rod that is the same as the front so there is no way for it to come apart, which is a scary scenario.
 
I believe the root cause here may be a subtle but important deviation from the instructions. You state the safety wire was being held by being compressed in the [rod end] assembly.

I don't think there was any real compression. That was just conjecture in the original post because the break wasn't seen until I started moving the nut. It's possible but it's hard to imagine how it could actually be compressed. It may have just been the "springy-ness" of the wire holding it in place.

I dug up some old pics of the build. The root cause is that I routed the safety wire perilously close to the top of the tube. There was no interference at that point but it's not hard to imagine after a lot of vibration that the "loop" inside the tube changed position and along with the normal few degrees of play in the bearings that it worked itself over the top.


I share this picture as an example of how not to do it. The wire is too tightly wound and it's routed too much "over the top". I don't know why I did it that way and not just along the left side. Also the nut looks like it's not as much on the flat as the plans say, but the weird angle might make it look worse than it is. I know there was no interference at the start, but you can see that with a little shifting of the wire working out of the hole at the top (and in at the bottom), etc, and the normal play in the bearings we could get the problem I found.
old_safety_wire_pic.jpg


This, I'm sure, is what caused the problem.
 
That?s a good photo of the issue, and I agree with you. If the plans say the tolerance there is so tight that you need the nut?s flat up, you don?t want anything (like safety wire) above the nut either.
 
Safety wire break

I found my safety wire broke during my condition inspection and reported it on VAF. I did a better job getting the wire away from the rotating/moving parts and have had no further problems.
 
Scott, you probably are correct on the Vegas one. I heard it second hand, and could absolutely have heard it incorrectly. I wish I could remember more details right now on the other one. It's probably been 3 - 5 years since the discussion. Time flies by so quick, it's hard to believe I've already put over 12 years on my -10. If I find it, I'll pass it on.

That said, I do have an idea for a permanent fix if someone wanted to think it though. Picture a T-shaped strap that goes on the bottom of the pushrod tube, riveted on. Maybe about .030-.040 thick. It comes forward as a strap, under the rod end and threads, and up by the whole connection point you have the ears of the T, that get folded up and bolted between the rod end and the other control mechanism, basically right where the safety wire sits today. The issue I see is that there would have to be enough gap for this thing to act as a bushing, but if there were a way to get it in there, it could prevent the tube from ever rotating at all. It could be done on either end of the tube, so long as the tube is never allowed to rotate. Anyway, that's just thinking out loud.
 
Last edited:
Does no one make a rod end with a longer threaded length that could be used in this application?
 
Does no one make a rod end with a longer threaded length that could be used in this application?

It's not just the length of the threads that matter. It can't be allowed to turn at all or the push tube may become the new elevator stop.
 
Spacers?

It's not just the length of the threads that matter. It can't be allowed to turn at all or the push tube may become the new elevator stop.

One solution to prevent a disaster if the tube did rotate would be to make custom length thin walled tube spacers over the exposed rod end threads.

Cut to length after the push rod is adjusted for final length and do it at both ends.
 
One solution to prevent a disaster if the tube did rotate would be to make custom length thin walled tube spacers over the exposed rod end threads.

Cut to length after the push rod is adjusted for final length and do it at both ends.

Yes, I agree! Can't think of anything more simple and almost fail proof.

Not sure what the other end looks like since I don't have my plans yet but if it has no other constraints to prevent it, I would install extra nuts on the other rod end bearing so that the push tube could not physically rotate enough to unscrew that end. I would also consider putting a sleeve on that end between the bearing head and the nut to prevent it from unscrewing the other end as well. A low profile sleeve should clear F-1065 if the nut does.

It is hard to tell from the photo, but it also looks like the safety wire was twisted a bit tight. AC 43.13 states 6-8 twist per inch are acceptable.

Glad you caught this!
 
One solution to prevent a disaster if the tube did rotate would be to make custom length thin walled tube spacers over the exposed rod end threads.

Cut to length after the push rod is adjusted for final length and do it at both ends.

+2
I was thinking of this too. Just not sure I can get thin walled tubing thin enough to fit inside the clevis fitting but thick enough to not crack.
 
+2
I was thinking of this too. Just not sure I can get thin walled tubing thin enough to fit inside the clevis fitting but thick enough to not crack.

1/16" wall tubing should be OK and would be of a lesser diameter than the jam nuts.

It does not need to be torqued against - I was actually thinking of the spacer being a little loose. It's just there to prevent the push rod from turning (a lot) and letting a rod end fall out, the "disaster scenario".

But I'm not sure what you mean when you say "fit inside the clevis fitting".
 
But I'm not sure what you mean when you say "fit inside the clevis fitting".

Maybe that?s the wrong word. I mean the forked fitting that the rod end bolts to. I think you want the tubing to butt against the rod end, not against that forked fitting. That means the OD has to fit inside the fork.
 
They used to have spacers like that on the aileron pushrods between the sticks and the torque tubes in the wing root. Then they lengthened the hex rods and eliminated the spacers. I wonder why they didn?t do that with the push rod discussed in this thread.
 
Maybe that?s the wrong word. I mean the forked fitting that the rod end bolts to. I think you want the tubing to butt against the rod end, not against that forked fitting. That means the OD has to fit inside the fork.

Good point.

Looking at the Aurora specs a .032 tube spacer would work. It's a straight push-pull so no angular deflection to the side.

Looks like 7/16 x 0.058 6061 tubing from spruce reamed out to 3/8 ID would work - or 7/16 OD 4130 steel if you prefer that.
 
The K&S hobby tubing rack at your local Ace Hardware will have lots of thin wall brass and aluminum tubing.
 
Back
Top