What's new
Van's Air Force

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

Trailing edge fun...

WVM

Well Known Member
Hi,

I am doing one of the last steps of my empennage... riveting the trailing edges of the elevators and rudders. So far the good news... :D

Following VANS instructions I glued the wedges in with Pro seal. They cured for more than two weeks during my vacation on a flat surface. Yesterday I cleaned the dimples to finally rivet the edge and close the chapter. But... When I turned around the piece I noticed that it didn't lay flat on the table top on once side. This is both true for the rudder and the elevator. The trailing edge itself looks straight to me. Not sure that this is normal and that it is a good idea to continue like that.

Rudder:
edge-1.jpg
edge-2.jpg
edge-3.jpg


Elevator:
edge-5.jpg
edge-4.jpg
edge-6.jpg


Just to be clear this is before riveting!
 
It is very hard to tell what is going on with the rudder on a bench. The rudder sides are not a plane and won't lay flat on a table. If the corner lifted is the same for both sides you are ok. If you want to measure more precisely, take a couple pieces of string in the end holes for the rudder trailing edge and suspend it from the ceiling. Adjust so the pivot is close to level, eyeball is good. Then hang a string line near each end that passes through the center of the TE. You can then tell exactly what you have for twist, if any.
 
Last edited:
It is very hard to tell what is going on with the rudder on a bench. The rudder sides are not a plane and won't lay flat on a table. If the corner lifted is the same for both sides you are ok. If you want to measure more precisely, take a couple pieces of string in the end holes for the rudder trailing edge and suspend it from the ceiling. Adjust so the pivot is close to level, eyeball is good. Then hang a string line near each end that passes through the center of the TE. You can then tell exactly what you have for twist, if any.

Hi Bill, Thanks for the advise.
Wouldn't that result in the same view a the picture I posted straight over the edge? This is with the counterweight on a table, the other side in my hand. With the camera straight over the egde. It looks straight from that point of view. Still I found it a bit strange that it lays flat on one side and not on the other. Logically it should not really lay flat on any side... it worries me a bit.
 
Hi Bill, Thanks for the advise.
Wouldn't that result in the same view a the picture I posted straight over the edge? This is with the counterweight on a table, the other side in my hand. With the camera straight over the egde. It looks straight from that point of view. Still I found it a bit strange that it lays flat on one side and not on the other. Logically it should not really lay flat on any side... it worries me a bit.

Right, It should not lay flat on either side. Think about the geometry. (Edit - I am WRONG - it should lay flat on both sides - here is a reference from RVBuilder 2002 http://www.vansairforce.com/community/showpost.php?p=858345&postcount=11) Imagine the vertical spar as a trapezoid. Now, looking at it with the small leg on top, draw a bisecting vertical line on it. That is the rear view of the rudder and it's trailing edge. If one side lays flat, then that vertical line must become parallel to one side of the trapezoid. You can see that the TE can be straight and yet the rudder become twisted. Think of this definitively and with hard geometry. Logical thinking without quantification will carry one in circles. I would urge you to quantify the twist then send a note to Vans about the result. They can tell you what effect it will have on flight characteristics.

To measure twist, you can either hang it, or lay on a table with the spar in contact and measure the difference between sides. The latter is most common. From the geometric example, you can see that it was necessary to lift the top of the spar from the table 1/2 the difference in width from top top to bottom during the gluing of the TE in order to make it perfect. Does this make sense now?

Screen%252520Shot%2525202015-09-14%252520at%2525206.38.09%252520AM.png
 
Last edited:
Hi,

If I understood it correctly, the misalignment is 35/64 inch... doesn't sound good.
 

Hi ShortSnorter,

Thanks for pointing me to this post. It overlooked this comment from RVBuilder2002 in post 11 of that thread: http://www.vansairforce.com/community/showpost.php?p=858345&postcount=11

"The thickness / cord ratios on the RV-10 and RV-9 rudder (and RV-7 as long as it was built with the riveted T.E. rudder), have flat sides and can be riveted with them weighted down flat to a work table."

Do I understand this correct as being good news for me? Or in other words, that it is normal to have a flat side for the RV9?

Still waiting on a reply from VANS for the mail I sent Saturday. I am not looking forward to redo it.
 
Trigonometry

It turns out that the 9 rudder does not lay perfectly flat with the spar weighted down, but only by a tiny bit. In order for the rudder to lay flat, the taper of the spar and the taper of the trailing edge of the rudder skin would need to extend out to the same point in space. They, unfortunately, don't (but, again, only by a tiny bit). If you take really careful measurements, because the angles are very small, and do some trigonometry, you find out that the spar tapers out to a point short of where the trailing edge tapers out. Working the math backwards to answer the question, "How much wider would the top of the spar need to be to make it taper to the same point as the rudder skin?" gives you an answer of 1/16" [at least, that's what my measurements and calculations came up with - I encourage others who want to do this to take measurements to see if it comes out differently]. However, the top of the spar already has an additional .032" skin for the counterweight, so it is really about 1/32", at least according to my measurements and calculations. What that means is, if you slide a 1/32" shim between the top of the spar and the table, the trailing edge of the rudder will lay flat, if your table is flat. Honestly, when I was standing at the table, poking at the rudder with and without the shim, I couldn't tell any difference.
 
Hi,

I got a reply from Sterling. He confirmed that the RV9 rudder need to lay flat on the table on both sides.

First I double checked that the workbench was 100% level which was not. I now messure a different of 13/64 on one side. According to Sterling it is perfectly safe to fly.

When I push it down on the side that is off the table I get a small bump in the upper tip at the wedge. The probably cause is that the rudder was not flat on the table during the match drilling (and later sealing) of the wegde.

I will give it a try to remove the pro seal from the wedge and glue it again with sand bags on the rudder. It is still unsure that I will be able to correct it as the holes in the edge will probably misalign when the skin is pressed down.

Anyone had success removing proseal from the wedge? What technique would you use? Or would it be a bad idea to try removing it?

Thank you
 
Hi,

I finally found the courage to remove the wegde from the rudder. The proseal is mostly cleaned out. Before I glue it again. Does anyone remember what came pre-drilled? It looks like both L & R skins of the rudder are pre-drilled. I cannot find any good picture for the wegde.

The reason for asking it because Sterling told me to put weight on the rudder skin in an attempt to correct it. As this may cause a misalignment at the edge, I want to understand what could be corrected and what not.

Based on some logs that I follow I see nobody putting weight on the rudder and I also don't find it in the procedure. Still a bit strange that this offset happened...

Thanks
 
Rudder skin

I'm quite sure only one skin trailing edge is predrilled.
I used a peice of angle to keep the edge straight and match drilled
Through the wedge, other skin and angle. Removed everything, deburred, counter sunk the wedge and then dimpled the skins. Clecoed and glued with prose all before riveting. Not sure what else to tell
You regarding the rudder.

I feel your pain, I had to buy a new skin for one of my elevators.
One elevator I completely trashed because I did not put enough
Down pressure on the pre-bent leading edge. It resulted in a very wonky
Trailing edge.
 
VAN'S responded:
All the part I have on the shelf have holes punched and drilled in them, the R-901 left and right as well as the R-916 trailing edge.

In that case it doesn't make any sense to me that this error could have happened. And it especially makes no sense that both sides of the rudder should lay flat on the table. The reason that I don't get it is because the corner on the top and bottom are having a different size as the trailing edge is closer to the spar on the top compared to the bottom...
 
Rudder

Looking at the plans. Yes your correct, my bad.
Is there another way you can figure if you have twist
In your trailing edge? Mount the rudder verticle measuring center of forward spar either with a level or plumb bob. Then measure the trailing edge to see if it is also verticle or plumb?
Or if you weight it down on a table it should be one flat plane and lie flat without it rocking or lifting.
 
Last edited:
And it especially makes no sense that both sides of the rudder should lay flat on the table. The reason that I don't get it is because the corner on the top and bottom are having a different size as the trailing edge is closer to the spar on the top compared to the bottom...

It is possible depending on what the taper ratio of the spar to the taper of the rudder cord is.

You can prove it to your self by making a couple of equal sized triangles similar to THIS, made form some stiff material.
Tape the two long edges together to make a hinge (this simulates the trailing edge).
Hinge it open and you will see that next longest side will have a tapered opening (where the spar would go) but the sides will still be flat.
 
It is possible depending on what the taper ratio of the spar to the taper of the rudder cord is.

You can prove it to your self by making a couple of equal sized triangles similar to THIS, made form some stiff material.
Tape the two long edges together to make a hinge (this simulates the trailing edge).
Hinge it open and you will see that next longest side will have a tapered opening (where the spar would go) but the sides will still be flat.

You are right.

Still the problem remains. This is the trailing edge now. It looks straight to me:


rudder1.jpg
rudder2.jpg


Still, when I lay it flat on table, one side is 100% flat, when I turn it I can push the rudder down in the corner where the angle is for the rudder cable attachments and it comes of the table in the other corner at the trailing edge. I am not sure that if I put sandbags on the rudder to bring it down all the way and glue it, that this will make much of a difference when I remove the sand bags again. I am afraid it will not.
 
Hi,

There is a difference in the angle of the rudder on the bottom and top. At least on my rudder. I know this because I put the rudder upside down on a paper and made a template. When I use this template on the bottom of the rudder, I have a difference of about 0,236 inch on the top on about 12 inch from the trailing edge. This can not be something I did, it comes from the preformed ribs. I deleted the second photo by accident else I would be able to show it to you.

rudder-angle.png


When I now lay down the rudder on the table I have one side laying 8/32 inch of the table in the bottom corner of the trailing edge and 4/32 on the other side. A difference of about 4/32 between both sides.

rudder-twist1.png

rudder-twist2.png


With a perfectly leveled table I've checked the center of the spar rivets with a plumb (where it attaches to the vertical stabilizer) and that is straight, maybe slightly off (the thickness of the rope). The rudder is tilted backward a bit to avoid that the robe touches the spar. Without touching the rudder I then move to the trailing edge and also that appears to be straight with the plumb (once again maybe 0.039 inch off on the top).

rudder-level.png

rudder-trail1.png

rudder-attach1.png

rudder-attach2.png


Weighting the rudder may push it straight on the table, but I have doubts that it is a good idea. When the center of both sides are (nearly) 100% the same, I suppose this is what is important for ensuring there is no twist, isn't it?

Thanks
 
Last edited:
I saw this thread right as I was about to leave the house this morning. My proseal had been on for 2 days, so took the last ten minutes I had and removed trailing edge clecos and laid the rudder on the kitchen table for a quick double check. Sure enough - flat on one side, way off on the other (may be 3/4"???). Did not have time to do any additional checks and won't be back home for a couple days.
 
I saw this thread right as I was about to leave the house this morning. My proseal had been on for 2 days, so took the last ten minutes I had and removed trailing edge clecos and laid the rudder on the kitchen table for a quick double check. Sure enough - flat on one side, way off on the other (may be 3/4"???). Did not have time to do any additional checks and won't be back home for a couple days.

I am surprised that we are the only builders actually having (or noticing) this issue. VAN'S support tells me that both sides should lay flat on the table for the RV-9. Who am I to not agree? But... based on the paper template I made, the angles both bottom and top are different, and it will not be possible that they lay flat on both sides without forcing it.
It is my personal opinion that if the cord of the spar and trailing edge are straight, and in line of each other, your rudder is not having a twist and is fine.
I am off to OSH in a few days and going to talk to VAN'S engineering about it. If I know more I will let you know. I glued mine as is on the pictures. And will continue with it.
 
I am surprised that we are the only builders actually having (or noticing) this issue. VAN'S support tells me that both sides should lay flat on the table for the RV-9. Who am I to not agree? But... based on the paper template I made, the angles both bottom and top are different, and it will not be possible that they lay flat on both sides without forcing it.
It is my personal opinion that if the cord of the spar and trailing edge are straight, and in line of each other, your rudder is not having a twist and is fine.
I am off to OSH in a few days and going to talk to VAN'S engineering about it. If I know more I will let you know. I glued mine as is on the pictures. And will continue with it.
I will be at OSH as well. When I get home tomorrow, I will check on my rudder in more detail. Mine is a -7. If chords are different I think then a "good" rudder will have equal "twist" on both sides, and having it be flat on one side and not the other is a bad thing I think. kinda annoyed at myself for not checking this first before prosealing...
 
It is normal that it is coming of the table, but it need to do so equally on both sides. This appears to be different for the RV7 compared to the RV9. RV9 should be flat on the table surface on both side. At least this is what I was told.
 
you know... I wonder if this problem goes hand in hand with oil canning. I'm not home right now, but I bet the side that oil cans is not the flat side.
 
alright, so I was exaggerating a bit with how twisted my rudder is. Total deflection is 3/8, so it has about 3/16" twist. I think I will live with it.
 
before riveting my trailing edge, I put some weight on to undo the twist a bit. Not much. Basically one 5 quart oil container on opposite ends of the twist. I'm half way done with riveting, and total deflection is at 3/16" now, so 3/32" twist.
 
What is the correct way to remove a bow in the trailing edge? The instructions from VAN's note that you should monitor for possible divations but how to act when it is happing?
 
alright, so I was exaggerating a bit with how twisted my rudder is. Total deflection is 3/8, so it has about 3/16" twist. I think I will live with it.

I don't think you will notice it. The rudder will trail with the forces from the top and bottom twist canceling each other out.
 
Straight edge?

I used a piece of 2 inch angle aluminum as my straight edge. Clecoed each hole with T88 epoxy. I was pleased with how straight it ended up as it was allowed to align vs. be weighted down which may cause mis alignment.

Just my 2 cents.
 
I used a piece of 2 inch angle aluminum as my straight edge. Clecoed each hole with T88 epoxy. I was pleased with how straight it ended up as it was allowed to align vs. be weighted down which may cause mis alignment.

Just my 2 cents.

I think you're referring to straightness. That was not the problem. My edge is straight, but not 100% parallel with the leading edge of the rudder. Both are straight lines however. I also used a 2" angle aluminum clecoed to each hole, alternating sides, bonded and dried for 3 days, before removing clecos in a pattern and riveting. You only really notice it if you lay the rudder down on each side on a flat surface.
 
I closed the rudder trailing edge. The result is not superb, but I expect it to be within limits. After all the trouble with this rudder, I re-glued it and made sure that both skins where perfectly aligned in the edge. This resulted in a less perfect straightness on the edge, but a better balance between both sides.

Please provide your advise. I this indeed within limits? Would you continue with it?

You see a little bow in the middle:
rudder-edge2.JPG


On the other side, the ends are pointing up:
rudder-edge3.JPG


General picture over the edge:
rudder-edge1.JPG
 
Back
Top