What's new
Van's Air Force

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

Goodbye wooden gearleg stiffeners

rocketbob

Well Known Member
About 5 years ago a friend of mine installed 380 tires on his RV7. I fly this airplane regularly. With the 380 tires a shake would develop right around 19 KTS GS which required firm braking to slow thru that speed fairly rapidly to get the shake to stop. Lived with it but it was pretty annoying. Things tried to fix the shake: carefully balancing the tires, new tires instead of recaps, new wooden gear leg stiffeners, various tire pressures, tried one gearleg with a stiffener without the other having one thinking a resonance was developing between the two. Nothing really worked but there were varying degrees of effect on dampening the oscillation. That is, until the new idea we came up with and implemented a few weeks ago.

Wrap the gear legs with 3K 5.8oz carbon fiber. No stiffeners.

Using a 48" roll of carbon fiber, squeegee wetted a 4ft or so square, 8" wide strips were cut using a rotary knife which were then rolled up and wrapped around clean gearlegs bottom to top in one pass overlapping 50% which amounts to two layers of carbon.


No peel ply or vacuum bagging was used as this was considered an experiment.

The layup was done with mains on the floor.

After about a dozen or so hours still no shimmy or shake at all and it seems to be holding up fine.

The gear feels absolutely perfect now with no tendency to rebound or skip. Tires are 35psi. In no way does the gear feel stiffer, but feels well dampened. For this reason I don't think any additional stresses are placed on the engine mount.
 
What was the fiber orientation?

Since carbon can contribute to dissimilar metal corrosion, how did you protect the bare steel? Or did you use some sort of primer between them, and if so, what kind?

Given that carbon has a Young's modulus that's generally considerably lower than steel, can you give us a comparison of the EI with and without the carbon?

Thanks very much!
Dave
 
We didn't cut in a bias direction off the roll as it wasted too much much material. Discussed it, didn't do it. So whatever orientation it is cutting across the 48" width of the roll.

The gearleg was cleaned bare. I have never had any problems with carbon fiber corroding steel or aluminum when bonding directly to each other. I believe this concern is overblown.

As far as any engineering analysis goes, none was done, nor is anything going to be done. Consider this an experiment that is working.
 
carbon over metal

carbon bonded directly to aluminum will corrode the aluminum rapidly if exposed to an electrolyte. Don't dunk it in sea water. Exposed to atmospheric moisture, the aluminum will corrode slowly and degrade the bond over time (years) because moisture (as vapor) can diffuse through the epoxy.

carbon and steel are not as galvanically dissimilar as carbon and aluminum. a bond to bare steel and not exposed to electrolyte will last a long time. But not indefinitely.

An easy way to prevent problems is to first cover the bare metal with a very thin ply of fiberglass, then put on the carbon. The fiberglass ply will provide an electrical insulator/isolator. But I still would not expose a bond like this to electrolyte.

The gold standard would be to apply AC-130 SolGel, followed by BR-6747 primer to the bare metal.
 
carbon bonded directly to aluminum will corrode the aluminum rapidly if exposed to an electrolyte. Don't dunk it in sea water. Exposed to atmospheric moisture, the aluminum will corrode slowly and degrade the bond over time (years) because moisture (as vapor) can diffuse through the epoxy.

carbon and steel are not as galvanically dissimilar as carbon and aluminum. a bond to bare steel and not exposed to electrolyte will last a long time. But not indefinitely.

An easy way to prevent problems is to first cover the bare metal with a very thin ply of fiberglass, then put on the carbon. The fiberglass ply will provide an electrical insulator/isolator. But I still would not expose a bond like this to electrolyte.

The gold standard would be to apply AC-130 SolGel, followed by BR-6747 primer to the bare metal.
Plus one
Obviously Steve has the credentials, my info is just based on readings.
 
Bob, just to be clear, did you spiral wind the 8" x4' length of carbon on the legs, overlapping 50%?
Thanks for sharing this discovery/experiment!
 
Bob, just to be clear, did you spiral wind the 8" x4' length of carbon on the legs, overlapping 50%?
Thanks for sharing this discovery/experiment!

We had 3 or 4 layups, can't remember. You can use whatever width you want as long as you wrap 50% overlap. Might be easier to order carbon fiber tape to do this as the edge unravelling can be a pain.
 
I would have a difficult time doing this to my beautiful titanium F1 gear legs because they look so nice!
 
Should be ok

If you are wrapping it over the powder coated gear leg there shouldn't be a problem with the carbon over the steel leg.

I think an ideal situation for this is carbon fiber sleeve available from CST Sales. Of course this would have to be done before you mount the brakes. With this material you pull it over the gear and tape it to the gear just below the mount. Run it down as far as you need and pull it tight. Tape the bottom. Wet it out with epoxy squeezing the epoxy into the carbon with gloved hands. Wipe off excess with a paper towel.

Interesting concept using carbon as a dampener.
 
About 5 years ago a friend of mine installed 380 tires on his RV7. I fly this airplane regularly. With the 380 tires a shake would develop right around 19 KTS GS which required firm braking to slow thru that speed fairly rapidly to get the shake to stop. Lived with it but it was pretty annoying. Things tried to fix the shake: carefully balancing the tires, new tires instead of recaps, new wooden gear leg stiffeners, various tire pressures, tried one gearleg with a stiffener without the other having one thinking a resonance was developing between the two. Nothing really worked but there were varying degrees of effect on dampening the oscillation. That is, until the new idea we came up with and implemented a few weeks ago.

Wrap the gear legs with 3K 5.8oz carbon fiber. No stiffeners.

Using a 48" roll of carbon fiber, squeegee wetted a 4ft or so square, 8" wide strips were cut using a rotary knife which were then rolled up and wrapped around clean gearlegs bottom to top in one pass overlapping 50% which amounts to two layers of carbon.


No peel ply or vacuum bagging was used as this was considered an experiment.

The layup was done with mains on the floor.

After about a dozen or so hours still no shimmy or shake at all and it seems to be holding up fine.

The gear feels absolutely perfect now with no tendency to rebound or skip. Tires are 35psi. In no way does the gear feel stiffer, but feels well dampened. For this reason I don't think any additional stresses are placed on the engine mount.

How timely. Same here. I glassed on some wooden stiffeners that were made from door molding, from a forum vendor. Same size tires, carefully balanced, tried Michelins, Dessers, blah blah. The shimmy would start at around 28 knots on deceleration, get awful, then diminish below 19 knots on rollout. Rattled fillings out of teeth. Messed with tire pressures, no help. The stiffeners shattered after awhile. Door molding is made from cheap wood, spliced together. The stiffeners failed at the joints, as in, splintered. Currently with several spiral wraps of glass clockwise and counterclockwise things are holding up, but as soon as the carbon cloth and some carbon bars I ordered show up, I'm going to try something different. Will report back.
 
Curious . . .not challenging . .

What was the fiber orientation?

Since carbon can contribute to dissimilar metal corrosion, how did you protect the bare steel? Or did you use some sort of primer between them, and if so, what kind?

Given that carbon has a Young's modulus that's generally considerably lower than steel, can you give us a comparison of the EI with and without the carbon?

Thanks very much!
Dave

I thought the carbon fibers was around 77(million), with steels in the 28-30 range? Are you talking about the composite modulus with weaving and resin in a finished state?

Ok, I looked it up, plain weave with resin is around 6-7. Dragon plate with perfectly aligned fibers is claiming 33.

Regardless, it is good to have options of things that work.
 
Question

About 5 years ago a friend of mine installed 380 tires on his RV7. I fly this airplane regularly. With the 380 tires a shake would develop right around 19 KTS GS which required firm braking to slow thru that speed fairly rapidly to get the shake to stop. Lived with it but it was pretty annoying. Things tried to fix the shake: carefully balancing the tires, new tires instead of recaps, new wooden gear leg stiffeners, various tire pressures, tried one gearleg with a stiffener without the other having one thinking a resonance was developing between the two. Nothing really worked but there were varying degrees of effect on dampening the oscillation. That is, until the new idea we came up with and implemented a few weeks ago.


The layup was done with mains on the floor.

Why did you do the layup with the gear on the floor?
I am assuming with the weight on the floor the gear was somewhat bowed.
And most likely pinned in that position permanently.
Thanks for sharing your experiment!
 
+1 for carbon

I wrapped my 3B legs about 200hr ago. It works. Some details:
1- per Dan H suggestion, wrapped the bare steel with one layer glass.
2- to maximize bending moment of inertia, I made a carbon plate (3or4 layer) cut in a shape to match the rear dimensions of Vans gear fairing. Glue it to the back of the gear leg ( unloaded on the bench).
3-wrapped leg/rear former with several layers of carbon to form an integral fairing/stiffener.
I thought this was a worthwhile experiment after reading all the shimmy discussion. The carbon slightly ripples on the ground due to gear flex, but has not delaminated at al. Not perfect ( very law amplitude) on some runway crack spacing and speeds, but basically eliminates concern for shimmy.
 
I thought the carbon fibers was around 77(million), with steels in the 28-30 range? Are you talking about the composite modulus with weaving and resin in a finished state?

Ok, I looked it up, plain weave with resin is around 6-7. Dragon plate with perfectly aligned fibers is claiming 33.

Regardless, it is good to have options of things that work.

some typical woven properties

typical%20woven%20composite%20properties_zpsvkk7yyor.png
 
Why did you do the layup with the gear on the floor?
I am assuming with the weight on the floor the gear was somewhat bowed.
And most likely pinned in that position permanently.
Thanks for sharing your experiment!

I felt there was no need to stress the carbon layup where it spends 99.999%+ of the time. Didn't have to mess with lifting the airplane either.
 
it may not be a damper. bending stresses the outermost edges of a part. by laying carbon around the outside edges of the gear leg the gear stiffness is increased. the stiffness increase may push the natural frequency of the gear leg outside of the tire frequency input. then resonance is eliminated. the extra stiffness of the gear leg doesn't appear to cause adverse affect on landing. I don't know how long the bond joint from the gear leg to carbon will hold but it may be good for quite awhile.

thermal expansion coefficient of carbon is -1 to 1. the gear leg is probably about a 6 to 7. since the gear leg is tapered I expect the carbon to separate from the gear leg at the upper edge over time, but the wrap may serve it's purpose for quite awhile.
 
Last edited:
Indeed. Anyone care to speculate about how it works?

It alters the resonant frequency of the gear.

We lowered the tire pressure also. 35psi where we were running 45psi before. Harder to roll the airplane around but it lands like a RV-8 now. Much better mannered.
 
Last edited:
Thanks Bob

FWIW to others, Bob?s solution makes sense for several reasons. Stressing the carbon bond while flying only, rather than in the hanger is logical. Also, Any surface anomalies will only occur on the bottom side in flight if you choose to make a complete fairing shape of carbon like I did on my 3.

Natural frequency change raises the question, how much carbon is enough for simple wrap like Bob did. Bob, could you make a suggestion on carbon weight spec and how many layers ?
 
Different Potential Ways To Eliminate Shimmy

It's interesting that there are several approaches to eliminating shimmy -

1. Increase the stiffness of the gear leg, like the OP did.

2. Change the moment of inertia in one axis to decouple the fore/aft and lateral modes of vibration. Since the wheel and brake are off the axis of the leg, the motion might be a combination of those modes. This is what I think the wood additions do. I think Larry took this approach in a different way.

3. Actually add dampening, which I haven't seen yet.

4. Reduce or eliminate the forcing vibration that starts it, like AntiSplatAero does with their wheel balancing/bearing mod.

5. Adding balance weight at the forward end of the wheel pant. I'm not entirely clear on why this would work but several people have said it does.

Dave
 
Last edited:
About 5 years ago a friend of mine installed 380 tires on his RV7. I fly this airplane regularly. With the 380 tires a shake would develop right around 19 KTS GS which required firm braking to slow thru that speed fairly rapidly to get the shake to stop. Lived with it but it was pretty annoying. Things tried to fix the shake: carefully balancing the tires, new tires instead of recaps, new wooden gear leg stiffeners, various tire pressures, tried one gearleg with a stiffener without the other having one thinking a resonance was developing between the two. Nothing really worked but there were varying degrees of effect on dampening the oscillation. That is, until the new idea we came up with and implemented a few weeks ago.

Wrap the gear legs with 3K 5.8oz carbon fiber. No stiffeners.

Using a 48" roll of carbon fiber, squeegee wetted a 4ft or so square, 8" wide strips were cut using a rotary knife which were then rolled up and wrapped around clean gearlegs bottom to top in one pass overlapping 50% which amounts to two layers of carbon.


No peel ply or vacuum bagging was used as this was considered an experiment.

The layup was done with mains on the floor.

After about a dozen or so hours still no shimmy or shake at all and it seems to be holding up fine.

The gear feels absolutely perfect now with no tendency to rebound or skip. Tires are 35psi. In no way does the gear feel stiffer, but feels well dampened. For this reason I don't think any additional stresses are placed on the engine mount.

Any chance you could post some pictures (worth a thousand words)?

My RV7 has the same problem. Tried running lower tire pressure, which significantly help the vibration, although had started having problems with my tires going flat way to often.

Thanks!
 
My take

I think this is a great idea. Seems it makes the gear behave more like the rv8 flat gear legs where they are stiff(er) fore and aft, except the RV7 gear is now stiffer in both directions, I aint picky.
Anybody want to wrap mine for a small fee?

Any chance this will trap moisture between wrap and leg? Would solid carbon rod be better?
 
Last edited:
some material expansion numbers:

spring steel 6150 coefficient of thermal expansion = 6.8

woven fiberglass / epoxy coefficient of thermal expansion = 10

unidirectional fiberglass / epoxy coefficient of thermal expansion =7

woven carbon fiber / epoxy coefficient of thermal expansion = 3.4

unidirectional carbon fiber / epoxy coefficient of thermal expansion = -1

wooden gear leg stiffeners wrapped with carbon or glass would also provide stiffening in the fore / aft direction. the build up would be a stiffening beam.
 
Last edited:
gear leg vibration

Just a thought on vibrations. It would appear that the gear leg is acting as a mechanical amplifier to the wheel vibration. So, there are two solutions. First, eliminate the source of the vibration by balancing the tires (best method) or limit the ability of the gear leg to act as a mechanical amplifier. Whatever you put on the legs to change the resonate frequency needs to be firmly attached. I would think that you could add weight in selected spots that would change the resonate frequency or harmonics of the rod.
 
Just a thought on vibrations. It would appear that the gear leg is acting as a mechanical amplifier to the wheel vibration. So, there are two solutions. First, eliminate the source of the vibration by balancing the tires (best method) or limit the ability of the gear leg to act as a mechanical amplifier. Whatever you put on the legs to change the resonate frequency needs to be firmly attached. I would think that you could add weight in selected spots that would change the resonate frequency or harmonics of the rod.

If I was back in grad school this would present quite an interesting engineering mechanics derivation challenge. Natural frequency calculation of a tapered rod with a spinning mass at the end. Combining both bending and torsion. Now (42yrs later) I am puzzled and mentally challenged in a different way. :eek:

It is curious, though, that some planes have the issue and some do not. It would seem that the tires are the primary damper, and the pants are an additional spring mass part of the system. It can't be just the forced vibration element, as all planes would eventually have the issue, even if transient.

Dave: couldn't the mass center of the pants be off axis thereby converting a linear vibration into a torsional element?

With all the fresh brain power on VAF it is amazing no one has ever identified (posted) the precise root cause(s).
 
Last edited:
It's interesting that there are several approaches to eliminating shimmy -

1. Increase the stiffness of the gear leg, like the OP did.

2. Change the moment of inertia in one axis to decouple the fore/aft and lateral modes of vibration. Since the wheel and brake are off the axis of the leg, the motion might be a combination of those modes. This is what I think the wood additions do. I think Larry took this approach in a different way.

3. Actually add dampening, which I haven't seen yet.

4. Reduce or eliminate the forcing vibration that starts it, like AntiSplatAero does with their wheel balancing/bearing mod.

5. Adding balance weight at the forward end of the wheel pant. I'm not entirely clear on why this would work but several people have said it does.

Dave

I thought number 5 prevents flutter like counter balancing a control
surface. It would also not magnify shimmy.
 
If I was back in grad school this would present quite an interesting engineering mechanics derivation challenge. Natural frequency calculation of a tapered rod with a spinning mass at the end. Combining both bending and torsion. Now (42yrs later) I am puzzled and mentally challenged in a different way. :eek:

It is curious, though, that some planes have the issue and some do not. It would seem that the tires are the primary damper, and the pants are an additional spring mass part of the system. It can't be just the forced vibration element, as all planes would eventually have the issue, even if transient.

Dave: couldn't the mass center of the pants be off axis thereby converting a linear vibration into a torsional element?

With all the fresh brain power on VAF it is amazing no one has ever identified (posted) the precise root cause(s).

The difference between them might be observational. We all have threshholds of what "vibration" means. Airplanes are basically one yuuuge collection of vibrations. My 6A has always had a main gear resonance at about 18 knots. Solution is simple - don't taxi at that speed. Since it takes a second or two to develop, passing through 18 knots up or down is not an issue. Temperature is another variable - flat spots take longer to get ironed out in cold weather (providing inputs to the mass-spring-damper system), but the damping effect of the tires is greater. I've had slower speed resonances in winter, not sure what that exactly is from, since the mass is unchanged. Might be the flat spots exciting a different mode.

Additionally, there is the wheel fairing pitch oscillations that are easy to observe when watching an RV taxi. This can be pronounced on nose gear systems. The bearing friction (see thread on nose gear lessons learned) seemed to be a major contributor to this.

The wood (Van's suggestion at the time I built) adding stiffness in the fore/aft direction is a good solution since wood has a lot of internal damping. Not sure how it compares to composites of epoxy/glass/carbon. Building a couple slim sticks from various materials, and then qualitatively comparing them for vibrational characteristics might be interesting. Think of a ruler held near the edge of a desk and plucked.
 
Last edited:
Most mechanical assembly have a natural resonance at some frequency or other. Many things will affect that. In our case even the aerodynamic covers will have some effect. That this shows up in a single rod spring with no effective damping is hardly a surprise.

All we need to do is change the natural harmonic resonance. Changing the tyre pressure, sticking a bit of wood on, sticking a bit of Carbon cloth around the outside, changing the balance with a weight, adding a bit of toe in (or out) intentionally or more likely not during build, changing the tyre wear. Anything like that may work on one and not on another.
 
It alters the resonant frequency of the gear.

We lowered the tire pressure also. 35psi where we were running 45psi before. Harder to roll the airplane around but it lands like a RV-8 now. Much better mannered.

Ummmm. That would be TWO changes, negating any sure conclusion about the wrap. Have you experimented before and after at the same tire pressure?

If there is damping, it could be from carbon fiber moving slightly within the epoxy. Just a guess.

My guess too, if there is damping, i.e. removal of energy from the system.

I kinda lean toward Dave's note regarding torsional frequency. It's a spiral wrap, so fiber orientation is right for increased torsional stiffness. I'm not skilled enough to calculate the actual stiffness increase.

Just gut, but I don't see the wrap changing bending frequency very much.
 
Last edited:
Ok, so before and after the wrap were both at 35 psi. Cool.

Actually we're going to go back and try 45psi now that this seems to work. Just as another datapoint. 45psi makes the airplane easier to roll around. Have tried various pressures over the last few years to get the shake to stop.
 
Still working great! Did a few landings earlier this evening and they were all greasers. Zero shake.
 
On the side of reducing the driving forces in the shimmy, has anyone tried the beads people load into tires to make them self-balance?
 
Mark suggests wrapping (only if needed!) the titanium gear with fiberglass for a similar damping effect. He used several layers.

YMMV! Experiment on.
 
Confirming some points with Bob

Newbie to carbon fiber here:

1) do I understand that you put the wraps on wet? Does wrapping dry cloth (or tape) around the gear legs and then wetting out just make no sense?

2) total number of layers a minimum of 6? One pass with 50% overlap is two layers and there were "3 or 4 layups".

3) presuming the answer to #2 above is "yes", did you let the a layup totally cure before the next wet wrap?

Thanks,
Larry
 
Carbon

It's my understanding that you put the wraps on wet..and apply additional wraps on top of wet ones. The reason is that if you let it cure, additional layers are only adhering mechanically, not chemically bonding. As for starting wet, my limited experience with CF is that you can't easily tell when you have saturated the cloth like you can with Fiberglass. . Fiberglass goes clear when wet. CF is so black, you can't easily tell. Also, you would basically be packing air into the layup if you added epoxy over a dry wrap. Look how the original poster wet out a sheet, then applied the wet strips..
 
Newbie to carbon fiber here:

1) do I understand that you put the wraps on wet? Does wrapping dry cloth (or tape) around the gear legs and then wetting out just make no sense?

2) total number of layers a minimum of 6? One pass with 50% overlap is two layers and there were "3 or 4 layups".

3) presuming the answer to #2 above is "yes", did you let the a layup totally cure before the next wet wrap?

Thanks,
Larry

1. You could do it dry, no overlap, and wet in place but wet the gearleg first. You would then have to do the same thing a second time reversing the direction.

2. There were 3 or 4 layups because the strips were cut from the width of the roll. Only two layers of CF.

3. Did it in one pass.
 
Landing Gear

Wittman Tailwind, 100# plus lighter than typical RV, Landing gear is 13/16" diameter at small diameters as opposed to 7/8 for most RV's. Sam James wheel pants. Goodyear tires with wheel/tire statically balanced.
hardwood stiffners taped in place with filament tape. Shake was minimal without wheel pants, pretty bad with pants installed. Seems to be mostly on the left side.
Balanced the left wheel pant which improved things considerably but did not eliminate the shake.
Shake starts at 25 statute on the GPS, I use moderate braking at that point and the shake quickly stops.
I can see the left wheel pant and before balancing the "tail" of the wheel pant was moving plus and minus 1" maybe a bit more. Tire pressure in 30-35 range, higher than 35 is worse.
Most recent landings have been on paved runways I would describe as poor. Lots of seal coating over numerous cracks.
I am going to do a few more flights with and without the wheel pants as well as a couple of landings on smooth runways.
Then I will try the carbon wrap.
One interesting point- similar gear installed on biplanes but with no "sweepback" does not seem to have this issue.
It has been an interesting test of my welding on the mount. No issues so far.
 
Did not work for me - another wrap, or?

Was really excited to try this. RV-7, 380 tires (since day one) and had shimmy that was mostly corrected with wooden stiffeners and 40 psi, but I would like it totally gone. I used 3K 5.8 oz CF 6" tape, West Systems Epoxy, 50% overlap, gear on the ground. The CF was installed wet over lightly sanded and acetone cleaned powder coated gear legs, that were lightly brushed with epoxy immediately prior to the layup. The shimmy is now much worse than with the stiffeners and about two thirds of what it was without anything.

Did I do something wrong?

Does anyone think another two layers of CF would help?

Right now, I'm going to reinstall the wooden stiffeners over the CF and see what that does.

Thanks,

Merrill
 
would the entire thing made out of fiberglass also do the trick?
Carbon Fiber is about 3 times stiffer than fiberglass and also stronger. Stiffness is the key here. Could fiberglass work? May be.

The down side with wrapping the gear regardless of material is you make the gear leg fairing fatter and not as streamlined.
 
Idea

I've got quite a ways before even first flight but this is an interesting thread.
Is it possible to combine the harmonic dampening of the wrap with the functionality of the gear leg fairing into one component? Thinking outside the box, but maybe a fairing filled with foam conforming to the gear? Just a crazy thought.
 
Directional stiffness

It seems to me that one would want the stiffness in the fore and aft direction, but not in the side to side direction. So would it be possible to lay some carbon fiber in the seam of the wood stiffeners? This would produce a thin web side to side, but a thick web fore and aft, selectively stiffening the fore aft movement of the gear leg.
 
Back
Top