testing a wing

[JLSalas]

Hi, I am building a new wing for a plane I've crashed. I am trying to test the wing as follows:

I put a chair under each wing tip and then I start to put a load in the middle of the wing.

The plane will weight about 5 lbs, and until now my wing can resist about 13 lbs without breaking.

The wing weights about 0.8 lb.

It's a pattern plane. How much weight does my wing have to support? It's enough with 13 lbs?

thanks,


JL

[Majortomski]

How heavy is the pattern plane when it is ready to fly?

Now multiply that by 20 to handle any possible G force you will encounter.


HTH

T

[JLSalas]

Thanks for your answer. The plane will weight about 5 lbs.

I know aerobatic planes can support about 9 to 10 g so your suggestion is a very valid one.

My test puts all the load in a very small area of the wing (the center) and the "lift" is also in a very small area (the tips of the wing). In real flight the forces are spread on all the surfaces of the wing. Due to this, I think (correct me if I am wrong) my wing can resist a heavier load in real flight. I am trying to find a way to calculate how heavier.

Also, the wing flexes a little bit while I am performing the test.


JL

[Andrew McGregor]

Flex is OK, it's going to flex in flight too. As for spread load, it doesn't matter... everything is taken up in that 20g safety factor anyway.

[Ollie]

Wing spar need strength estimation is a cookbook approach that has some simplifying assumptions and aproximations built in.

Take 1/4 of the span in inches and multiply it by half the G's in pounds. This is the maximum bending moment at the center of the wing in inch-pounds.

Divide the maximum bending moment by the (top to bottom) spar depth in inches. This is the force at the center of the wing required of the the spar caps, in pounds.

Find the strength of the spar cap material in pounds per square inch. Most fibrous materials like unidirectional carbon fiber in epoxy and wood are about twice as strong in tension as in compression.

Divide the force in the spar caps by the strength of the sparcaps to get the required crossection of the sparcaps.

The required crossection of the spar caps decreases as the square of the distance, to zero at the tips.

The shear force at the center of the wing is half the G's and decreases linearly toward zero at the tip

[ronmeister]

JLSalas, your going to need the wing to take more than that definitly, either make the plane lighter, or put a carbon spar on the wing.
I think the carbon spar will help, and will be a lot easier too.

[vintage1]

Remember that the lift a wing has is not at the tips. Never seen a wing fold except near the center, yet.


On average its about halfway to the tips,so its better to do a static stress/strain test with the wing supported mid-span.

20g is a safe target to go for..the wing should fleax, but not excessivly.

The actual stresses on a wing are IIRC parabolic..from zero at the tip to a maximum at the center section.

You can make a lighter weight structure - especailly if you can laser cut ribs - taking note of this by using a tapered top and bottom spar and varaiable sized holes cut in the outer shear webs.

[Sparky Paul]

Most of the bending load on a wing is from the 12% span point to the root.
Supporting the wing at the tip and loading it gives a poor indication of what load the wing could support.

[avianaut]

This wing is from a World Models Super Stunts .40. I modified it by adding spar webs of two ply balsa with diagonal grain, similar shear webs in the trailing edge and two servo bays each side. I also added hardpoints at mid-span to carry drop tanks, or whatever. I wanted to be sure, so this is how I tested it.

Supported at mid-span on two chairs, with four jugs of fuel suspended on the wing attachments. Total weight is about 32lbs. Fuselage weight about 4lbs. Should be good for 8 G's. The wing didn't look like breaking, but my nerves did!

I also tested the hardpoints by doing the same test with the wing supported at root and tip and the jugs at the hardpoint. No problems.

First pic shows the mods, second pic the test. Flew fine, still in one piece.

Cheers.

[JLSalas]

Thanks to all. I am learning a lot here.

Some comments, questions:

The wing has 2 carbon fiber tubes as wing joiner and a "I" beam spars. I've put 13 lbs and the wing didn't break, but I liked to have some advice here to confirm or not if my test were valid.

Ollie: Thanks for your method. I was trying to find something like that. Is there any table with the pounds per square inch for the most common materials used, like balsa, ply and carbon fiber?


Vintage1: I made the same test with the wings supported at the middle of each wing half, and I put 23 lbs without breaking anything. The bending is little.

Sparky Paul: Thanks for your graph. Your comment about the test method confirms my initial thought that the test wasn't a good one.

Avianaut : "The wing didn't look like breaking, but my nerves did!" I understand you absolutely!! I made my tests with milk boxes (1 kg or 2.2 lbs each one).

JL

[Ollie]

JL,

The wood spars depend height between sparcaps, webs between the sparcaps on the type of wood and density of the wood, especially the density balsa. Balsa at density 10 pounds of cubic foot, has a compression strength of 1700 pounds of square inch (PSI). Sitka Spruce has a compression strength 5610 PSI.

For a maximum lifting wing is (5 - 0.8) weight x 10G's = 42 pounds. One side of the wing's lift in load is 21 pounds. I need the span to calculate bending moment and dimensions of spar caps and web between them.

I am quessing a span of 48". So the bending moment 21 x 12 = 252 inch-pounds.

I am quessing a spar height of 1.5 inches and 1.25 inches between the spar caps. So, the maximum bending moment at the center of the wing the compression load of spar cap is 252/1.25= 202 pounds! The crossection area of the spruce spar cap is 202/5610=0.036 square inches. So, the Spar cap might be 5/16" X 1/8" = 0.039 square inches. The web maximum load is 21 pounds. The web with balsa has about 200 PSI shear strength. So, the crossection area is 21/200 ~ 0.1 square inches and 3/32" sheet.

If you wanted balsa spar caps, use rock hard balsa (16 pounds per cubic foot density) with 1/2" x 3/16".

[avianaut]

One thing I wonder about - just how many G's does a model really pull? If I knew that I would use the information above, and add about a 50% margin, as the full scale guys do, to allow for variables such as glue and wood quality, lack of precise stress data etc., and design for that. And then test it as I do.

Thanks Ollie, for running the numbers, I find it usefull.

Just a note. If your spar caps are joined at the centre, as many are, the joint must be able to take at least the same stress as the spar caps. Obvious, but it's a weak point in many I've seen. Then if the spars join at an angle, such as dihedral or sweep, there will be additional stress on joints, centre rib and spar web at the root. Even more margin is prudent.

Cheers.

[Ollie]

If the wing has two panels jointed at the center, then the joint detail needs rational design. The joint carries the bending load.

I like to design to standards rather than testing to failure.

See:
http://www.woodbin.com/ref/wood/strength_table.htm
http://abcdpittsburgh.org/kids/kids.htm
http://www.paulowniasupply.com/paulownia_vs_balsa.htm

[Sparky Paul]

G in flight....

[avianaut]

Ollie, good links, thanks again. "Design to standards" - I agree. Then test, to be sure! ;o)

FWIW, the Airbus A380 wing recently failed it's static test. I'm sure it was well designed to very specific standards. I guess they cut it a bit fine.

Cheers.

[avianaut]

Sparky Paul, thanks for responding to my question. It's usefull, but I don't have the ability to measure what speed my models fly at, rate of change of direction, etc.. I don't even have scales to weigh them accurately. With so much guesswork, my calculations will have a wide margin of error.

I wonder if I could rig up a 'G' meter of some sort and fly it in a model? Surely that would remove some of the guesswork. A little project for another day.

Cheers.

[Sparky Paul]

Sparky Paul, thanks for responding to my question. It's usefull, but I don't have the ability to measure what speed my models fly at, rate of change of direction, etc.. I don't even have scales to weigh them accurately. With so much guesswork, my calculations will have a wide margin of error.

I wonder if I could rig up a 'G' meter of some sort and fly it in a model? Surely that would remove some of the guesswork. A little project for another day.

Cheers.
.
Naut, your plane won't be much different than any others...
Extreme g manuvers aren't all that common.
I've measured a lot of airspeeds...

http://www.angelfire.com/indie/aerostuff/inflights.htm
Calculating this stuff is more an exercise in self-fulfillment than anything useful, as we generally have no way of knowing the airspeed, which is the first performance parameter.
Most of us just figure that "the last plane built in this manner didn't break, the next one built the same won't either."

[avianaut]

Paul, I sometimes wonder if models are built more for crashworthiness and abuse than for structural efficiency. Or maybe for economy, ease of assembly, or even just for appearance? I guess it doesn't really matter so long as it holds up, unless you really have a specific requirement. I am just curious. Thanks again.

Cheers.

[Sparky Paul]

Models are built for crashworthiness more than anything else.
Real planes don't land on a wing tip and a prop... and then fly again in 5 minutes. :)
A model built to the exactness of a full-scale structure-wise is rare. I believe Dr. Mark Drela does his planes to that precision.. but that's too much work for me.
I trust his designs will survive anything I might do to them.
Workable structures.. those that will survive a typical model's environment are no doubt overbuilt.... but the lack of major structural damage in normal use makes up for that.. as does adding power. More is better! :)
IMAC aerobatics probably stress a plane more than any other. The "Blender" can rip the fuselage off the wings. The next plane should be built better when that occurs. :D

[JLSalas]

This thread is becoming a very useful one.

Ollie: Thanks for your example. My spar height is 2.5 inches, and my wing span is 59 inches, so I calculate the bending moment as 275,3 inch-pounds.

Avianaut: My wing has the spar caps joined at the centre, and is joined at an angle, due to sweep. I joined the wing with carbon fiber tubes.

Sparky Paul: You web site is very interesting.

JL

[peterangus]

JL

How do you intend to test the torsional stiffness?

[JLSalas]

My wing is not covered yet, but it bends a little when I apply a torsional force.

Really I don't know how to test torsionally the wing.

Can anybody tell me how to test it and what values are good for the test?

JL

[peterangus]

Really I don't know how to test torsionally the wing.

Can anybody tell me how to test it and what values are good for the test?


I can't tell you. I doubt that anyone can. And there's no need.
--------------------------------------------------------------------------
Look at the enormous diversity of successful models now flying, with an equal diversity in wing structural forms, each one well suited to its task. From microfilm aircraft to 200mph soarers.

Natural selection in action.

Do what everyone else does; copy success.
--------------------------------------------------------------------------
I don't doubt that there are many '40 powered models flying well with wings similar to yours. But there could be trouble in store if you fit a larger motor and fly faster. Not only might you pull higher "gs", but speed-related troubles could show-up. In this case, watch out for a decrease in aileron effectiveness at speed. This would indicate wing twist due to aileron loads. Ailerons can become zero effective because of wing twist; their effect can even reverse.
--------------------------------------------------------------------------
With an open structure [as yours], torsional stiffness depends mainly on the covering. Film and fabrics have limited reliability. The usual enhancement is to cover with sheet balsa forward of the spar to form a closed torsion cell [a D box].
--------------------------------------------------------------------------
The point I'm trying to make is that wing bending, though important, is only one small [and relatively easily understood] aspect of wing structure design.

[JLSalas]

Peterangus:

Thanks for your post. It shows another very important aspect of wings nature.

I found very useful your description about wing behavior under twist loosing aileron effectiveness due to speed.

"Natural selection in action."
I hope my wing can survive ;)

JL