Not-so-solid massive core wings: Lightening the core foam

[wsimpso1]

No, I'm just joking. But I did think about ways to get a little reinforcement onto the other side of that foam (and it wouldn't take much). But I will speak no more of it, now.

Mark

I was not joking. The materials and processes exist. Whether you would end up saving any weight is another matter... Grin.

 

[Vigilant1]

If you are determined to play with it empirically, might I suggest making full scale models of your cutouts same foam and composite skins, and pressurize them internally with a water column or bicycle pump so you can ramp pressure up and down accurately, then record deflections and first failures

Thanks, Billski. As you guessed, I've been thinking about this caveman approach. Three observations if I may:
1) On the real wing nose, we'll have very high negative pressures (I.e below static pressure) on the front and top of the outer skin, possibly positive pressures underneath. This produces an overall upward bending moment transferred through the skin to the spar. It would seem I'd need to actually pull on the top skin to replicate this in a physical model (both the local pressure on the foam and the overall stresses through the entire section to the spar). I would replicate the forces with weights mechanically attached to the skin. It would be much easier, though, to do as you describe and just add positive pressure to the void area, if that would capture the significant loads.
2) section size: I've been envisioning a fairly moderate model with a full scale wing chord (36-48"l but just a 4-6" thick slice.

And the hammer tests...

3) Happy Thanksgiving to all. Lots to be thankful for, including the abundant patience of so many folks.

 

[wsimpso1]

In Rutan solid core wings the foam is only there as buckling support for the skins and spar flanges. The characteristic for the foam to do its job is rigidity not strength. Unlike a classic sandwich where the foam/core takes the shear load the Rutan wing has BID wrapped all around, and this takes the shear loading leaving the core as only support.

In a Rutan wing the core is particularly effective in supporting the skins and as such enables both the facings and web to take virtually its full ultimate tension and compressive load without prematurely buckling or wrinkling. I have back analyzed to verify this.

Now as you hollow out the foam the buckling support becomes compromised. At some point when the thickness of foam between the skin and hole is small enough the skin/ facings will wrinkle (before its ultimate load) either inward or outward into or above the holes in the foam. This is a well understand phenomenon in composite structures called wrinkling. The same thing happens when you choose a very low density core. The facings will pull out or dig it to the foam causing wrinkles or ripples. In honeycomb there is somewhat related phenomenon called dimpling where the facings dimples into the hole.

So the question is how big can your holes in the foam get before wrinkling occurs. The only way to know is testing. There is no analytical solution to this. FEA can give you an answer but is of little value if you cannot validate with tests. Botton line test.

PS. by the way wrinkling in classic sandwich structures has a pretty simple formula (Hoff formula) but does not apply to Rutan type wings (with or without holes).

Oh, there is ALWAYS modeling that can be done. I suggest that it can be useful once adjusted by a test or three. For uniform thickness, you can go to Roark's table 11.4, cases number 8a and 8d seem most appropriate, but you have to back out of homogenous plate theory and into your sandwich plate to get there. Knowledge of plate/shell theory as applied to composites is required. FEA is also possible. Both have their limitations. I sure would want to apply the results conservatively and do full scale tests for both strength confirmation and for fatigue checks. Pioneering..

Billski

 

[stanislavz]

I would replicate the forces with weights mechanically attached to the skin. It would be much easier, though

It will create a lot of point loads and will peel skin from foam at 1/10 of actual force... You will have to build you negative pressure chamber

 

[wsimpso1]

What would happen if the planes wing mass center was at 18-20% position of the chord ?

Why would you want to do that? I suspect it would take ballast to do it...

Once you get to 25%, the foil becomes stable in pitch (no divergence) and resistance to flutter. AFAIK driving it further forward does not have pay value. It will most likely take additional weight to do so, and will likely add to the mass moment of inertia about the shear center - both lower the twist mode natural frequency as well as the flap mode natural frequencies, all of which is usually directional wrong for avoiding flutter.

Billski

 

[Vigilant1]

It will create a lot of point loads and will peel skin from foam at 1/10 of actual force... You will have to build you negative pressure chamber

You just want to make this crazier!
I'll provide a sketch later. Spread the load over the skin, hanging weights-- it will look like something from the Spanish Inquisition, but at mouse scale.

 

[blane.c]

If you are not concerned with cost and want low drag and low weight, you can build in graphite fiber. Maybe even lighter than the wooden wing and can be made as laminar as any other airplane ever built.


Billski

Would it really be that much more expensive materials wise? Or another way to put it, would most of the expense be in the man hours?

 

[stanislavz]

You just want to make this crazier!

I think we need a video call to settle down this problem

 

[Speedboat100]

Why would you want to do that? I suspect it would take ballast to do it...

Once you get to 25%, the foil becomes stable in pitch (no divergence) and resistance to flutter. AFAIK driving it further forward does not have pay value. It will most likely take additional weight to do so, and will likely add to the mass moment of inertia about the shear center - both lower the twist mode natural frequency as well as the flap mode natural frequencies, all of which is usually directional wrong for avoiding flutter.

Billski

I could have batteries in the wings. To load the wing "correctly".

 

[ragflyer]

Oh, there is ALWAYS modeling that can be done. I suggest that it can be useful once adjusted by a test or three. For uniform thickness, you can go to Roark's table 11.4, cases number 8a and 8d seem most appropriate, but you have to back out of homogenous plate theory and into your sandwich plate to get there.

Billski

Sure your point is valid and there is always modeling to be done. My point was more for the wrinkling issue than the membrane stresses. As a research uni project I would first build a FEA cross check with any available theory and then embark on testing program and then create an empirical theory/curves for practical design. For a one off average amateur design testing is more fun and realistic.

btw I have the sixth edition of Roarks and 11.4 does not map. I have a 12.3 that refers to combined membrane and bending in thin pressure vessels. Is that what you were referring?

To recap there are two issues with holed core: membrane stresses due to pressure differential and wrinkling of skin due to limited foam support. The former (membrane stresses), I do not believe, is an issue. The KR series airplane flew with one sided faced foam about 1" to 2" thick. In other words there was a giant hole as wide as the distance between the spars and a few feet span wise between the ribs. As long as you leave about an inch or two between the facings and the holes and leave the spar area clear you should be fine. The latter will need testing to be sure, though frankly if you make the holes aft of the spar I think you should be fine.

 

[Vigilant1]

This might be a two-part testing problem:
1) "Inflation": measure skin deflection and determine foam integrity due to pressure differential between the voids and the outside. This might be done with positive pneumatic pressure in the voids.
2) Sheet Buckling: Determine effect of various foam thicknesses on surface buckling of simple panels and simple 2-D shapes when loaded as they would be in a wing. This might be done be pulling or pushing from outside the wing. Probably pulling is best. Attach stationary spar (timber) firmly to wing foam and skin.

Oh, and 3) Ball peen testing for toughness.

 

[stanislavz]

This might be a two-part testing problem:
1) "Inflation": measure skin deflection and determine foam integrity due to pressure differential between the voids and the outside. This might be some with positive pneumatic pressure in the voids.
2) Sheet Buckling: Determine affect of various foam thicknesses on surface buckling of simple panels and simple 2-D shapes when loaded as they would be in a wing. This might be done be pulling or pushing from outside the wing. Probably pulling is best. Attach spar side firmly to big timber.

Oh, and 3) Ball peen testing for toughness.

3) Panel buckling due shear load - torsion load of wing.

 

[wsimpso1]

Sure your point is valid and there is always modeling to be done. My point was more for the wrinkling issue than the membrane stresses. As a research uni project I would first build a FEA cross check with any available theory and then embark on testing program and then create an empirical theory/curves for practical design. For a one off average amateur design testing is more fun and realistic.

Testing is fun, but I have a strong preference to make one test article instead of a bunch.

Wrinkling/buckling/crippling. Hmm sounds like a trip to Bruhn and then take apart the plate theory for homogenous material, then put it back together again with composite theory to get even close. I am not doing it, but it might be useful to folks talking about doing it.

btw I have the sixth edition of Roarks and 11.4 does not map. I have a 12.3 that refers to combined membrane and bending in thin pressure vessels. Is that what you were referring?

Nope, Roark's Seventh Edition Chapter 11 is about flat plates. Chapter 13 is Shells of Revolution, and could well be useful for this same purpose. Chapter 11 is for flat plates, not the gently curved plates in one direction that are our wing skins, but they are thin (3/8" on my roughly 12" x 30" panels) and constrained on the edges, so it seems a reasonable approximation. Chapter 13 is for shell of revolution, so they are curved in one axis similar to our wing skins, but there are no appropriate edge constraints. I picked Chapter 11.

To recap there are two issues with holed core: membrane stresses due to pressure differential and wrinkling of skin due to limited foam support. The former (membrane stresses), I do not believe, is an issue. The KR series airplane flew with one sided faced foam about 1" to 2" thick. In other words there was a giant hole as wide as the distance between the spars and a few feet span wise between the ribs. As long as you leave about an inch or two between the facings and the holes and leave the spar area clear you should be fine. The latter will need testing to be sure, though frankly if you make the holes aft of the spar I think you should be fine.

I had forgotten that the KR's were not skinned on both sides of the foam. Now that I am thinking about it, I recall that the KR wings were wooden structurally - wood spars, wood ribs, then filled in with foam and finally glassed with 4 oz cloth. The originals were not even glass, but Dynel, which is acrylonitrile and vinyl chloride copolymer of modest strength and stiffness, and is a sponge to resin. Many suppliers recommend using it only with vacuum bagging to keep weight down Think of the KR's as a fabric covered wing, but with thick foam backing up the fabric. It does give us some hope that a reasonable amount of foam can be left on the ground without being glassed on the inside...

Billski

 

[Vigilant1]

3) Panel buckling due shear load - torsion load of wing.

I would think I could include normal twisting moments (due to Cm) in the second test. Now, tip-to-root twisting ("divergence") is a different animal and requires a whole wing. But, those loads are taken by the skins, and if the foam has stayed stuck through both other tests and is still providing support against buckling, and if the magnitude of the twisting forces is less than the other in-plane compressive forces on the skins, then perhaps all is well. .

 

[ragflyer]

Wrinkling/buckling/crippling. Hmm sounds like a trip to Bruhn and then take apart the plate theory for homogenous material, then put it back together again with composite theory to get even close. I am not doing it, but it might be useful to folks talking about doing it.
Billski

Wrinkling is well covered in more specialist composite texts. On standard Rutan full core wings wrinkling is not an issue but in rib and composite sandwich skin wings its entirely another matter. The sandwich skin (2lb density foam and Rutan bid) will wrinkle well before you get close to the ultimate compression and shear shear stresses of the facings.

 

[stanislavz]

But, those loads are taken by the skins, and if the foam has stayed stuck through both other tests, and if the magnitude of the twisting forces is less than the other in-plane compressive forces, then (I think) we'd be okay.

I am out of this question at the moment. I know that it exists, and is one of the primary loads for skin on cantilever lightplane. For my not-correct composite wing with V strut with jury struts - top skin will have 0/90 fibers, for less deflections due to Cl, bottom 45/-45, to act as drag/antidrag wires.

 

[BBerson]

I copied the KR-1 wing specs exactly from my plans in post 93. (I guess nobody noticed?)

 

[ragflyer]

I copied the KR-1 wing specs exactly from my plans in post 93. (I guess nobody noticed?)

Sorry... you deserve the credit
now that you have the plans for KR.... is the rear wood spar pin joined or joined exactly as the fore spar? To clarify my question is at the fuselage bulkhead not the stub wings

 

[BBerson]

is the rear wood spar pin joined or joined exactly as the fore spar? To clarify my question is at the fuselage bulkhead not the stub wings

It has a center section. Both spars attach at outer wings root. The rear spar is single bolt.

 

[cavelamb]

Just a quick and dirty idea, but basically attache the internal structure to the spars yes. ... that doesn't mean that the outer skin cannot or does not have structural carry through to the spars especially if it is necessary.

Take a look at KR2 wing construction.
It's pretty much what you are describing, although through different techniques.

Foam is attached to the primary structure and hotwired to shape.
The inner ace of the foam core is just raw foam.
The foam supporting the skin has to be thick enough to pass skin loads to the spars.

It's easy to build, but not ideal structurally.