Composite wing spar web design - composite webs, foam webs, etc.

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

User27

Well-Known Member
Joined
Apr 5, 2015
Messages
106
Location
England
This is a great thread from my perspective. I'm an aero (handling qualities) guy who moved into autopilots and then systems development, so structures is something I last looked at when I graduated in the mid 80s! I'm now looking at building a 300kg single seater based on an existing wooden design. I need to save some weight to be sure of achieving 300kg TO weight with a useful payload.

It occurs to me that loads estimation is important in designing an optimised structure and is something of a mystery for me. The basic bending & shear cases appear (to me at least) to be straight forward once the n-V diagram (with gust loads) is drawn and a lift distribution assumed. But the other less intuitive load cases, torsion, control surface loads, u/c loads and so on seem much more difficult to estimate. Is this an appropriate subject for this thread? Should I start a new thread? Is there a text book anyone can recommend?

My current scheme is a wooden spar with carbon uni reinforced caps, foam ribs and plywood skins, partly for the ease of construction. Full composite spars seem more difficult to build than a wood/carbon composite. I'm also looking at a 3 piece wing, with the joint outboard of the undercarriage mount, to allow the outboards to fold and reduce hangar costs. Still very much at the outline design stage.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,945
Location
Saline Michigan
I took these to be elastic modulus and if so GPa would be the correct unit. However, these convert to 24,600,000 Psi (24.6 Msi) and 20,300,000 psi (20.3 Msi) and are high numbers for modulus ..... well high for standard carbon uni but they are in the range that might be expected for a high modulus carbon. We would need to know what material they are for and how they were measured to comment further.
In the MPa range, these would probably be strengths, not modulus.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,945
Location
Saline Michigan
I think we can at least trust Drela's test data.

Sky Pup spars use blue Styrofoam as the sheer web in most of the spar and also for some of the fuselage. If silicone was common in Dow building insulation foam, we'd have heard about Sky Pups falling apart, I think. OTOH, if I was building one I'd want to know that silicone in the foam was extremely unlikely. I guess one could compare adhesive strength or something. Or ask Dow with a properly worded question.
Trust Drela on structures? Not me and for good reason. When I read his writeup, I was able to forecast that he was going to have "surprise" failures, which he did. Same thing with Hollmann's early test failures at Lancair. Drela's specific beef-ups to make up for his "surprise" failures may or may not be scalable from his RC models to your airplanes. Scaling his fixes correctly will require that you know enough about beam design to do your own anyway.

Does the Skypup specify insulation board or flotation billet? The difference is significant. Dow flotation billet (the large cell stuff) as used for airfoil cores was known back when Rutan was selling plans to not use any silicones. Might still be true today. Insulation board appears to be an unknown.

The issue is that foam has to stay well attached to the fiber-resin facings or it is not supporting the facings against buckling. Silicones and mold waxes that *might* be used in processing extruded foam can cause delamination and then other really-bad-day events.

A good plan when preparing to use an unknown foam is to build some test pieces, laminate samples with your intended resin and cloth, cure it, post cure it, then peel the lamina from the foam. If it takes parent foam everywhere, good. If the resin comes off the foam clean anywhere on the test pieces, AVOID THAT FOAM. I have tested batches of flotation billet and insulation board before making airplane parts with it. I have also tested Divinycel - no problems there. One stack of light blue insulation board that I got as a gift is "tooling only" for poor adhesion. Consider yourselves warned.

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,945
Location
Saline Michigan
This is a great thread from my perspective.
Glad you like it!

It occurs to me that loads estimation is important in designing an optimised structure and is something of a mystery for me. The basic bending & shear cases appear (to me at least) to be straight forward once the n-V diagram (with gust loads) is drawn and a lift distribution assumed. But the other less intuitive load cases, torsion, control surface loads, u/c loads and so on seem much more difficult to estimate. Is this an appropriate subject for this thread? Should I start a new thread? Is there a text book anyone can recommend?
Loads are the starting point for everything structural. The old version of CFR14 Part 23 had a whole section defining load estimation for structures. It is available on line and in searchable forms too. Sorry, do not have a pick for it. There are also LSA and Euro standards for this...

Torsion is distributed along the span like lift is, usually elliptical and sums from tip toward root or paired struts like shear does. Control surface design loads based upon wing loading and g factors for the airplane are part of the CFR14 Part 23 or the Euro standards.

My current scheme is a wooden spar with carbon uni reinforced caps, foam ribs and plywood skins, partly for the ease of construction. Full composite spars seem more difficult to build than a wood/carbon composite. I'm also looking at a 3 piece wing, with the joint outboard of the undercarriage mount, to allow the outboards to fold and reduce hangar costs. Still very much at the outline design stage.
We have talked about the wooden spar with carbon reinforcements many times. Trouble is that if the graphite is carrying much load, the wood becomes a heavy core. If you need to save weight, this is not usually a beneficial scheme. In rag and tube airplanes, the base design is usually pretty hard to beat for weight. If you are to save much weight, a composite spar with graphite rod caps (Marske Graphlite rods) is probably a good path..

As someone who is well on his way with a self designed airplane, I urge you to review your goals, the materials you prefer to work in, and the available airplanes that suit your mission and regulatory environment, and then build that airplane to plans. You will be flying far sooner, with more joy, and with far less headaches. Self-design and then construction of the whole airplane or even of the wings is a huge add to the project.

On the other hand, if you want the intellectual challenge and don't mind that you may never fly the thing, have at the design-your-own work.

Billski
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I said we could trust him on test data, not "on structures", but, apart from the test data, I'd have to have what he wrote in front of me to decide whether I trusted it. Mark is one of tbe most thorough people I know, with a bunch of successful, built projects to his credit, so I'm not going to dismiss his work without careful examination.
---------
In the case of the Sky Pup, insulation board. I'd have to check the build manual, but I think maybe Dow in particular. Certainly a good idea to check adhesion. Was the silicone on the faces or all the way through? If I'm not mistaken, in some applications the foam is meant to be glued.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
14,675
Location
Memphis, TN
I think it’s overthinking the wing on a SkyPup. You may consider the ribs as composite on the SkyPup. I’m sure the designer was thinking shortcut to truss ribs; no more. Got out of that job, while bushing hands off. If it lasts ten years before you run a wing into a barn or ground loop, you got good life out of it. It was not designed to be a 100 year airplane and then be worthy of restoration. I think it’s apples to oranges to compare a SkyPup rib to a Lancair. You are free to but not supposed to.
 

User27

Well-Known Member
Joined
Apr 5, 2015
Messages
106
Location
England
We have talked about the wooden spar with carbon reinforcements many times. Trouble is that if the graphite is carrying much load, the wood becomes a heavy core. If you need to save weight, this is not usually a beneficial scheme. In rag and tube airplanes, the base design is usually pretty hard to beat for weight. If you are to save much weight, a composite spar with graphite rod caps (Marske Graphlite rods) is probably a good path..

As someone who is well on his way with a self designed airplane, I urge you to review your goals, the materials you prefer to work in, and the available airplanes that suit your mission and regulatory environment, and then build that airplane to plans. You will be flying far sooner, with more joy, and with far less headaches. Self-design and then construction of the whole airplane or even of the wings is a huge add to the project.

On the other hand, if you want the intellectual challenge and don't mind that you may never fly the thing, have at the design-your-own work.
Thanks for that - I have become aware that as retirement is not that far over the planning horizon I have only around 15 years to build and enjoy flying an aeroplane I have designed myself - or at least designed a major part of. Clearly there is no point in spending 5 years designing, 5 years building, 5 years tweaking and zero years enjoying, although the design, build and development will be fulfilling in their own right. I am trying to figure out exactly what my goals are to give me the best chance of achieving them. This site is a huge help in that regard.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I think it’s overthinking the wing on a SkyPup. You may consider the ribs as composite on the SkyPup. I’m sure the designer was thinking shortcut to truss ribs; no more. ...
Not just the ribs, but the spar as well, and other parts.
 

Exian

Well-Known Member
Joined
Jul 26, 2018
Messages
53
Location
BORDEAUX
Tension 170 GPa (60% fiber in volume)
Compression 140 GPa
Yes, those are Elastic modulus.

From Toray web-site :
Typical "HR" (high resistance) fibers are T700S, modulus 230 GPa, strenght 4900 MPa. At 60% fiber content your composite should have a modulus close to 140 GPa, strenght of 2900 MPa.
What I use is indeed "IM" (intermediary modulus) fibers, modulus 290 GPa, strenght 5500 MPa.
Modulus and strenght (tension) of composite we get are consistent with the fiber content.

But few talk about compression properties. As said, compression modulus is a little lower, and compression strenght is a lot much lower than in tension.
For the fiber I use, compressive strenght is only in the 700 MPa range for the composite (60% fiber content).

If we go back at my design practice :
For + G loads, the upper spar cap (compression) has 1,2 the cross section of the lower spar cap (tension). I do this because of the modulus difference, not the strenght!
Max elongation strain in the spar caps it thus the same (and minimised for the shear web), the neutral fiber is at the middle of the spar.
Because of the difference in modulus, max stress in the spar cap that is in tension is 1,2 that of the one in compression.
My design criteria is only the compressive strength : 700 MPa. It means that max stress in tension is only 840 MPa when this material could probably reach 3300MPa (5500 MPa x 60%).
Of course, I check - G loads and deflections, but it all works out with what I do for the + G loads.

This is at ultimate load = 2,625 limit load in my case. So at limit load, stress are far under the 400 MPa for fatigue according to CS-VLA.


I heard of some sailplaine manufacturer using now "HM" (high modulus) fibers.
HM fibers have lower strenght in tension, but not so much in compression, and high modulus is a big help to reduce deflection and minimise the effect on the shear web. So spar caps an shear web can be thiner with this material (for gliders, probably no gain for low aspect ration planes)
High aspect ratio / very thin profile wing are possible at lower weight (GP gliders if I recall correctly, so light that they are eligible as French ultralights).
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
Billski: A bit of fiddling with a calculator suggests that the carbon part of a spar cap can be made to fail at about the same strain as the wood part, if the wood species and the type of carbon are selected carefully. However, I can't see an advantage in doing so. I could understand using a wood shear web, though. A bit of homework and/or testing would be prudent to make sure the web wouldn't fail prematurely. I imagine the species of wood is important here.
 

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
3,825
Location
KBJC
I could understand using a wood shear web, though.
I've looked into this a little bit (box beam, plywood webs & pultruded caps). One of the challenges is that the interface between the carbon and the plywood web is very small (because the caps can be so tiny). This very small area, then, experiences quite high shear stress when you take a look at the shear flow. There's just not much area to spread out the force as it's fed into the web and it ends up being too much for the plywood. Then I looked at going to multiple shear webs (up to four of them at the root in my case) and arranging the rods that form the caps to maximize the area touching the webs, but concluded it probably wasn't worth the effort and complication compared to just using a single CF web. (Also I am unsure how the glue joint between the carbon and would behave due to difference in strain.)
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,200
Location
US
I've looked into this a little bit (box beam, plywood webs & pultruded caps). One of the challenges is that the interface between the carbon and the plywood web is very small (because the caps can be so tiny). This very small area, then, experiences quite high shear stress when you take a look at the shear flow.
I've thought about a similar situation with pultruded caps and a web of CF with a foam core. At the interface where caps meet the web, it would seem pretty important to assure the foam at the top and bottom of the web doesn't see significant compressive or shear loads. I'll need to study a few of these spars to see how successful designs work.
IIRC, in the UK, the PFA or LAA had some reservations about the design of a popular plane's foam/CF spar.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,945
Location
Saline Michigan
Billski: A bit of fiddling with a calculator suggests that the carbon part of a spar cap can be made to fail at about the same strain as the wood part, if the wood species and the type of carbon are selected carefully. However, I can't see an advantage in doing so. I could understand using a wood shear web, though. A bit of homework and/or testing would be prudent to make sure the web wouldn't fail prematurely. I imagine the species of wood is important here.
Hmm. I challenge your thinking here. For spars to work, plain sections stay plain. That is the fancy way to say that the entire section bends together, the axial strain at the neutral axis is zero, and the axial strain increases linearly the further from the neutral axis you get. The material chosen for various parts does not change this. At the junction of the cap and web, the axial strains are the same in both pieces. The shear strain in the cap however is much lower than the shear strains in the web. The web sees more strains there.

Then there is the whole issue of getting load across the interface between cap and web. See mcrae104's comments. In composite beams the shear webs are wrapped around the outside of the caps and web cores or wrap on to the interior surfaces of the caps (bottom of the top cap, top of the bottom cap) or intertwine the web between cap structures or combinations to achieve the load transfer within the strength of the resins used.

So, to use a wooden web with a nice slender cap, the glue line will have to be supplanted with tapes with cloth on the bias and picking up load in caps and wooden web and transferring loads between them. These tapes would have to be as thick as with a composite web, extend as far onto the rest of the spar as with a composite web, and extend over much of the wooden web too. In the end, you have your caps, a relatively heavy wooden web, and most of the composite web you would have had if you had no wooden web. Lighter (WEIGHT IS THE ENEMY) and more direct to just build with a composite web and a foam core to support the web laminates against buckling.

Billski
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,945
Location
Saline Michigan
I've thought about a similar situation with pultruded caps and a web of CF with a foam core. At the interface where caps meet the web, it would seem pretty important to assure the foam at the top and bottom of the web doesn't see significant compressive or shear loads. I'll need to study a few of these spars to see how successful designs work.
IIRC, in the UK, the PFA or LAA had some reservations about the design of a popular plane's foam/CF spar.
The PFA/LAA like rag and tube airplanes and plywood airplanes. They should have dispensed with concerns over the materials sets and gotten into the actual design details long ago...

The foams we usually use have strain at failure in the 5-8% range, while our fiber-resin lamina are in 1-2% range. Make sure the laminates do not fail, and the foam that is fully captured inside the laminates CAN NOT FAIL. Fail the composites, and yeah, the foam will go too.

Go to the cases where the foam is used as the shear web between spar caps, and yeah, I would want a lot of verification, preferably from both analytical approaches and from some well done load tests. These too have long histories in the ends of wings and in winglets of Rutan designs and derivatives. Like every other thing in the design, they must be done correctly...

Billski
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,200
Location
US
IIRC, in the UK, the PFA or LAA had some reservations about the design of a popular plane's foam/CF spar.
The PFA/LAA like rag and tube airplanes and plywood airplanes. They should have dispensed with concerns over the materials sets and gotten into the actual design details long ago...

The foams we usually use have strain at failure in the 5-8% range, while our fiber-resin lamina are in 1-2% range. Make sure the laminates do not fail, and the foam that is fully captured inside the laminates CAN NOT FAIL. Fail the composites, and yeah, the foam will go too.
Thanks, Billski.
For completeness, here's the reference I was remembering:
The SD-1 was being developed around the same time as the Luciole . . . .. both utilise simple foam slab ribs. Perhaps the most radical of their shared design features though is the use of carbon fibre pultrusions (pre-formed lengths of uniform cross-section) in conjunction with wood and ply to add a high degree of strength where required for minimal additional weight.

However, unlike the Luciole, the SD-1 main spars do not use the carbon pultrusions in association with wood to form the spar caps, which are then glued to a ply web. Instead the carbon pultrusions form the entire spar caps and a foam web is used, the whole then being encased in glass. Ply reinforcements are included at structural locations such as the root ends, which contain the bushings for the retaining pins that hold the wings into the fuselage. These spars, which weigh just 2kg each, are stressed to 7.5g ultimate, according to the designer. However, when LAA Engineering looked at the design a year or two back, the engineers had reservations about the spar’s detail design, as well as with the tailplane attachments.
Whatever the issues were, apparently they have been resolved to the satisfaction of the LAA since many SD-1s are now flying in the UK. I haven't heard of any spar failures. either.

FWIW, it was intersting to me that these very light spar webs use fiberglass rather than CF over the foam.
 
Last edited:

dino

Well-Known Member
HBA Supporter
Joined
Sep 18, 2007
Messages
696
Location
florida
MC30 seems to have optimized spruce-pultruded CF in the spar. The marriage of foam-wood-CF pultrusions in the wing are efficient, proven and according to the attached analysis extensively documented. Hard to beat that weight. A good read even for non French speakers.
 

Attachments

Last edited:

daveklingler

Well-Known Member
Joined
Jan 22, 2013
Messages
128
Location
Albuquerque
Does the Skypup specify insulation board or flotation billet? The difference is significant. Dow flotation billet (the large cell stuff) as used for airfoil cores was known back when Rutan was selling plans to not use any silicones. Might still be true today. Insulation board appears to be an unknown.

The issue is that foam has to stay well attached to the fiber-resin facings or it is not supporting the facings against buckling. Silicones and mold waxes that *might* be used in processing extruded foam can cause delamination and then other really-bad-day events.
Accepted practice in the surfboard and trailer industries for using Dow XPS blue board seems to be to either wipe it down with isopropyl alcohol or sand lightly with 150 grit to avoid delamination. That's not a bad practice anyway, given that it's good to be sure the bonding surface is clean.

This is one of those possible issues that would be good to nail down, because it's obvious that the material would be "good enough" in a lot of cases. Divinycell foam is both highly superior and much more expensive. I wrote a note using Dow's technical question portal. I'll post here if I get an answer.

A good plan when preparing to use an unknown foam is to build some test pieces, laminate samples with your intended resin and cloth, cure it, post cure it, then peel the lamina from the foam. If it takes parent foam everywhere, good. If the resin comes off the foam clean anywhere on the test pieces, AVOID THAT FOAM. I have tested batches of flotation billet and insulation board before making airplane parts with it. I have also tested Divinycel - no problems there. One stack of light blue insulation board that I got as a gift is "tooling only" for poor adhesion. Consider yourselves warned.
That's good advice.

There's a natural law that says if you opt not to test, the results never go in your favor. The universe has a nasty sense of humor.
 
Last edited:
Top