Spar Webbing

[James Frolik]

I'm building a balsa model from plans with a 1.64-meter wing span. It consists of three wing sections, one mid-section and two outer panels. All three sections are about equal length.

The mid-section calls for a hardwood [vertical] spar with parallel top and bottom balsa runners. So it essentially has built-in webbing.

The outer panels, on the other hand, only call for top and bottom balsa runners (5mm x 5mm) and no webbing, so it's open between the two -- and between the ribs. The front third of the entire wing surface is sheeted, top and bottom.

What prompts my curiousity is that ALL other balsa models I've ever built of this size had webbing along the entire span. But this one doesn't.

It's a high-wing, twin-engine model with ailerons and flaps. The plan and pictures in a magazine building review of this model reveal no webbing on the outer wing sections.

What's the general thinking on this? I'm tempted to add 1mm (or about 1/16-inch) balsa webbing in the outer panels which I'm currently constructing. It shouldn't add too much weight, if at all noticeable. But for all eflight purposes, "save on weight" is a builder's mantra -- yet the plan doesn't show this typical webbing.

This post belatedly added to Builder's Workshop discussion.

[iva]

James,

having webbing on the central section only is not uncommon. There is a good reason for that. Wingtips are exposed to a lesser load than the wing center. If I remember correctly, that function is exponetional. Having wing tips or outer portions a bit weaker will allow for a bit of a wing flex which will in turn reduce the total load to the wing center.
In your particular case, I don't think that adding webbing to the outer panels would make much of a difference either way.

Zoran

[tim hooper]

I'm moving this to Modelling Science.

tim hooper

[vintage1]

Iva has it about right. The bending moment on a long end supported beam is maximum in the middle, and zero at the ends. Its a little different o a wing, where lift is distrubute along the wing, but essentially it is still maxiumum in the wing center., You almost never see a wing fold anywhere except in the center.

Use of spar webs at the tips wastes weight for no gain.

[avie]

two ideas for you. first, don't use hardwood webbing. that is a waste weight. use 3mm balsa, it will do, especially if the spars are 5x5 mm balsa. on the outer panels, I would put web but from 1.5mm balsa, and if you really want to keep it down then place the web every other bay.


Avi

[vintage1]

Putting webs every other bay is useless IMHO. One of the non webbed bays will fail before the webbeed ones get anywhere near collpase.

It may increase stiffness a little, but will not increase strength. It may actually reduce it by concentrating stress on the unwebbed bays.

Short of a lot of tedious calculation, the way to get the best is to build a wing and stress it till it breaks. If it takes more than 10x model weight when tip supported, its too strong/heavy. If it starts to go at 5, look where its beginning to fail and add webs there. Not a lot between liteply and hard balsa to choose for material.

[James Frolik]

This is a picture of the model I'm building taken from the French magazine Aérotech. (Photo by magazine editor, Laurent Berlivet.) I got the plans from the magazine. Here you can pretty much see where the wing's stress will occur. The center section naturally receives the most stress. And the rest will follow just outside the motor nacells where the outer panels begin, (The struts may really help here.)

I suspect this original model might have webbing in the outer panels; a second model in the review, which was constructed for the purpose of drawing the plans, probably doesn't have webbing as the designer felt they weren't really needed. Hence the plans don't show any.

Nevertheless, I constructed the first outer pansl (left side) with 1.5mm balsa webbing. With glue it probably added about 5 grams. Negligible weight considering an AUW of roughly 2.15 kilograms.

My decision is better safe than sorry.

I will begin a thread of my construction in the Scale Modeling column as soon as the model is pretty much noticeable as a Shorts 360 (in raw balsa form). Right now the construction is quite scattered, although all that's left to really put together is the right wing's outer paner and the fuselage.

[Ollie]

If the wing strut is under tension then it reduces the shear load between the wing strut attachment point of the wing and the fuselage. You can roughly conservatively size the shear web on the cantilevered portion of the wing as follows. Assuming the strut attachment point is half way between the fuselage and wing tip then the maximum shear force is 1/4 of the weight of the model times the maximum vertical acceleration (G's) that you expect the model to experience in an extreme maneuver. The required shear web crossectional area in the plane of the wing rib is half the shear strength of the balsa wood, with the grain, divided by the shear load at the wing strut. It is safe to decrease the thickness of the shear web along the span linearly to zero at the wing tip.

[James Frolik]

Right, copy that.

I'll try to remember the physics the next time I build a balsa model -- which will be in about another 5 or 6 years -- where the plan possibly ignores any webbing.

Most certainly it won't be a French plan, though. You wouldn't believe the strange, error-suspected and lack-of-diagram plans I'm working from. For instance, balsa thickness is printed as 15/10 for 1.5mm whereas it's 5x5 for things like the caps; the main landing gear is drawn from three different angles, but no matter how I try and align them, they're all slightly different bends or lengths by 3 to 5 millimeters -- too much when in the wheel pant you have only 2 to 5 millimeters tolerance; and there are no frontal veiws of the entrie model, much less any part except the main center-section spar.

[James Frolik]

While coming home from work this morning (where I made the last post) I thought about Ollie's last sentence:

It is safe to decrease the thickness of the shear web along the span linearly to zero at the wing tip.

Although my last statement about "remember the physics next time I build a balsa model" was in humor, his last sentence is quite easy to remember and actually makes a lot of sense.

It would take some additional work sorting through whatever balsa is in [a modeler's] possession to get adaquate decreasing thicknesses, it wouldn't be difficult and would certainly spare an additional few grams were a person really counting [i]every speck of weight.

I will remember that for the next time! Thanks for the tips.

[Weisse Luft]

If you really want to get to the root of the science, pick up any text on mechanics of solids and look up "beams of uniform strength". An internally braced wing is a cantilever beam with a uniformily distributed load (lift).

If the spar is constant width, the distance from the bottom to top caps (flanges) can uniformly decrease. If the depth is constant, the thickness of the spar decreases parabolically.

Instead of using singly ply balsa for shear webbing, you can get by using 2/3's the thickness of two-ply balsa, laminated so the grain runs at a 45 degree angle to the spar axis. Laminate with polyurethane as its the lowest weight.

[avie]

in theory, a web made of two strings creating an X shape will do, much like a bipe's wings are cross braced. if the spur presses the ribs hard enough, they will colapse to one side, and the rectangle between the spurs will turn to a parallelogram. this would increase the diagonal's length and stretch the string, which (hopefully ;)) would resist.
that is also the idea behind a web every other bay. the web will also keep the 2 spurs parallel, while without any web a spur under pressure would become wavy.

my 2 cents,
Avi

[RSCherry]

The typical failure mode of a wing spar is by compression failure of the top spar cap. The compression strength of most materials is lower than the tensile strength. Analysis of failed wings will usually show that the top spar failed by buckling.

To properly make a shear web, it must be glued to the top and bottom spars full length, and must not skip any bays or the buckling failure will occur in an empty bay or where the shear web is not glued to the spar. Balsa shear webs are usually designed with the grain vertical, because wood is much stronger in compression and tension parallel to the grain than perpendicular to it.

A shear web also has the additional benefit of connecting the top and bottom spar together preventing shear between the two, and thereby greatly stiffening the wing structure in bending. The X-bracing discussed above is essentially a truss, and while it will significantly stiffen the structure, it does nothing to prevent the buckling failure mode.

[Chris True]

It seems you really like Shorts airplanes James! I seem to remeber you had a smaller model a few years ago.

If those struts are functional you basically don't need webs on the inner panels. On the outer panels the load is already reduced a lot, it is probably fine without webs. To be safe webbing the first 1/2 of the outer panel couldn't be faulted. Carrying the webs all the way to the tip is fairly meaningless, the constant cross section spars are already way oversize that far out on the span - no way they will buckle short of dropping the model on a wingtip!

[James Frolik]

Chris,

Yes, I built a Shorts Skyvan five years ago from free plans provided in a German modeling magazine. That was my first build from plans, and those plans were excellent -- absolutely everything was noted and there was no doubt about construction. When I noticed a French magazine plan for a Shorts 360 that's virtually the same scale as the Skyvan, I thought it would look good in my hangar as a sistership.

But for this I paid about USD$30 for the plans. And contrary to the German exactness of the [free!] Skyvan plan, this French one leaves too much to question -- at least for me. It's a real slog to complete. Yet I wouldn't have begun building without a plan (of any kind).

When it's nearly completed, but not fully assembled, I'll box it up and ship it to Oregon before my summer vacation. There I'll finsih it and see what happens at the flying field. If it's still in one piece after my summer -- and I fully expect it to be -- then it'll hang next to the Skyvan in the hobby shop where many of my models are suspended as eflight displays.

Also, I only started work under the condition that Christian Ramoser, make of the VarioPROP, would machine me two 5-blade 6A VarioPROP hubs -- he already makes a 5-blade 8B hub, but it's too bulky for this model. I wanted to experiment with some extra scale design, so this model is a propeller testbed.

[vintage1]

Originally posted by RSCherry


A shear web also has the additional benefit of connecting the top and bottom spar together preventing shear between the two, and thereby greatly stiffening the wing structure in bending. The X-bracing discussed above is essentially a truss, and while it will significantly stiffen the structure, it does nothing to prevent the buckling failure mode.

That is completely true, with one proviso. If the ribs are weak in compression the cross bracing will stop them collapsing, and strengthen the overall structure. The whle purpose of shear webbing in terms of strenth, rather than stiffness, is to reduce the span of each spar between ribs, or at least between places where it is realtively held rigidly, below the critical euler limit.

The whole structure tends towards teh calssic I beam, where the compressive and tensile strength of the material is fully utilsed by having e.g. the top and bottom spars much broader then deep, to control lateral buclking, and then controlling buckling between them by the ribs, shear webs or cross bracing. Or any or all of the above.

Cross bracing works best where the top and bottom webs have sufficient stiffness in themselves to resist buckling between the braces: This is true of square section spars, less true of shallow ones.
In the limit, probably the best construction for a wing is to have shallow top and bottom spars tapering in width, but of relatively constant depth, and taper the thickness of the wing, and use vertical grained shear webbing all along. For a contsant airfoil section, this leads naturally to an elliptical planform, or nearly so.

If leading edge sheeting and cap strips are used, it also makes sense to run the grain vertically on the ribs as well. They become just more 'shear webs' in the overall structure with the sheeting and cap strips providing most of the strength.

Needless to say, this is a complex way to build a wing. Its easier to shove in square spars and run webbing most of the way long. But it is possible to do better if ultimate light weight is an issue. Or if CNC milling or laser cutting is available to produce an ultra light interlocking jigsaw that can be assembled in seconds prior to being locked with thin CA....

[Dereck]

If you really want to complicate matters - make the spar caps from Spruce the same thickness as the LE D box sheeting, and taper the spar from root to tip.

Then use thicker inter-spar webbing for the first couple of rib bays - grain vertical for the webs - thinning each subsquent bay until reaching 1/16" balsa from around 1/3 span.

Yes, I did it once! The wing didn't collapse - just like my Four Star 40's wing, with untapered spars, no D box sheeting and constant thickness webbing glued on the backs of the hunky great mainspars hasn't collapsed either!

Sometimes, I suspect we really overdo all manner of things and it might be easier to make our lives easier! I suspect my big Sig Cub's wings are close to as strong as the full sized Cub's!

Regards

Dereck

[Ollie]

Dereck,

Right on!

A typical spar of constant crossection and height, is structurally inefficient in the sense that 2/3 of the spar caps and 1/2 of the shear web are just along for the ride because they are overdesigned everywhere except at the wing root. Spar caps that are tapered in width and thickness will have strength that more closely matches the tension and compression associated with bending load along the span. Spar webs that are tapered in thickness will have shear strength along the span that more closely matches the shear associated with the bending load.