Nightowl, I think we're still not on the same page.

Back to Structural Engineering 101:

A spar is essentially a beam that carries bending loads.

In a typical spar under a positive-G load, the top edge (or upper spar cap in a built-up spar) carries a compression load, and the bottom edge carries a tension load. The principle fibers in the wood of the spar caps (the "grain" that most of us think of) needs to run spanwise to carry these loads.

However, these loads have to be connected to each other. Each one needs something to "pull against". That's the job of the shear web, and it does this by carrying a horizontal (spanwise) shear load.

In addition, there is a vertical shear load in the shear web along the length of the spar. If you generate a pound of lift part-way out on the wing, somehow that has to be carried down the length of the spar to the wing root so it can lift upwards against the fuselage's weight. That's the function performed by the vertical shear load.

The third function of the shear web is to support the upper spar cap so it doesn't buckle under its compression load. (Note, under negative G the same thing applies, except it's the bottom spar cap that's now carrying a compressive load.)

In order to carry these loads effectively, you need to have wood fibers running in the direction of these loads. For shear loads, ideally you need fibers running at 45 degrees to the load, on both of the 45-degree orientations. Plywood with a +/-45 degree orientation to the span would be ideal for both of the shear loads.

For the job of supporting the upper spar cap against buckling, we need as much stiffness as possible in the vertical direction. This is why I-beam spars with a balsa shear web in the middle have the grain of the shear web oriented vertically. Note, to prevent buckling, it's the stiffness in the load direction, not the strength, that matters. "Flexibility" is your enemy!

On a solid spar made from a single rectangular piece of wood, the top and bottom edges act as the spar caps, and the wood in the middle of the rectangular cross-section acts as the shear web. This is where my comments in my previous post apply.

The "spar cap" functions of the top and bottom edges of the spar stock require that the principle grain of the wood run spanwise, just like we're all used to. However, that's not the question that I was addressing.

If you look at the end of the piece of spar stock, you can see the tree rings of the log that the spar stock was cut from. The question is whether we should orient these rings, as viewed from the end, vertically or horizontally.

To answer this, we need to consider that in addition to the principle wood fibers that run along the length of the tree trunk, and the length of the spar stock, there are also some other fibers, the "medullary rays", that run radially outward from the center of the trunk, like spokes on a wheel. Wood that is "quarter-sawn", what we refer to as "C-grain", has the rings running perpendicular to the width of the piece, and therefore the medullary rays running parallel to the width of the piece (vertically in the case of our hypothetical spar). This gives it properties akin to plywood, with most of the fibers running lengthwise, but some fibers, the medullary rays, running through the height of the spar as well. A spar cut from C-grain wood will have the principle, lengthwise fibers running in the direction of the compression and tension loads, and the medullary rays in the direction needed to carry the shear loads and provide anti-buckling support.

OTOH, if we cut the spar from the log so that, viewed from the end, the annular rings run vertically in our spar, parallel to the faces of the spar ("A-grain"), the medullary rays are now running in the chordwise direction, where they can make no significant contribution to carrying the shear loads, or supporting the top edge of the spar against buckling. The spar is more prone to splitting in the middle along its length, and more prone to buckling failures of the compressively-loaded top edge. The spar contains the same amount of wood and weighs the same, but because we are not allowing the medullary rays to make their contribution, the spar is weaker. The structure is less efficient.

Note, there is obviously still some "cross-grain" strength and stiffness in A-grain wood, but it is not as strong in that direction as C-grain. You can make an A-grain spar work, but it will have to be bigger and heavier than an equally strong C-grain spar. Depending on the wood species and the details of that particular piece of wood, the difference could be anywhere from insignificant to substantial.

Now, in the case of an I-beam spar made by gluing strips of wood to the front and back faces of the top and bottom edges of a vertical shear web, or the case of a box spar with top and bottom spar caps, and vertical shear webs located at the front and back edges (typically plywood), we have a slightly different consideration. Consider the tension and/or compression loads in the fibers of the spar caps furthest away from the shear webs. Somehow we have to carry that tension or compression in shear across the face of the spar cap to the shear web, where the shear web can then carry the load to the other spar cap to complete the load path. As before, we want the medullary rays to run in the direction of the shear load, but in this case that is in the horizontal direction, chordwise. As before, the grain in the shear web itself should run vertically if the main concern is buckling of the upper spar cap, or at +/-45 degrees in the case of a plywood shear web in a situation where the shear loads in the shear web are the biggest concern.

Note, in a practical, produceable design there may be other considerations, such as efficient, low-waste use of available wood sizes, that force us to compromise on less-than-optimum grain orientations.

A lot more going on in these spars than most folks realize. Kinda makes you have a lot more respect for the lowly shear web.

God bless you, Don. I for one will have to reference your words at some stage so I can find them again down the track, thats great info to think about.

Originally Posted by Don Stackhouse

Nightowl, I think we're still not on the same page.


If you look at the end of the piece of spar stock, you can see the tree rings of the log that the spar stock was cut from. The question is whether we should orient these rings, as viewed from the end, vertically or horizontally.

OTOH, if we cut the spar from the log so that, viewed from the end, the annular rings run vertically in our spar, parallel to the faces of the spar ("A-grain"), the medullary rays are now running in the chordwise direction, where they can make no significant contribution to carrying the shear loads, or supporting the top edge of the spar against buckling. The spar is more prone to splitting in the middle along its length, and more prone to buckling failures of the compressively-loaded top edge. The spar contains the same amount of wood and weighs the same, but because we are not allowing the medullary rays to make their contribution, the spar is weaker. The structure is less efficient.

A lot more going on in these spars than most folks realize. Kinda makes you have a lot more respect for the lowly shear web.

By the way, Don, I hope you realize what this is. Try to imagine this as you and me sitting in a coffee shop (although that would be better, because we could draw pictures!), and just discussing our undertanding of how this all works together. This is NOT an argument. If anyone thinks it is, stop us now, because that is not what I am trying to do. I don't "have" to be right; I'm just saying this is the way I understand it, and there is a possibility I can be dead wrong. In which case, I would like to understand it correctly! I'm trying to figure out if I am right, and if I'm not, to learn something.

This is where my understanding is in disagreement. If you use spars so that the C-grain is oriented effectively "vertically" in the spar, it is stiffer, but what are we talking about here? If you're using a spar cap that is 1/16 x 3 inches, then stiffness may be a consideration. But if you're using a 1/8 x 1/4 spar, all you're doing is making it less resistant to splitting in a direction that is NOT subject to much stress to begin with. And in that case, what you do is reduce the flexibility of the spar longitudinally, which means it will break and transfer it's load to the vertical shear webs that have no strength in supporting the then torsion/compression load that will pull the vertical grain of the shears apart.

If you lay a sheet of 1/16 x 3 C-grain wood across two points and put weight on it, it will not deflect as far before breaking as an A-grain sheet. That's why we use A-grain, or long-grain, in the direction we want it to bend. C-grain will resist splitting across the 3-inch dimension, but that isn't where the load on a spar is. Even if the spar splits longitudinally, it doesn't appreciably reduce the strength of the wing. Two parallel 1/8 inch square A-grain spars laid next to each other will stretch between two points and support the same weight as a 1/8 x 1/4 inch A-grain spar. It seems to me, all you're talking about doing is effectively using the grain direction to tie them together. But it's not going to affect the strength, only the longitudinal flexibility through one of the narrow axes.

Think of it this way; a unidirection carbon fiber capstrip is always "A-grain." Why would you replace a unidirectional cap strip with carbon fiber weave? It adds reinforcement in a direction it's not really needed because it's not subject to stress.

Nightowl

Nightowl,

The stresses Don talks about are not bending loads. A bending load is always comprised of the 3 actual stresses that Don describes. The quarter-sawn "C" grain wood has been proven to take more tension and compression stress that "A" grain wood. For leading edge sheeting, A-grain balsa is the first choice because it bends much better across the grain without splitting. But for spar caps, ribs, and other members that take loads and don't have to be bent, C-grain has been proven to be the stronger of the two.

Don's post was spot on. For instance, I am using a piece of quarter-sawn Englemann spruce right now for a sound board. It's .090 thick and 9" wide. It's grain is very even, and about .050 on centers, running vertically through the wood, and it is heavily "cross silked" (luthier term). The cross silking is the radial wood fibers that Don mentioned. It will go a long way to keep this thin spruce from splitting.

UD CF is not A-grain, nor C-grain. It's constant filament fill is completely different than wood where there are softer portions between harder, darker grains.

Again, no argument... just adding to the discussion... ...

Jack

Face it guys.... Ya gotta love this thread. Where else on the web can you go to learn so much for so little? I love these "discussions" like this because I can learn a lot without having to go out and do the research myself. Keep up the teaching gentlemen!

Amen Jeff!

Originally Posted by dwells

T Where are the seven sisters in relation to the Nebula? I always thought it was a cool story about Orion's pursuit of the ladies.

Sorry for being slow on the uptake, Don.

M45, the Pleides- seven sisters is between the constellations Perseus and Taurus. Take a line between Orion and the const Taurus and Pleides will be just to their right for you folks in the northern hemisphere, they are almost eqi-distant apart. For us they are the other way around.

here's a pic i stole and annotated from http://spaceweather.com/archive.php?...h=12&year=2009 taken from Ca, US.

May its owner forgive me.

Don?

Did I read your post correctly that the "perfect" shear web would be composed of two webs oriented at 45 degree angles? If so, I think I'll try that on the next woodie I build. I haven't broken a wing in years, but I like the extra insurance. Now I just have to get the wife to move all the sewing stuff and extra junk out of my building room. She INSISTS on calling it her spare bedroom and does not realize the danger we are in if I can't sand some balsa!

Jack, thanks for the input. Sounds like you and I have some common interests and activities in addition to model sailplanes!

Nightowl, we were indeed on different pages, and now that I see where you're coming from, I'll try to respond. However, it requires an involved response, and you might have to wait a bit until I can fit it in. Right now I have three fiddles here getting major structural reconstruction work, a violin needing a complete setup along with a rehair and some repair work on its bow, a 1957 Gretsch guitar here for a complete refret (all five of those projects hopefully done by the end of next week), the design for my new mandolin to finish, a prototype wooden sailplane to finish and designs for several others to work on, and several other projects that need work. I'll have to pace myself on this thread a bit.

Hossfly, +/-45 degree orientation is a term from advanced composites. It usually refers to orienting the fibers of woven ("bidirectional") cloth so the fibers run at 45 degrees to either side of the principle direction. When we apply that idea to wooden structures, the best "fit" of the term is with plywood, the wooden equivalent of bidirectional woven cloth.

What you describe amounts to making your own plywood from two individual layers of wood. That would work, but the resuting shear web would be "unbalanced", so that when you apply a bending load to that sort of spar assembly, it will want to twist.

By going with an "isotropically balanced" arrangement (the plies are layered so that starting from the middle of the stack, the layers to one side of the middle are the mirror image of the plies on the other side) we can eliminate this problem. In the case of your shear web, the typical solution is to make the shear web from plywood with an odd number of plies, and the ply thicknesses varied so that the amount of fibers running one way is the same as the other way (for example, a three-ply, with the middle layer twice as thick as the two outer layers). Orient that so the plies run 45 degrees to the span.

This optimizes the shear-carrying ability, but hurts the anti-buckling support function of the shear web. It also is difficult to fit on standard-sized plywood sheets without having a lot of waste. Running the fibers at a 0/90 orientation doesn't lose too much shear capability in comparison to what most models require (the situation on full-scale aircaft might be more critical in this regard), helps the anti-buckling support and makes much more efficient use of the typically available raw materials.

Originally Posted by Don Stackhouse

Jack, thanks for the input. Sounds like you and I have some common interests and activities in addition to model sailplanes!

Nightowl, we were indeed on different pages, and now that I see where you're coming from, I'll try to respond. However, it requires an involved response, and you might have to wait a bit until I can fit it in. I'll have to pace myself on this thread a bit.

Not a problem, Don. I don't have anything critical going on right now where the answer is going to make a difference. I am actually very interested in learning more from your perspective on this.

In a way, Jack's post pointed out the same thing I said, where his .090 x 9 inch sheet will resist splitting along the grain better than A-grain, but if it was .090 x .180, I just don't see the stresses that would be trying to create a split.

Nightowl

AtmosSteve

http://apod.nasa.gov/apod/ap091217.html

This was also the Astronomy Picture Of the Day (APOD) yesterday! That is the webpage that I have my browser, Firefox, set to open to.

Thanks to all for this structures discussion, it has helped clarify several things for me

http://www.auf.asn.au/scratchbuilder/joints.html Another cool webpage

One of the interesting things to note is the wood joint being glued should be FRESH. Lightly sand the surfaces with fine sandpaper before gluing if they have had any appreciable time to oxidize.

Scraping is better, ideally a carpenter's (or a luthier's) scraper, but the edge of a very sharp knife or a razor blade will do.

Sanding is better than nothing, but the dust tends to clog the pores of the wood and makes it harder for the glue to get a good grip.

Originally Posted by atmosteve

Sorry for being slow on the uptake, Don.

M45, the Pleides- seven sisters is between the constellations Perseus and Taurus. Take a line between Orion and the const Taurus and Pleides will be just to their right for you folks in the northern hemisphere, they are almost eqi-distant apart. For us they are the other way around.

here's a pic i stole and annotated from http://spaceweather.com/archive.php?...h=12&year=2009 taken from Ca, US.

May its owner forgive me.

Ah ha! The Pleides, that's it Steve. I knew there was a name for the group other than the descriptive "Seven Sisters". I didn't remember them being so far from Orion though. The pic of the big Geminid chunk exploding is very cool. I'll bet the photographer is still cuttin gun patches over that .

Sick with the crud and fever but building .

Be careful of that creeping crud, Dwells. All the rest of you, too.

I'm doing way too many reports on respiratory illnesses that are really laying people low, lately. Lot of people winding up in hospitals with pneumonias this year.

Nightowl