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

### Help Support HomeBuiltAirplanes.com:

##### Well-Known Member
For you novices, some warnings. Epoxy typically carries some sensitizing agents, that can make you allergic to your airplane. Yes, some more recent epoxies are lower sensitizing, but they are not non-sensitizing, while others are even worse than the epoxies of old...

The guy leading you through the processes wore "invisible gloves", a barrier material, on his hands to allow what looked like bare hand work with the epoxy. Many folks have used it, others are just religious about nitrile gloves. Either way, protection against absorption of epoxy through the skin is a wise precaution.

Next, some small amount of the epoxy will evaporate, and usually some amount is also a sensitizer. Folks, either really good ventilation or a respirator with filters for organic vapor capture is a good idea, or you can again become sensitized (allergic to) your airplane.

There are airplane projects out there that were partially built,and then abandoned, sold, given away, etc, because the builder became allergic to it. Do not be that guy. Please protect yourself from breathing epoxy vapors or absorbing epoxy through your skin.

Billski
Excellent point, and thanks for bringing the subject of industrial hygiene into perspective! As a former military aviation safety officer, I should have made these points below from past and present experiences...

I hated latex gloves, so at the beginning of my experiences with epoxies, I never used anything. Then I got wise to the dangers of the materials and eventually became religious about shop safety, ventilation, PPE and the like.

I am not a fan of Ply-9, but my go-to are nitrile gloves, respirator and eye protection at all times.

Also, in that older video, I darn near cringed when I saw that hotwire power source setup without any protection around it. What you saw in the video is a setup for electric shock!!

Here are some tips on epoxies for those not in the know!

#### Attachments

• 114.5 KB Views: 19

#### Hot Wings

##### Grumpy Cynic
HBA Supporter
Log Member
I darn near cringed when I saw that hotwire power source setup without any protection around it.
Then you don't want to know how I cook hot dogs at the shop sometimes....................

#### wsimpso1

##### Super Moderator
Staff member
Log Member
Clear! [except two minor typos marked in square brackets thus] I ordered Moldless Composites: cost $18~ good: shipping$12 ~ not so good. Thanks
Brian
The first is not a typo. It will make a strong part to skip the micro. But it will be heavier than it could have been by simply mixing micro and epoxy about 50/50 for this job, squeegeeing it on, then squeegeeing the excess off.

The plastic foam is closed cell, but there are MANY little broken cells with air in them on the surface of the foam. These broken cells have to get filled with something. Three approaches are possible, and they will all work if you have enough time before the resin gels and locks your progress:

The first is to just apply cloth, work resin into the cloth, then as the resin slowly works down into the broken cells, it displaces the air from the cells which float up into the cloth. If you leave this field of little bubbles in the cloth, you will have a weak part, so you must worj the cloth to free the bubbles from the cloth. Maybe you will have enough time to get rid of the bubbles, maybe you won't. This practice can result in ruined parts if the laminate is shot through with bubbles when the resin gels;

Second process is to wet the foam with neat resin, wipe it around with your squeegee until the surface is filled, then put on cloth, wet with resin and squeegee, repeat until your laminate is complete. This works, but tends to be heavy in open wet layups. In vacuum bagging, this actually works pretty nicely as the cloth is very nicely compacted against he foam, and all excess resin gets squeezed out into the breather layer;

Third process is to fill the surface with slurry, which is a little over half the density of neat epoxy, squeegee all the excess off the part, then do the cloth and resin. This produces a lighter laminate that with neat resin on the foam, and is preferred in open wet layups. If you are vacuum bagging, well, we do not want any micro in the cloth - it is just about as bad as air in the laminate - so most of us just skip micro entirely. My vacuum bagged parts have such low resin content, I stop there...

In vacuum bagging, some folks get cranked up enough about weight to go after the weight of the neat epoxy in the foam cells. How? Well, you do not want to use wet micro that can end up in the cloth, so some folks hard shell. This takes three bag sessions instead of one or two, and includes applying micro resin mix to the foam for separate cure cycles. It does replace neat epoxy with micro-resin and avoids getting micro into your cloth.

Billski

Last edited:

#### User27

##### Well-Known Member
This includes body filler too. Just don't. Polyester is weak and shrinks forever.
Billski,
[Sorry for slight thread drift] I have a huge amount of respect for everything you post, but on this small point I don't agree. I would suggest it is acceptable to use good quality (expensive) polyester filler otherwise progress is slowed drastically. I work in a sailplane repair company, we use polyester filler all the time when a thin wipe of filler is required. We only use one type of filler (Seilo 80, made in Germany) and apply it thinly - and usually sand most of it off - but we cannot wait for epoxy & micro to cure. If thicker filler is required we will use epoxy & micro and plan ahead to allow for curing time. We use 285 & H286, it also shrinks when post cured so we cure before painting.

#### tdfsks

##### Well-Known Member
But how do I design a composite web ?
Getting back to the original question ... more specifically the question about how to design a web.

I have had a look through some old stress reports.

Some typical numbers for allowable shear strength for a +/- 45 deg wet layup of Rutan Style 7715 Uni E glass / epoxy on two core materials are:

Blue Styrofoam (2 lb/cu ft): 4000 psi ultimate shear.
PVC Foam (4.5 lb/cu ft): 5000 psi ultimate shear.

Note: these are ultimate stresses. Just so there is no confusion: if you have a normal category aircraft with 3.8 g limit and 5.7 g ultimate, these are the allowable stresses at 5.7 g.

I do not have a shear modulus or allowable shear strain data for this material.

Be careful applying these to anything but 7715 / epoxy and note that these are for a 45 deg layup (i.e. fibres at 45 deg to the shear load). A 0/90 laminate will have lower shear allowables.

Also, these allowables are based on a ply thickness of 0.009". If your laminate has thicker plies (measure a test laminate) you have a higher resin to glass ratio than used in deriving these allowables and you will need to reduce these values.

For simplicity just calculate the shear stress in the web as V / (t x d) and that will work OK with these approximate allowables. V is the shear load, t is the thickness of the shear web laminate (no core) and d is the depth of the web.

You will need to read up on how to calculate the shear stress in the bond line between the cap and the web laminate (you will be headed in the right direction when you see the equation Shear stress = V Q / I t).

OK ...... I hope that helps.

Last edited:

#### UAVGuy

##### Active Member
The stress corresponding to 4000 microstrain is stress = E x Strain = 7,000,000 x 4000 x 10^6 = 28,000 psi or 193 MPa.
That seems incredibly low.

#### UAVGuy

##### Active Member
Thanks for all the replies and discussion.

The thing that I don't understand is how foam between 2 web sections helps to prevent bucking of the web material. 1/2" or 1" of construction foam has a compressive strength (not to yield) of ~ 50 PSI, depending upon the variant. And the contact strength between a laminate and foam is not great, so it adds almost zero tensile strength between the webs.

So how does placing a weak material like construction foam between 2 layers of composite materially help prevent bucking or bowing of the webs ?

From a materials point of view, I can't get my head around a mental model that predicts how the foam material adds any strength to the structure. As far as I can tell all it does is change the geometry of the composite part by moving the individual webs out further from the centerline of the spar caps and create a miniature box. And the wider you make that box the more stable the structure becomes in the longitudinal direction. That makes sense to me.

But the question then becomes why even make an I beam type spar, why not make a box spar ? And if you can, why not leave out the foam entirely and make it hollow ? Am I missing something ?

Any comment on this spar design ?

Here is its strength test.

Here is the second version. Note the lack of a continuous web.

Keep in mind the all up weight of this airframe is 350 ish pounds, fully loaded.

#### UAVGuy

##### Active Member
I would also appreciate comments on this:

I'd love to see the analysis behind this:

Last edited:

#### UAVGuy

##### Active Member
Some quick back of the envelope calcs for the KR2.
Wing span: 24 feet
wing area: 93 ft^2

Make the gross assumption that the wing is a rectangle. Ignore that the wing starts outside the cabin. Chord = 93 ft^2/ 24 feet = 3.875 feet
Spar load per foot = 3.875 x 14 = 54.25 lbs/ft
Moment at center fo spar = 1/2 12^2 x 54.25 = 3906 foot pounds = 48,872 in lbs.
Spar height at center = 7 inches
Cap forces = 6,981 pounds, roughly

Each main spar cap has 6 rods in it. The Secondary spar caps (front and back) have 10 rods in the top and 7 rods in the bottom. So the top has 16 rods in total and the bottom has 13 rods. The force in each top rod is 6981 pounds/16 = 436 pounds. The force in each top rod is 6981/13 = 537 pounds.

All loads at max weight, 1g, no safety factor.

The rods are 0.070 x .437 Graphite, solid rectangle. Area = 0.0306 in^2.

The static 1g stress in the top rods is 436 pounds / 0.0306 in^2 = 14,250 PSI.
The static 1g stress in the bottom rods is 537 pounds / 0.0306 in^2 = 17,500 PSI

Does this look correct ?

What is the G rating on this airframe and what safety factor should be applied at that rating ?

Anyone care to show me a similar analysis of the web stresses ?

Last edited:

#### poormansairforce

##### Well-Known Member
The thing that I don't understand is how foam between 2 web sections helps to prevent bucking of the web material.
So how does placing a weak material like construction foam between 2 layers of composite materially help prevent bucking or bowing of the webs ?
From a materials point of view, I can't get my head around a mental model that predicts how the foam material adds any strength to the structure.
Think about it like a strut with jury struts. The jury strut cuts the effective length of the strut in half allowing much more load before buckling. If we add hundreds of jury struts we could theoretically achieve compressive strength that is almost equal to tension strength. So the foam acts as an almost infinite number of jury struts bracing each fiber against buckling failure. It doesn't need much strength to do its job.

Last edited:

#### BrianW

##### Well-Known Member
Thanks for all the replies and discussion.

The thing that I don't understand is how foam between 2 web sections helps to prevent bucking of the web material. 1/2" or 1" of construction foam has a compressive strength (not to yield) of ~ 50 PSI, depending upon the variant. And the contact strength between a laminate and foam is not great, so it adds almost zero tensile strength between the webs.

So how does placing a weak material like construction foam between 2 layers of composite materially help prevent bucking or bowing of the webs ?
/snip/
Man in the street explanation: if a tall thin tube under compression will fail in buckling, then keeping it straight with just your hands will extend its compression strength to failure.

#### claude.bouzerand

##### Active Member
I am sorry, I didn't completely read all messages, but your thread reminds me a book I bought some years ago, and I want to share with fellow amateurs here : Composite Design Manual from Jim Marske !! This book has a special place on my bookshelf . Jim has a very practical method and a little unique approach with pultruded carbon rods ... He gives a bit simplified but very effective calculation formulas , with good and understandable samples with mathematical comparaison with other materials. And I must add that Jim is a very good man ,always ready to reply any questions. I had a question about buildind my MWB10 with composite and he replied with kindness even with a very understandable strength calculation for my spar's tip with the rudder !
Regards
Claude Bouzerand

#### Hot Wings

##### Grumpy Cynic
HBA Supporter
Log Member
Am I missing something ?
Maybe not missing something just not looking at it from the right viewpoint?z As noted above the foam acts like a lot of little jury struts. Taking a look at it from another perspective that you can demonstrate easily:

A long cable tightly strung between 2 points, like a tow chain or power line. Lots of force on the cable, but you can go up to it in the middle and pull up on the center and deflect the center a fair amount.* The inverse math is identical.

* Don't actually do this or you may end up with an over-stressed chain wrapped around your neck.

#### Tommy222

##### Member
Another practical informative post. The terminal sentence is tantalizing. I guess it meant to signify the opposite? And what was it that did NOT work well? <g>
Sorry for not being more complete in my reply. The company shot from the hip as they designed bigger and more powerful aircraft. There fiberglass work was really good and I don’t know of any failures in it. As an example they fastened the end fittings in the elevator control tube with four aluminum pop rivets and I think they had failures with them. The owners were boat builders and lacked in the airplane build side. The doors were prone to leaving the airplane so they designed a fix and that kinda worked. The elevator bellcrank was made of a cut out of 1/8” aluminum and had a bearing pop riveted in . Now as a IA I look back and shudder at.

#### wsimpso1

##### Super Moderator
Staff member
Log Member
I would suggest it is acceptable to use good quality (expensive) polyester filler otherwise progress is slowed drastically. I work in a sailplane repair company, we use polyester filler all the time when a thin wipe of filler is required. We only use one type of filler (Seilo 80, made in Germany) and apply it thinly - and usually sand most of it off - but we cannot wait for epoxy & micro to cure. If thicker filler is required we will use epoxy & micro and plan ahead to allow for curing time. We use 285 & H286, it also shrinks when post cured so we cure before painting.
Well, that does make a point for discussion - The reasons behind generally not recommending polyester resins and polyester fillers:
• Polyester resins tend to shrink on curing and keep on shrinking over time. If you fill and fair a fiberglass surface that has the texture of the glass cloth on it, the polyester will have different depths at different places along the weave of the underlying cloth, and when it shrinks, the weave or other underlying shape will "print through" down the road;
• When polyester resins shrink while on top of an underlying structure that is not shrinking, some will also be peeling themselves loose from things underneath;
• Polyester resin generally has lower strain at failure than the fibers it is holding together, while vinylester and epoxy resins generally will hold together until after fibers begin failing.
None of this means that there are not places where it can work for you, but know that those places are few and far between. User27 has given us a place where a special polyester is used in thin layers when speed is helpful. If they were to put on a ticker layer of filler, they use epoxy. And us homebuilders can probably find some other stuff to work on while we wait for our general purpose epoxy to cure. I am sure that their repairs work well, or word would get around and their business would dwindle. The rest of us would have to figure out where we would slip some polyester in, then test and validate it, before making it part of our airplane...

Most of us would STILL do well to limit polyester to temporary fixtures and to tooling, although if you just gotta try it out, Seilo 80 sounds like a good bet.

Billski

#### wsimpso1

##### Super Moderator
Staff member
Log Member
That seems incredibly low.
I think that they are VERY conservative in their shop. Earlier, a FOS of 6 was cited. I thought that I was conservative with many of may parts around 3 on a 6 g airplane.

Billski

#### wsimpso1

##### Super Moderator
Staff member
Log Member
The thing that I don't understand is how foam between 2 web sections helps to prevent bucking of the web material. 1/2" or 1" of construction foam has a compressive strength (not to yield) of ~ 50 PSI, depending upon the variant. And the contact strength between a laminate and foam is not great, so it adds almost zero tensile strength between the webs.

So how does placing a weak material like construction foam between 2 layers of composite materially help prevent bucking or bowing of the webs?
Buckling in any form is predicated upon slight misalignment between the load and the column section. The stiffness the column has against this misalignment is everything in setting the critical load where this buckling and collapse occurs. Please go through any Mechanics of Materials or Machine Design book on the topic for the derivation, which was done entirely in theory by Euler in the 18th century and fits reality really well. In any event, if we increase the bending stiffness of the column, the critical load goes up. Taking a solid fiberglass web, splitting it in two and bonding it to a foam core greatly increases the bending stiffness of the plate that is acting as a column.

From a materials point of view, I can't get my head around a mental model that predicts how the foam material adds any strength to the structure. As far as I can tell all it does is change the geometry of the composite part by moving the individual webs out further from the centerline of the spar caps and create a miniature box. And the wider you make that box the more stable the structure becomes in the longitudinal direction. That makes sense to me.

But the question then becomes why even make an I beam type spar, why not make a box spar ? And if you can, why not leave out the foam entirely and make it hollow ? Am I missing something ?
Bending stiffness of any plate works out to being exactly proportional to EI of the plate. This was worked out from purely theoretical means and fits reality really well too. E is Young's Modulus and I is the second area moment of inertia of the plate. If you have more than one material, then you do EI for each of the pieces and add them together. Think of a steel plate a certain thickness, set it on two supports some distance apart. It sags some amount between the supports. Now split the plate into two plates half as thick, separate them by the distance of the original thickness with a bunch of little webs. With a weight gain of 10%, you just increased the bending stiffness of the plate by 7.1 times. This is just like an I beam or a wide flange beam. Way stiffer for very little extra weight. Do this with composite facings on foam and huge increases can occur with very modest weight increases.

In my shop I have pieces trimmed from the edges of vacuum bagged wing skins with one ply each of 22 oz TRAIX, 8 oz tape, and 18 ox BIAX. These 1-1/2 wide, four foot long, 0.052" thick pieces are flexible. You can bend them in a circle a foot in diameter with your hands and without much effort. Set one end on the floor and its own weight will buckle it. I also have some 3/8" foam with a ply of 18 oz BIAX vacuum bagged on one side and 22 oz TRIAX on the other side. The foam is open on all sides. This piece is a brick compared to the uncored part. Stand it on end and it just sits there. You can load it as a long column and yeah, it will eventually crush the foam and collapse, but it will take some doing compared to the other piece collapsing of its own weight...

Seriously, open a mechanics of materials text book, look in the chapters on columns and beams, and learn. The topic is entirely practical and works beautifully in the real world. Increase the bending stiffness of the web by a couple orders of magnitude with a little foam, then watch the buckling strength go from almost nothing to some usable amount.

Billski

Last edited:

#### proppastie

##### Well-Known Member
Log Member
That seems incredibly low.
yes but what does it weigh?.....of course looking at certified composite aircraft such as the Cirrus .....it is not lighter in weight than the Mooney. SR22 G6 Empty Weight: 2,270 lbs vs 1986 Mooney M20K 1800 lb.....The Cirrus has a parachute....but I am pretty sure it does not weigh close to 500 lb.

#### wsimpso1

##### Super Moderator
Staff member
Log Member
Any comment on this spar design ?
He did not talk about it much, but the scheme is easy to deduce.

He built two webs that he calls L brackets. Looks more like two channels back to back to me, but maybe he assembled them from 4 large L shaped angles. Anyway, the vertical sections of the channels or L shapes comprises the web of the spar and will carry the bulk of the shear load. The horizontal sections serve to connect the web to the caps, and is a common way of doing this.

The caps appear to be unidirectional tapes that he wets out and then transfers to the part. His methods are curious to me. The first ply went from the close end to a spot a couple feet away. The second ply went from the end of the first piece to a spot further away from the camera. This makes a weak spot in the cap at the butt join between the two plies. Perhaps he had good reasons to do this, but I have not viewed the rest of his videos in order to find out.

Then he appeared to have applied several full length plies over the first two. This is unusual. Usually, we like to leave cap plies dead straight. In this case if I was applying a cap with tailored thickness, I would make the web flanges dead straight, then apply the full length plies, then the shorter and shorter ones, leaving no steps for the longer plies to make as they cross from a thicker to a thinner layer beneath. The reason for maintaining ply straightness is that carbon fiber structures are generally VERY sensitive to any lack of straightness in fibers. He appears to have disregarded this point in his construction.

Another issue of interest to me is that he was doing his caps with open wet layups. This is generally viewed as reducer of strength in these caps, and he may know this.

Good thing he is testing and then beefing up as he goes.

Here is its strength test.
His test uses an unusual load build up. He did mention that they had already proven the wing structure and were then working on the attachments. I presume he was mainly aiming for specific loads at the attachments, which accounts for the unusual load distribution.

Anytime something breaks at a load less than the design intent ultimate load, I call that a surprise of the bad kind. It means that the design was inadequate to carry the intended load and his design failed its test. In my lifetime, testing is meant to prove that we did know what we were doing back when we designed the thing. Why not have it break and beef it up? Because you have a bunch of design cases that the structures must meet, and you can only check one or two of them. If things break early, then you have to wonder how it will do in all of those other cases. If instead, it goes to ultimate load and beyond a little, and you forecast that it would, then you know that you knew what you were doing in the first place, and the other cases are probably OK. In this case, he broke carry-through that is not per design, and now he has to wonder if the rest of the carry through is OK...

I may look at the remaining video another time. Ultralights, etc, are not of huge interest to me.

Billski

#### tdfsks

##### Well-Known Member
Anyone care to show me a similar analysis of the web stresses ?
OK ... here is an example analysis for a composite spar generalized for any tapered wing planform. I have included both bending and shear stresses.

I have simplified it by assuming that the loading on the wing is constant over the surface (i.e. no spanwise lift distribution which is conservative).

The hand written notes derive the equations in general form.

I have then programmed these into a spreadsheet to do the calculations for each station along the wing (see last image).

I have also provided a sample manual calculation for one station.

This is about as simple as you can get to do a half decent job but there are still quite a few assumptions in this analysis. Good enough for a homebuilt if you build in some margins but more work would be required for a certified aircraft.

Also note that this analysis should be preceded by a more detailed analysis of the loads on the airplane to understand the critical case for the wing.

If I were doing this for real, I would run the calcs on the spar for each of dozens of load cases for all the corners of the envelope, different CG positions etc, aileron deflections or flap deflection etc as it is not always obvious which case is critical for which station of the wing. However, you need to include the spanwise distribution of lift in the analysis in order to do this and that is a lot of work.

As I have explained above in earlier posts, your biggest challenge is establishing the allowable stresses and it is hard to help any more with that other than the rules of thumb I provided earlier.

I have not posted the actual Excel file and I don't really intend to because I would prefer that someone make the effort to understand the maths (high school level only) and make their own spreadsheet (an hour or so work) rather than just blindly plugging numbers into my spreadsheet without understanding how it works. Also I haven't really checked it to the point where I would want it being used for real ... all this is just for illustration to help understand how to do it. Also please check my maths, I just wrote this out quickly and I cannot guarantee that there is no errors ... but it passed some simple checks.

Yep - I was looking for something to do during COVID19 lockdown ....

OK ... I hope this helps and feel free to ask any questions.