Carbon reinforcement

Hi all, I am a new member.
This follows a discussion with Gary Baigent about wing masts.
In several places structures, but especially wing masts are built with composite structure, that is several layers of different materials, frequently wood and carbon used frequently as local reinforcements.
I think that carbon reinforcement are risky solutions, I try to explain my mind.

Any elastic material, to take a load, must change dimension, like a spring. If you load a spring it reduces the length and produce one elastic reaction that compensate the load.
Wood and carbon layer are glued togeter, then forced to move sincronously.
Wood and carbon have very differen elastic modulus, this means that wood, to take a load, must reduce the length much more than carbon. (I know, all this is obvious....)
The consecuence is that wood with carbon reinforcement, under load, follows the elastic behavior of the strongest material, that is carbon in my example, which shortens very little, wood do not take almost any load because, with it,s low elastic modulus, to produce any meaningfull elastic reaction, must shortens much more.

The consequence is that carbon take almost all the load, if there if enough of it, no problem, but wood is useless weight, else, carbon can break and leave all the load to wood, wich as been reinforced because can't take that load...


The question is: I am right or not? If not why?

Paolo

 

You are right, but there are four things to also consider:

With spars the Gougeon approach was to put the carbon inside. So it is as if you had a bungy cord paired up with a slack piece of wire rope. The bungy takes the initial load, then stretches out until it meets up with the wire rope. In the spar case it is just cheaper to pair the carbon with wood and get the wood to do all the work that it can within the dimensions that spar aero will allow. If you would need so much wood to acheive a structural need that it impairs some performance parameter, then you need to transition to carbon - inside the tube. This is for bending loads to the tube.


In some case, probably even including spars, the carbon is cross grain.

In some of the loading the carbon is just helping tie the not terribly elastic personality of wire rope, deeper into the spar to spread out it's effect. At some point some nasty metal stuff is going to dump it's loads into some frail wooden stuff. The carbon just pushes this further in and farther around.

The Gougeons always said that carbon was a match for the fatigue rate of wood, not some much in other regards, but they work around those. The mariage issues occur with all materials so carbon isn't special in that regard.

You are on to a very important point. Sometimes people have some carbon kicking around and add it as a sellective reenforcement to wood, because you can do that with glass. Not only can the carbon be insufficient to carry all the loads, but it can also form a stress riser.

When carbon was cheap in the early 2000s, it was starting to rival wood on a cost performance basis. At which point the Meade Gougeon switched to an all carbon spar. That is obviously the future. But where multis are concerned the wide staying base really makes rotating aluminum hard to beat if cost is in the picture at all.

 

Not really correct

Gday Paolo

You have the correct idea about stiffness and such but you can easily combine carbon and wood. You just have to be clever.

Think of it this way. It you get a small piece of wood and pull on it it will stretch. If you make it twice as thick it will stretch half as much (with the same load). If you have a piece of carbon that you want to put on a plywood piece then work out the amount of wood that is in tension (or compression) (be very careful about plywood - only about half of the plies go along the line of force) and times this by the stiffness of the wood. This gives you a stress X modulus value. Then make the carbon thickness X modulus equal them same value. This means they will both take the same load and deflect linearly. You can't do this with carbon and rubber but it works well with wood and carbon. I did it with some beams in my cat and it is done all the time in wing mast construction.

cheers

Phil

 

I think both Paolo and Phil are correct. Phil is correct when the whole structure is designed for equal elasticity of the two components. I think Paolo is correct in most cases where carbon is added to a wood structure.

 

Phil, Tom, In the time from my first post I have spent some time with math.
The test case is one elliptical mast, simply because I have one excel sheet that do the math work... the mast under eval is elliptical, 400 x 160 mm, two layer, the inside is okume ply, 3 mm, the outer layer is unidirectional carbon, high modulus, (I know, the schedule would require some extra layer with cross fibres, but excel don't know this...).
Ok if I compress the mast 0.2 % of the total lenght and it takes about 232000 Newton of load, 193000 N are sustained by carbon, 39000 sustaned by ply.

Two points:

1) one stayed mast, can't go over the stability limit, and that limit is usually much less than the maximum load for the material, then there is no real problem in carbon reinforcement, taking into account Phil suggestions.

2) if we strech the pole (in this way the stability isn't a problem) to 1% of total lenght carbon continue to take five times the wood load, but is close to it's break limit, while wood can't go close to it's limit simply because wood can elongate 5% of total length before failure.

My conclusion: stayed mast can safely reinforced with carbon, taking into account Phil suggestions, to reinforce other shorter structures maybe better to use a more elastic material (glass?) or take into account that only 1/4 or 1/5 of maximum wood resistance is available...

Paolo

 

Young's modulus of Douglas fir 13GPa
Young's modulus of Carbon uni 130Gpa (all very approx)

So the fir will stretch 10 times more than the carbon uni. (Or it its strain will be the same with 10 times less stress) So you need 10 times more thickness of timber to be equivalent to the carbon.

That is about right. A 4 mm piece of 5 ply (2.4 mm going up and down) will need a carbon laminate of about 0.24 mm all around to help it take the load.

Then things get interesting. The mast is going to fail in Euler bending so any way of increasing the second moment of area will help. Therefore putting lots of carbon at the edges of the mast will be of great assistance. You need to make sure you can feed the stress into and out of the carbon unis. Simply putting lots on may overload the shear capabilities of the base laminate. But ply is good for this so this helps.

So you end up with a carbon laminate with some around the spar and some in very localised places - mainly sides and maybe fore and aft.

Maybe Eric Sponberg is watching. He would sort us right out.

cheers

Phil

 

Paolo,

Your findings and conclusions are all correct. Due to the modulus differences between wood and carbon, if you try to mix the two, you will find that you need a lot of carbon to take the total strength, because the wood is not stiff enough to assume any significant portion of the load.

You equate stiffness on the basis of E x I as in beam and column deflection equations. E = modulus of elasticity, I = geometric moment of inertia. Section shapes of the same E x I will have the same bending characteristics. But carbon has large E, wood has small E. Therefore, to be equivalent, carbon can have small I (small wall thickness) but wood needs a large I (large wall thickness) for the same size section. So mixing the two materials, you end up with a required thickness of carbon to take most of the load, and a lot of wood, too, to take its portion of the load. Therefore, you end up with a really heavy mast where most of the weight is in the wood. This is not efficient construction. May as well leave the wood out if you can.

This is why in my wood/carbon wingmast designs, I use the only enough wood to create the shape of the mast. It is a minimal amount just to form the mast section shape, which is easily done with wood and at low cost. All the required carbon fiber for total strength and stiffness is then laid up over the wood, and the wood stays captive in the mast. The weight of the minimal amount of wood is a small price to pay weight-wise for the ease of construction--everybody knows how to work with wood. Wood is also very good for relatively light weight local reinforcement, like where fastenings have to go, or sheaves and axles. I am designing a 29.5 Meter long free-standing wingmast right now for an 80' sloop with just this technique.

Eric

 

Eric, tank you for your comments.
About building mast... one viable solution is (I tink...) to use wood (or other easy to use material...) to support carbon, as you say, but after leave it on the bench...

That is build a U shaped mould, laminate the proper amount of glass or carbon in it, extract it from the mould, place and glue one I beam (carbon or wood? maybe carbon?) near the lower end of U, force the upper end of U to close against a narrow strip of wood (or C profile in carbon) and you have the wing mast.

Never tested on a mast, but works for model airplane wings....
My only concern is the large thin surface without support, it can collapse making waves (don't know the correct english term)...
Hope that you can understand my mind, the language don't help me...

Paolo

P.S. Merry Christmas to you and all the forum...

 

Paolo,

I have seen the technique you suggest, and it does work, although surprisingly, the final shape is not necessarily as accurate as you may think. In the final closing process, you have to be pretty careful with closing the laminate perfectly. It takes care.

You are correct to be concerned about thin laminates. You can make them with cored skins, but this adds unnecessary extra weight. In boat hulls, the loads are perpendicular to the surface and the panels work in bending where the core adds "I beam effect" to resist buckling. In masts, the loads are all in plane with the laminate, so a core does not really help with carrying the load until you get close to buckling. But you should never be close to buckling, so you should not need a core. A core in a mast is just plain extra weight. Masts are supposed to be light.

This means that every mast design should be optimized for the minimum section and planform size. A larger section mast will have more drag and more windage particularly when moored or in very heavy air than a smaller section mast. You want the minimum about of planform size and section shape that will do the job without buckling and without too much windage. So pick the smallest section that will work with a solid laminate--don't use a core. This means that there is basically an optimum mast size for every boat.

Eric

 

The wood has gotta help

Thanks for the heads up Eric

I can't quite understand why wood is of no use. I get that it is heavier on an EI basis but is there not more to this. I would be very worried about light laminates developing localised buckling and also being prone to problems of notch sensitivity.

If Omohundro or Halls make the spar on a nice polished mandrel I think there would be few problems but if the mould has a slight wobble then you could get a large stress concentration in a thin carbon laminate and possible failure. Surely a wood section would help support the thin carbon laminate just like spreaders do to a mast. Wood has high shear and deforms less catastrophically than carbon. On top of this would the calculation of localised buckling be much enhanced by a large section of timber and carbon laminate? (Vastly higher I when thinking of the panel). I remember talking to Nigel Irens about B and Q. It had lots of balsa and less high tech stuff to give it toughness. Too much carbon had given the previous generation of tris much fragility.

I don't like thin carbon laminates. They just fail without warning. Wood fails rather nicely compared - lots of sounds and cracks that may allow you to ease the loading. So what thickness does carbon become less worrisome and more like a normal laminate? Is wood/carbon a good choice for the amateur because wood does give a backup in some hard to determine load conditions?

cheers

Phil Thompson

 

This post is half off-topic, but you wise heads may know if there's another thread somewhere - I cannot find....

I want advice on foam sandwich construction for a a rooftop box for a 4WD - to carry tent etc. Thus it has point loads and vibration to deal with, rather than slamming forces etc as in a hull. I am wondering what is the best type of core - any clues? (I have searched Foam and Sandwich in the main Design pages)

 

Yes, the carbon spar makers have really good tooling that have no hollows or humps in them. Such features are taken care of before a new mandrel goes into production. Obviously, you have to have lots of mandrels for a variety of mast sizes. The same applies to female tooling if such is used. And, obviously, if you are building masts on mandrels, usually using hi-tech prepreg carbon tape, vacuum bags, and autoclave cure, you do not need nor want any wood in the structure to "back things up". That is just unnecessary weight. But it is a very expensive process--expensive materials, and autoclaves cost hundreds of thousands of dollars. In such masts, the reinforcements for spreaders, tangs, and exits, where there might be stress concentrations, are all molded in multi-layered and multi-oriented carbon fiber on the outside surface of the mast where it is easy to keep the laminate in sight and under control during lay-up and cure. Properly engineered and made, any buckling problems should be engineered out of the laminate, as are stress concentrations.

Yes, there are certain important ratios that you have to follow in carbon fiber mast design. For instance, the overall wall thickness should be a certain percentage of the diameter and will vary from shop to shop depending on the care that they exercise in engineering. For my free-standing masts, I generally make sure that the wall thickness is about 3% of the minor diameter--at that level I know that I have just about the right amount of carbon to prevent local buckling (and I can check this by calculation), and I know that if I have way more than this, I am making the mast too heavy and too small in section, I am not at optimum for weight and cost.

Another ratio that you have to be careful of is the amount of unidirectional carbon that is parallel to the mast axis versus the off-axis carbon (+/-45 and 90 deg orientations). Some masts may require a 60/40 split between the axial and the off-axis carbon, and some require an 80/20 split, or other levels in between. It is a mistake usually to go beyone an 80/20 split because you always need some +/-45 to sandwich the laminate inside and out to stabilize the unidirectional carbon and keep it in column. The fabric properties are important here--width and thickness of laminate, etc. They figure into making the actual laminate schedule, which is a bit of a trick to do itself.

So all of this stuff is taken into account in professionally built masts. For the amateur who does not have the tooling or resources to build an all-carbon mast, the wood-epoxy technique is worthwhile where the wood makes up for some benefits that the amateur cannot achieve. This makes the mast easier to build. The price to pay is a heavier mast that is perhaps not as optimum to the boat.

Eric