Composite mast questions

Hi,

I'll be building a mast this winter
Basic dimensions are 45 foot long and 7.5" diameter.

Original specs call for 1.25" walls in spruce.

This should produce a mast weighing (45'(L) x ((2x(22/7)x3.5)/12) (W) x (1.25/12) (thick) * 28lbs/cu ft) = (45 x 1.96 x .10 x 28) = 252lbs

It' won't be my first mast, but....

I'd like to make the mast lighter using thinner walls and judicious use of carbon


Now I'm not made of money so the price of spruce being what it is and the fact that I have a load of marine ply available I am considering using that in place of the spruce.

Now Doug Fir (plywood) is 18% heavier at 33lbs/cu ft but scantlings can be offset due to Doug Fir's superior strength.

I realise the orientation of 40% of the grain is not optimal ie offset 90 degrees.


So, my thoughts are -
if I reduced the wall thickness to 0.75" I'll have a basic spar weighing 252 x (0.75/1.25) x (33/28) = 178 lbs - good but not great
Going down to a wall thickness of 0.5" would result in a basic spar weighing 252 x (0.5/1.25) x (33/28) = 119 lbs - much better

But either of these sizes would be pretty limp noodles.

So 2 layers of 320gm (which orientation +-45/45 , uni, ??) carbon would stiffen it up at a cost of (2 x 8.5sqm x 320gm/sqm x 2 (epoxy weight)) = 10.9kg or 24lbs


Question 1 is - is this sufficient carbon to restore original stiffness or will I need more?

I saw an article that stated that carbon is 1300% stronger and 660% stiffer than Doug Fir for a given weight.

Question 2 - So 24lbs of carbon/epoxy should equate to similar stiffness of 24x 660% or 158.4lbs of removed Doug Fir - Correct?

Mast will be postcured.

Am I on the right track?


Cheers

 

TS,

You have a number of problems here. First of all, to my mind, 252 pounds for a mast on a 45' boat does not seem unreasonable. However, you have to consider that in a spruce mast, the grain runs with the mast axis, properly carrying the compression load, and in a plywood construction, much of the grain will not. In general, in engineering with plywood for strength and stiffness, you have to assume (according to scantling rules) that you get only 22% of the virgin strength of the species when it is combined in plywood. You might be able to use a higher ratio if you have lab testing results that show what the actual strength and stiffness of the plywood really is, particularly in compression.

Also, when combining carbon with wood, carbon is very much stronger and stiffer than the wood simply in raw numbers (ignoring for a moment the specific strength and stiffness when divided by the weight), and the loads on the mast will be carried by the stiffer material in direct proportion to their relative moduli (material stiffness). That is to say, if you put only a little bit of carbon over the wood, the loading in the mast will want to carry through the carbon much more readily than through the wood, because of the differences in their stiffnesses. There is a chance that the thin carbon laminate could be quickly overloaded and fracture. The load would then transfer to the wood, which if undersized, would also fail quickly.

This type of problem is exactly what happened in the 1979 Fastnet Race when so many carbon/aluminum rudder stocks--undersized aluminum reinforced with carbon fiber--failed. The key is if you use carbon with wood, you better use a lot of carbon to make sure it can carry the WHOLE load, and use only a little bit of wood, just enough to make the shape of the mast.

I have on occasion used carbon and wood in the way I describe. They are a good combination if used judiciously. The carbon always carries the whole load, and the wood is there to give the mast its shape, and can also be used for internal reinforcing and to anchor fastenings. If you go to my website, www.sponbergyachtdesign.com, go to the Free-standing mast section, and look at the mast for the yacht Copernicus. The mast building process is described there using prepreg carbon. With wet-lay-up carbon, the process could be similar.

Eric

 

Eric,

Thanks for feedback. It was just an idea. I had a bunch of spare ply and wondered if it was feasible.
I recalled whilst at Gougeons technical workshops, I had seen/heard that some wingmasts had been built in 1/4" ply and e-glass (for 40' boats), but these had obvious greater cross sectional area than simple circlular shape and thus more stiffness.

I've used Ken Hankinsons book in comparing different materials in the past and have had good success, but swapping to carbon is another (and more expensive) matter.

Does the ply/carbon relationship apply to all composite relationships?

Many thanks for your clear and constructive reply,

Dave

 

TS,

The stressform wingmasts from the Gougeons indeed were all wood. I am not sure for what loads they were designed, but they came in three sizes and were intended as stayed rigs for multihulls. Indeed, they were larger in section to handle the loads. They were stayed masts and so were designed to carry load in compression, and presumably there is enough wood in them to handle the intended loads.

Yes, the ply/carbon relationship applies to all materials of different stiffness modulus. The stiffer material will always carry more of the load. They way you check this is to compare the product ExA for each material (for linear loads like compression or tension) or ExI for each material (for bending). E is the modulus of elasticity of the material (in composites, the finished laminate), A is the cross-sectional area of the section for that material, and I is the moment of inertia of the cross-sectional area for that material. To have equal (50/50) load carrying capability in tension or compression, the ExA of each material must be equal. To vary the amount of load between materials, change it by changing the ratio of ExA. For bending, the same applies when comparing ExI.

So you can see, if you have a very stiff material like carbon, which can have a very thin wall thickness, it can be molded with another less stiff material like wood, but the wood has to be very much thicker to make up for its low modulus of elasticity. Adding thickness adds weight, so you don't really gain anything by mixing the two in most circumstances. And when you consider plywood, you only get to assume that only 22% of the modulus is working for you, so you are at a particularly bad disadvantage with plywood. It would be better to use regular timber with all the grain running in the direction of the load and the composite laminate.

In my free-standing mast designs that use wood and carbon fiber, I use only the minimum amount of wood to simply form the shape of the mast. I ignore its contribution to strength and stiffness, and assume that all of the load is being carried by the carbon.

Having said that, I am having a boat built right now in Michigan whose free-standing wingmast will be a combination of wood and carbon based upon equal ratios of ExI. I did not design this mast laminate, the owner did (he is an engineer), and I collaborated with him on his engineering to make sure he was doing it right. You can see this boat on my website (Boat Designs/Sail Boats/Saint Barbara). You can see the boat itself under construction at www.vandamwoodcraft.com. Click on the center photo, go to currently building, then to sailboats--Saint Barbara.

Eric

 

Eric -

I think selecting materials with matching elongations-to-failure is critically important when designing a wood composite spar. If both materials have matching elongations-to-failure, the need to have each material section carry an equal portion of the load is not mandated.

If I remember correctly, ETF for uni-directional carbon fiber laminates in epoxy and softwoods such as spruce, parallel to grain, are on order of 1.5%, whereas uni-directional E-glass is over 4%. Local stiffening in the form of uni-d spar caps over strip planked spruce masts is common in DN ice boat construction, and was used in a strip planked mast design by Gougeon (not Stressform). The DN masts are often built in wood strip, tested, and additional carbon tapes are added until deflection is optimized. The E x A ratios of carbon fiber to wood are not necessarily 50/50. The wood strips are thick enough to prevent local panel buckling while making a significant contribution to resisting bending and compression loads, with the balance of the bending loads taken up by the carbon fiber. The spar is also wrapped with light biaxial glass cloth for torsional strength.

Uni-directional E-glass would be a bad choice here because although it's tensile strength in a hand-laid laminate may be close to the tensile strength of the uni-d carbon, it's elastic modulus is so much lower than the carbon, the wood strips it is bonded too would likely fail in tension or compression long before the E-glass.

There was an old Gougeon Epoxyworks article where JR Watson built a new centerboard for a Searunner trimaran using Western red cedar and carbon fiber, all wrapped in glass cloth, along these same lines. Apparently the original glass cloth wrapped plywood board had fatigued and broken, with the failure from rolling shear between the plies near the neutral axis of the board.

I wonder if using very thin occume marine ply to build a stress skin mast with a carbon overlams would work? Use enough interior bulkheads to insure local panel stiffness. Discount the plywood's contribution in bending or compression (but maybe it helps with torsion?). Is that your approach, or are you orienting all your wood parallel to the long axis? Thanks, and any critique or commentary is welcome.

Chris

 

Chris,

Yes, as I show on my website with Copernicus' mast, the wood inside is meant to give section shape stability and local reinforcing, as well as provide the male shape for the mast itself. The carbon fiber laminate takes all the load and the bending strength and stiffness of the wood is ignored. Considering the fact that the factor of safety is about 3, the error in mast bending performance by ignoring the effect of the wood is negligible.

The interior bulkheads were in fact plywood (I don't know off-hand if it was Okoume ply, but could have been). On the sides of the mast, I specified spruce veneer with the grain running parallel to the mast axis. This is because the veneer has to bend around the shape of the mast sides, and plywood would have been too stiff to take the bend easily. Yet the veneer is stiff enough to support the carbon fiber during molding.

Eric

 

Eric -

I looked through the wingmast articles on your site, and will take time to read them more thoroughly. Very cool. At risk of steering this thread off course, I'd like to ask about you r re-design considerations for the boom system on Wobegon Daze:

1) By eliminating one arm of the wishbone, would you be concerned about complicating (and making much heavier) the goseneck and boom arm configuration to deal with assymetrical loadings, new couples and bending moments at the fittings, etc?

2). If the pivot points for the wishbone gooseneck were moved forward of the mast track onto tangs or brackets on the mast side wall, and a corresponding clew plate was used with holes an equal distance apart, couldn't you in effect get a parallelogram that would allow the wishbone arms to pivot independently of the mast without binding or pinching the mast track?

3) With a loose footed sail that has camber or draft built into the foot, would removing the lower boom cause the clew to lift? I'm thinking the "boomlet" isn't just a place to furl the sail, but also a tension member that reacts against the wishbone lifting/compressing.

Chris

 

Chris,

Answers to questions:

1) Eliminating one half of the wishbone reduces boom weight by nearly half (remaining wishbone would have to be made heavier) and the gooseneck fitting would be a universal joint affair similar to standard boom goosenecks where the boom can go up/down/left/right. Using a universal joint gooseneck reduces all compression to a point load on the gooseneck pins. You have to guard against over-rotation of the wingmast relative to the boom. This is handled by stop lines on top of the wishbone that are attached to the leading and trailing edges of the wingmast and run back along the wishbone to the clew end where they are stopped on cleats. This is a very simple settup. The wishbone also has to be held level, because otherwise its weight causes it to fall with the bow of the wishbone down. A keeper line attached to the wingmast up high and to the wishbone at about 40% of its length back from the gooseneck holds it up.

2) We looked at the setup you suggested and realized that all of the fittings would bind and jam, and they would wear quickly. It was too much hardware to deal with. I built a model of a wingmast to try various ideas, and that one got scrapped early on.

3) The mainsheet at the clew end of the boom prevents it from riding up, at least when sailing on the wind with the sails relatively close hauled. Granted, when the boom is swept well outboard and the mainsheet cannot follow, then some kind of keeper line or wire from clew to the base of the mast might be in order. The attachment of this line/wire can be moved forward of the clew somewhat so that it can clear the lifelines as the boom swings forward.

In the redesign of Project Amazon's boom when Reidl sold it to Stricker, the conventional booms were changed to wishbones with a tension wire from the clew to the base of the mast at deck level in order to keep the clew low. Near the mast there was a plate of aluminum on which were mounted gear for handling the reefing lines.

Eric