Need help with trimaran beams

Hey All,
Making good progress on the 23' trimaran. After toiling with box beams and aluminum beams I wound up building laminated beams from juniper bonded with epoxy. Photos are attached. The beams measure 3.5 inches tall by 2.5 inches wide. The beams deflect about 3 inches over 96 inches with a load of 250 pounds which seems excessive based on what I've read (perhaps not?).

I've also read that adding carbon tow on the juniper makes the juniper dead weight for the most part since the modulus of elasticity are greatly different. This makes sense. Would adding a layer 10oz glass (with epoxy) be the right call here? Or wat would be to stiffen them up.

Funny thing is that a doug fir 2 x 4 (dimensional lumber) is lighter and deflects less. Makes me think about using a simple 2 x 4. But the shape is wrong.

The photo below is before the 6mm shear web was added to both sides.




Any suggestions??

Thanks,
T

 

How have you loaded this?

Is the beam a cantilever, and thus the load applied at the free end?...or...have you supported both ends and applied the load centre span?

The supports, is this simply a small support the ends are free to rotate, or have you constrained them?

The applied load, is this a weight that is applied via some means, to a small location, like a wire wrapped around the beam and then hanging the weight on the wire...or have you applied the weight in another way?

Roughly what thickness is the finished laminate of the beam you tested..or have you applied the load on bare wood frames (as per your pic) which are simply "glued" together with epoxy. This is not 100% clear.

 

Clarity

AdHoc
The first beam had two 1 inch holes drilled through the entire beam. One hole was 3 inches from the end and the other hole was 30 inches from the end. 9" x 1" lag bolts with fender washers were screwed through these holes into a section of 10" x 10" lumber used as a post for a house that was laying horizontally on the ground. This lifted the beam 10" off of the ground. 250 pounds were then applied to the other end of the cantilevered beam which was applied between 80 " and 95" from the cantilevered end.

The weight was applied quite easily. I sat on the end and measured the deflection which was between 3" and 4".

I'm a novice enjoying the challenges having learned more during the build than I knew in total when I started. The idea is to have the amas pivot at points on the beam supports which you cans see in the photos below..

 

The finished laminate is what you see in the photos + 6mm shear webs on the outside. SO the total vertical height is 3.5" and the total width is 3". Material is North Atlantic Ceday (Juniper). Total length is 96".

 

Hmmm..this is not making much sense. It does to you, but not to me. Do you have a picture of your set up..or a drawing showing what you did?

Are you saying that you used 2 holes, spaced at 30inches apart to fix/secure the beam to something solid? And once “fixed” this was your support for the beam to be loaded at the opposite free end?

What is holding the 10x10 post and how? Was there any movement, in any direction, of the 10x10" once the main beam was loaded?

You sat on the end…..is the 250Ibs weight you, not a known fixed dead weight?

Between 3 and 4..is a lot, it is a 30% increase in deflection, thus you need to be accurate.

You still haven’t answered my question about the beams and the thickness, or not, of the laminate on the wood beams.

 

On the left end of your beam, you need to place the spacer blocks like you have on the right end.

1. I think you have the grain strength running width wise of your beam. Turn a beam on its side, spacers in the middle, and see if you do not have less deflection.

2. Your overall 'design' would be the method I would use over a 2x4 - forget the weight, 2x4 will fail once, the ladder like design will (should) be stronger, and would allow you to limp home - more often than a single 2x4.

3. Design sideways from your current design, lighter wood if you want. Steam it and get the curve you want. It seems you want a long bow curve to the Aka, not a bad idea. Then place the two beams together with a spacer in the middle.

I would go for a single spacer, but I do not think a double spacer would hurt you, I just like the look of a single spacer, once you have the shape right, wrap in fiberglass, carbon, or other appropriate material.

To keep everything in the beam in place - it will also greatly strengthen it overall.

wayne

 

Are you saying that you used 2 holes, spaced at 30inches apart to fix/secure the beam to something solid? And once “fixed” this was your support for the beam to be loaded at the opposite free end? Exactly Correct!

The 10" x 10" did not move at all (at least anything to the naked eye). It is 30 feet long and the lag screws were on one end.

I am a fixed dead weight at 248 pounds. The deflection was measured 4 times. 3.5", 3.50". 3.6", 3.5". It probably changed a bit since I was not sitting in the same exact position each time

Each of the beams are made of 4 pieces of 7/8" x 2.25" x 96" juniper laminated around a form with 2 pieces on the top and bottom taking the curve. Separating the two lengths on top and bottom are 4 sets of spacers that are 3" long by 1 3/4" wide. See photo. The hole at the 30" point is drilled through another spacer that is 12" long (point of greatest point load I believe) and the end for the first mount point is drilled through a 6" spacer.

Covering both sides of the frame that you see is 6mm (.25") okume plywood. Everything bonded with epoxy and temporarily clamped with drywall screws.

 

Can you set it up the same way you have it in this photo?

See if your get the same deflection 'sideways' that you get 'right side up.'

It is hard to tell in the photos, but I think the wood grain is stronger 'sideways' than it is 'right side up.'

 

Here's the detail of the beams..

There is another block fitted in the end which will be the connection point for the ama. This is not in the photo since it is not in place yet. I had a test spacer in place for the deflection test. Building the connection spacer is a bit tricky since I have to build some angles into it and design for a bearing/bushing for the pivot.

Right now the beams weigh 9.6 pounds each. If I add expensive carbon is the 9.6 pounds of mass doing anything other than creating a form for the carbon? Or does it provide something as well?

 

It would be both, you cannot 'remove' the strength of the beam by adding a wrapper ....

But, did you try it sideways?

 

I haven't tried it yet.. Dark and raining tonight.. But it seems like the grain is going the same way either sideways or vertical and I have a greater dimension and more material running in the vertical direction.. I'm most likely not understanding your theory..

 

You will have greater strength running along one side of a board. There can be exceptions, but not often.

In the pictures, it looks like the grain is wider along the widest side of your boards (2.25"), if that is so, I think that is your strong side, in other words, the thin 7/8" sides which you laminated together look like they are the weaker direction of your wood.

I am guessing you are comparing the weaker side of the grain in your beam to the stronger side of the 2x4.

I am not real good at the grain, yet. But, grain direction is important when designing your structural components.

 

I would think you want the strength in the vertical, so why are do you have the 2.25 side horizontal instead of the 7/8 side?

 

The moment of inertia of a beam is bxd^3/12 this determines its stiffness together with the modulus of elasticity of the material.

If this beam was laid on its side the M of I would be 2x ( 7/8x 2.25x2.25x2.25/12) = 1.66
I would not bother testing it sideways on, as the beam is bent, you will have a lot of torsion. You would need to make a new beam.

For this beam as you have tested it, M of I would be 2.25x3.5x3.5x3.5/12 -2.25x1.75x1.75x1.75/12 = 7.03

M of I for a 4x2 vertically is 2x4x4x4/12 = 10.67

For simplicity I am ignoring the contribution of the ply in the calculations.

The cantilever length for calculation would be from where the support to the beam stops to your centre of gravity, not the total length of the beam.

Also Richard Woods on his beams has the same orientation as you.

I can not help you on how much deflection is too much. Some flex may be helpful unless they are supporting the mast.

 

Latestarter,

You are correct that the beams are built and this curve is oriented in a way where I cannot lay t sideways and building new beams would be the course of action if I were to change.

The more I dwell on it the more I think some flex is a good thing since the weakest point in the systems is most likely the beam connection point on the main hull. Pretty safe to say the all of the forces are not ideal for each other between sail area, ama flotation, beam connection and beam strength. Something is the weak link and it is most likely the pivoting connection point. Building uber rigid beams will simply increase the force at the connection point which might not make sense.

I'll most likely sheath in a light glass and call it a day. If for some reason I find I need greater stiffness I can either add carbon or start over and try and create a much lighter beam. For now the project moves forward.

Would adding in the shear webbing simply be a matter of adding in .25 x 3.5^3? Is that how that would work?

Nimble. The 7/8" is on part of the lamination. It is actually 7/8 x 4 on the vertical direction for a total of 3.5".

El G. This was a simple way to build the beams. Increasing the vertical dimension gives a cubed function for stiffness so that's why this is the greater dimension. I played around with 4" x 4" box beams that were incredibly stiff and much lighter however they were straight and would have been in the water too much of the time. Also, they looked lame. There is a way to build curved box beams but that was getting a bit too laborious for now. Maybe in the future..