Need help with trimaran beams

Discussion in 'Boat Design' started by FirstLight, Oct 8, 2013.

  1. FirstLight
    Joined: Sep 2007
    Posts: 78
    Likes: 6, Points: 8, Legacy Rep: 21
    Location: North Carolina

    FirstLight Junior Member

    Latestarter, By the way the standing rigging will not be attached to the amas but to the ends of the beam supports so the flexibility will be ok in that respect..
     
  2. latestarter
    Joined: Jul 2010
    Posts: 402
    Likes: 51, Points: 28, Legacy Rep: 233
    Location: N.W. England

    latestarter Senior Member

    Given the way you have it now is 7.03/1.66 =4.23 stiffer than on its side it would be a waste of time to change.

    If the webs were made from the same material and the grain in the same direction you could add .25x3.5^3/12 for each web (you missed out the /12 ) with ply it gets more complicated as half the plys are pointing in a good direction and the other half bad. Very roughly you could count half the added plywood as being effective.
     
  3. Skyak
    Joined: Jul 2012
    Posts: 1,462
    Likes: 145, Points: 63, Legacy Rep: 152
    Location: United States

    Skyak Senior Member

    Beam deflection is not a good thing in a multihull. Yes something must break first and that is generally the rigging. Flexy akas can result in impact loads as they snap back in wind and waves.

    Your beam has two flaws that I can see. First it's a constant height but load is much greater at the hinge to the main hull. Who designed this? The second is your use of short blocks as shear members on a curved beam. The blocks in shear can rotate slightly bending the tension and compression members locally. This is why your beams are not as stiff as a similar dimension solid. Even if the design was sufficient with a boxed beam your blocks would be dramatically weaker.

    I suggest you box in the beams at least with thin plywood, preferably inside the compression and tension members rather than outside shearing the glue. If you don't box with wood, thin fiberglass might have a thin plate defection mode. If you end up adding carbon as you say, start at the main hinge. If you apply it at 45 it will more closely match the wood deflection in the tension/compression members and add maximum stiffness to the shear.
     
  4. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Sorry was busy yesterday, now I have a little time to review again.

    In terms of the beams per se, i couldn't get exactly the same values you posted, in terms of deflection, from the data you supplied. Thus the data is slightly incorrect, but "consistent".

    A dwg/diagram of exactly how/where you loaded this is soooooooooooooooooo much easier, avoids misunderstanding and confusion.

    So, if i take "mid values" since i can interpret you "words" into actual locations of loads, (applied loads and locations), several ways, since the values are also inconsistent, however, I get "similar" values, that being a deflection of 65mm, or 2.5" in your funny units.

    The thing to note is that the stress in the beam is roughly a FoS of just 2 in compression.

    It appears from your beam that the only thing separating the two 7/8 x 2.1/4 sections are the spacers which are glued to the 96" lengths?? Therefore the shear strength of your beam is the strength of the epoxy glue in shear. Which is similar (relatively, but actually not very good) but less than that of the compression strength of the Juniper. Thus the beam shall fail in shear first. So what shear loads will the beam experience? See below.

    Depending upon the values I use, to get my answers, the shear stress in the epoxy glue is approaching the failure of typical epoxy.

    However, all this is moot. Since what loads will you outriggers experience?...you need to define the loads then see if the beams are suitable.

    BTW..flexibility in structure, can, if designed correctly, be a good thing. However it can also be a disaster in the wrong applications. You don't want your outriggers moving independently from the main hull!
     
    Last edited: Oct 9, 2013
  5. FirstLight
    Joined: Sep 2007
    Posts: 78
    Likes: 6, Points: 8, Legacy Rep: 21
    Location: North Carolina

    FirstLight Junior Member

    Thanks Fellas

    Guys,

    Thanks for the great feedback.. Helpful and educational.

    I liked the idea of bonding the 6mm skin to the inside of the frame however we already had the 6mm skin on the outside. Bummer.

    In retrospect I would have built them tapered for sure. That would have been so easy to do I'm kicking myself a bit on that one. This is a learning process for sure. Enjoying the ride. Designing was fun and building great. Making mistakes is part of any great adventure.

    As Skyak suggested I'll add carbon tow in 45 degree orientation with a couple extra layers at the vaka side. How much to add I'm not sure. Any thoughts?

    I sort of think of these beams as an experiment and if they do not work I'll most likely purchase 4" OD x .125" wall 6061 tubing. I feel like I'm far enough along on these to see them through and see how they work. They are getting heavy for sure and will wind up being 12-13 pounds each.. 50 pounds of beam total. Not sure if that ts good or bad..

    It made sense to hear that the rigging would be the first thing to go. That would make sense having the rigging be the constraining factor instead of the beam mounts.

    Below is a photo of one of the beams with the skins. Maybe I should just laminate the three together. That would be stiff..

    Anything else.. Hit me with it!
     

    Attached Files:

  6. Richard Woods
    Joined: Jun 2006
    Posts: 2,209
    Likes: 175, Points: 63, Legacy Rep: 1244
    Location: Back full time in the UK

    Richard Woods Woods Designs

    I am catamaran sailing in Greece right now, so have missed this thread. I don't know anything about your boat except its 23ft long

    I have to say I think your beams are way too small. "use your eyes and plagiarize' and do some reverse engineering

    A typical 23ft trimaran might have an aluminium beam 5in x 3in or a 4in dia tube with waterstay. That would certainly be a safe starting point

    The 250lb load doesn't mean much. You have to assume 4 men on the windward float, split between both beams that is 500lb loading on each beam

    or assume the lee beam is being deflected by water pressure. You would expect at least 100% buoyant outrigger, so say 500kgs total displ, or 1000lbs say. Way more than your 250 lbs (x2 for 2 beams)

    typically the fore/aft loads are about 1/3rd the vertical loads

    you say "building new beams would be the course of action if I were to change."

    That assumes they don't break when you are out at sea......If you are lucky you only lose the boat. Build new ones now, not after they break

    Richard Woods of Woods Designs

    www.sailingcatamarans.com
     
  7. FirstLight
    Joined: Sep 2007
    Posts: 78
    Likes: 6, Points: 8, Legacy Rep: 21
    Location: North Carolina

    FirstLight Junior Member

    Great Points..

    This application is probably a bit atypical as we are doing flat water sailing in a protected sound and the boat is designed to have 300 pounds of moveable ballast (crew weight). The idea is to be able to row the boat with amas in a folded position since we have a dock crammed up a tidal creek about 1 mile form the sound. Being able to row it to the sound is the goal. Although I think I overachieved on the size. I'd love to try the Watertribe event as well if she rows ok.. Richard, your Strike 18 is what got me so damn excited about trimarans! I love that design! I was unsuccessful at contacting you for a few weeks.. Sounds like you've been having some great sailing!

    Just the same I'm switching from laminated beams to alloy beams. Reading a bit more about flexible beams I get nervous about the shock loading. Also there are many less inherent construction issues with alloy beams than my laminated beams. We'll finish them up on a rainy day down the road and see how they work out. Still nervous about the pivoting beam supports.

    Here is a photo of the design with the alloy beams. I'm thinking of using a 4" x .125" section for the horizontal component and 3" OD x .125" for the vertical component that will be welded to the 4" section 12" from the end. Sound about right?

    10-11-2013 7-15-26 AM.jpg
     
  8. Skyak
    Joined: Jul 2012
    Posts: 1,462
    Likes: 145, Points: 63, Legacy Rep: 152
    Location: United States

    Skyak Senior Member

    So your loading is lighter than a typical 23. The watertribe EC 300 is not protected waters -you will want full strength (I am a fan of the watertribe events too).

    The 4 inch beam height was for water stays with the beam in compression. You are still showing a bending beam configuration -NOT COMPARABLE. The ama end makes me queasy too, welding, a hinge... if the beam stays round you are wasting material. If you squeeze it oval...

    While we are waiting for Richard have you considered a sliding beam? If you want to stick with swing beams my mind drifts to a tapered wood/fiberglass compression member with a metal water stay triangle.
     
  9. FirstLight
    Joined: Sep 2007
    Posts: 78
    Likes: 6, Points: 8, Legacy Rep: 21
    Location: North Carolina

    FirstLight Junior Member

    Different Beams

    Guys,
    Thanks for all of the responses. Based on your comments and reality checks I've redesigned the beams and putting the laminated beams on the shelf and going for a simple aluminum extrusion with water stays.

    The thought process has always been to be able to sit on the weather ama and steer and it seems like using the suggested safety factors the laminated beams were way to whimpy and needing lots of carbon. In the end it would be more efficient (time and money) to buy 4" aluminum beams than play the carbon game.

    Hoping for some feedback on the beam mounting and pivoting idea. The thought was to fillet weld a 1/8" aluminum plate to the top and bottom of the 1/8" wall 4" tubing. This should mitigate the point loading and at least create line loads and partial face loads where the beam\plate intersect with the 3/16" x 36" aluminum channel mount. There would be a 3/4" stainless pivot pin running through the assembly 36" from center line of the boat.

    If there is a better way to do the beam mount that 3/16" aluminum I'm all ears. And also curious what the best way is to attach water stays to a beam? I would guess a strap around the beam to avoid drilling holes?

    Any and all thoughts greatly appreciated.. Blow it up if it needs to be blown up.

    Below are three links that illustrate the design...

    1. The first image is the ama in the extended position.
    2. The second image is the ama in an intermediate position.
    3. The third position is a close-up of the rear side of the beams showing the plates on the beams.

    Thanks,

    T
     

    Attached Files:

    • P1.jpg
      P1.jpg
      File size:
      92.1 KB
      Views:
      565
    • P2.jpg
      P2.jpg
      File size:
      95.6 KB
      Views:
      454
    • P3.jpg
      P3.jpg
      File size:
      53.7 KB
      Views:
      389
  10. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    If I understand your no.1 image correctly....you're heading for failure.

    As the outrigger (ama) rotates and displaced vertically, what will happen to the beam at the outrigger end? This shall follow the outrigger. At the other end what do you have resisting this....?

    As the beam takes up the same displacement as the outrigger all the load shall go into the very thin 1/8 wall thickness at a distance of 4". So the couple shall create a very high localised load, in other words the pn shall tear through the thin walled material.

    For example, if you outrigger weighs 100kg and the lever from the outrigger to the connection is say 2.0m, and accounting for a simple 2g load, you have a bending moment of 1.96kNm. Taken over 2 beams = 981Nm. With just 4" separating the 2 skins (of the bolted joint) that means you have 981/0.101 = 9710N into the 1/8 wall. With a pin/bolt of 3/4 that gives a bearing stress of 170MPa. It wont be long before the pin starts to tear through the thin 1/8 plates top and bottom.

    This is all very quick simplistic calc's and takes no fatigue into account, which is a must. And since your welding these 1/8 thick plates to the tube, you've lost the un-welded strength. And then with a weld in SW environment, the allowable stress, in a fatigue is circa 10-2% of yield. And that is with no FoS too!

    Bottom line is, you need a more secure arrangement and taking the method of fabrication and environment in which the joint is exposed into account.
     
  11. FirstLight
    Joined: Sep 2007
    Posts: 78
    Likes: 6, Points: 8, Legacy Rep: 21
    Location: North Carolina

    FirstLight Junior Member

    AH,

    I think I understand what you are saying.. You are suggesting failure when the beam is being retracted.. Correct.

    Does this suggest you think it is ok when being used in extended mode?

    The total wall thickness would be 1/4" on both sides since I am gaining 1/8" from both the tube and the plate..

    There are a couple other things I can do to mitigate the stress points when being rotated. I can increase the plate area at the end of the beam support which could spread the load out (you can see in the diagram below I have added the angles feature to increase surface area. Additionally, I can add thickness at the end of the beam support on each side by 1 to 2 inches to increase the span from 4.5" up to 8.5" or something like that. A couple diagrams below..

    I will have everything anodized for corrosion resistance.
     

    Attached Files:

    • P4.jpg
      P4.jpg
      File size:
      98.6 KB
      Views:
      391
    • P5.jpg
      P5.jpg
      File size:
      159.3 KB
      Views:
      385
  12. Stumble
    Joined: Oct 2008
    Posts: 1,913
    Likes: 73, Points: 48, Legacy Rep: 739
    Location: New Orleans

    Stumble Senior Member

    If you a re using aluminium save the trouble of anodizing it. It will have no real effect on the corrosion resistance of aluminium parts in this application.

    As for the design... Way out of my league.
     
  13. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    No.

    You only have one pin and a thin walled connection. See below:

    Tube joint.jpg

    As the outrigger moves up and down, it will be resisted by the pin's contact with the tube. There is nothing else. The force is just a simple couple. But the force on a thin wall is problematic. One joint and a thin wall is insufficient to do what you are attempting. It'll fail.

    As Stumble says, no need to anodise. Bear ally is fine on its own. The only issue will be the dissimilar metals you sue for your connection. So be careful there, as that'll be your source of corrosion.
     
  14. FirstLight
    Joined: Sep 2007
    Posts: 78
    Likes: 6, Points: 8, Legacy Rep: 21
    Location: North Carolina

    FirstLight Junior Member

    ??

    AH,
    Thanks for the diagram. I understand what you are saying. Am I missing something. In my mind I see that this type of moment will not occur at the pin since the plates which are welded to the round beam would take the load from the inside of the beam mount box.

    In the diagram below if the blue plates that are welded to the purple beams are in contact with the surfaces of the yellow box wouldn't that prevent the type of rotation to which you refer?
    \
     

    Attached Files:

    • p7.jpg
      p7.jpg
      File size:
      116.1 KB
      Views:
      327
    • p8.jpg
      p8.jpg
      File size:
      114.4 KB
      Views:
      341

  15. Richard Woods
    Joined: Jun 2006
    Posts: 2,209
    Likes: 175, Points: 63, Legacy Rep: 1244
    Location: Back full time in the UK

    Richard Woods Woods Designs

    In general its not a good idea to put holes in the flanges of a beam, you should try to put them in the neutral axis where there is no load.

    http://en.wikipedia.org/wiki/Neutral_axis
    http://en.wikipedia.org/wiki/I-beam

    How do you effectively attach a flat plate to a round tube? There should be a better way. I'd extend any reinforcing outboard of the main hull wing to help spread the loads. The pivot bolt as shown is at the max load point

    I'm not sure of your geometry. Will the waterstay slacken as you fold the beams? What about the trampolines? How are you attaching them? The Dragonfly way?

    Richard Woods of Woods Designs

    www.sailingcatamarans.com
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.