Need help with trimaran beams

[Ad Hoc]

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.

The question is, how does the load go from the beam connecting to the outrigger, to the beam connecting to the main hull?

What you are describing is simply that the pin is a "locator" and to "hold" the beam in position. In order for that to work, you need to establish whether the beam is capable of transferring the load from one to the other without failure, in this method.

This is exactly the same as a recent question noted here:
http://www.boatdesign.net/forums/bo...aran-genius-deathtrap-47223-2.html#post632260

The only difference between the on on that thread and yours is that yours is inside, otherwise the calculation is ostensibly the same. So you need to work out the moment and can the moment be transferred in the manner shown, but from inside to outside tube.

 

[FirstLight]

Lost the Pivot Points

Well guys,

After much tinkering in the shop, researching and blowing through paper in sketches I've decided to change things up away from the pivot point and go with something a bit more tried and tested.

In the end it was difficult to get an elegant solution for the pivot point. So I decided to change it up to the sliding beams with a waterstay. 4" OD x 1/8" wall.

I played around with beams that nested into each other. That was the most elegant solution with different OD tubes on port and starboard with one telescoping into the other. Getting one of the sides nailed down was easy as the smaller tube could slide all the way through. However, figuring out how to move the larger OD tubes through the opposite side ama was trickier and more experimental than I could sort out.

So I've gone the way of others here.. Trying to determine if I should use larger OD aluminum tubes for the mounted components with some sort of ring or if I should simply use the 4" OD aluminum tube as a mold (with a .010" or so?) offset to laminate fiberglass tubes. Any ideas? Here are the new 3D models...

Thanks for all your input.. I look at where I started and am pretty sure it would have been a disaster..

 

[FirstLight]

One Last Idea..

OK Fellas,

2am a few days back there was an epiphany of sorts regarding the pivoting beams for the amas.. Pulled out the trusty Solidworks and mocked up a couple of the ideas to see how things might play out. I added a 1.5" aluminum tube/strut (this be red in the images- is this considered a water stay?) to the 4" tube and brought that to the lower side of the pivot pin (blue in the diagrams)... The pin from top of mount to bottom of strut would be 12".

I read the information Richard Woods posted regarding neutral axis and moved the pivot point to this axis (off by a little bit).

And then the irony of it all was seeing a Strike 15 trimaran on line with something similar...

Other question is that the main beams would not really be in compression in this scenario unless I preloaded the lower strut somehow. Would this be ok in this scenario???

Please let me know if you think this has any merit??

We got the main hull skinned last night... Felt good to get that done.. Photos are at www.mildrice.ellavickers.com...

Thanks All!!

 

[FirstLight]

Giving the Beam Mount a Shot

Building up one beam mount after tinkering with the software a bit more.. I figured a bit of empirical data would be a good thing. So we are building the mount according to the following image and will load up a 4" beam x .065" wall with 3/16" waterstay and apply 275 pound load to the end and see how she goes..

From reading what others have done I planned on .10" of carbon for skins on all horizontal surfaces and .05" carbon on vertical surfaces. The "core" is .75" juniper \ cedar. I'm not sure how this plays out math wise. If this sounds way off please let me know..

More to follow as we move forward.. Thanks again for all your input and help..

 

[SamSam]

I haven't read the thread, but it looks like the loads will be concentrated most at the top of the end of the square tube.

If load handling capabilities of the round tube are exceeded, that will be where the top of the round tube would crimp and then collapse. That might be mitigated somewhat by having the last few inches of the square tube end with a round hole that would give circular support to the round tube.

For the square wood tube, an overload would put all the stress on the top board, the top glue joint and the last screws in the end. That force might be spread and transferred to the sides and bottom and length of the square tube by a metal or carbon fiber collar wrapped around the end, somewhat like the iron bands on this wooden maul that keep it from splitting.

 

[John Perry]

Your drawing with the telescoping tubes is a fairly well tried solution used by various designers/builders, so it should be possible to make it work. I would suggest stock aluminium tubes with fairly thin walled plastic sleeves (e.g. delrin) turned and mechanically fastened inside the outer ends of the central tubes and outside the inner ends of the outer tubes. We did something similar for a retractable bowsprit and it worked ok. I wonder how you would make the nice looking curved ends to the outer tubes - do you have a supply of suitably thin walled large diameter ally pipe bends?

You have now gone back to pivoting beams with vertical pivot axis. I guess your arrangement with the 'waterstays' could work in principle provided that you can provide a suitably strong mounting for the inner ends of the waterstays - your drawings dont seem to be at all clear about how that could be done. Assuming that the waterstays are intended to carry both tension and compression, the inner mounting needs to carry loads in both directions athwartships and both directions fore and aft. A point to note is that the worst situation may well be when the beams are slightly folded in. As the beams approach the fully folded position the pivot points at the inner ends do not need to carry large loads since vertical loads on the floats can then be reacted against bending moments in the outer parts of the swinging beams. I agree with SamSam that it could be well worth while to provide strong 'end stops' for the swinging beams so that the pivots for these beams do not need to carry the compression load in the beams while the boat is sailing.

I find your Solidworks drawings look a bit messy - I wonder why you have black lines all over your curved surfaces. There is a document properties option not to display lines where surfaces meet with common tangent planes, perhaps it would help to set this option?

If you have the version of Solidworks with Cosmos built in you should be able to model all this with FEA and check for yourself that all the parts are strong enough for the intended usage.

Looked at your webpage - looks like you are making good progress on the hulls etc. I absolutely disagree that the place to draw a part is on the actual part during the course of construction, but we all have different ways of working!

 

[FirstLight]

Beam Collar and plates.

SamSam, Thanks for the thoughts there.. I was going sort of the same direction you proposed with Thor's Mallet although a bit different.. The loads will be the highest at the outboard end of the "u" section for sure and closing the "u" seemed like a good thought to distribute the loads and avoid hinging .. At first I was considering a hinged "clasp" that would connect from the top of the "U" to the bottom but then went to the idea of a sliding collar (similar to your thought) that would fit over the end of the "U" when the beam was extended and slide over the aluminum tube when retracted.. In the photos below the collar is colored black (as you mentioned in carbon or aluminum). I was almost thinking of wrapping some visqueen around the end of the beam mount when the tube was extended and using it as a mold to create the collar. Not sure yet. I would imagine the name of the game is to transfer the loads from the "u" to the collar and having a close tolerance fit is important. The collar is out of position in the second photo since it is concentric to the beam mount in the CAD drawing and not the tube (which it should be).

Another thought I had to avoid the point load at the end of the beam was to weld (checking with my welding guy today) 1/8" flat plates (blue in the photos) to the top and the bottom of the tube. In theory this would create more of a surface load (or a line load at least) and assist with torsional loads. If I cannot weld the plate to the tube then perhaps I can use a healthy fillet?? Any thoughts here appreciated...

I like to think that you are boat building with the maul you have in the photo??

Here's a few pics.. Flipped the boat yesterday.

Thanks for the great thoughts!!!

 

[FirstLight]

FEA, Solidworks and Waterstays.

Hey John.. Thanks for the input.

Learning Solidworks as I go. . I've looked for the "black line removal" and found something that seemed to remove most of the lines. Although not all of them.. However this is a vast improvement and seems to refresh a bit faster.. Thanks!

I've tinkered with the waterstay mounts and am leaning toward laminating in a 1" x .5" aluminum bar (yellow in the photo below) across the front of the bulkhead against the floor with epoxy and 3" tape.

The stay would then be attached to the end of the waterstay bar. This bar would be continuous athwartships.

I'm not sure how I would mount the stay to the bar however there seem to be several options. I would imagine low drag would be the best way to go to reduce splashing and noise?

With a bit more experience with this software I might try the FEA stuff. I've played with it a bit however I'm not sure how to model up carbon laminated to wood with epoxy..

Right now I'm going to experiment with 7 oz x 4" carbon tape on the inside corners and 17 oz glass on both the inside and outside for stiffness of the wood.. Is this enough? I have no idea.

I suppose this is what sets the men apart from the boys. The ability to create good structures. I'm a bit of a fraud in this department mostly copying what others have done in the past. However, there are not a lot of examples of beam mounts like this. So I'm hoping to build them and test them. I've made the distance between the upper and lower parts of the beam mount 4.5" using a tube with a 4" OD. So that gives me .25" on each side of the tube for additional re-enforcement if necessary. If at the end I have an adequate structure and a bit of a gap I'll laminate in a thin plate of G9 or something..

I have no idea if I'm on the right track here.. Any thoughts, fire away!!

Thanks..

T

 

[tspeer]

Here's something to think about. The elastic axis of a C channel lies on the outside of the web. If you load it inside the C, it will twist. That twisting needs to be taken into account in the design of the beams and the calculation of the stresses.

Also, the top flange of the open-ended C channel must take the load from the outer beam pressing on it. The bottom flange won't be contributing very much.

It would really help is there was some sort of locking mechanism that came down around the end of the fixed beam to carry the loads around the pivoting outer beam. That would help to engage the bottom flange and turn the C channel into more of a box beam.

 

[FirstLight]

Loads

Hey Tom.. Thanks for the info on the C-channel.. Never thought about how the torsional loads were created.. Makes sense..

I was thinking of doing what you suggested and adding a collar around the end of the C-channel (end of the fixed beam) to close the load at its highest point. In the image below the black ring represents a sliding collar.. Not sure how that will work yet but should probably be rigged in a way where it can be tightened to compress the ring against the flanges...

Is that what you were suggesting or am I off here??

Cheers,

Tony