Connection between cross structure and demihull

Not sure where you found that diagram, but i'd ask for my money back!!

The shear is constant, in the absence of any other applied load for that simple cantilever analysis. That curves shows the slope/deflection, but is also wrong way around. The built-in end is always zero.

 

AH,

Sorry, you are not going to get your money back. Equation 1 is when it is a pure cantelever and the shear is constant, I use equation 2, slightly modified with a partial load on the far end because I think the hull does not get lifted out of the water. It just tend to rotate on the center of gravity.

I use equation 1 on cat sailboats when one hull gets lifted out of the water due to the lever acting on the mast, rotating the two hulls along the middle of the crossbeam vcg.

A third equation which seems to be more appropriate for a sailing cat is a couple when there the lever twist the midspan of the mastbeam due to windload acting on the mast. This twist the mast into an "S" shape. Not much progress though. One equation is for a fixed beam, the right one written in Russian.

 

RX

There is no 'increasing' load in the diagram of the demihull being 'fixed' to a built-in connection as you show here. The weight of the demihull does not go up and down, as that would imply weight being add then removed and also from the demihull to the fixed built-in end. It would thus imply the draft is constantly changing too.

You need to be clear what load case it is you are attempting to analysis, rather than applying what looks like a similar situation when in fact it is not.

It matters not if the hull is totally or partially lifted out of the water. There is a simple bending moment, the force x distance. The distance from the built-in end to the centre of the demihull (assume the CoG remains constant on the demihull) is constant. The force is the hulls weight. The max is when the hull is out of the water, and lesser when partially, as it is still supported by buoyancy. Thus either one designs for the partial load case or the worst load case. It is a no brainer, one must always design for the worst load case. Thus assume the hull is totally out of the water.

As for the rotating of the two hulls. Again what load case are you referring too? There is the classical transverse bending moment, then the pitch connecting moment and the torsional connecting moment. These are different from when hull is out of the water and the other demihull is the "built-in" end as a simple cantilever.

 

In ABS rule (attached note),
1. Bending Moment they assume is simple; one hull out of water and one hull in (in bending moment equation is Displacement*Bcl).
2. For Torsional Moment, they first calculate the moment from Displacement*L . And, then calculate the deflection in each beam. And they calculate the bending moment which would cause this deflection. Then, we have to check if our section satisfy stress requirements.

There, I didn't see which end support they assumed in calculating the deflection. (I twisted my eyes in element stiffness they use. I think I should later look into basic mechanics to understand this.)

Therefore, I assume what RX is pointing out, shear force and bending moment are covered by the stress checks at the end of the document. So, if I chose sections good enough to pass those stress checks [section modulus check for transverse bending moment, also], I think the only reason why I should give it a radius is the structural continuity. Did I get it wrong? Thanks,

And, again, Thanks, both RX and AH, for your time and patience. , Plz keep on correcting my wrong ideas.

 

Sorry for the late reply, I was out busy with something else. Attach is a diagram of what I use for beam analysis. This is based on numerous articles available and uses the outside influencing forces (wind and wave) to upset balance. If there are any AH, I would be more than happy to incorporate it or change the method if faulty. I always stand to be corrected.

In sheet 1, partial vertical displacement, I use integration. This is difficult since I have to know the hydrostatic properties of the hull.
In worst case condition, one hull out of the water, I use the full vertical displacement.

In all of these, I verify it using the LR rules on MMDW (dynamic bending moment), Mb (transverse bending moment), Mt (tortional bending moment), and the combinations of all three to arrive at Primary Load considerations of Head Seas, Beam Seas, and Quartering seas, all of which has been fed to an Excel spreadsheet. Coupled to this is a spreadsheet which analyzes, the composite stresses on the web and flange of on a non constant cross section member. This includes the end connection.

In sheet 2 for sailing yachts for one hull out of the water, I break down the diagram into two parts and use the standard method of calculations. I use Skeene's method (and two others) to arrive at forces for the mast load and heeling moment at 18-20 degree max.

 

I'm sorry to dig this thread up again. But, I'm reading this thread again now, but, I don't think I get an answer yet, how to analyze this radius. Now I realize this radius is connected with structural continuity. So, how can I know my radius is enough for continuity? In which WAY can I play and see? by FEM? by Trial constructions?

 

The closest thing I found in the various regulations, to give an answer to your question, is the attached figure.
I do not know, really, if that can help in your problem.

 

Why apologise? This is the point of the forum, neh!

So, as yourself this Q. How do you analyse a simple bracket joint between two structural members at right angles to each other? What rules are there to say which is right or wrong, in terms of radius etc?

Once you realise that one..then you see that the only way, to satisfy yourself, the radius is not an issue is to use a generous radius (a factor of safety if you like), or if you wish to have a more quantifiable answer, use FEA as a metric for measure.

 

ktming

Have you tried working out the rules in LR? Because I work in composites, I made a graph of how it works in accordance with the rules, attached.

Basically, the primary member and the beam is sized to handle the load and the brackets are added to ensure stabilty.

Page 1 shows the reaction, page 2 shows the rule, page 3 shows the sizing of the brackets.

The rule says when adding a bracket, the edge must touch the 1.5x dw. Using a straight bracket at 45 degree (in blue), I graphed it. This happens to be at 1/3 of the base width or height of a 45 degree right triangle. The 1/3 l/h is the centroid of area.

For a soft toe, curved bracket (CE like to use spandrel), I graphed the soft toe using the width and height of the straight bracket (red line). This appears too small.

Next, I graphed a much larger radius for the soft toe (green line) so that the le touches a point in the radius. The rule also says that the size must be increased when using soft toe.

To find the section stiffness, I used proportionate dimensions, no units. Given an E, I calculated the EI.

With a preliminary radius of 5, it appears I have not satisfied the rule of 2X the section stiffness of the primary stiffeners. It also tells me that the centroid of area of the spandrel is much lower than the unit x=1, y=-2 intersection. I could use a thicker web and cap thickness but since I am simplifying, I just increased the radius a bit.

The calculations shows that I have satisfied the rule.

There is a similar rule for aluminum, maybe slightly different but I suggest you work out the rule and post it here.