Cat Scantlings

Ad Hoc,

I understand the reasoning, but not allowing the hulls to twist over-estimates the total deflection in the bridgework. Over estimates mean the structure will be designed heavier.

I built (not designed) a catamaran row boat - very short, and the twist in the hulls was very evident. The twist in the beams was also evident.

Real world there is twist and bending in the beams and at least twist in the hulls. You can't really establish the raft structure (cross beams) without considering the hulls, if you want light weight. If you don't care about weight its OK.

 

Perhaps you didn't appreciate this part of the reply:

This related to the main hulls scantlings and hence its stiffness when compared to the raft structure. A row boat will have an "open" deck as such. That is to say it is not a closed cell, like in the FEM picture I posted.

A closed cell (hull bottom sides and a deck to close it) compared to an open cell structure (hull bottom and sides only) both being equal, save for the closed deck, has a difference in stiffness of 250,000%. So it comes as no surprise that you found an appreciable twist.

 

This row boat was a catamaran.

Each hull was a closed structure - typical small catamaran style. Nothing open and flexible about it. The cross beams were a square hollow beam. Not as stiff torsionally as a piece of equivilent sized tube, but reasonably stiff in torsion and bending. The seat is attached to the crossbeam in the center of the boat, the sliding rigger goes between the two crossbeams. I don't sit in either hull, they are 6" wide, each.

I do appreciate your SWAG about the 250,000% difference, wouldn't disagree with the situation you describe.

 

This I dont follow?

A RHS is always stiffer that an equivalent size/scantling tube.

Here, an example:

A RHS of 100x100x5mm versus a tube of 100mm diamater with same 5mm thickness.

RHS
I = 833.3cm^4
Z = 166.7 cm^3

The Tube
I ( J ) = 168.8cm^4
Z = 33.8 cm^3

(For ref: http://www.engineeringtoolbox.com/area-moment-inertia-d_1328.html )

Perhaps you under estimated the loads and hence the stiffness requirements?

 

The S-type loading is not the primary deflection mode, it seems rather to be the diagonal folding (see sketch below). Of course, depending on hull shape and "landing procedure", there is a sideforce to the hulls as well, but from my experience, the primary mode (diagonal folding) is the determining mode.

If I may be a bit old-fashioned and grumpy here, but the nice coloured pictures from computerized structural analysis can be extremely misleading if you are not understanding what limitations were introduced in the calculation. Here, for instance, Ad Hoc limited the freedom to twist/rotate in the hulls. Take that limitation away, and you have another picture.

So: first ALWAYS go back to basic!

 

I see what you're getting at. However, I beg to differ. Since if the line of symmetry is that diagonal line, only one hull is pitching and the other hull remains stationary. That implies a wave of very short length to only affect the one hull and not the other, this is not a realistic loading scenario. It only occurs in a “mathematical” sense, but can never actually occur. To have just one hull affected by a wave and not the other is not plausible, as this implies one hull is in flat calm water whereas the other, is not. But assuming this to be the case that they do, both hulls shall rotate at the same rate, which means the ends are being pulled apart, just pure tensile load, no bending nor shear.

This is with the raft butting into the inboard side of the hull. Because of the raft form the main deck too, then this is restrained by the inboard and outboard sides of the hull, and having an axis of symmetry in only one location, the inboard and outboard ends would not deflect the same amount. There would be a straight line between the inboard ends, the axis, but the outboard ends are out of plane, thus must conform to a different deflection. This would cause a curve in the ends, just like the image of the Classic S shape.

As an oblique wave passes, perpendicular to the diagonal line, the fwd end it lifted up. As the wave crest continues on its fwd path it intersects with the diagonal line of symmetry and since both the fwd end (of the other hull) and the aft end of the same hull are coincident with the wave crest (being perpendicular to the waves direction of travel) these must now lift. But in order to do that the fwd end of the hull that originally had the wave crest now has a trough. Therefore the wave length is the perpendicular distance between the 2 bows, along that diagonal line. Such a short wave will also be small in amplitude, to the point of having almost no effect on the global structure.

This is why the torsional moment of axis (in this pitching scenario) is roughly amidships and transverse and allows each hull to rotate against each other, thus the classis S shape at the ends and is a more onerous loading condition.

 

I should have indicated the hull position in the wavetrain, I guess. Place the hull diagonally with, say the fwd part of the port hull on one crest, and the aft part of the stbd hull carried by the next wave crest. Then you have the loading that corresponds to the "diagonal folding", with the center of mass "hanging" over the through.

The onboard acceleration levels were also worst in this sea/heading condition, so both the structure and the crew were tortured equally.....

 

I write some simplistic comments about (sailing) multihull crossbeams on the FAQs page of my website, copied in part below. Power multihulls suffer lower loadings

"There are two stages in creating a successful catamaran crossbeam solution.

First you must DESIGN the structure, only then can you CALCULATE it. The former is usually the more important and certainly the one most people get wrong.

There are several factors to consider when designing crossbeams:
First, you need stiff crossbeams, not just strong ones. Fortunately stiff beams are nearly always over-strong. By stiff I mean one without any obvious deflection. Engineers normally consider that to be 1-2% of length.

Why a stiff beam? Well imagine crossbeams made out of rubber. They would never break, but would be so flexible you could never keep the two hulls in line and the mast would fall down as the rigging flexed.

How can you make a stiff beam? Well, actually it’s not just the beams that you want stiff, rather it’s the boat as a whole. I’ve found that the best way to do this on an open catamaran is to have two crossbeams plus a separate one to take the mast loads. The actual positioning of the beams is also very important.

Although crossbeam size and placement is often complicated by rig and accommodation considerations, the beams must take priority! If they are too near the middle of the boat then the bows can flex up and down and you cannot keep the rig tight. If too close to the ends (especially to the bows) there isn’t enough boat to take the loads and, furthermore the beam cantilever is longer. "

Other comments

I have always thought that the biggest loading will be when sailing diagonally into waves. There are many photos available showing beach cats twisting under the rig loads, even with trapezing crew well aft. I have sailed beach cats with the windward shroud slack, so the mast is being held up by the trapezing crew

I have seen a number of full bridgedeck cabin cats that have cracked bulkheads because they rely on the hull/deck alone to prevent twist

Most full bridgedeck designs use very similar hull laminates, I have found them available on line. With no keel or rig loads multihull hull skins can be very light

The biggest loads will be the underdeck bridgedeck slamming loads and again those can/should be minimised by designing a low-slamming structure

Richard Woods of Woods Designs

www.sailingcatamarans.com

 

Apologies to the OP if this thread isn't going where he wanted it to. But I'm grateful to the forum engineers and architects for helping me to understand catamaran stresses.

I can intuitively "see" the bending and torsional stresses in an open-deck, two-beam catamaran configuration, such as in the sketches that Ad Hoc and baeckmo have posted. It is less clear to me how those stresses are managed in a planar-beam pod cat such as Gougeon's Strings. Could someone draw up a cartoon sketch of such a pod cat, showing where the stresses are in the beams, hulls, and pod?

baeckmo, could you further explain what you meant in your pdf by "longitudinal shearing in deck plane"

 

I understand what you’re saying, but I still disagree.

Here, take 2 monohulls side by side.



An oblique wave passes them, hull #1 at its stern has a crest and hull #2 has the crest at its bow. Then looking in profile, with the crest at the bow, and a trough at the stern. The buoyancy up at the bow and the weight down at the stern. The vessel shall rotate to maintain equilibrium, thus resulting in a pitch up attitude.

Waterlines along the hull are shown in sections A to C. No torsion longitudinally.

Now the 2 hulls are joined together, lets say by a very thin rubber membrane.



Thus the same shall occur as if 2 monohulls, the result being a rotation about the transverse plan as no other forces present. A classic pitch connecting moment. This then relates to the differences in stiffness between the hull as a girder and the raft structure to how much deflection is experienced.

Since if you now look at a classic monohull torsion in an oblique seaway, the crest of the wave is at the bow and the stern, thus two opposing moments to maintain equilibrium and no rigid body rotation.



You cannot have this with a catamaran in your scenario, the hull twisting, because the crest is not at the bow and stern, just bow, or stern. Thus it is a pitch connecting moment about transverse plane approximately at amidships, not a diagonal from bow of one hull to stern of the other.

 

Good thread.

See my over-simplified cartoon attached.
Note I've left out the stresses in the hulls, which will have bending loads, and may have significant torsion depending on the beam-hull joint and beam geometry.

edit:Forgive my non-standard beam load symbols. I changed my mind halfway through about which way round to draw the loads, and couldn't be bothered changing the arrows - hence the legend.

 

Thanks Autodafe. For some of us a sketch is worth a thousand words . It seems that the two large sliding doors in your pod will make it more challenging to carry the torsional loads between the forward and aft beams.

 

You're not wrong.

 

The Pod doesn't need to have anything to do with torsional loads. A regular day sailing cat has all the structure you need. Put any Pod between the beams, just make it light.

If you want the structure to be stiff - keeping the hulls from moving independently - the issue is not the beam strength, its how the beam is attached to the hull.

 

Wharrams have certainly proved that flexible can be durable, but in my opinion stiff is simpler to design and build.

I agree that a pod cat could be designed to have all the torsion in the hulls and none in the pod, but my feeling is that the pod might as well be used to stiffen up the boat.
A pod designed to withstand wave strike will have scantlings comparable to the hulls, and a greater overall size, and will hence be stiffer than the hulls and take a good proportion of the torsion load provided all structures are rigidly connected to the beams.

So far we have only been talking about a single load case - the simplified cross beams I showed in my sketch would be totally useless for dealing with shearing between hulls, or with hull rolling moments that can be introduced by other load cases.