SCALING in Boat Design and why the twin towers fell down so easily

[tom kane]

Scaling is a principle familiar to structural engineers and physicists.In proportion to their size, small things are stronger than big things.How do boats and ship design cope with this problem.

[Guillermo]

From MARIN:

"Model testing of ships is usually performed at equivalent Froude numbers to obtain equal wave patterns at model scale and at full scale. However, once the Froude number is fixed, one can not also obtain equivalent Reynolds numbers. Typically, in ship design projects the Reynolds number at model scale is a factor one hundred smaller than at full scale. To account for the so-called scale effects caused by the difference in Reynolds number, extrapolation procedures are used. The primary scale effect on the flow is a decrease of the boundary-layer thickness on the hull and a corresponding decrease of the width of the wake behind the stern with increasing Reynolds number. This results in a considerable change of the velocity distribution in the propeller plane. A secondary, and therefore less pronounced, scale effect is caused by the interaction between the viscous flow and the wave pattern. Wave effects on the viscous flow around the hull may be significant for all cases with substantial wave making, as the wavy surface affects the development of the boundary layer all along the hull. On the other hand, viscous effects on the wave pattern are generally insignificant and only substantial in the stern region, as is confirmed by many validations of non-linear panel methods. Viscous effects are expected to result in a reduction of the height of the stern wave system and an upstream shift of the first wave crest behind the stern. The level of this reduction depends on the amount to which the flow itself is affected by viscosity. Since the width of the boundary layer and the wake decrease with increasing Reynolds number, so does the reduction of the stern wave system. This secondary scale effect on the wave system is not specifically accounted for in extrapolation procedures. However, with the development of solution methods for the free-surface viscous-flow problem, it has now become possible to calculate these effects."

Model test results may suffer from such scale effects related to the large difference in Reynolds number between model and ship. This difference is particularly significant for full-bodied ships, and for ships operating in shallow water. This suggests that realistic form and numerical values for the hydrodynamic efforts should be based on full-scale tests.

Cheers.

[tom kane]

I did try to add an attachment I will try again

[Guillermo]

"Despite the undeniable progress made in putting hydrodynamic hull design and performance predictions on a rational basis through the use of computational fluid dynamic (CFD) techniques, the design of hulls in commercial ships is still partly an art mainly supported by experimental evidence and with no inmediate perspective of being reduced to a fixed procedure. Present methods for design of ship hulls typically involve the following steps: 1) Definition of speed range and operational conditions, 2) Definition of preliminary hull shape using empirical methods or simple potential flow based hydrodynamic codes, 3) Manufacturing of scaled prototype, 4) Experimental testing of hydrodynamic, manouvering and seakeeping performance, 5) Empirical scale-up process, 6) Modification of prototype and repetition of steps 3-6 until the desired tolerance in ship's performance is satisfied. This is a costly and expensive process which generally does not suffice to achieve optimum cruising performance (due to the difficulties in the scaling process) giving the highest level of hydrodynamic efficiency characterized by prescribed speed and stability requirements."
http://www.cimne.upc.es/cimnel_new/a...ntent.asp?id=1

Something on scale effects on propellers:
http://books.nap.edu/openbook.php?re...10834&page=744

More info can be found through the ITTC:
http://ittc.sname.org/documents.htm

Cheers.

[BWD]

Limited, but not really OT...
Re: the towers
As I recall they did not come down until after application of around 389M kcal heat, each.
This may be an obnoxious example to introduce a topic, but it's apt as it illustrates an important problem in modern design:

The more complex, as well as the bigger, a structure is, the more things can happen to it. Doubtless the WTC engineers never said, "what if we burn a fuel tanker on the 80th floor," any more than they considered the kinetic energy of a jetliner, as has been argued before by many.

Thus other aspects of complexity need special consideration when designing a large structure, as well as those derived from size.
It is something to think about when adding another system to a boat, of any size.
Making something large or complex that will last a long time pushes the usual assumptions about probability as hard or harder than other aspects of design. This challenges computability and leaves us with the art of design, said another way in Guillermo's post above refering to the design cycle of ships.

[Leo Lazauskas]

Originally Posted by Guillermo

From MARIN:
Model test results may suffer from such scale effects related to the large difference in Reynolds number between model and ship. This difference is particularly significant for full-bodied ships, and for ships operating in shallow water. This suggests that realistic form and numerical values for the hydrodynamic efforts should be based on full-scale tests.

And if one candidate (full-size) hull is not quite right then crumple it up like a piece of paper and throw it in the bin. Then build a new one and hope it performs better.

The quote from Marin was excellent, by the way.

Regards,
Leo.

[Guillermo]

Originally Posted by Leo Lazauskas

And if one candidate (full-size) hull is not quite right then crumple it up like a piece of paper and throw it in the bin. Then build a new one and hope it performs better.

You know it's not like that. Tank tests and the best of CFD analysys still have to be compared to the real thing, to check their accuracy relating many aspects still not well known, i.e. the loads at ships' bows when encountering very high waves. Then you learn, refine the methods and go one step upwards in the spiral. Isn't it this way?
Cheers.

[SamSam]

Originally Posted by Guillermo

" the design of hulls in commercial ships is still partly an art mainly supported by experimental evidence and with no inmediate perspective of being reduced to a fixed procedure.

That's what I was going to say, "an educated guess". Sam

[Raggi_Thor]

Originally Posted by tom kane

Scaling is a principle familiar to structural engineers and physicists.In proportion to their size, small things are stronger than big things.How do boats and ship design cope with this problem.

In small boats you use the local stiffness as criteria for dimensions, if the deck is stiff enough to be walked on, it's also strong enough of you look at the whole boat as a beam supported in both ends and the keel and mast pushing down.

In larger boats and ships, global stiffness, the hull and deck as a beam becomes the critical issue.

Interesting side effect: Wood is the best material for small boats, steel for larger boats :-)

[Leo Lazauskas]

Originally Posted by Guillermo

You know it's not like that. Tank tests and the best of CFD analysys still have to be compared to the real thing, to check their accuracy relating many aspects still not well known, i.e. the loads at ships' bows when encountering very high waves. Then you learn, refine the methods and go one step upwards in the spiral. Isn't it this way?
Cheers.

It certainly should be that way, Guillermo. I'm not very
confident about many CFD predictions and extrapolations,
nor do I trust many towing tank tests as being indicative
of what happens at full-scale. So yes, full-scale verification
is the ultimate test of the predictions and they should be
used to refine CFD techniques and towing tank methods. But
it's a very slow slide up (or down) the spiral!

You mentioned wave loads as one poorly understood phenomenon.
There are, of course, many others. In a real sea the effective
turbulent eddy viscosity can be quite different to that in a
towing tank. What this does to skin-friction and wave-making
is still something I'm trying to understand.

In the attached picture (similar to that produced by Fred Stern
in some of his papers), five regions of the flow at midships are
shown. Region I is the "potential" zone. Region II is the boundary
layer (BL) on the free-surface, Region III is the BL on the hull,
Region IV is the zone of interaction of the free-surface BL and
the hull BL, and Region V is the meniscus clinging to the hull.

Only Region I is considered to be reasonably well-understood
because viscosity can often be ignored. The BL on the hull is
understood for laminar flow, but this is not of much help for
anything bigger than a duck. How the extent and effects of
the five zones scale in a real sea with waves, currents and
surfactants is going to take a very long time. Add hull flex
and vibration, propulsors, and, in shallow water, complex
topography and it's almost enough to make one find a less
complicated pursuit. Almost.

All the best,
Leo.

[tom kane]

Many ships,airships,bridges,buildings,infact many big things have failed because of a design change often.. by accident.There is something missing in the design of big things.It is not murphys law, or What ever can happen will happen?

[PI Design]

I'm not sure what point you are trying to make?
Big things don't always break. Little things sometimes do.

[kach22i]

Originally Posted by tom kane

Many ships,airships,bridges,buildings,infact many big things have failed because of a design change often.. by accident.There is something missing in the design of big things.It is not murphys law, or What ever can happen will happen?

Good point, the cable suspended bridge that people were dancing on in a Kansas City hotel failed (ten years ago) because the contractor decided he didn't want to build it the way it was correctly drawn. The small but critical design detail change which was made by the structural engineer (a lesser in his office actually) to make it easier to build for the contractor and approved by the city could not take the rhythmatic motion set up by a 100 + people dance in step on it at the same time (who would of ever guessed?). The structural engineer lost his license but did not go to jail. Something like 30 people died in that one.

[water addict]

Originally Posted by tom kane

Scaling is a principle familiar to structural engineers and physicists.In proportion to their size, small things are stronger than big things.How do boats and ship design cope with this problem.

Quantify the expected loads as best you can. Quantify the material and structural behavior as best you can. Use this knowledge to design something that will not break.

[RatliffFranklin]

Originally Posted by BWD

Limited, but not really OT...
Re: the towers
As I recall they did not come down until after application of around 389M kcal heat, each.
This may be an obnoxious example to introduce a topic, but it's apt as it illustrates an important problem in modern design:

The more complex, as well as the bigger, a structure is, the more things can happen to it. Doubtless the WTC engineers never said, "what if we burn a fuel tanker on the 80th floor," any more than they considered the kinetic energy of a jetliner, as has been argued before by many.

Thus other aspects of complexity need special consideration when designing a large structure, as well as those derived from size.
It is something to think about when adding another system to a boat, of any size.
Making something large or complex that will last a long time pushes the usual assumptions about probability as hard or harder than other aspects of design. This challenges computability and leaves us with the art of design, said another way in Guillermo's post above refering to the design cycle of ships.

No, the WTC engineers did consider the KINETIC effects of the largest airliner flying in the late sixties, the Boeing 727, but not the thermal effects. When you compare a 727 against a 757 the gross weights are not that different, so even if it had been a 727 instead of a 757 that hit the towers, the final results may have been the same.