center of flotation calculation and implications?

[Eric Sponberg]

Originally Posted by Fanie

From Fanie: Hi Erich,

If I have a slender displacement hull like that of a cat with a LWL of 10m then the hull speed is going to be around 13.4kn or 24,8km/hr...
Ok I see, if the CP is 0.62

If the CP is 2 for the same hull then the speed for the same hull would be 20kn or 37km/hr

Is this correct ?

Fanie, keep consistent units. For Lwl = 10M, length in feet is 32.81'. At hull speed, V = 1.34*32.81^0.5 = 7.68 knots. If your Cp is 0.62, the hull resistance will be about as optimum as possible. Cp cannot be more than 1.0. If you want to go at speed-length ratio of 2.0, your Cp should be 0.70 for the least resistance and, therefore, the least power required. The 10M long boat then would have a speed of 2.0*32.81^0.5 = 11.46 knots.

[/quote]From Capt. Vimes: if we have a hull working in displacement mode only and we want to reach hull-speed - i.e. speed/length ratio = 1.34 (2.43 in metric) - we need a Cp of 0.63 to even reach hull speed...
if the hull has a Cp lower than that and is not of a long and narrow persuasion - we would end up with a highest achieveable speed less than this - am i right?

to fully comprehend this and understand what happens there is really giving me troubles...

Cp defines the fullness of a hull - so the more the hull 'fills out' the block of water between the perpendiculars and beam at WL the faster it could go in displacement mode...

could it be that with a higher Cp the ratio between wetted surface and displacement is getting lower (less wetted surface for a given displacement) and thus reducing frictional resistance enabling a 'fuller' hull to reach higher displacement speeds?[/quote]

Capt. Vimes, Sorenfdk's description is correct. All boats will be able to exceed hull speed given enough power. But not everyone wants to go that fast, and you pick your Cp for your intended sailing or operating speed, which is usually less than hull speed, maybe speed-length ratio of 1.0 or 1.1. At 1.34, you can expect resistance to go up exponentially in most cases, and so really need a lot of power to exceed that speed.

You are correct in that with higher Cp the ratio of volume to wetted surface is lower, but you are not reducing frictional resistance. Frictional resistance is directly proportional to surface area, and even with larger Cp, surface area increases slightly. It happens that with higher Cp, the wave-making resistance does not increase as fast, and so you can achieve a higher speed before you hit "the wall".

Some years ago, a fellow naval architect that I know spent some time in a model tank testing the model of a 30' round-the-buoys racing sailboat (which his client paid for) in order to optimize the Cp to a range between 0.53 and 0.54. That was the speed that they were shooting for (between 1.0 and 1.1 speed-length ratio) and so wanted to minimize volume and wetted surface for the best all-around performance. They felt that the effort was worth it, although I did wonder at the time. The boat certainly performed well and it was nicely built. However, the success of the boat depended largely, too, on the ability of the crew. Bad crew meant mediocre performance. But when they had a rock-star crew on board, the boat did really well.

Eric

[capt vimes]

thank's soren and eric...
it is getting brighter... yet another corner of my dark-room you all call "boat design" illuminated...

[Leo Lazauskas]

Originally Posted by Eric Sponberg

Some years ago, a fellow naval architect that I know spent some time in a model tank testing the model of a 30' round-the-buoys racing sailboat (which his client paid for) in order to optimize the Cp to a range between 0.53 and 0.54. That was the speed that they were shooting for (between 1.0 and 1.1 speed-length ratio) and so wanted to minimize volume and wetted surface for the best all-around performance.

Eric,
Are you saying that they only tried to minimise skin-friction and not the total hydrodynamic resistance (i.e. wave + friction + form)?

Or was wave resistance assumed to be implicitly accounted for by fixing the range of Cp?

Thanks for the many insights so far,
Leo.

[Eric Sponberg]

Originally Posted by Leo Lazauskas

Eric,
Are you saying that they only tried to minimise skin-friction and not the total hydrodynamic resistance (i.e. wave + friction + form)?

Or was wave resistance assumed to be implicitly accounted for by fixing the range of Cp?

Thanks for the many insights so far,
Leo.

They were trying to minimize both friction and form resistance particularly at different angles of heel and yaw. And, with software that they had available which came out of the America's Cup in the mid 1990s (they hired another consultant to do this), they were able to do model tests with just one model, and then do variations of form in the computer. And these were fairly slight variations of form with Cp, one of the controlling factors, limited to a small range. That can be fairly reliable on a computer analysis. For larger variations of form, such computer analysis becomes less reliable, and you should anchor it with multiple models so that you have good, fixed data points. Interpolating between data points is OK, extrapolating from a single data point beyond small variations is dangerous. Model testing is still more reliable than CFD (Computational Fluid Dynamics) but they are getting closer and closer all the time.

I found the effort interesting in that this was for a 30' sailboat. The client had some money to devote to that, which is very rare for a boat that size. Also, the budget did not allow any testing on different keel designs, which on a racing sailboat, has tremendous influence, particularly in this case where the keel was very long and narrow and with a bulb at the bottom. What they were able to do, however, was derive a very good map of where the CE and the CLP travel as the boat heels (they move back and forth, the rig more so than the keel, obviously) and so had a more precise analytical way of establishing the balance between the two. The boat in the end balanced very nicely.

You are right in that wave making resistance was not an important factor, really, because the design speed was well enough below hull speed that it became of secondary importance. Form and friction resistance were the particular elements to deal with, and most importantly at different heel angles, therefore different wind conditions.

Eric

[Leo Lazauskas]

My first hydrodynamics contract (with E.O Tuck) was on a 1990's America's Cup yacht. The boat was by far the fastest in a straight line, straight down into the mud of San Diego harbour

Originally Posted by Eric Sponberg

Interpolating between data points is OK, extrapolating from a single data point beyond small variations is dangerous. Model testing is still more reliable than CFD (Computational Fluid Dynamics) but they are getting closer and closer all the time.

I still can't help but think of voodoo mathematical models (Volker Betrtram's term) being matched to very uncertain experimental data.

Experimental resistance with small hulls (< 2.0m) is particularly uncertain. The type of boundary layer trip used to promote turbulent flow can result in variations of in residuary resistance of up to 18%. See for example,
"From model scale to full size. Investigation on turbulence stimulation in resistance tests of high speed craft", Bertorello et al, FAST 2003.

Thanks,
Leo.

[Eric Sponberg]

Originally Posted by Leo Lazauskas

I still can't help but think of voodoo mathematical models (Volker Betrtram's term) being matched to very uncertain experimental data.

Experimental resistance with small hulls (< 2.0m) is particularly uncertain. The type of boundary layer trip used to promote turbulent flow can result in variations of in residuary resistance of up to 18%. See for example,
"From model scale to full size. Investigation on turbulence stimulation in resistance tests of high speed craft", Bertorello et al, FAST 2003.

Thanks,
Leo.

Yes, you really have to know what you are doing when it comes to model testing or CFD. These are tools for analysis and cannot be expected to give results that are 100% true to reality. They help us most of all with our intuition.

Eric

[Leo Lazauskas]

Originally Posted by Eric Sponberg

Yes, you really have to know what you are doing when it comes to model testing or CFD. These are tools for analysis and cannot be expected to give results that are 100% true to reality. They help us most of all with our intuition.

Eric

The ITTC are presently conducting the "International Collaboration on Benchmark CFD Validation Data for Surface Vessels". This involves testing two geosims at about 30 towing tanks around the world.
There is quite a deal of scatter in the results submitted so far. If all the tank operators "know what they are doing" why are there such large differences between tanks?

Experienced CFD practitioners, who also should know what they are doing, often tweak their models to agree better with experiments. Picking the right set of experiments seems a bit problematic to me.

As you said, experiments and CFD should be used as aids to intuition, and not some absolute, independent, infallible measures of reality.

Leo.

[Eric Sponberg]

Originally Posted by Leo Lazauskas

If all the tank operators "know what they are doing" why are there such large differences between tanks?

For the same reason that if you go to a bake sale looking for brownies, you will find many different varieties on the same idea. All the bakers know what they are doing, but they all make them just a little bit differently. Frankly, my brownies are filled with a whole bag of chocolate chips because that's the way I like them.

Now everyone--please don't steal this thread with brownie recipes! This simply makes a good metaphor.

Remember, we are discussing basic naval architecture. Next week I am taking up Displacement/Length ratio which is the next step on the way to the relatively new concept, The S Number (it is actually over 20 years old, but much newer than the other design ratios).

Eric

[ancient kayaker]

Originally Posted by Leo Lazauskas

... testing two geosims at about 30 towing tanks around the world ... quite a deal of scatter in the results submitted so far ... Leo.

Different tank depths and widths and different trim conditions could be the cause of the scatter. Perhaps tank configuration has more effect on results than hitherto suspected.

[Landlubber]

....just shows actually how a good designer earns his money doesn't it.
Boat design is NOT just getting a design program and playing about for a few days, it is about UNDERSTANDING the principles and USING them in association with your own experiences to create what is known as a "good" design....just how many designers do you know that are capable of writing and explaining to us like Eric has been doing....he understands......that is the difference.

Once again, thanks for your comments.

[MikeJohns]

So much effort goes into tank testing in smooth water for small boats that will operate in a seaway. Like Erics example of refining the Cp and testing numerous variations. As soon as the vessel starts its dance on the waves a lot of that drag reduction refinement is worthless.
Winged keels and pitching come to mind.

Powered Ship hullforms are a lot easier to refine in the tank.

I still get a thrill though watching the generated waves flow away from a hull I designed the first time the model is run, I think I intuitively get more from observing that than I do from the resistance curves.

Tanks are very usefull for general comparisons but as Eric implies you need a lot of gelled knowledge and experience as to how you interpret the results.

I am trying out 1.5m models, Others have used 1.2m models for refining hullforms for a variety of craft with apparent success.

Optimized through computerised CFD is just a line to scare the opposition isn't it ?

[Leo Lazauskas]

Originally Posted by ancient kayaker

Different tank depths and widths and different trim conditions could be the cause of the scatter. Perhaps tank configuration has more effect on results than hitherto suspected.

Certainly, and that's why this study is so important. Unfortunately, someone has lost data from 5 tanks so it will take a while to finalise.

Leo.

[Leo Lazauskas]

Originally Posted by MikeJohns

So much effort goes into tank testing in smooth water for small boats that will operate in a seaway. Like Erics example of refining the Cp and testing numerous variations. As soon as the vessel starts its dance on the waves a lot of that drag reduction refinement is worthless.
Winged keels and pitching come to mind.

I'd go further and say that tank conditions (e.g. the eddy viscosity) are nothing like what real ships experience at sea. The eddy viscosity can vary by up to 6 orders of magnitude in the ocean, whereas tank tests are usually conducted in very still conditions. That can make a lot of difference in estimating wave resistance and wave decay.

There are also a host of incompletely understood issues to do with tank boundary layers on the bottom and the side-walls, and the finite records of wave elevations.

Originally Posted by MikeJohns

I am trying out 1.5m models, Others have used 1.2m models for refining hullforms for a variety of craft with apparent success.

And others haven't, spectacularly!

The FAST 2003 paper I cited earlier used a Telfer-like procedure with three geosims of length 1.805m, 2.347m and 4.494m.
The authors' conclusion was that L < 2.0m was inadequate for model-ship correlations for Fn < 0.6.

The other conclusion was that the type of turbulence stimulator had a very large effect on the results. So, how do experienced naval architects interpret the effect of that? Or do they ultimately just accept what is given to them by the tank operators because there is nothing better available?

The International collaboration is using L=3.048m and L=5.72m, and the shorter hull seems barely adequate, IMO.

Originally Posted by MikeJohns

Optimized through computerised CFD is just a line to scare the opposition isn't it ?

I agree. CFD = Colour For Dollars

All the best,
Leo.

[Leo Lazauskas]

Originally Posted by Landlubber

....just shows actually how a good designer earns his money doesn't it.
Boat design is NOT just getting a design program and playing about for a few days, it is about UNDERSTANDING the principles and USING them in association with your own experiences to create what is known as a "good" design....just how many designers do you know that are capable of writing and explaining to us like Eric has been doing....he understands......that is the difference.

Yeah, but Naval Architects are just a spork in the engineering cutlery drawer.

I'd agree that they are probably best-placed and best-informed to combine the myriad of factors entering into a good design and to make educated, experienced compromises. But they are not necessarily the best tool (and I mean that in the nicest possible way) for investigating specific design factors in great detail. At some stage they have to rely on the advice of specialists, or just accept the data that is supplied to them on trust.

Cheers,
Leo.

[MikeJohns]

Thats why we collect and cite papers, lifes too short to be anything but a Spork.