Two hulls, which has least drag?

I modelled hulls with this B/t ratio in Michlet, ok so its only a theoretical computer modelling, but it still held true down to this range also...

 

I think visually and have a hard time following the mathematical progression of developing the hull geometry given so many variables. I reprogrammed my spreadsheet to demonstrate what is being discussed here. I disabled the L/B, B/T, Cp constraint but kept the error trap in Cm. If the Cm value goes up more than “1”, error message will prompt because the midship section ratio cannot be more that unity. The Cb is a derivative of Cp, displacement. Again, it cannot be more than 1. Anyways, the recommended limits are in the upper cell of the parameters.

There are many resistance prediction methods but my collection is limited. I used the Newman method to find the resistance of a displacement/semi displacement hull. I also have two choices for the WSA calculation. MARIN is good for L/Disp analysis while Denny is good for B/T ratio analysis.

I added the notes from Principles of Naval Architecture. Unfortunately, the SS cannot demonstrate form factor and wavemaking analysis. I need to use Havelock resistance calculation for form and Holtrop formula for wave resistance demonstration. Both method is hard to program. Has anybody have a much simpler resistance prediction formula that uses form and wavemaking that output total resistance that perhaps I can use?

The yellow cells in the SS is the inputs, the blue ones are the outputs so don’t change those cells. The check boxes allow the principal entry to increment the rate of progression inputted above the cell. Change one parameter at a time in order to get a better spread. Changing two or more parameters will dilute the population or create errors. I pasted the graphs posted by AH to demonstrate that the output follows the graph. The SS takes care of computing other values that need to change by using valid NA formulae.

Sheet number 2 demonstrate the comparison of two hulls. It shows which hull exhibit the lesser resistance plotted on a range of Fn. It demonstrates that ship with higher L/Disp ratio has less resistance.

Sheet number 3 is for B/T comparison. It shows that B/T or underwater form has little effect on drag but should not be discounted. At low Fn, there is a difference but at higher Fn, it is almost negligible.

Have fun.

 

look at sheet nr 3:

IMO you can not set free the value of disp,length, beam, draft and Cp al together

disp = length*beam*draft*Cp (mayby * spezific whight of salt water)

or

draft= disp / (length*beam*Cp)


Cm could be = 1
-----> =WENN(P6/G6>1;"RANGE";P6/G6)
you used ............. >=


Wsa denny is much more plausible than mumford


I am not shure if your calculated forces are plausible
if I hold the values of disp beam and Cp, but double the length and set draft in halve

Disp . Length Beam Draft Cp
65,21 32,00 2,29 1,14 0,78
65,21 64,00 2,29 0,57 0,78

then at 22kn the RT,N is only about 10% for the longer hull compared to the first one......???

 

yes you can increment the value all at once but I have cautioned the user it will create errors. Even the initial input can be wrong if it conflicts with other parameters. Follow the guide.

You will have to look deeply to find where I used the salt water density. It is there.

Cm cannot be 1 otherwise you get a box shaped midship. Should "<1" , not ">=" to be accurate. 0.98 is about the practical limit

Denny/Mumford are two persons. I do not know which one is more plausible but enough to let them into textbook material. The WSA formula is known as Denny-Mumford. There were two revisions.

I did not calculate the forces. The formula did. Is there a wrong formula cell entry? Newman is not the best method for high speed hulls but works great at lower end.

 

ok, my mistake, confused with the name marin and mumford

for the question what B/T is more optimal the setting to marin helps me not so much because if I try to compare two hulls using your sheet and use same disp,length,Cp but dubble draft and set beam to half, there is no difference in the result Wsa.....Rt,N ....

so I mean if I use the Wsa setting to "danny/mumford", the predicatet result seems more plausible to be near the reale world on the water as if I set the switch to "marin"

nothing against the persons behind this two possible settings in your exel sheet, and all my apologies for my crude use of the english language

 

I was in a hurry and was not able to post accurately.

You are right about the density thing. I switched off the water density option but kept the Sea Water density in the FnV and Reynolds number. The input is in displacement weight so 1 ton of weight will displace 1/998.9=1.0011 m3 of FW at a temperature of 16 degree C. Thee decimals off in my calc if I use FW. It will not matter on boats but it will in ships.

MARIN uses simple Disp and Length as input to calculate WSA while Denny-Mumford uses T, B, and Cb as inputs. The second Denny-Mumford formula uses the L,T,and disp volume input. Other formula,such as Taylor's and Baier-Bragg is much more complicated. So for L/D analysis MARIN uses the right input while for B/T analysis, Denny uses relevant inputs. There is a difference in the WSA using either one. There are about six WSA approximation formula with each applicable to certain hull type. Accuracy is better if the "correct" approximation formula is used,

Again, the resistance prediction Rt is also an approximation. There are about twice as much formula (12) for speed prediction that I have encountered, each one best suited for each hull type. Even commercially available software takes this into account. No one formula fits all. I find de Groot seems to be the best approximation as it has the humps and hollows in the curve which indicates it is tracking the wavemaking drag coefficient.

So it is best that this SS not be treated as an optimal resistance prediction method unless specific formula is used for the hull type. At best, it is an approximation.

I have used this SS for only for three days but I find it helps me learn which parameters to tweak to search for the desired results. I find quite revealling what AH has posted. Before, I used to concentrate only on WSA and the forms of the boat. I will continue expanding the choices on the resistance prediction for better accuracy.

I am postin again the different WSA method for you to see how it differs.

 

Here is the power prediction link as posted by Alik. The thread is active again.
http://www.boatdesign.net/forums/props/power-predition-formula-45639.html

 

but mik the stick is on the "Voyaging under Power by Beebe" way.... a not to slender displ. monohul with fn(L)<0.4

this boat needs for stability quite different B/T than a hull for multihulls

 

Not that. It is about the graph posted by Alik. Look at the number of methods of resistance and powering and how the outputs differs from each other.

 

you mean (.....-power-predition-formula-p550_comparsion_of_methods_v2.jpg) ?

there are as x-coordinates speed[kn], did you understand what Fn(L) may be behind this speed (what length 9m?.....10m? what displ....etc)

my 2 ct. feeling is that he compare quite different flow-typs and that there will be a good chance of skalar mistaks.... and alone with this jpg we have no sweet chance to sort it out in the moment,

3D flow is not easy, eaven for the friction part of the drag you need the correct aerea for every speed and the flow speed (what is this exact?) for every part of the surface, the use of the calmness-WSA and boat-speed is not more than an first approach, the margin for this resulting error rate is enough to explain the difference of a lot of the different methods in the picture above

but with good will we can start to understand the idee (the physik) behind a calculating like de groote or NPL and why there are so much lines just climb slower and linear and no longer more ore less with fn² as before for higher speeds(fn>0.55 ?)

 

i think you missed the point haribo...the point being, there are many different mathematical approaches that attempt to predict a boats resistance... each can provide a very different result for a given problem, the key is understanding which method works more reliably for a specific application.

Yes, flow dynamics is a very complex subject, but these simple approaches are deliberately kept simple, so that reasonable predictions can be made quickly, easily and for little cost. A full CFD simulation can only be afforded by very expensive projects, even then the results are often quite close to what these simple methods produce so even then its difficult to justify. As computers get faster and cheaper we will no doubt see more CFD used in boat design, and probably less traditional tank testing for reduced costs and immediate results.

 

To answer your headline question "Two hulls, which has least drag? " - - THE ONE OUT OF THE WATER

 

RIGHT! And the one with the lesser resistance is the one nearly out of the water!

Haribo...... Alik's resistance prediction was based on only one hull with the same dimensions, ratios, coeficients, ect. and was plotted on progressing Fn.

I believe he is using a very expensive Wolfson software.

As pointed out by Groper, different mathematical formulas will give different results with some being more accurate than the other. The more complex the formula, the more input you need. Parameters such as LCB, entry angle, Form factors (1+k) and analyzing from two dimension to three dimension will give it more accuracy.

The SS, simple as it is, quickly demonstrate the basics. It can be made complex but why reinvent the wheel? Use Michlet with the basic constraint derived from the SS.

 

ok, but where is the discription of his reference hull? and exist a real world measurment of his power vs speed (or Fn)

 

Why don't you PM him about the details of the graph? He is a nice guy really and would be glad to give out the details of the hull used.