# Two hulls, which has least drag?

Discussion in 'Hydrodynamics and Aerodynamics' started by John Perry, Oct 26, 2012.

1. Joined: Nov 2003
Posts: 272
Likes: 25, Points: 28, Legacy Rep: 129
Location: UK

### John PerrySenior Member

A question for the naval architects / fluid experts here:

Below are pictures of two hulls being towed by a symetrical beam arrangment which will indicate, in quite a sensitive manner, which hull has more drag than the other. The attachment of lines through cross beams on the hull is only to avoid the possiblity of the hulls snaking under tow and to avoid the tension in the lines having an effect on the pitching of the hulls. I have shown the above water parts of the hulls in yellow, the underwater parts in grey. The first picture shows the assembly from above, the second is from beneath the water so as to give an idea of the underwater hull shapes.

Both hulls are relativly long and narrow, the narrower of the two would need some ballast to float vertical. Both hulls would suit a multhull yacht rather than a monohull. I would envisage them being a medium sailing yacht size, say 10m length.

The hulls are the same length, the same displacement, the same wetted area when at rest, the same pitching moment of inertia, the same surface finish. The hulls are towed far enough apart that their wave patterns do not interact.

Both hulls have half elliptical underwater cross sections all along the length, however the beamier hull has the width of the half elliptical sections twice the depth, the narrow hull has the width of the half elliptical sections equal to the depth.

Which hull would have the most drag:
a) Calm water at low speed with negligible wave making
b) Calm water at higher speed, making wave patterns.
c) Rougher water, making wave patterns.

My guess is that for a) there would be little difference, perhaps not enough to be measurable.

For b) I would expect the narrow hull to have less drag because the centre of bouyancy is further below the water surface. You could say that it is part way towards a SWATH hull, the idea of a SWATH hull being to place most of the bouyancy volume well below the surface so as to reduce wave making. Having said that, I note that the author of this article:
http://www.boatdesign.net/forums/at...-40-ft-multihull-sailboat-offshore-racing.pdf
suggests that a wider shallower hull will provide dynamic lift even at large length to beam ratios. Is that right? I would be a bit sceptical since I have not seen long narrow displacement style hulls (not water skis!) lifting at speed, if anything my impression is that they tend to get slightly sucked down.

For c) I really dont know the answer. The wider hull has twice the hydrostatic pitching stiffness and also considerably greater pitch damping. It will have a higher natural frequency of pitch motion, and heave motion too for that matter. I think conventional wisdom would suggest that a higher natural frequency of pitch motion will reduce drag in rough water, isnt that why race yacht designers try to keep weight out of the ends of a yacht? On the other hand, although the wider hull may pitch less (note the may!) for the same pitch angle it will immerse larger transverse cross sections of the normally above water hull (the yellow part) and I would have thought that is going to cause some extra drag. I realise tht the answer may depend on the wave frequency spectrum, possibly I would need to do the experiment with a variety of waves and on different courses relative to the principle wave direction. If the drag of the two hulls in rough water turns out to be about the same then I would prefer the one that pitches the least because it would be more comfortable at sea and shake the rig around less.

I had thought of getting a quote from a local machine shop to carve out these hulls from blocks of wood at model scale on a CNC milling machine, so that I could do the experiment albeit not at full size. If I did that, how large should I make the models, would something like 1.2m long be large enough to represent a 10m long full size hull?

If someone can tell me the answer to the above questions, or if someone can tell me that the experiment has already been done, then that would save me making the models! Indeed, if I had access to some up to date CFD software perhaps I wouldnt need to ask the question in the first place.

#### Attached Files:

File size:
98 KB
Views:
2,390
• ###### Two hulls from below.png
File size:
65.6 KB
Views:
1,416
2. Joined: Jan 2002
Posts: 2,696
Likes: 146, Points: 63, Legacy Rep: 2229
Location: Adelaide, South Australia

### Leo LazauskasSenior Member

As a first attempt, try Michlet:

3. Joined: Oct 2008
Posts: 6,269
Likes: 441, Points: 83, Legacy Rep: 2488
Location: Japan

### Ad HocNaval Architect

John

Curious, what is your motivation for this?

No, you won’t be able to tell, other than visually if at all, which is better. Since you’ll need to know the tension in the line. Different speeds will also have an influence too.

Confused. Here you say:

Yet here you say:

Can’t be both, thus the 2 are not the same?

That’s not correct. SWATH’s don’t do that. SWATHs have high WSA and more drag.

4. Joined: Nov 2003
Posts: 272
Likes: 25, Points: 28, Legacy Rep: 129
Location: UK

### John PerrySenior Member

Leo:
Yes, I knew I should try that first! Mind you, I didnt think Michlet would deal with motion of a hull in waves and I dont think the story is complete without that. I have just downloaded the Michlet files and have been looking at the manual.

Motivation is to gain an insight into a fundamental aspect of naval architecture.

A visual indication based on the angle of the towing beam is fine for me, at least to begin with. I realise that carrying out an experiment like this could be more difficult than one first imagines, there may be various dynamic effects that mess things up and the real test rig could look quite different to my simpified sketches. Later on I might consider measuring the hull drags, I have to hand an electronic balance which weighs to 5kg accurate to +/-0.0005grams, perhaps I could use that somehow. Unfortunately it needs to be kept still, level and dry! I would really prefer to do this with CFD, at least to start with.

Re your second and third points, if you were to take the underwater part of the wider hull (the grey part) and slice it in two down the vertical centreplane then rejoin the two pieces with alternative surfaces in contact, you would get the form of the underwater part of the narrow hull, so I believe that what I said is correct.

Re SWATH, I am not trying to make any argument about how the drag of a SWATH boat compares with other boats or its wetted surface area, I am simply suggesting that concentrating displacement volume further away from the water surface might reduce wave making drag. Michlet should at least tell me if that is correct thinking.

5. Joined: Oct 2008
Posts: 6,269
Likes: 441, Points: 83, Legacy Rep: 2488
Location: Japan

### Ad HocNaval Architect

John

I assume you mean hydrodynamics rather than naval architecture, per se?

Hydrodynamics is essentially down to two factors

1) Wave making
2) Friction.

These are the principal components that are broken down when analysing resistance. They are broken down into further ‘subsections’ but that shall just confuse matters for now.

If we look at #2) this has an influence. Since a SWATH, has so much volume below the waterline, the WSA is high, to carry the “volume”. This means high frictional resistance. It does not as you are suggesting, since a body needs to be significantly far away from the free surface to have zero or very low wave making. Far more than many realise, many times the number of ‘diameters’ of the body.

So looking at #1, this is influence by the length-displacement ratio. (Other facts can come into play, but only marginally).

This graph below indicates the influenced that the length displacement ratio has on wave making or residuary resistance.

In other words…a short fat boat, of displacement X…compare to a long slim hull of the same displacement X, the long slender hull shall have lower residuary resistance.

The corollary is that as the residuary become very low, the frication resistance begin to dominate again. Since if there is low or little wave making, the frictional resistance is what is left!

1 person likes this.
6. Joined: May 2004
Posts: 5,372
Likes: 239, Points: 73, Legacy Rep: 3380
Location: Italy (Garda Lake) and Croatia (Istria)

### daiquiriEngineering and Design

Hi John,
Regarding the method you intend to use for the measurements, I don't think you will have enough time to perform any useful observations with that towing configuration. As soon as there's an asymmetry of forces between the two arms of the balance, it will rotate fairly quickly to the position in which the vessels will be in line, one behind the other. In that moment you will have the interference due to wake, waves and struts used for towing.
The only information you will get from that towing rig is which hull is more draggy at that given speed. But you won't have any quantitative data about how much draggy. So you won't have enough time to observe and evaluate any of characteristics indicated in your list (a through c)
Cheers

7. ### FrostyPrevious Member

If that were to happen then the answer would have been found.

8. Joined: May 2004
Posts: 5,372
Likes: 239, Points: 73, Legacy Rep: 3380
Location: Italy (Garda Lake) and Croatia (Istria)

### daiquiriEngineering and Design

Yeah, that's what will happen, and he will know which hull has more drag than the other. But I understood (wrongly?) that he wanted to observe these hull's behaviour in waves. IMHO, he will not be able to.
Cheers

9. ### FrostyPrevious Member

Oh I see,-- well yes then in that case the end of the experiment will be very quick.

10. Joined: Jan 2002
Posts: 2,696
Likes: 146, Points: 63, Legacy Rep: 2229
Location: Adelaide, South Australia

### Leo LazauskasSenior Member

It is certainly not enough, but it is a start.
After you have the calm-water resistance, I would run it through something like
the US Navy's Ship Motion Program (SMP) to get the behaviour in waves
including an estimate of the added resistance.

11. Joined: Oct 2009
Posts: 4,488
Likes: 236, Points: 63, Legacy Rep: 1485
Location: Midcoast Maine

### DCockeySenior Member

The rotation of the beam could be limited to keep the hulls close to abreast while still seeing which one has greater drag.

12. Joined: Nov 2003
Posts: 19,133
Likes: 473, Points: 93, Legacy Rep: 3967
Location: Eustis, FL

### PARYacht Designer/Builder

Nat Herreshoff used a similar rig to test resistance, though his data was purely empirical. A couple of load cells would offer a lot more information about the two hull forms, though judging by the image, the answers, should be fairly obvious.

13. Joined: Jun 2006
Posts: 2,199
Likes: 150, Points: 63, Legacy Rep: 1244
Location: UK, USA and Canada

### Richard WoodsWoods Designs

Pat Patterson did those sorts of tests maybe 20 years ago. In practice it proved quite a bit harder to do than expected. The hulls were about 1.5m long (from memory). So they had to be 2m apart at least and that much again off the side of the towboat. So the towing arm got very long and the loads were higher than expected (the first arm broke)

And you need calm still water so dawn was the best time, which meant setting up in the dark

And then the results you get just confirm what you know already.

From my limited time in (good) test tanks I know that you cannot get meaningful results unless you use a very broad brush approach, and everything happens faster than you expect, so you have to do multiply runs and then you wonder about repeatability

You also have to decide on the most realistic place to attach the towrope which for sure isn't at the bow, as you already know

I doubt if you'll get anything meaningful from wave tests. Interestingly, radio controlled model yachts (especially trimarans) behave very differently from their full size counterparts, great fun to test them though!

Richard Woods of Woods Designs

www.sailingcatamarans.com

14. Joined: Nov 2003
Posts: 272
Likes: 25, Points: 28, Legacy Rep: 129
Location: UK

### John PerrySenior Member

Daiquiri: it does look as though the three pivot points in my diagram are in line, but in fact the centre one is a little ahead of the two end ones. That way, in principle, I can get an idea of the drag of one hull relative to the drag of the other by measuring the angle of the beam from a photograph or video playback and using a bit of trigonometry. I drew the three pivots almost in line, thinking that the arrangement may need to be sensitive to very small drag differences, but they could easily be made more out of line if necessary. Since the angle of the beam gives the relative drag of the hulls, if I wanted absolute drag measurements for the individual hulls I only need a single load cell. But really, if I could get relative drag measurements for a range of conditions that would be quite enough to start with.

Ad Hoc: Thanks for the interesting graph. The graph covers a range of slenderness ratios but does not differentiate between hulls that have the same slenderness ratio but different transverse cross section shape, hence this is not relevant to the question I am seeking to answer. Without the text that probably accompagnies the graph I cant say for sure, but I guess this graph is for a set of hulls having the same transverse section shape, e.g. a semicircular shape.

You say "a body needs to be significantly far away from the free surface to have zero or very low wave making. Far more than many realise, many times the number of ‘diameters’ of the body." That is useful information, thank you. Based on that, it may well be the case that the calm water drag of these two hulls will be very similar even at speeds where there is significant wave making. Or, to put it another way, if there is a significant difference under these conditions, then I would need to think hard about why that is!

Leo, thanks for the sugestion, as yet I have only just started to look at Michlet. A couple of years ago I met one of the technical guys from an engineering consultancy that amongst other things does CFD work for aerospace companies etc. During a lunch hour he played back for me an animation he had produced simulating a new lifeboat operating in rough seas. Obviously I have no way to tell how accurate the simulation was, but it did look pretty realistic. The waves on the sea looked much like the ones I can see on the English Channel a few hundred yards from my house, the boat was rolling and pitching, throwing up bow and stern waves, spray coming over the foredeck and all. This kind of thing needs to be verified, but I guess that is being done, it does seem to be the future.

PAR: are the answers really so obvious? If I had to guess, without any analysis or experiment, I would say that under most conditions the deep narrow hull would have a bit less drag than the wide shallow one, but it would be more prone to rythmic pitching in certain wave conditions, so might have more drag in that particular situation. But that is just my 'gut feeling', I have heard that gut feelings can be wrong when it comes to fluid mechanics.

Richard: I am not surprised this kind of thing has been done before, it is an obvious thing to try. I dont know where to look for the results though, if indeed they were ever written up and made public. Agree the tests are likely to be a lot harder to carry out than one might at first imagine, which is why these days I lean towards modern computer simulations.

My overal aim was to answer the question 'If hydrostatic roll stability of an individual hull is irrelevant, as it is for a multihull, should we stick with semi-circular underwater transverse cross sections, which offer minimum wetted surface, or is there an advantage in slightly flattening the semicicle vertically or squeezing it in sideways' I realise that for sailing multihulls it is a bit more complicated because of the shift of displacement from hull to hull as the craft tacks.

15. Joined: Feb 2002
Posts: 2,252
Likes: 226, Points: 63, Legacy Rep: 1673
Location: Port Gamble, Washington, USA

### tspeerSenior Member

AFAIK, the added resistance in waves is correlated with the heave response. For example, http://shipmotions.nl/DUT/PapersReports/0428-DUT-76.pdf, http://www.shipmotions.nl/DUT/PapersReports/0440-DUT-76.pdf

So, in the absence of any calculations, I would expect the wider hull to have a greater added resistance in waves by virtue of its larger waterplane area. As you say, the slender hull is closer to a SWATH.

As for actually doing balance beam tow tests, Frank Bethwaite talks about these in some detail in "High Performance Sailing", in which they did this with dinghy hulls - including in waves.

Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.