Wave interference in catamarans

I was reading a debate(argument) about wave interference in catamarans on this site.

Someone made the point that the finer the hulls are the less of an issue wave interference between the hulls is. By this I think he meant less wave means less wave interference and thus less potential resistance. This makes perfect sense to me but Ad Hoc wasn't having it. Can someone make sense of this debate?

 

Dustman, I think that the main criteria is probably the spacing of the hulls, rather than how fine the hulls are, relatively.

This research paper by Mustafa Insel and Tony Molland is one of the early classic references on this topic - there are many more if you ask Google to find some for you.
https://www.researchgate.net/public...ponents_of_high_speed_displacement_catamarans

If the maths is a bit too complicated, just read the conclusions at the end.

I would not dispute what Ad Hoc says about this subject - he has had a lot of experience in this field.

 

You might want to take a look at what Ernest O. Tuck and Leo Lazauskas say about it in Optimum hull spacing of a family of multihulls. Search around these forums and you might find Michlet, the program they used to generate those results. These are all displacement hulls, not planing hulls, however.

 

Correct.

 

Not having any scientific or modeling experience with this, but I grew up watching my father's 65 foot catamaran fishing boats. I use to look down between the hulls when we were cruising at around 18 to 20 knots and watch the bow waves come together between the hulls. They threw up a much bigger rooster tail aft from the stem waves coming together.

My take on it is the interference between the hulls is a function of spacing, hull length and speed. In any boat, the bow wave radiates off the hull with various amounts of disturbance due to hull design. Some boats throw bigger waves than others, but they all throw out a wave. As the hull moves faster, more water is forced out at a faster rate, but the forward movement also rakes the radiating wave backwards more. The arrow shape of the wave gets pointier. This means the wave interference from the two hulls will move aft as the boat goes faster, but, until it gets up on a plane, it will also get bigger. The length of the hull will have something to do with how much interference is trapped between the hulls, just as distance apart will.

A flat, planing hull with a sharp bow won't tend to throw as big a bow wave as a blunt bow on a hull with a lot of rocker. Catamaran hulls are pretty skinny, typically, but width will make a difference in wave size.

Just the observations of the son of a fisherman.

My father with a couple of test models designed and built by Clark Mills for the Double Eagle fishing boats. Sometime around 1967.

 

Something I've occasionally wondered: what's the effect of tunnel hull type asymmetry, like the interior surfaces are parallel flats?

Does it significantly reduce the bow wave on the inside, or is that a myth? Intuitively it seems like it should, but it also seems like if you could radically reduce the interfering bow wives simply by designing planar inner surfaces, that would be extremely common.

But I only see it on skater-type extreme performance hulls, not hulls trying to avoid low speed interference, so empirically I would have to assume it doesn't work that way.

 

Re the effect of tunnel hull type asymmetry. in terms of pure overall efficiency this only becomes an advantage at higher Froude numbers - at slower speeds a symmetric hull form will be 'better' generally.
The 15 metre aluminium cat in my avatar photo has assymetric hulls simply because they were much much easier to build than symmetrical hulls - we knew that symmetric hulls would have been a bit more efficient, but we couldn't justify the extra time and expense at the time.
I cannot remember off hand at what Froude Number the transition is, but Ad Hoc should know?

 

Well, like most things in design - it depends!

Since you're attempting to put all the different types of hulls into one basket and get a "better-than" type of result. It is not really that simple.
Since if you take a normal cat hull and split into 2 sections - like you did to create an asymmetrical hull - the beam is now different..and thus the displacement is now different.
So if you adjust for the displacement, it is not the same like for like as the beam is still different. If you adjust for the beam, the draft will be different.

Thus it is very difficult to get exactly the same L/B ratio or B/T ratio of L/Disp ratio... to act as a true like for like.

However, "in general", you can expect a reduction in the total resistance in the Fn 0.50 - 1.00 range, as a rule of thumb.

 

In retrospect this maybe should have been more obvious: a tunnel hull with planar inner surfaces must push water almost exactly the way a monohull would.

You might not get the wave interference in a tunnel hull, but if I took my monohull boat, cut it in half lengthwise, stuck plywood in the cutouts and set the new half hulls six feet apart, maybe all I've done is make it heavier, with more initial stability, but presumably a higher center of gravity.

The sponson shapes haven't been optimized for catamaran performance, so I guess of course it behaves more conventionally, with a planing hump etc.

 

From Molland and al. experiments on catamaran (as quoted here by Bajansailor)+ Oosannen report on small high-speed displacement vessels, I did a synthesis with figures in adimensional form to help estimate the residuary drag of a high speed catamaran including the wave interference.
** Adimensional meaning that the residuary drag Dr, in % of boat weight Mg, is given in function of the Froude number and of adimensional parameters : S/Lwl (S = spacing between hull axis) , Lwl/Bwl and Bwl/Tc (Tc = hull body draft).
** The validity domain of this approach is Froude 0,3 to 1,0 for slender hulls of Lwl/Bwl > 7 .
** The extra drag due to the wave interference is given in the form of an amplification factor (1+K) of the residuary drag of the two hulls if alone. The Fig. 2 gives an average of the K values where, as predictable, the greater the S/L, the smaller the K value, with a hump in the Froude 0,4 - 0,5 zone. But actually these averages cover a wide variety of cases which are showed in Fig. 3 to 12 and I suggest to use one of them (the one closer to your case) instead of Fig. 2 for a better estimation.
Here attached the synthesis in pdf and in ods formats, and the 2 sources mentioned.

 

That is a good selection of sources Dolfiman, thanks.