# Drag for Small Shallow Draft Displacement Boats

Discussion in 'Hydrodynamics and Aerodynamics' started by Rurudyne, Jan 2, 2015.

1. Joined: Mar 2014
Posts: 1,170
Likes: 40, Points: 48, Legacy Rep: 155
Location: North Texas

### RurudyneSenior Member

As with many folks around here I play around with hull designs and daydream, even plan.

I've been using the free version of DelftShip which provides resistance per Delft Series '98 and John Winters - KAPER.

I also have access to, from The Design Ratios essay, such power equations as provided by Gerr etc.

As some may have noticed in other threads my interest is in paddlewheel design and I've actually managed to cobble together a number of useful equations from such sources as Saunder's Hydrodynamics in Ship Design or Seaton's rules (the 1894 version actually seems best). These depend on an estimate of the thrust needed to overcome the resistance and with them I can manage, in a round about way, to relate that thrust to an IHP value and from that try to work over to the scantling rules for wheel construction (which are in Seaton all related to the diameter of the boss, itself related to IHP in steam engine design ... which I also have other references for that I can cope with).

By using these I would hope to avoid such things as having a badly sized wheel, poor reduction ratio selection, wheels built too lightly (which would suffer vibration and fatigue issues) or built to heavily.

But of course the available resistance or power prediction tools I have are somewhat all over the place. Generally it seems: KAPER < Delft < Gerr (when I try to relate Gerr back to a value of R for a given V) is the rule of thumb for how these fall.

The difference between them is pretty large too.

For one sternwheel hull (LOA = 38.25', LWL=32.023', BWL = 8', Draft = 9", Displacement = 5266#, Cp = 0.63, S = 208.63 ft^2, midship = 4.531 ft^2, etc) at 7.9 kt, R due to friction remaining or assumed to be constant, KAPER gives an Rr of 59.22#, Delft Series '98 271.03# and Gerr seems to indicate about 460# respectively.

These boats are either flat bottomed or have small amounts deadrise at midsection, they are shallow draft, have a LBR of 1:3 (26' LOA including wheel) or 1:4 (larger). The one picture in my gallery is for the present 26' LOA example.

What I want to ask about is, predictably enough, improving resistance predictions.

I'll now leave the floor to you nice people.

2. Joined: Oct 2009
Posts: 5,229
Likes: 634, Points: 113, Legacy Rep: 1485
Location: Midcoast Maine

### DCockeySenior Member

KAPER, Delft series based formulas, and Gerr's formulas are based on regression, which is essentially curve fitting using math. Two questions to consider in making estimates with a regression formulas:
1) How close is your design to the designs used in the regression?
2) What is amount and distribution of deviation between the test results used to create the regression and the regression formula results for the tested designs?

KAPER is based on a series of sea kayak tests. Sea kayaks typically have a length to beam ratio between 6:1 to 10:1 with rounded bottoms. I would not expect it to be useful for wider vessels with flat or shallow V bottoms. That is before taking into account differences in displacement to size ratios.

The Delft series designs are round bottom sailboat designs typical of late 1980's or early 1990's racer/cruisers. These are are presumably considerably different in shape than your shallow draft designs with flat or shallow V bottoms. While somewhat closer than the KAPER designs I wouldn't expect good agreement with reality for your designs.

Gerr's formulas are generally quite simple, and the basis for formulas in terms of the designs they are based on may not be readily available. It is difficult to guess how accurate they might be for a particular design.

What speed range in terms of Froude Number or Speed/Length Ratio are you interested in? For Fn below 0.2 (SLR below 0.7) wave making drag will generally be small with viscous drag predominate. Viscous drag can be estimated with wetted surface area, waterline length and speed. For higher speeds you need a regression method based on designs similar to your design.

3. Joined: Oct 2008
Posts: 7,789
Likes: 1,688, Points: 113, Legacy Rep: 2488
Location: Japan

DC has pretty much nailed it.

All you can do is find existing published data on a hull form that is very close to one that you propose/design. In the absence of that...suck in some air through your teeth and think of a value. It is about as accurate as anything else in the absence of such quantitative data.....unless you do your own tank testing of course

4. Joined: Jun 2006
Posts: 2,209
Likes: 175, Points: 63, Legacy Rep: 1244
Location: Back full time in the UK

### Richard WoodsWoods Designs

It's always a good idea to have a feel for the likely resistance before you start. Fortunately you tried three different prediction methods and so realised something was wrong.

Suppose you had taken the easy way out and only tried one method? "The computer says" is always a risky approach

Apart from that, as Ad Hoc says "DC has nailed it"

RW

5. Joined: Mar 2014
Posts: 1,170
Likes: 40, Points: 48, Legacy Rep: 155
Location: North Texas

### RurudyneSenior Member

Now I know why KAPER seemed to produce such screwy (for my purposes) numbers.

As for the desired S/LR for anything I can afford to build I'll need to press the limits of displacement speeds to get anywhere only sorta slow.

I do have the option of bringing the hull into Hull Form and seeing what Savitsky or Savitsky and Brown might produce. Though for planing craft they are meant for hulls closer to what I'm considering and I could ask to return a values just in the neighborhood displacement hull speed.

There may at some point be another option based on what I've read in Hydrodynamics in Ship Design, V2, Saunders 1955, concerning things like the Telfer Merit Factor ... yet the graphs referenced are in the unavailable to me 1st volume.

One other check I've attempted is to use the rules of thumb given for paddlewheel design to "assume a wheel" (insert joke about economists and can openers here) and work backwards from A = T / 2 * p * Vo * (Vo - V) given in HiSD and A = IHP * 33000 * K / N * (DxR)^3 from the older references.

For example, in the wheel were assumed to be 80% maximum beam and the dip factor 20%, I seem to recall those being like proportions advanced somewhere, then the area of on float ends up as 577.5 in^2. 7.9 kt and 20% slip assumed.

From the equation from Saunders, T ends up being 341 lbs ... 25% higher than Delft Series.

Since the older equation is for sidewheels, some fudging (cutting the area to half that) gives an IHP of 12.5.

Of course, all that is pulling a rabbit outta the hat and we all know how well that worked for Bullwinkle, right?

6. Joined: Oct 2008
Posts: 7,789
Likes: 1,688, Points: 113, Legacy Rep: 2488
Location: Japan

You also need to consider what is the objective of your "exercise".

Is it:

1) Purely for fun/interest
or
2) To produce an actual design.

If #1, then more research will help..just a purely 'academic' investigation and no more.
if #2, then you first need to 'design' the boat, from an SOR, and then see what it will weigh and thus select your hull form to suit the design and SOR.

The two routes above are both different. Thus what is your principal objective in this investigation?

7. Joined: Oct 2009
Posts: 5,229
Likes: 634, Points: 113, Legacy Rep: 1485
Location: Midcoast Maine

### DCockeySenior Member

Savitsky, Savitsky and Brown, and related mehtods assume the boat is fully planing. The basic physics are fundamentally different than at "displacement" speeds. Trying to use them is equivalent to trying to use a map of Dallas to navigate around Washington DC.

8. Joined: Mar 2014
Posts: 1,170
Likes: 40, Points: 48, Legacy Rep: 155
Location: North Texas

### RurudyneSenior Member

Build would be my goal.

The SOR is for a trailerable shallow water paddlewheel day cruiser / picnic boat. No illusions about being fast (the 38'3" version mentioned above could just reach 9 mph, though a somewhat slower 25'11" version is more likely doable). Riverine and lake use. I could, for example, take it 60 miles south of Dallas and cruise up and down the Muddy Buddy, er, the Trinity where I wouldn't even have to hold my breath or carry a cargo of air fresheners. Other rivers could of course be open. I could even cruise around what's left of Lake Travis and gawk at Texas' biggest bathtub ring.

I don't really want steam power ... it's just my scantling rules for the wheel construction are related to the boss that would have been part of a steam engine back when they were written so I gotta figure that out even if a small diesel or gas engine is how I'd like to go.

Oh, and I'm dead set on a horn that goes "baa-boo-ba-booooo" like in Ray Steven's Erik the Aweful.

The underwater lines of the hull are based off the Alligator design that appeared many years ago in The Rudder: the whole great artists steal thing. Scantlings would be determined per Gerr. The sides would be mostly (marine) plywood construction though the plumb bow with flare could be cold molded in one piece. The bottom, if lacking deadrise, would be plywood, or if with deadrise then herringbone planked.

Light weight superstructure. Very possibly not unlike that shown for the paddle wheeler design presented in this book: https://archive.org/details/ninemotorboatsho00motorich

About the superstructure: in past posts I've commented about maybe borrowing SOF construction and surrounding a frame with a padded skin (the padding to provide insulation). Also, given that it is to be trailerable, I'd not be adverse to a "pop up" camper cabin ... potentially reducing cost to tow quite a lot ... if it could still look nice when up.

As it's shallow draft with a lot of windage there would be a (off-center) centerboard used.

In whatever superstructure there would be I think you would likely encounter mosquito nets and zippered plastic window panels rather than glass.

The wheel supports form the aft part of a continuous hull line and rub rail to within a few feet of the bow. No version of this will be sleek and sporty ... but I want reasonably attractive.

9. Joined: Mar 2014
Posts: 1,170
Likes: 40, Points: 48, Legacy Rep: 155
Location: North Texas

### RurudyneSenior Member

Yeah, I'd figure it'd be an act of desperation.

10. Joined: Jun 2008
Posts: 161
Likes: 7, Points: 18, Legacy Rep: 55
Location: Vancouver BC

### cmckessonNaval Architect

Barge data

Due to the shallow draft, I suspect that your river boat is a lot more like a barge hull than it is like, say, a kayak. To that end I then suggest you look at collected barge data, such the collection published by Stuart Cohen in 1983:

I only did a brief search, so I haven't found any nice regression formulae or the like, but I suspect that these are "out there" in the literature as well.

Best of luck - I too have a hankering for paddlewheeler, and have filled many pages with idle sketches of what she might look like!

Chris McKesson

11. Joined: Mar 2014
Posts: 1,170
Likes: 40, Points: 48, Legacy Rep: 155
Location: North Texas

### RurudyneSenior Member

Thanks.

Makes a lot of sense.

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.