designing a fast rowboat

Discussion in 'Boat Design' started by nordvindcrew, Oct 13, 2006.

1. Joined: Aug 2006
Posts: 3,497
Likes: 147, Points: 0, Legacy Rep: 2291

ancient kayakeraka Terry Haines

Anyone who has ever rowed, cycled, ran or swam competitively knows that Newton's entire concept of equal and opposite forces in a linear universe was utterly wrong.

2. Joined: Sep 2006
Posts: 541
Likes: 13, Points: 18, Legacy Rep: 231
Location: Marshfield massachusetts usa

nordvindcrewSenior Member

linear universe

the graph on hull speed vs effort is all you need to know. It's the inverse of the graph of how fast you go into the crapper relative to the severity of the mistake that precipitated the action

3. Joined: Jun 2003
Posts: 512
Likes: 40, Points: 28, Legacy Rep: 506
Location: Kotka, Finland

terhohalmeBEng Boat Technology

4. Joined: Aug 2006
Posts: 3,497
Likes: 147, Points: 0, Legacy Rep: 2291

ancient kayakeraka Terry Haines

Ugh! No wonder. Given that skin drag predominates in shells (it's typically 50% for my canoe designs which are less extreme) so that power it proportional to the cube of the speed, then a 1% increase in speed -which you can scarcely measure - is going to need 3% more power - which you can definitely feel after a short time! I suspect it has an even higher power for non-shell boats - it gives the phrase "hitting the wall" a whole new meaning.

5. Joined: Jun 2003
Posts: 512
Likes: 40, Points: 28, Legacy Rep: 506
Location: Kotka, Finland

terhohalmeBEng Boat Technology

Rowboat resistance

Here is some curves of rowboat resistance and effective power. This boat is a Finnish wooden raceboat, LH = 6.5 m. The speed is in Froude number which is defined as: Fn = V / (g * LWL)^0,5. Froude number is unitless speed-length ratio. Curves are calculated from Michlet output file.

You'll see that the viscous (friction) resistance is predominant and the problem going faster occurs, when wave resistance start to grow at Fn=0.35. In power plot, needed effective power start to grow very rapidly at the same speed. Effective power is what the boat needs to go, you must swet almost double that much.

A fit rower can row about Fn = 0.4 on long distance (over 30 km) and Fn = 0.45 on short distance (2 km).

Terho

Attached Files:

File size:
16.2 KB
Views:
627
• Rowboat power.png
File size:
12.2 KB
Views:
628
6. Joined: Aug 2006
Posts: 3,497
Likes: 147, Points: 0, Legacy Rep: 2291

ancient kayakeraka Terry Haines

I suspect the power-speed curve is much closer to linear than the majority of hulls

7. Joined: Jun 2010
Posts: 983
Likes: 32, Points: 28, Legacy Rep: 300
Location: Australia

NoEyeDeerSenior Member

Closer to parabolic would be a better way of putting it. Since skin friction increases roughly as the square of the speed even shells don't have a linear resistance curve.

8. Joined: Jun 2010
Posts: 983
Likes: 32, Points: 28, Legacy Rep: 300
Location: Australia

NoEyeDeerSenior Member

By the way, have you ever actually tow tested the full size boats to see how well the drag in real life matches the predicted drag from Michlet?

9. Joined: Jun 2003
Posts: 512
Likes: 40, Points: 28, Legacy Rep: 506
Location: Kotka, Finland

terhohalmeBEng Boat Technology

No, too expensive. How much the constant (average) speed tells about the real resistance? The test should include all the speed variation during the strokes.

10. Joined: Jun 2010
Posts: 983
Likes: 32, Points: 28, Legacy Rep: 300
Location: Australia

NoEyeDeerSenior Member

Sure, in theory it should. That could be dealt with by towing at different speeds to see what the resistance was at each speed. I'm not sure why this should be expensive either. The Bethwaites in Australia did a lot of this sort of testing when developing the NS14's that preceded the Tasar and 49er. All that you need is calm conditions, a launch that can hold a steady speed, and a spring balance or similar to measure the tension on the towing line to the boat being tested.

Of course the boat under test should really be towed from an arm off to one side to keep it clear of the wake of the towing boat, but that's not hard to arrange.

11. Joined: Jun 2010
Posts: 983
Likes: 32, Points: 28, Legacy Rep: 300
Location: Australia

NoEyeDeerSenior Member

Given that the amount of effort to travel at the peak speed is a lot more than the amount required to travel at the minimum speed (with power required varying nearly as the cube of the speed), the boat should not be optimised for the mean speed around the course. It should instead be optimised for a speed somewhere between the mean speed and the peak speed. This would give the lowest total expenditure of energy around the course.

If the variation in speed under race conditions can be measured accurately (which shouldn't be too hard) and if the change in resistance with speed is known fairly accurately, then it wouldn't be hard to calculate the speed the boat should be optimised to.

Admittedly this is getting into the territory of chopping large rabbits in half (ie: splitting hares) but if it gets you an extra boat length or two over the course of a race it'd be worth it.

12. Joined: Aug 2006
Posts: 3,497
Likes: 147, Points: 0, Legacy Rep: 2291

ancient kayakeraka Terry Haines

I did a quick bit of math for this one: my math is a bit rusty but hopefully I got it right.

As I expected it turns out that the power required to maintain a given average speed is lowest when the speed is constant; i.e., no speed variation. This is true of any hull and is a direct consequence of the non-linear drag/speed relationship. I didn’t work out both cases but intuitively it would apply both to a displacement with an upward curving graph and a planing hull with graph that curves downward at one point.

To get some actual numbers I assumed a sinusoidal speed variation and a square law for the drag/speed relationship to simplify the math. For that case the power to maintain an average speed doubles when the speed variation is +/- 26% of the average speed. That is purely a “what-if” example not real life, just for illustrative purposes.

Conclusions seem to be, design for the most linear drag/speed ratio possible (no surprise there), and keep the stroke rate high - other human factors being equal. Strangely, it also suggests increasing boat or crew mass, but the benefit from that would have to be weighed against the increased drag. There is also the cyclic exchange of momentum between crew and boat to be considered if sliding seats are used, which perhaps could be used to reduce the peak speed where the power requirement is increased disproportionately. Overall, an interesting design challenge: as always, the task is to optimize the combination of crew and boat.

13. Joined: Jan 2002
Posts: 2,696
Likes: 151, Points: 63, Legacy Rep: 2229

Leo LazauskasSenior Member

I've attached a very cut-down report that might be of interest.

Leo.

Last edited: Aug 12, 2015
14. Joined: Aug 2006
Posts: 3,497
Likes: 147, Points: 0, Legacy Rep: 2291

ancient kayakeraka Terry Haines

Wow Leo, that's really new! So new it's dated tomorrow - must have tripped over the dateline

It certainly provides an insight to how detailed competitive rowing (and other boating) has become. The acceleration and velocity graphs are what I was trying to predict although I used the simplest possible model. With that much detail it should be possible to account for virtually all of the rowing energy. Is the document still being completed or is the rest available?

15. Joined: Jan 2002
Posts: 2,696
Likes: 151, Points: 63, Legacy Rep: 2229