Rowboat - Improvement to Rowing a Boat, Canoe, Scull, or Crew

A relatively simple modification could increase the overall efficiency and speed of a racing shell or scull or crew. Simply mounting the oar fulcrum on the opposite side of the craft (rigging) from where the oars enter the water, allows the oar blades to keep better angles with the water for better mechanical efficiency and thrust. With given athletes, the craft is likely to go several percent faster as a result. This represents several boat-lengths in a race! The mathematical proof of this is included.

Public Service
Self-Sufficiency - Many Suggestions

Environmental Subjects

Scientific Subjects

Advanced Physics

Social Subjects

Religious Subjects

Public Services Home Page

Main Menu

In a normal rowboat or crew shell, the oarsmen must have their backs toward the front of the boat in order to get maximum power from each pulling stroke. If a rower faced forward, he would have to "push" on the power stroke, which happens to be physiologically a less powerful action. A SIMPLE improvement to a boat can improve the mechanical efficiency of rowing several percent and solve this situation at the same time, making it so a rower faces forward, with additional performance benefits.

Text Font Face
Text Size
(for printing)
A standard rowboat has an oarlock attached to the side (gunwale) of the boat, between the person and the oar blade. The oarlock has the effect of being a fulcrum of a lever. In Engineering, this arrangement is often called a "First Order" lever. The fact that the fulcrum is between the force (the rower's power stroke) and the load (the oar's blade pressing against the water) reverses the motion. The handle of the oar must be pulled FORWARD in the boat to make the blade move AFT to propel the boat forward.

On a rowboat, the oarlock is mounted right on the top of the edge of the boat. This causes the oars to rotate through a significant angle during the course of a stroke. Also, length of time the oar is in the water is limited, unless very long oars are used, and in that case, the lever arrangement requires enormous pulling effort. In the middle of the stroke, the oar is pushing the water directly aft, the idea situation where the greatest mechanical efficiency is possible. At all other times, however, there is an outward (early in the stroke) or inward (late in the stroke) force applied to the water. This effect pushes water sideways instead of astern, which is both wasteful of energy and also the cause of extra turbulence drag in the water. So, for two different reasons, if the oars are at an angle in the water, the boat will go slower.

On a racing scull or crew shell, a metal framework (called rigging) is attached to the sides of the craft to allow a location for the oarlock that is often two feet outboard of the boat. This allows the use of much longer oars. This allows longer periods that the oars are in the water, without further increasing the disadvantageous angles or requiring excessive strength by the paddlers. The primary advantage of using the rigging is that the oars have consistently more efficient angles in the water.

This is a wonderful enhancement, and the following improvement extends the concept even more.

For this discussion, consider the distance from the oarlock to the rower's hands to be 3.3 feet and the oar is 10 feet long. The oar blade therefore moves at two times (6.7 / 3.3) the speed of the rower's hands. The rower must then also provide two times as much pulling power for a given force applied to the water.


Consider a similar sort of rigging structure that is used on the shells, to move the oarlock out to the side of the boat. Consider making this structure even about a foot longer, so that the oarlock location would be 5 feet out from the centerline of the boat, and upward around one foot above the gunwale. Now, imagine a 15-foot long oar, but with its very (conventionally, the handle) end attached at the oarlock. The oar crosses over just above the boat, with the oar tip in the water 10 feet on the opposite side of the boat. The overall width of the oar space would therefore be the same as before, roughly 20 feet in overall width. The specific geometry described would have the oar move at three times the speed of the rower's hands, and require three times the power by the rower for a given force applied to the water. For this particular geometry, it would make sense that the in-water area of the oar be reduced to around 2/3, such that the same power load is on the rower.

This particular geometry provides an oar-tip speed around 1.5 times the conventional oar-tip speed. This is considered reasonable because the boat should be able to travel at significantly higher speeds.

A (taped?) handle area would be provided on the oar five feet from the oarlock, which is therefore centered over the middle of the boat.

The fact that the oarlock supports are a foot higher allows the oars to slope slightly downward to allow the oars to enter the water over the craft body. It also ensures that the structure does not interfere with the motions of the oars on the opposite side of the craft.

Why do this?

First of all, the rower, being on the same side of the fulcrum as the load, which is called a "Third Order" lever, will pull the oar rearward to make the boat go forward. That means that he will face forward. This will make it easier for rowers and crew to accurately stay on the desired line, without drifting laterally which represents wasted effort and requires additional correction.

A more important benefit also exists. Because the distance between the oarlock and the blade is so very long, the blade moves through a much smaller angle of arc during the stroke. It therefore remains at much more efficient angles (to the water) throughout the pulling stroke. This represents a SIGNIFICANT improvement in efficiency!

When a conventional crew's oars first enter the water, say 4.7 feet forward of being abeam, for discussion sake we will say that they are angled at around 45 degrees forward of being directly abeam. When the rower pulls, the water is actually being pushed aft AND OUTWARD. At that instant when the oar first starts to pull (at 45 degrees forward) the geometry is not very good. The rower is actually pulling at a disadvantageous angle, since the handle starts out at 45°, so his effort is at that angle rather than being straight along the line of the boat. The FORCE applied to the water acting directly aft is therefore somewhat reduced, but the worst part is at the other end of the oar. The force he is creating against the water is outward as well as rearward. The actual rearward force he is applying to the water is dependent on the cosine of that angle (0.707). The POWER transferred to the water goes as product of Force * Velocity. Since both the rearward Force component and the rearward velocity component are reduced by that same factor, the Power is therefore multiplied by the square of the cosine (0.500). Therefore, only 50.0% of the rower's (horsepower) effort (at that instant) goes into moving the craft forward, with the other 50.0% of his effort going into attempting to slide the craft sideways. (The oar on the other side of the boat counteracts the sideways effect, but the total effect is a slight wiggling of the motion of the craft, which results in a minor amount of turbulence, and a lot of wasted energy.)

During the power stroke, the geometry continually changes, with 100% of his pulling force going into the desired thrust only at that instant when the oar is directly abeam. If the rower was able to apply constant pulling force throughout the whole stroke, the total power benefit would be the mathematical integral of the continuously varying geometrical disadvantage integrated over the entire arc of the length. In this case, the overall net efficiency (due to this geometrical factor) is about 81.82%. We are considering a total in-water stroke length of around 9.4 feet in this example.

Because the proposed improved boat has its oarlock on the opposite side of the craft, oars that are 15 feet long would be used, and this geometrical angle situation is greatly improved. With the same length of rowing stroke (9.4 feet total), the oar's blade will now never be more than 17.4 degrees away from being abeam. This improves that 50.0% initial efficiency all the way up to 91.06% (as the WORST case situation!) As before, for an entire stroke, it is necessary to mathematically integrate the situation over the entire stroke. In this case, the overall net efficiency (due to this geometrical factor) increases to about 96.45%.

That's about a 14.6% increase in net power transmitted to the water with NO increase in effort done! In crew and scull races, a fraction of one percent better application of power to the water often makes the difference between winning and losing. Imagine how much difference an added 14.6% of additional applied power would do!

The same technology would work for a conventional rowboat. The rower would face forward, and the blade-water angle would give comparable added efficiency. In the case of a rowboat, it is unclear whether 14% of added power transmitted would be worth the trouble. Switching a single oar from one side of the boat to the other would be difficult, so two oars would certainly be appropriate. An adjustable linked-handle mechanism could pull both together, while still allowing steering.

There are two minor negatives regarding usage on shells, and two more possible advantages.

The disadvantages are the greater wind resistance (due to the larger structure of the oarlock supports and oars) and the potentially greater weight and wind resistance of longer oars. The wind resistance consideration can be minimized with aerodynamic designs of those strut members and the oars. The fact that such craft travel at relatively low velocities makes such effects much smaller than the advantage of the more efficient thrust angles' benefit. The greater length of the oars (regarding weight considerations) might even turn out to be an advantage!

If the oar is made of two parts, with a rotating joint connecting them, two new advantages could result. The oar could be mounted in the oarlock with only a single degree of freedom, essentially a flat plane of movement just above the craft's gunwales. In that case, since the main body of the oar could not rotate, that third of it could be made much lighter than any conventional oar, but with a flat airfoil shape. The width would allow great strength and rigidity, forward-to-aft, while a light composite construction might even be able to be completely hollow for lightness. The aerodynamic shape would reduce wind resistance, so the resultant wind load would be slightly less than for a conventional round oar.

The remaining two-thirds of this oar's length would be rotatably mounted to this described portion. Therefore, the blade end of the oar would be rotatable during each stroke, with a fixed stop at exactly vertical for the pulling stroke and another fixed stop at 90 degrees rotation, so the blade could have the lowest possible aerodynamic drag in the air during the return stroke.

The 14.6% overall power efficiency benefit from the better geometry is far in excess of the very minor increase in aerodynamic drag due to the slightly larger oarlock support structure. The fact that a portion of the oars could be aerodynamic airfoils, and the fact that the oar blades could consistently be rotated for maximum applied power and minimum aerodynamic drag should more than counteract that effect.


The net advantage of this improvement is so great that it seems worth investigating. A small shell with a one-man scull setup or a two-man crew should be able to be modified pretty easily in this way. Since the shell already has the attachment points for the oarlock support, only the modified oarlock support structures would need to be made, and the different oars.

I would be available to assist or consult on the necessary changes.

A 14.6% improvement in net applied power should result in an increase in speed of around 7.3%. Even a moderately competitive Crew or Scull team may be able to establish new world records in races! In a 1500 meter race, that represents around 100 meters, many boat lengths gained, with the same effort on the part of the athletes!

Whether or not these modifications would be permitted by the sanctioning bodies is an entirely different matter. However, the hull is not changed AT ALL, and the rigging is simply slightly different. The oars do not have to be very different, except longer.

In the same way that the Fosbury Flop revolutionized the high-jump (because of incrementally better geometrical advantage), this invention might do the same for Crew and Scull racing.

I first invented this improvement in May 1996. This presentation was first placed on the Internet in August 1999.

This page - - - - is at
This subject presentation was last updated on - -

Link to the Public Services Home Page

Link to the Science Projects Index - Public Service

E-mail to:

C Johnson, Theoretical Physicist, Physics Degree from Univ of Chicago