Entries in Power (1)


Net Power Production & Performance at Different Stroke Rates & Abilities During Sculling 

Constanze Loschner - New South Wales Institute of Sport, Sydney, Australia

Article link: Net Power Production & Performance at Different Stroke Rates & Abilities During Sculling   


During on-water rowing, power developed by the rower may be delivered to the oars through the hands and to the foot stretcher through the feet. The proficiency of the rower will be partly determined by the effectiveness with which this power is coupled to boat propulsion. Velocity cost, the average power required to maintain the boat velocity divided by that boat velocity, describes how expensive a rower's technique is in terms of the power delivered in moving the boat through the water. Propulsion is defined as any action that directly affects the forward progression of the boat. For example, the transverse component of the handle force is a necessary accompaniment to the longitudinal component of the total handle force and requires power but has no effect on propulsion. This paper examines the relationship between patterns of power production and absorption and the velocity of the boat.


Three female international level scullers rowed an instrumented single scull at steady state cadences of 20, 24, 28, and 30 strokes per minute. Scull velocity was measured with a magnetic turbine and pickup coil, pin force with multi-component force transducers, stretcher force with strain gauge transducers, and oar angle with servo potentiometers. This information was sampled at 100 Hz and telemetered to a laptop computer on the shore. Approximately twenty strokes for each rower were time normalised and averaged at each stroke rate. Power delivered to the boat by the rower was calculated as the product of boat velocity and the pin and stretcher forces. Power delivered to the oar handle was calculated as the product of the handle force and handle velocity. Handle velocity was the result of the angular velocity of the oar and linear velocity of the boat. The oar was modelled as a simple lever with the water acting as a fulcrum to calculate handle forces from the pin forces. Motion in the horizontal plane only was considered.

Results & Discussion

For all three rowers power was absorbed by the rower just after the catch but rapidly changes to power generation for the boat, reaching a peak at about 30% of the stroke (Figure 1). The power reaches a second peak during the recovery phase as the rower exerts a propulsive force on the foot stretcher. Power generation during the recovery phase was comparable to power generation during the drive phase.
The pattern of propulsive power for rower A is relatively smooth and is generally of lower magnitude in both generating and absorbing modes and especially in the recovery phase, compared with that for rowers B and C.
The pattern of power production for rower A resulted in less variable boat velocity (Loschner and Smith, 1999). The smaller the range of the boat velocity the less power required for a given average boat velocity (Dal Monte and Komor, 1989). The mass of Rower B was considerably larger than rower A or C causing a larger drag force on the boat and thus requiring more power output for a given velocity. However rower C was slightly lighter than rower A and a different cause(s) must be found for the poorer performance as measured by velocity cost.

Figure 1 - Total propulsive power delivered to the oar handles and stretcher at
28 strokes per minute for rowers A, B, and C.

Rower C had a relatively irregular propulsive power curve with the largest power absorbing components near the catch. The outcome of this irregularity was a more highly variable boat velocity which in turn required a higher power output for a given average boat velocity. This relativity among the three rowers as described here for a stroke rate of 28 was consistent over the other three stroke rates with differences in the amplitudes of the variables only.

The consistency over stroke rates can also be seen in the values of work efficiency (WE). The values of WE varied among rowers but were constant over the stroke rates. Rower A, although scoring low on velocity cost (VC), obtained the lowest score for WE. WE was the ratio of propulsive work done (the integral of propulsive power over time for one stroke) to the total work done. As the transverse oar handle work done was similar among the rowers this lower value for WE for rower A was due simply to the lower propulsive energy expenditure that was most evident in the recovery phase.

It could be expected that VC would increase with higher BV since the drag force increases exponentially with boat velocity. The power output required for each boat velocity did have a tendency to increase exponentially (Figure 2 below).

Rower A is a world champion junior women's sculler and some of the characteristics of her rowing revealed in this analysis may provide some insight into which techniques make for efficient and effective rowing.


The rowers studied in this project were highly consistent in their work effectiveness over four stroke rates spanning cadences of 20-30. The lower velocity cost achieved by rower A was partly due to a less variable boat velocity produced by a smoother power production. Velocity cost increased with boat velocity.

Studies with larger numbers of rowers in homogeneous groups are required before firm generalisations to other rowers can be made.


Dal Monte A, Komor A (1989) Rowing and Sculling Mechanics. In Vaughan C (Ed) Biomechanics of Sport. Florida: CRC Press. 54-117.
Loschner C, Smith R (1999) The relationship between pin forces and individual feet forces applied during sculling. Proceedings of the Third Australasian Biomechanics Conference. Griffith University, Australia. 31 January-1 February.