Articles & Press


A Discussion of Rowing Ergometer Use

Ivan Hooper M.Phty.St (Sports Phty), B.Sc (HMS)
APA Sports Physiotherapist
Australian Institute of Sport

Rowing Australia
Sports Science / Sports Medicine Coordinator

Following on from my recent email regarding injury and illness statistics, I would like to raise some comments and considerations regarding the use of ergometers for training. I have noticed that there seems to be a trend towards increasing use of the ergometer in training, particularly doing low rate work down to ratings as low as 12. I understand the benefits that this type of work can produce, but I would like to make you aware that this form of training is not without risk of injury.

In some of his regular newsletters, Valery Kleshnev highlighted the fact that the kinetics and kinematics of ergometer rowing are different from that of on water rowing. On an ergometer, the handle force has a higher peak and develops later, the stroke length tends to be 3-5% longer and the curve of foot stretcher force is considerably moved towards the beginning of the stroke. An important point is that the legs:trunk:arms proportions of power development on an ergometer are 37%:41%:22% compared to 45%:37%:18% for on water rowing. This means that the trunk is doing a larger proportion of the work on an ergometer. I believe all of these factors lead to an increased load applied to the structures of the trunk, and particularly the spine. Greater work done by the trunk could produce earlier fatigue of the trunk muscles, placing the spine at risk.

Holt et al (2003) studied the effects of prolonged ergometer rowing. Over a 60 minute piece there were significant changes in the way the athletes moved. Lumbar spine range of motion at the catch and total lumbar spine range of motion increased during the piece. The gradient of force production decreased, and the ratio of drive to recovery time increased, over the piece. The authors attributed these changes to fatigue of the trunk muscles during the piece, reinforcing that fatigued trunk muscles may lead to low back injury.

Teitz et al (2002) conducted a retrospective study of 1632 US intercollegiate rowers. By the use of detailed questionnaires they established that 32% of these athletes had experienced back pain of at least one week's duration during their rowing careers. The use of rowing ergometers for greater than 30 minutes per session and free weights were the variables most consistently associated with back pain.

In my experience, I feel that athletes often pay little attention to their rowing technique when on an ergometer. The level of coaching supervision is often limited as well. The result is that athletes spend time on the ergometer under greater trunk load than when on the water, with poor technique and poor postural positions. The end result is an increased load on the spine which can increase the risk of injury.

It is very common for athletes to report that they feel that the ergometer was highly related to their back pain. And those athletes with current back pain regularly report that ergometer rowing aggravates their pain more than on water rowing. When this feedback occurs over a significant number of athletes over a number of years it is difficult to dismiss.

Unfortunately I believe that we are seeing an increase in the number of low back injuries amongst rowers. The three month injury and illness statistics that I recently sent to you highlighted the fact that back injuries are having a significant effect on team preparation, both at an individual and crew level. Even though I am sure that there are many causes of this increase in back pain, evidence and experience suggests that ergometer use is a significant one.

While I am the first to acknowledge that the ergometer is a powerful training tool, I ask coaches and athletes to give due consideration to the risks involved. Please consider the time spent on the ergometer, the rates that training is done at, the supervision provided and how diligently athletes concentrate on their technique.

I hope that we can all work towards a reduction in low back injury rates. For every back injury that we avoid, that is an extra 30 days (on average) that the athlete can spend training properly! Any feedback regarding this subject would be most welcome.

Holt P J E et al. Kinematics of spinal motion during prolonged rowing. International Journal of Sports Medicine 2003; 24: 597-602.
Kleshnev V. Rowing Biomechanics Newsletter;  :April 2001.
Kleshnev V. Rowing Biomechanics Newsletter;  :October 2003.
Kleshnev V. Rowing Biomechanics Newsletter;  :January 2005.
Teitz C C et al. Back pain in intercollegiate rowers. The American Journal of Sports Medicine 2002; 30 (5): 674-679.


A Discussion of Fixed vs Dynamic Ergometers

Ivan Hooper M.Phty.St (Sports Phty), B.Sc (HMS)
APA Sports Physiotherapist
Australian Institute of Sport

Rowing Australia
Sports Science / Sports Medicine Coordinator

Since I sent out some comments regarding ergometer use, I have had quite a few emails back regarding the use of the Row Perfect ergometer, or putting the Concept II ergometer on sliders. I am aware that there is some work underway investigating this issue, but currently there are not a lot of papers that have been published.

In working through some of the literature I came across a website that goes some way towards explaining the physics of ergometer rowing (Dudhia, 1999). It discusses that a fundamental difference between the linear mechanics of a ‘static’ ergometer (such as a Concept II) and a boat can be illustrated by the following test:

  • If you sit at front-stops on an erg and then push your legs down you move backwards relative to room by an amount equal to your leg length

  • If you sit at front-stops in a single and then push your legs down (oars out of the water) you only move backwards relative to the bank by an amount ~20% of your leg length - the rest of the motion is taken by the boat moving away from you.

This is not just a matter of the frame of reference: in the static case (ergometer) you are actually performing more work accelerating your whole body weight up and down the slides, thereby creating high levels of kinetic energy. In the dynamic case (boat) your body weight is relatively stationary, creating much lower levels of kinetic energy and thus requiring less work to be done to reverse this kinetic energy. It results in an athlete needing to put in six times more energy just accelerating and decelerating their own body weight, compared to on water rowing.

A ‘dynamic’ ergometer, such as the Row Perfect, attempts to simulate the mechanics of on water rowing by having the stretcher/flywheel also mounted on a rail. Attempts have been made to simulate the same effect by mounting the Concept II on sliders.

Most of the literature that I have read was performed examining the Row Perfect ergometer in a mobile and fixed state. The weight of the Row Perfect mobile power head is approximately 19kg, which is not that dissimilar to the weight of a single scull. This is the weight that an athlete’s leg drive is moving every stroke. Hence the manufacturer’s claims that the mechanics of the Row Perfect and on water rowing are similar.

The weight of a Concept II is nearly 28kg. When you include the mobile component of the sliders, the weight is around 35kg. If you consider the mechanics discussed earlier, when a Concept II is mounted on sliders there would be more motion of the rower and less motion of the ergometer when compared to the Row Perfect. Hence, my thinking is that sliders probably go a long way to replicating the mechanics of on water rowing, but still involve forces nearly double that of the Row Perfect.

There are two recent papers that have both described the mechanics of static versus dynamic ergometers, using the Row Perfect in both a dynamic and fixed state. Bernstein et al (2002) found that average stroke length on the static ergometer was 53mm longer. They discussed that this is due to the higher kinetic energy associated with moving the whole body mass, as was discussed earlier. Colloud et al (2006) also discussed the higher inertial forces generated during the transition between the recovery and propulsive phases, especially at the catch.

This kinetic energy, and / or inertia, has to decrease to zero for a change in direction to occur, thus something has to exert or absorb forces.  Coming forward this force is absorbed by passive tissue structures of the knees resulting in an 8-10% increased leg compression (Kleshnev, 2005). It is reasonable to assume that the lumbar spine also absorbs some of this kinetic energy, creating an increase in lumbar flexion. Holt et al (2003) supported this when studying the effects of prolonged ergometer rowing. Over a 60 minute piece there were significant increases in the lumbar spine range of motion at the catch and total lumbar spine range of motion.

At the finish it is the large hip flexors that act to decrease and reverse the kinetic energy of the trunk (Rekers, 2006). This places very high loads on the lumbar spine, equivalent to doing prolonged sit ups. This places large sheer forces across the structures of the lumbar spine, potentially contributing to injury (Stallard, 1994).

Both Bernstein et al (2002) and Colloud et al (2006) found higher maximum stroke forces and power when using the static compared to the dynamic ergometer. They suggest that the passive structures of the rower’s joints could be loaded more at the catch on the static ergometer when the lower limb joints and trunk are fully flexed. They both propose that these higher forces, imposed over a longer stroke, may be associated with injury.

Undoubtedly, higher forces applied over a longer distance means more work done by the body’s muscles. More work done means earlier fatigue. Fatigued lumbar spine muscles may allow even more lumbar flexion, transferring higher forces to the passive tissues of the spine. The combination of lumbar flexion and muscular fatigue has long been identified as a cause of lumbar spine injury amongst rowers (Reid & McNair, 2000).

After repetitive motion, protective muscle activity has been shown to be reduced, often for a number of hours after the exercise is completed (Gedalia et al, 1999) The ramification for rowers is that, during this period, the athlete may be more vulnerable to injury, even when they may not be experiencing high loading on the spine (Reid & McNair, 2000). Ergometer use and weight training are two modalities that are likely to load the trunk muscles more than on water rowing. Based on the findings mentioned above, placing these two training modalities in close proximity is likely to increase injury risk.

In discussing ergometer versus on water rowing, Kleshnev (2005) noted several differences. He stated that the legs execute more work on a stationary ergo, but in a slower static motion. On the water the legs work much faster at the catch, when the force is not very high and therefore execute less power. In this aspect a dynamic ergometer stands somewhere between a stationary ergometer and on water rowing.

This may be an aspect that coaches wish to utilise if they are looking to enhance leg training, but I question the value of this when the load and contraction speeds are significantly different to on water rowing. The other issue is that once the legs fatigue, the trunk then becomes a greater contributor to total work performed. As mentioned above, this leads to a fatigue of the trunk muscles, placing lumbar spine structures at higher risk of injury.

In conclusion, the information that is currently available supports the idea that ergometer use is a risk factor for lumbar spine injury. It also suggests that the Row Perfect places much lower detrimental forces on the rower than the Concept II. It seems that placing the Concept II on sliders is also a way of reducing these detrimental forces, but this is probably not as effective as the Row Perfect.

At this point in time, the Concept II is the standard for conducting physiological testing of the elite rower. I do not propose that this change immediately, but I do think that what machine we test on in the future needs further examination and evaluation. Issues such as injury risk and physiological specificity need to be considered when selecting the most appropriate way to test our athletes.

In summarising the information that is currently available regarding ergometer use and its effects on injury, I would like to make the following recommendations:

  • Reduce the volume of work done on Concept II ergometers in the stationary setting.
  • Keep the maximum length of a piece on an ergometer less than 30 minutes. If more than 30 minutes is to be done in a session, make sure that the session is broken up into shorter pieces with appropriate rest and stretching in between the pieces.
  • Where appropriate, use either the Row Perfect or Concept II ergometer on sliders.
  • Where appropriate, use other forms of cross training. Consider using cross training in conjunction with ergometer training in order to achieve the necessary training volume.
  • Endeavour to place ergometer sessions and weights sessions on separate training days, or at least several hours apart.
  • Provide good supervision of technique while athletes train on an ergometer. The level of attention to technical detail on an ergometer should be no different to when training on water.
  • Ensure that athletes understand that the need for good technique while training on an ergometer is as important as when on water.
  • Be aware that some people will never have problems on an ergometer, while others may have significant problems. Coaches should be prepared to individualise training programs to suit each athlete.

The recommendations made in this article are based on a balance between possible injury risks, and the acknowledged benefits of ergometer training. Ideally these recommendations are designed to stimulate thought when devising training programs. I would encourage coaches to consider both the potential benefits and the potential risks of all forms of training.

Finally I would like to remind everyone that coaches have a duty to make their crews as fast as possible, without causing damage to the people for whom they are responsible (Stallard, 1994). An ongoing challenge for all coaches is to minimise the potentially detrimental aspects of their training programs.

Dudhia, A (1999) The physics of rowing: dynamic versus static ergometers.


Road Testing the Rowperfect


The RowPerfect Ergometer: A Training Aid for On-Water Single Scull Rowing

Bruce Elliott, Andrew Lyttle, and Olivia Burkett

The purpose of this study was to compare rowing technique on the dynamic RowPerfect ergometer with a single scull. 

Eight national-level rowers performed on both the RowPerfect ergometer and in a single scull over 500m, at rates of 24, 26 and 28 strokes/minute. Click here for the PDF of article.


Verification of Rowperfect weight correction software

Cas Rekers

Recently I did another check to test the validity of my calculation method for conversion of boat speeds from one boat type to another. The weight adjustment factor in Rowperfect software is based on the known relationship between variations in displacement of particular boat types and the resulting variation in wetted surface area.

In the calculations, the ratio of wetted surface area of a given boat type at different displacements is approximated considering the boat as a semi-cylinder of a given length. To further validate this approximation I made a comparison between a coxed (2+) and a coxless (2-) pair with the same calculation procedure, and comparing the outcome to actual results. Input data for 2-: length 10.30m, weight boat 27kgs, weight oars 2.5kg each. For 2+: length 11m, weight 32kgs, weight oars 2.5 kg each, weight cox 55 kgs.

In Luzern at the 2001 World Championships racing conditions were very stable. The coxed and the coxless pair were both won by Cracknell and Pinsent (UK) in very tight races. Their times in these two races were: 6:49.33 for the coxed pair and 6:27,57 for the coxless pair respectively. In their coxed pair race, Pinsent and Cracknell visibly eased-up a couple of strokes before the finish, which I estimate, may have slowed them down by one or two seconds.

Assuming Pinsent & Cracknell’s weight at average 100 kgs each (pretty close to their weights at Henley Royal Regatta), the weight corrected speed ratio between the coxless and the coxed pair for this crew, I calculated from the above input at 1.0505. From this one can make the following comparison, using the time of the coxed pair as a base, and calculating the theoretical time of the coxless boat, based on wetted area/ weight correction factor as used in Rowperfect.















2 secs. 

Even without the "easing up" correction, I think this is as close as one would wish, and demonstrates clearly the validity of the "wetted surface" based weight correction factor. With the "easing up" correction of 2 seconds to bring the time for the coxed pair to 6.47.3 this would result in a calculated time for the coxless pair of 6.27,75 which is dead on the spot.


Ergo design and safety: fixed head or floating?


Australian Ultra-FIT — Aletha Mays tested this revolutionary ergometer for three months & she’s not giving it back!

Have you ever wanted to own a single piece of exercise equipment that would not only give you a great cardiovascular, but also a demanding strength training workout? Would you then want this machine to need very little in the way of maintenance, and also fit into your spare bedroom or living room, without being too obtrusive? Plus, would you like it to be able to monitor your performance and give you immediate and useful feedback? Oh, and have you ever wanted to row like the Oarsome Foursome?

You might not have answered yes to the final question, but if you nodded your head in agreement to the others, read on, because I reckon I’ve found the perfect fitness machine for the home. The Rowperfect—a rowing machine designed by a rower for rowers.

Having enjoyed using rowing machines in the gym environment for many years, I was pretty excited when asked to do a home equipment review on the Rowperfect.

Developed by Dutch rowing coach Casper Rekers in the early 1990’s, this is a unique piece of fitness equipment which has, up until fairly recently, been used almost exclusively by the world’s rowing fraternity as a training weapon for those aiming for Olympic Gold—like Great Britain’s Eight, or members of Australia’s own Oarsome Foursome, who use this machine for their land-based training. Rekers’ ideal was to build the perfect rowing simulator of on-water rowers to train on land during the winter months. And, from what I’ve seen, it looks like he succeeded.

Unique Construction
With its bare stainless steel facia and no fuss, wooden handlebar, seat and foot pads, the Rowperfect has a basic, industrial non-nonsense, utilitarian appearance. And, although it doesn’t look too dissimilar to other ergos (rowing machines) on the market, it’s wholly unique in that it has a dual action movement, ie the flywheel housing, as well as the seat, move independently along its length. The result of this unique, ground breaking design is a surprisingly sophisticated rowing machine which manufacturers claim puts comparatively less strain on your knee and hip joints and lower back, when put up against single action (fixed resistance housing) rowing machines. This has since been verified in tests conducted by Dr Richard Smith of the School of Biomechanics at the University of Sydney.

What is it like to use?
The fact that the Rowperfect was originally designed to simulate as closely as possible the action of a light racing shell on water, means that it’s inherently a little unstable, and relies on the user having some knowledge of stroke mechanics to get the best out of it. However, once you begin to concentrate on your stroke technique and get used to the flywheel housing instead of your body moving backwards and forwards along the main rail, the sensation actually becomes far more appealing than sitting and rowing backwards and forwards as you would on other machines.

This dual movement is also the secret to the Rowperfect’s comparative gentleness on your joints. How so? Well, because the fan housing weighs just 17.5kg (the approximate weight of a single scull), at the instant you take up the catch of the stroke, it’s the fan housing that moves away from you, and not the other way around. So, in that split second it takes to initiate movement (the moment where you have most compression on your knees), you’re only asking your body to move against 17.5kg, not the full weight of your body, which might be up to five times the mass of the fan housing. Also, at the finish of the stoke, where your hip flexors and lower back are put under pressure because of the reversal of momentum, again, it is merely 17.5kg of weight which is asked to be moved. This ultimately puts less strain on your knees, hips and lower back than single action ergos.

Feedback & Resistance
To complement its performance, and to help you get into your stroke, Rowperfect have developed a PC interface which, when attached to your home computer, gives you all the normal feedback (speed, stroke rate, time, distance, etc) plus it monitors your actual stoke technique by displaying an easy-to-understand power curve. And if you’re a serious rower (indoor or outdoor) Rowperfect have pre-programmed into the software the power curves of some of the best rowers in the world, so you can compare your technique to, say Steve Redgrave. The only negative side to receiving this comprehensive feedback is that the PC must be positioned nearby so that you can link into the software.

Resistance is provided by a covered air fan, and its adjustment is basic. By adding or removing concentric plastic discs to or from the open side of the fan/flywheel cage, available air is controlled from entering the cage, so a variable resistance is achieved. This operation is only possible before or after a workout, not during, which isn’t a major problem, really. Other than an occasional wipe down with some light oil every couple of months, the Rowperfect requires little maintenance.

Not so perfect
There are a couple of negative aspects to the Rowperfect, all of which are currently being addressed by the manufacturers. Firstly, the seat is somewhat small and hard, which can cause discomfort after thirty minutes or more of continuous rowing. Rowperfect has assured us that their ergos will soon be built with wider seats and more padding to reduce this discomfort. In the meantime, importers recommend you buy some rubber and glue it on to the seat, or simply fold a towel to use as a cushion. Secondly, the imminent development of an on-board computer, displaying similar motivational information to those found on all of its rivals, will eliminate the need of having a dedicated PC nearby.

The Verdict
After thee months using this machine, the verdict is that I loved it so much that I’m not giving it back! The Rowperfect has inspired both myself and my partner to get fitter, and we use it almost every day. When my partner first started using the Rowperfect he was on it for maybe ten minutes a day. Now his workouts have extended to rowing sessions fifty minutes—and he loves it! In my mind, there’s no doubt that it has improved our strength, endurance, coordination, muscle tone and has reduced our body fat levels. The Rowperfect is a machine for the serious rower looking to improve technique, as well as the individual who is looking for a machine that can provide a total body workout.

A growing trend
Because ergos work every major muscle group in the body, they are very effective fat burners. They’re also fun to use, create little stress on the joints, and so have excellent rehabilitative exercise qualities. All these factors have made the humble ergo, originally designed as an aid to ‘real’ rowing, a very popular form of exercise—so popular, in fact, that there are now numerous indoor rowing championships held around the world. In America and Europe, indoor rowing competitions are common. However, in Australia these indoor regattas are still a relatively new phenomenon. You may remember earlier this year Ultra-FIT sponsored the inaugural Zurich Australian Indoor Rowing Championships 2001, with Ultra-FIT readers making up over half the competitions! And more recently Rowperfect themselves sponsored an indoor regatta event in Sydney, which attracted over 200 competitors. Be on the lookout for more indoor regattas in the future. They’re a growing trend.


Sliding Rigger Physics

People often mistakenly believe that the Rowperfect simulates a sliding rigger boat. The explanation below was sent to me by Cas Rekers when I wondered whether fixing the seat would achieve the equivalent of a sliding rigger boat. (Mark Campbell)

As far as simulation of the sliding rigger boat with just the seat fixed, I’m afraid you are wrong. Remember it is all about masses moving relative to the common center of gravity of boat plus rower. In a normal racing shell the moving masses are: The rower (e.g. 87.5kgs on one side), and the boat plus oars (for the Rowperfect taken at 17.5kgs) on the other side, the ratio of masses then is 87.5/17.5 equals 5.This means that with a traveling length of 60cm on the slide, the rower moves 10cm relative to the common center of gravity and the boat 50cm. Assuming this movement is a perfect sinus at a stroke rate of 30 strokes per minute that would cause a fluctuation of the boat speed relative to the common center of gravity of plus and minus 0.7m/sec. In practice this is higher. Nolte measured at the 1981 World Championships on average of the 5 sliding seat finalists from -1m/sec. to +1.15m/sec.

In a sliding rigger boat, the situation is quite different; there we have the combined mass of rower plus the shell of the boat and the seat (roughly11kg in a single) versus the remaining 6.5kg of moving stretcher/rigger/oar, so a ratio of around 15. Assuming that the center of gravity of the rower is at rest with respect to the boat, this ratio would mean a fluctuation of the boat, (now coupled to the heavy rower) relative to the common center of gravity of only 4cm, generating a fluctuation in boat speed of plus and minus 0.06m/sec. In practice this fluctuation will be higher because by bending the knees and pulling the heels towards oneself, the center of gravity of the legs moves also with respect to the boat, so does also the center of gravity of the upper body. The total displacement of the center of gravity of the body relative to the boat however is definitely a lot smaller than in case of a sliding seat. In practice Nolte actually measured plus and minus 0.26m/sec. If you would just fix the seat of the Rowperfect, you would have an infinite mass of (rower + seat + mother earth) against a mass of 17.5kg, in this case simulating just the oar rigger combination. In view of the above, for a good simulation of a sliding rigger boat, one would need to increase the weight of the seat to around 11kg, and to decrease the weight of the stretcher/flywheel main frame to around 6.5kg.

Regards, Cas Rekers