Elliptical Chain Rings Don’t Work (Because You Won’t Let Them)

New research shows how biomechanics trump bike mechanics

Depending on how you feel about elliptical chain rings, a new study on them will either sound like a snipe hunt or a deep dive through cycling’s warehouse of obscure and useless novelties. The new study in question has reached profound conclusions on longstanding debates and generated new areas of study for how this magnificent mechanism we call the bicycle actually works.

The short answer goes like this: Elliptical chain rings don’t do anything to help you. Nor do they do anything to hurt you. Also, it’s not that simple.

The long answer is much more fascinating (if you’re a bike geek).  How we could spend so long debating one aspect of its operation or find new horizons of discovery pertaining to a contraption that’s been around since the 1800’s requires a little context.

Elliptical chain rings entered the mainstream of modern cycling in the 1980’s when Shimano introduced their “Biopace” model for racing bikes. They quickly faded back out of fashion when people began associating them with knee injuries. Yet the idea persisted in disparate corners of the cycling world, from garage-based engineers to pro tour competitors, that there existed a magical eccentricity that would yield exceptional power gains.

Research bent on discovering the “perfect oval” became something of a hunt for cycling’s holy grail. One of the most provocative studies emerged in 2008. Two of the most prominent biomechanical researchers in the cycling world, Jeffery Rankin and Richard Neptune, published their theoretical model of a human leg and the pattern of muscular contractions it performs during the crank cycle.

Computer simulations suggested that a chain ring of some eccentricity was the optimal shape. Even though the model had not been validated by real-world testing, advocates of elliptical rings took it as a sign that the grail was out there to be found.  

Manufacturers like O.Symmetric, Rotor, and AbsoluteBlack have offered their best-guess products to cyclists in the hope that they were close enough to the answer.

Putting  the product ahead of the research created immediate problems. Because they had to sell a product, the manufacturers had to offer an explanation for why that product was worth customers’ money. The only way to do that with a product so based on science was to offer scientific evidence of its effectiveness. Scientific “proof” of a product’s worthiness coming from the product manufacturer itself creates endemic bias.

Ask any two people to produce “the best elliptical chain ring” and they’ll make two different designs. How they design the product will naturally be based on what qualities they think are most important to being “the best.” The same rule applies to testing protocols. Why would you try to beat a competitor at their own game when you think they’re playing the wrong game in the first place? In effect, commercial competitors can’t reconcile the debate over who has the best answer because they argue about how to argue.

Much less advertised is the ongoing academic research. Academia pursues things along different lines, though. While product manufacturers test only to prove their product does one thing, academics look at multiple measures of effectiveness to determine everything a product might do. A new study published by researchers at the University of Utah did exactly that with surprising results.

Chee Hoi Leong is an Assistant Professor at Central Connecticut State University and former graduate student of Dr. Jim Martin, a renowned professor of exercise and sport science who has published numerous studies on cycling kinematics. Leong set out to understand the relationship between chain ring shape, the motion of the joints in the ankle, knee, and hip, and metabolic cost of cycling. In two studies using chain rings of varying shapes, he found no discernible relationship between eccentricity and cycling performance.

Leong’s first study used three chain rings: one circular and two of increasingly elliptical shape. His results found that cyclists did not experience an improvement in maximal cycling power or optimal pedaling rate. Because other studies have found different results, Leong’s research could be disputed on the grounds that perhaps he didn’t test enough subjects (a common problem with tests on cycling kinematics), that his lab conditions (size of bike used, nature of the resistance wheel) threw the subjects off, or that he didn’t give his subjects enough time to “adapt” to the elliptical rings. This is a claim often made by elliptical ring manufacturers. The problem is that the argument itself seems to contradict the original claim that the benefit is all in the rings.

Is it all in the mechanics or is there something to do with how the legs respond to them? That was what Leong set out to investigate in his second test.

6226092833_c0c7c0e6a4_b

In that experiment, Leong measured the angular velocity of subjects’ ankle, knee and hip joints while using the aforementioned chain rings. He found that velocity at the ankle joint changed radically as a result of elliptical chain ring use, but there was no change at either the knee or hip. Nor did power production at any of the joints change.

“The ankle negates the chain ring,” says Leong. “Nothing from the ring gets transferred to the knee or hip.”

In other words, the cyclists adapted their leg motion to the chain ring shape so that they could continue pedaling the way they’re used to. And because they kept pedaling the way they always do, they produced the same power they always do.

That inspired a third test to check Leong’s suspicions, which made his research truly unique. One of the difficulties of investigating elliptical chain rings is that the speed of the pedal varies throughout the crank cycle. If you pedal with a circular ring, your knee joint bends with varying rates of angular velocity throughout the crank cycle. However, that velocity is directly related to where the pedal is in the crank cycle and therefore easily modeled because the pedal moves at a constant speed throughout. But with an elliptical ring the pedal speed itself varies.

Without a constant pedal speed, it’s difficult to tell what influences what. Leong was able to work past this variation to determine exactly what was causing the changes in leg motion. He found that there were some changes to the joint kinematics, but there was no influence on the metabolic cost of pedaling.

Leong’s research is a direct contradiction to the old “function follows form” belief that an elliptical ring works by simulating a bigger ring during the power production phase of the crank cycle and a smaller one through the “dead zones” at the top and bottom.

“It’s an optimistic perspective that chain rings work they way they do based on what the ring looks like. Our view is based on what the legs do. Causing the leg to spend more time in the power production phase of the crank cycle does not increase performance.”

In short, it doesn’t matter how your legs turn the cranks. They always have to come full circle, and because of that there’s no escaping the associated cost of movement.

According to Leong, one of the primary challenges is the confusion between power and efficiency.

“These rings attempt to change the gear ratio, but their shape is only the equivalent of adding two or three teeth. Two teeth equals a 4-percent increase or decrease in pedal rate. It doesn’t change the metabolic rate.”

If the metabolic rate doesn’t change, then neither can the cycling efficiency, which is the ratio of power produced to the amount of energy spent to produce it. This ought to be intuitive enough to anyone who’s ever moved up from a compact 50/34 to a standard 53/39. The larger gear allows you to generate more power without spinning at a torturous rate, but only if you’re willing to invest the greater level of muscular force. There’s no free lunch.

So elliptical rings don’t work. If you survey the current equipment landscape it seems pretty apparent we already knew that. However, the implications of Leong’s research still hold value to anyone trying to go faster on a bike. To begin with, Leong says that Biopace may be due an apology.

“While elliptical rings don’t help, they’re not going to hurt either. It’s totally a matter of preference if someone wants to buy them or not.”

The bigger consequence is what Leong’s findings mean for another debate over power production. If gear size doesn’t change cycling efficiency, then neither should crank arm length. That’s exactly what researchers are coming to understand. But while a shorter crank arm doesn’t influence power production, it can improve bike fit and aerodynamics.

For as many answers as Leong has found, there are also new questions. A big one is why exactly our legs adapt. The fact that we’re still making discoveries about the relationship between something as simple as a mechanical gearing system and something as intuitive as the human leg means that we don’t know as much as we thought we did. Leong is excited about potential future discoveries.

“Our study had many limitations. There were some things we couldn’t look at, and so there are more places to look.”

Leong is most interested in combining his findings with the previous work done by Neptune and Rankin.

“The icing on the cake would be to use our findings to drive Neptune’s model. The data we have could refine his model and reveal new information about how the muscles work during the crank cycle.”

What does he hope to find? There’s no telling, and that’s what makes it exciting.

Follow Jim Gourley on twitter. Buy his books below.

photo credit: Bradley Wiggins via photopin (license)

photo credit: Thy Cykle Rings Masterløb 2011 via photopin (license)

About the Author

Jim Gourley is the author of Faster: Demystifying the Secrets of Triathlon Speed and The Race Within: The Story of the Ultraman Triathlon. He is a regular contributor to Tom Ricks' blog "The Best Defense." His work has been featured in Men's Health, Stars and Stripes, and several triathlon and cycling publications.
  • Chris Zacho

    I think, based on the scientific method of personal experience in the real world, that it depends on the person using it and it’s applicability to that person’s use of the product that decides whether it works any better or not. What’s good for the goose…