A few months ago GCN came to The Bicycle Academy to shoot a short series about frame building, covering each of the main steps of the process as Simon built his very own custom frame. During the shoot we had some long discussions about the different approaches that can be taken with frame design and the subject of frame stiffness was one we debated quite a bit. This isn't surprising as, after all, it is the main variable than can be altered during material selection, but interestingly it was clear that each of us had a different understanding of the actual result of stiffness and flex in terms of real life performance. The truth is that frame stiffness has a slightly more complex influence on the performance of a frame than might be first expected.
So, GCN came back to The Bicycle Academy a couple of weeks ago to attempt to illustrate this in a little more detail, as we knew that it would make for an interesting subject to debate...
In this blog post I’m going to try to explain the theory and our hypothesis, and also address the subject of the little experiment we created for the video.
So, what do we mean by frame stiffness?
In this instance we are specifically referring to the degree to which a frame will deflect laterally at the bottom bracket under a pedalling load. Essentially how much the bottom bracket area moves from side to side as you pedal; if you’ve ridden your bike on a turbo trainer before you will probably noticed this happening. This is what most people are referring to when discussing the ‘stiffness’ of a frame.
What is the commonly believed result of riding a stiff frame?
It is generally assumed that any energy that has been used to flex the frame is wasted as it does no ‘useful work’ in turning the rear wheel. The lateral stiffness of a frame is often used as one of the primary ways in which to judge its efficiency, the assumption being that stiffer is more efficient.
What alternative are we suggesting?
The first thing to clear up is that we are not suggesting that flexible frames are more efficient in terms of energy transfer. It simply isn’t possible for a frame to return all of the energy that was used to flex it. What we are suggesting is that there might be a way that the vast majority of this energy does, in fact, find its way to the rear wheel. As such the lateral stiffness of a frame may not be the benchmark of performance that it is often assumed to be. It might even be that there are some secondary benefits to allowing a frame to be a little more flexible.
Jan Heine, of bicycle quarterly magazine has also written about this phenomenon dubbing it ‘planing’. Riders often talk about favourite frames in terms of almost mystical qualities that they can’t quite put their finger on, perhaps some of these come from the somewhat grey area that links the rider to the machine. Perhaps this is where that magic ‘ride quality’ exists and that rather than riding frames that are overtly stiff (often to the point that it becomes the defining feature of the frame) we could benefit from focussing on the softer points: The points that might make a frame a bit more intuitive or ‘rider friendly’ safe in the knowledge that, at the very least, we aren’t actually incurring the performance penalty that is so often assumed.
How can we demonstrate this?
We rigged up a frame on a static trainer to illustrate just how energy stored in a frame as the bottom bracket flexes laterally is returned to the drive-train as the frame flexes back. This is a really great demonstration for highlighting a mechanism that is often missed at first glance
How does it work?
When the pedal is loaded the frame flexes laterally there is a slight lowering of the bottom bracket, and therefore the pedal, on the loaded side of the frame. In the demonstration a solid block is used to stop the cranks rotating so that any lowering of the pedal is due to frame flex. Out on the road this lowering would still happen but much more dynamically as the rider starts to exert enough force on the pedal to induce frame flex. As the frame flex is released the bottom bracket rises whilst the pedal stays on the block. This produces a useful moment as the bottom bracket momentarily rotates about the now stationary pedal. On the road your foot isn’t on a stationary block but continues to rotate, via a combination of exerted (muscular) and inertial forces, whilst the bottom bracket rises and contributes to the rotation of cranks.
Whilst this is happening to the loaded side of the bike the unloaded side is experiencing the opposite movement. First impressions would suggest that this effectively cancels out any benefit but the key is the movement of the bottom bracket relative to the pedal: On both sides of the bike the bottom bracket moves in the opposite direction relative to the pedal as the frame returns to its neutral position. It is this that produces a useful rotation of the crank and returns the strain energy to the rear wheel.
Are there any other benefits?
The key in all of this is the link between the bicycle and the human riding it: Now that we have seen how energy can be usefully sorted and released by the frame the interesting question becomes whether of not this can be put to good use. Consider here that a more flexible frame will deflect with less force and so the mechanism will be in action for proportionally more of the time that the bike is being ridden. Consider also that the more flexible frame will remain deflected for a greater portion of each pedal stroke and therefore return the energy later during the pedal stroke, at a point where the rider themselves is less able to produce useful work. The timing of the release of this energy could potentially be beneficial to the rider; altering the proprioceptive feedback that they get during each pedal stroke. Perhaps offering improved comfort or even pedalling biomechanics: In the same way that when we walk or run we do so a little differently depending on the surface that we are on.
Whilst the difference may not be quite as black and white as running on tarmac vs a mountain trail they are definitely perceptible and most riders will be able to tell two frames apart in terms of lateral stiffness. The initial perception of a very stiff frame is one of immediate responsiveness as the reaction force that we feel through our feet when we press on the pedals is instantaneous. A more flexible frame feels a little 'softer' under our feet as the reaction force as we push on the pedals ramps up and tails off a little more gradually. My own experimenting with this has found that, with a bit of adaptation, it is possible to use this flex advantageously making it easier to maintain that feeling of really being 'on top' of a gear. This probably isn't that surprising considering that we are very adept at using flex and spring in tools to our advantage; from the more obvious examples rackets, bats and clubs down to the tendons in our own bodies, all are used to reduce the actual load that our muscles have to take, 'softening' the reaction force that the muscle has to overcome to do any useful work.
How robust is the experiment?
It is important to point out that the demonstration used in the video is really nothing more than an illustration. We made no attempt to control any of the variables (to name but a few); how much force was put onto the pedal, the distance between the pedal and the block, the ‘stiffness’ of the frame itself were all nominal. It certainly wouldn’t be the best way to try and draw any quantitative conclusions as there are many elements of it that do not reflect the reality of riding the bike ‘out on the road’, not least the very static nature of the demonstration.
It does, however, allow us to separate displacement of the bottom bracket and pedal caused by lateral deflection of the frame from that caused by rotation of the cranks. This allows us to clearly see the contribution that lateral deflection can make. It does not give us a clear idea as to the extent to which this happens during actual pedalling load but it does demonstrate the potential.
There is a huge amount more work that could be done here. In the first instance it will be important to create far more controlled environments to look at what is going on dynamically. This would help us to better understand the extent to which this might happen ‘out on the road’ and some further modelling and testing might help to quantify under exactly what circumstances the frame behaves in a particular way.
It is also really important to get a greater understanding of the potential human benefits; exactly how much does this potentially alter the pedalling of the rider, what biomechanical or physiological benefits could there be, what are the adaptations needed to make best use of this?
What do you think ?
Do you prefer a stiffer frame? Can you identify with some of the points we make? Is the demonstration so far removed from reality that it’s giving us unrealistic feedback? Are we on to something or are we just talking rubbish? Let us know in the video comments.
Thank you for reading
Head of Education - The Bicycle Academy