High stiffness and performance gain

[LetsRide]

I have seen several tests showing how aeroness and weight contribute to performance improvement. How about tests to show if high stiffness really helps with power saving or speed increase, especially on a climb or in a sprint ? Has anyone seen some tests like that before ? I have not seen any so far.

On a very stiff bike, riders can immediately feel high power transfer and responsiveness. However, I have no idea how those stiffness figures are objectively related to performance gain. For instance, will a very stiff bike significantly transfer much more watts to the road than a flexy one ? climb faster when putting in the same amount of power ? go faster with 1200-watt sprint ? Is there a point for "stiff enough" or the higher the better ? Hope some tests can shed some lights on this. Thanks.

Figures listed below are some example of stiffness tests, not from third party though.

[eric]

I haven't seen any tests either. Since the frame is a spring, most of the energy that goes into bending it will be returned. The problem is that it's returned via the crank and pedals to the rider. That's where it gets difficult to understand, model or measure.

[WMW]

I haven't seen any tests either. Since the frame is a spring, most of the energy that goes into bending it will be returned. The problem is that it's returned via the crank and pedals to the rider. That's where it gets difficult to understand, model or measure.

There was a company a few years back who made a spring loaded hub to cause a lot of drivetrain flex. They claimed big power gains. Also Jan Heine has convinced himself that frames with a decent amount of flex are actually faster on climbs.

The amount of energy that the bike absorbs when it flexes is small to begin with, so there couldn't be much that fails to be returned. A very well calibrated pedal and Powertap pairing should be able to settle the matter.

And the thing that might (maybe) tip it in favor of flex, is the effect on muscle engagement.

[LetsRide]

I agree that a pedal based power meter with a hub based power meter can get the answers. Steep climbs and sprints may magnify the difference between the two power meters. With one power meter, riding a stiff bike vs a flex bike, with the same weight, uphill with constant power, say 200w & 300w, then record average speed and time may give some ideas. On high-power sprint, there should be air resistance largely involved so picking bike with the same shape with varying amount of stiffness may not be easy. I am still surprised I can't find a simple test about this topic on the Internet.

[kgt]

Maybe because such a test is not something simple at all.

[Pugrot]

What about putting it in terms if energy.

If we can assume that there is negligble foward momentum gained when the frame returns from its flexed state then the energy expended to flex the frame is all lost (turned to heat and frame motion)

If we know the bb deflection from a line drawn between the tires, and we know the force going in to the pedals and the duration of each pedal stroke, and the angle between the force and the deflection then the energy loss per pedal stroke could be calculated.

We might need some clever people in here for that though

[drchull]

I would think that the recoil of the BB would be coming back at the dead part of your pedal stroke as the pedal comes around the very bottom of the pedal stroke potentially providing some power where the leg is least efficient, especially when standing. Doubt there is significant real world gain or loss on the downstroke. In some really, really powerful sprinters though the more fleyx frames would start shifting or losing the chain if there was a lot of flex.

Just goes to show you that throwing up some comparative graphs is a great way to sell something even if there isn't a defined benefit. I was never faster than when I rode a wet noodle Columbus SL steel bike, then again I was 20 years younger.

[prebsy]

lots of limits to these potential "flex" benefits, you rub one brake pad and there go your "gains". Also let us not forget about handling. I think the discussion is a good one I know it has been discussed in depth with wheels. I don't know if there was conclusive evidence that a stiffer wheel has any benefits.

[LetsRide]

Does the returned energy of flexed frame actually help move the bike forward ? Or it goes somewhere else ? That's what I still doubt the benefits of flexy frame. If results from tests show no performance difference between a stiff bike and a flex bike, then this theory might make sense.

I tried putting some data into a bike calculator. Stiffness was assumed as a part of transmission efficiency.

9% grade, 1 km distance, rider weight 75 kg, bike weight 9 kg, tubular tires, no head wind
transmission efficiency Set A 94% Set B 98%

Case 1 : power 300 watts
speed : Set A 12.66 kph Set B 13.16 kph
time : Set A 4:44 mins Set B 4:34 mins
time difference 10 seconds

Case 2 : power 200 watts
speed : Set A 8.60 kph Set B 8.96 kph
time : Set A 6:58 mins Set B 6:42 mins
time difference 16 seconds

Anyone would like to share your test results is welcomed. Adding extra weights to the lighter bike to match the heavier bike should help. Same wheelset is recommended on both bikes.

[eric]

Transmission efficiency (drivetrain friction) is different from flex- 100% of the power lost to friction turns to heat.

The problem in modelling or even understanding the effects of frame flex is what happens when the force is returned to the rider.

[XCProMD]

The higher amount of flex by far happens at the tyres, and a great deal of it is dispersed. Next comes rear wheel and chain, but very far.

[bm0p700f]

If the test were simple it would have been done. There are so many variables to control any meaningful test is impossible hense I do not worry about how stiff something is. Does it feel right if so good.

[LetsRide]

The test might not be quite complicated. I know it's not a perfect test. Pick a long steep climb then ride Bike A and Bike B uphill with constant power on a calm day. Try using the same wheelset and matching overall weight on both bikes. Repeat the test by other riders. If frame stiffness did not affect performance or power transfer at all, bikes with the same overall weight should climb equally well with the same average speed and finish the same time. Perhaps, consumers could ignore all claims of high stiffness, and also hi-mod carbon fiber. Bike companies might save some materials to produce a sub-500g super light noodle frame that should climb superbly.

However, to me, how frame is flexed or twisted back and forth when force is applied does not seem to help rotate wheels forwards. (That's why I assumed it as power loss in a bike calculator) I am still skeptical with the hypothesis of springy frame until I see a valid proof showing the same performance for both a flexy bike and a stiff bike with controlled variables as I've pointed out before.

[DanW]

In their online blog, Cervelo claim frame stiffness is such a small component of overall performance even in a sprint so can be ignored towards overall performance. I interpret that as "it is too difficult to model and give a decent answer"

"Negligible" contribution to performance is a subjective thing though. In Spesh's Win Tunnel videos they show ~80s saved in a 40km TT @ 40km/hr from shaving legs (something most people would write off as a negligible change) yet when they compare an old school skinny steel frame to the latest Venge there was just ~50s saved... yet this is interpreted as an excellent demonstration of the advances in aero bike design over the last 30 years I'm sure similar comparisons wrt stiffness must also be true- if it is significant or not depends on what you are trying to prove/ sell

[dynaserve]

Using a Powertap, when sprinting above 1000W I consistently got up to 10% more power on my 'stiff' bike vs 'flexy' bike. At around 300W there was virtually zero difference.