High stiffness and performance gain

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

petromyzon
Posts: 781
Joined: Mon Apr 05, 2010 4:14 pm

by petromyzon

10% of 1KW is 100W --> if you are losing power due to flex something should be heating up, that is a lot of juice. if you are losing power due to the psychology of not pushing as hard on your bendy bike, that is another situation.
Powertaps are notoriously inaccurate for peak power at short durations - an aliasing effect due to power being generated at the crank and sampled per hub revolution.

DanW - Cervelo have top notch engineers and produce stiff bikes anyway. I agree there may be psychological/neuromuscular feedback benefits to having a stiff bike but those who claim significant power loss in the absence of brake rub etc. probably fall in to the same category as those who think ceramic bearings are a useful upgrade.
Unfortunately, aero is the real deal, however ugly or expensive it may be, it's just high school physics.

dmulligan
Posts: 314
Joined: Wed Jul 23, 2014 6:16 pm

by dmulligan

This is something that I've thought about a lot. My old bike, from now on just the "commuter," is quite flexy. I am not a good climber at all, so when my bike was flexing noticeably during hill climbs it was demoralising. Then I bought a Cervelo R3 as a sunny day and race bike. As you know the R3 is a very stiff bike. It doesn't flex much at all during climbs, so I am much happier while I suffer up climbs which makes me just a little faster. The change is a little bit about the bike being 3.5KG lighter, stiffer, as well as psychological. I'm being that stiffness is the smallest direct contributer to the gain but it feels better and I ride faster because of it.
Psychological gains are gains.

Sent from my Oneplus One using Tapatalk

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



petromyzon
Posts: 781
Joined: Mon Apr 05, 2010 4:14 pm

by petromyzon

Yup, I don't disagree....but it's not always clear if the things that bike designers optimise for are the things that make any individual enjoy a ride. I like a "soft" carbon handlebar, for example, for damping, which is not how you would typically see a handlebar advertised.

tinozee
Posts: 764
Joined: Wed Jun 01, 2011 7:53 am

by tinozee

I have an s-works SL3 tarmac and it is so stiff. I now race on evo hi-mod and it is so comfortable! I will have to test some short sprints on the trainer, maybe 10 zwift green jersey sprints on each bike in same form (using training peaks or strava). I think I could do a ballpark comparison of sprint efficience with that setup. But over 100 mile best effort the Evo destroys the tarmac for me.

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

dmulligan wrote:Psychological gains are gains.


It's easy enough to get these without spending money.
formerly rruff...

Franklin
Posts: 400
Joined: Mon Aug 04, 2008 11:09 am

by Franklin

petromyzon wrote:10% of 1KW is 100W --> if you are losing power due to flex something should be heating up, that is a lot of juice.

The question remains: where does the energy go? Does it go largely into the drivetrain? Or is it transmitted back to the riders body who dampens it?

The former seems highly optimistic, but I'm not an engineer :D

petromyzon
Posts: 781
Joined: Mon Apr 05, 2010 4:14 pm

by petromyzon

What I'm suggesting is you never generate it in the first place! For sure there will be losses, but a difference of 10% or 100W between bikes is a little bit much.

jano
Posts: 151
Joined: Sun Oct 21, 2012 8:15 pm

by jano

Elastic deformation of the frame means that energy is being used for something else then moving your bike forward.
A large part of the energy, not 100% of it, that was used to produce an amount of deformation will be returned generally to the rider's body, this will generally result in some fatigue due to the need for the core muscles to compensate for absorbing that returned energy.
The only time energy that generates deformation is returned and results in forward movement is when it happens in the drive train.

In the end, you want some vertical flex in the frame for comfort reasons, though you will want to avoid all lateral flex which makes the bike feel wobbly and influences negatively it's stability.

LetsRide
Posts: 42
Joined: Sat Jun 06, 2015 2:03 pm

by LetsRide

Thank you everyone for joining in. More tests are welcomed, at least I should be able to see the trend.

I tried imagining my hands as a frame holding around BB. I can't think of any hand movements that will help pedal forwards. Or imagine your hands as two chainstays holding a rear wheel, try to find any hand movements that will rotate wheel forwards. That's why I consider some energy to flex the frame back and forth as a loss which affects power transfer.

dynaserve wrote: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.


Thank you for info. Ten percent seems huge to me but it really depends on the flexy bike I believe. It's usual that the accuracy of power meter may be questioned. But if others who have tools ready do tests and also have the consistent results in favor of the stiffer bikes, then we may be able to draw a quick conclusion. 300W on flats may not be easy to notice the small difference. You may try 300W on a climb if you get a chance.

dmulligan wrote:This is something that I've thought about a lot. My old bike, from now on just the "commuter," is quite flexy. I am not a good climber at all, so when my bike was flexing noticeably during hill climbs it was demoralising. Then I bought a Cervelo R3 as a sunny day and race bike. As you know the R3 is a very stiff bike. It doesn't flex much at all during climbs, so I am much happier while I suffer up climbs which makes me just a little faster. The change is a little bit about the bike being 3.5KG lighter, stiffer, as well as psychological. I'm being that stiffness is the smallest direct contributer to the gain but it feels better and I ride faster because of it.
Psychological gains are gains.


Thanks for your info. Without a power meter, it seems hard to control and maintain your effort. In case you still would like to have some fun and find out more, you may put more weights, 3.5kg, on R3 to match with your old bike. And pick a climb that you think you can maintain the same effort until finish. Ride the old bike first when you are still fresh, take some rest and then switch to R3 later. Next time switch the order, R3 first then the old bike. Do this test 4-5 times to see results which one is quicker on the climb.

User avatar
djconnel
Posts: 7917
Joined: Wed Sep 26, 2007 1:57 pm
Location: San Francisco, CA
Contact:

by djconnel

A difference of 10% at 1 kW would be ergonomic: the body is producing more power. It wouldn't be dissipation as heat in the frame.

Consider if you were to ride a bike which is too small. You'd probably have less sprint power due to the poor fit. Similarly a bike which flexed in a disturbing manner would likely reduce peak power.

I agree "where does the heat go" is the key question. The obvious place for flex power to go is back into propulsion. It's the obvious power sink. But even if 100% does, it doesn't mean it doesn't help.

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

Franklin wrote:The question remains: where does the energy go? Does it go largely into the drivetrain? Or is it transmitted back to the riders body who dampens it?

I'm pretty sure that most of the energy is not lost. The bike will flex as the crank torque increases, and spring back as it decreases. The flex shifts the force/motion curve a little. And as I mentioned earlier, some people believe the right amount of flex shifts the force/motion curve in a physiologically favorable manner.

The rider isn't going to absorb it unless they aren't providing any propulsive force.
formerly rruff...

TheKaiser
Posts: 653
Joined: Thu Sep 05, 2013 3:29 pm

by TheKaiser

WMW wrote:
Franklin wrote:The question remains: where does the energy go? Does it go largely into the drivetrain? Or is it transmitted back to the riders body who dampens it?

I'm pretty sure that most of the energy is not lost. The bike will flex as the crank torque increases, and spring back as it decreases. The flex shifts the force/motion curve a little. And as I mentioned earlier, some people believe the right amount of flex shifts the force/motion curve in a physiologically favorable manner.

The rider isn't going to absorb it unless they aren't providing any propulsive force.


All, I am glad to see this topic being discussed seriously, as it is one that I have pondered many a time. Hopefully it leads to some satisfying data driven conclusions so I can stop mentally grappling with the topic.

As WMW says, the energy is likely not lost, however as several have mentioned, simply having near 100% energy return does not mean that the return is in such a way that drives forward propulsion.

There are also a bunch of different, distinct, components to this flex. In regard to lateral BB flex, it doesn't seem that any of the lateral flex of the frame will contribute to the pedal stroke, based on the simple consideration of the mechanics involved. We have all seen those spin scan charts that demonstrate that even people who think they are spinning smooth circles are really mashing the pedal almost exclusively downward, even down to the 6oclock crank position. Thus, the lateral deflection will be peaking during the dead point, and the frame will then return to a more or less neutral position at a point when little power is being transmitted through the drivetrain, rather than springing back against your productive pedal force, which might possibly work in the additive way some people surmise.

On the other hand, in my mind, wind-up in the rear wheel, crank, bb spindle, torsional flex in the crank arm, and compressive shortening of the chainstay all have the potential to return the power toward forward propulsion. In these cases, the flex would have the effect of "smoothing" power delivery, so that you would get a lower peak torque transmitted to the rear tire but (other than the minimal internal frictional losses of the materials involved) the power could be returned on the back side of the torque curve of each pedal stroke. This assumes a relatively smooth overall application of power though, as the "unwinding" of the spring elements would need to occur while power was still being transmitted through the drive-train to have the additive effect.

The physiological effects are a big ????, although I have seen people who claim that the smoother torque curve of oval rings leads to lower fatigue levels, so if the elements I mentioned in catagory 2 worked similarly then I could imagine them perhaps being helpful.

The oval ring issue though is a good comparison to this flex question though, as all of these variables may exceed the resolution of currently available power meters. Supposedly, oval rings can generate a 5-6% higher power reading, even with the same "actual" power output, simply because of the sample rates and assumptions made about the consistency of pedal velocity during each revolution. Frame flex could have similar effects, and one meter may have more trouble with this than another, so seeing a greater delta between a pedal system and a powertap on 2 different flex level frames does not necessarily indicate a true difference in power loss from frame flex.

Having said all of that, I think that anyone with the capability should definitely run some tests to get the ball rolling. In my mind, the ultimate setup would involve 3 meters, at pedal, crank, and hub, and trials should be run on rollers equipped with a resistance unit to allow testing a wide variety of outputs. Rollers eliminate the issue of fixing the rear axle to a trainer which would alter the frame flex picture. Additional points for anyone who tests the rollers on a variety of inclines.

Anyway, sorry for being long winded, but as I said, I have been pondering this for a while!

tinozee
Posts: 764
Joined: Wed Jun 01, 2011 7:53 am

by tinozee

That's a good point about the rollers vs trainer. Another thing is that in sprinting on a trainer, even a durable one like a kinetic, you can get that thing bouncing and sliding around pretty easily over 1300 or so watts it's hard to control that. Not sure how you'd max sprint on rollers, never done it, just tempo etc. Maybe those Tru Trainer rollers.

TheKaiser
Posts: 653
Joined: Thu Sep 05, 2013 3:29 pm

by TheKaiser

tinozee wrote:Not sure how you'd max sprint on rollers, never done it, just tempo etc. Maybe those Tru Trainer rollers.


Yeah, full-on sprint testing on rollers could be problematic. I had never seen those Tru Trainer models before, the flywheel design sounds interesting. I had thought that the Elite Real E Motion would be ideal because, besides magnetic resistance, they offer additional security to prevent accidentally disengaging from the rollers during violent movements, plus they feel more natural when standing than standard designs. I don't however know the max resistance wattage that they provide.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



ultyguy
Posts: 2332
Joined: Fri Aug 06, 2010 1:35 pm
Location: Geneva

by ultyguy

Just because the bottom bracket doesn't move very much when pressure is applied doesn't at all mean that all of the energy is ending up being translated to where the rubber meets the road which is the ultimate goal of increasing speed. On dynaserve's point of more or less watts going through, my real question would be what's the speed difference because THAT is the ultimate measure.

I'm pretty sure that this flex and how it's returned to the rider and the road is part of the black magic of a great frame builder. The natural frequency of the material involved also has an effect on this as different materials will return the energy at different rates in respect of the molecular framework.

Post Reply