zacster

We've all been debating endlessly about frame type, whether carbon, aluminum, steel, titanium, stiffness, configurations, and how efficient they are. So I thought of an experiment that could be done to answer some of these questions for once and for all, with equipment that has become commonly available. We have power meter hubs, such as the Powertap, and we have power meter pedals, Assioma, Vector, Powertap. Now pedals will record the power that is applied by the legs directly, with whatever efficiency there is in that interface, but power meter hubs record the power except with the frame in between. If we establish a baseline where pedal power readings = wheel power readings, we can then swap frames and measure the differences in the readings. If the wheel measures lower than the pedals, then the transmission through the frame is less efficient, if the wheel measures more, than the frame is more efficient.

Now I'm not suggesting a single reading, but over time and terrain and conditions. We would be using the same wheel and pedals on each to hold those constant. The change would be the frame and maybe the component but even the wheels would be the same. The best part of this is that the result is unit-less. There is nothing else being measured or held constant.

What do you all think?

Anybody in the Brooklyn area with a Powertap hub want to give this a try? I have Assioma pedals. It may have to wait until spring though unless we get some warm weather.

asgelle

Are you sure the frame is the only thing between the pedals and the hub?

rms13

To have any kind of accuracy you'd need to start with the same rider on the same course in the same conditions. You'd need a carbon, aluminum, ti and steel bike that are all the same size, geometry and aero profile, same wheels and tires and components. Then you'd need to do a number of runs on each bike and average everything

Dean V

I think the complete lack of data on this from the industry speaks volumes in itself.
You may of simplified the test a bit but it wouldn't be that difficult.
If any company wanted to provide data that shows their frame X transfers power with greater "efficiency" than frame Y I am sure we would of seen it by now.
But all we get is subjective comments and hyperbole no doubt because any real data would't back up the sales pitch.

asgelle

Or because getting useful data is a lot more difficult than you suppose. What the OP ignores is drivetrain losses, and it's likely that those losses so overwhelm power lost to frame flex that no one's been able to reliably tease out the signal of frame losses from the noise of power lost to the drivetrain.

Dean V

It is not difficult to duplicate drive trains, gears used, chain line etc.
If a company like Specialized can build a wind tunnel to test bikes, do you think this would be beyond them?

asgelle

Easy to say, much harder to do. As a gedanken experiment, you'll obviously have to transfer the entire drivetrain between bikes or the difference in drivetrain losses would overwhelm frame flex. Now, how closely do you think you can duplicate chain line between two frames and how will the power lost to chain bending depend on chain line? Then consider all other components.

rms13

Companies are in the business of selling expensive carbon frames that are cheap for then to produce. I'd imagine the results of these tests would that gains over cheaper bikes are negligible to the average rider. Most of our rational brains know that already but companies selling the expensive bikes aren't going to provide you with that data

Campag4life

To me, the frame stiffness versus speed debate is almost religion. Some believe it and some don't. I tend to be in the camp that stiffer is faster. Each pedal stroke being an entity comprised of a 'power zone' when the foot has leverage on each side of the crank. When a BB and bike flexes, the circularity of the pedal stroke is lost due to frame deformation. This displacement and eccentricity of the pedal stroke is like going on a wild goose chase. A direct circle is faster with 0 deformation.

Some believe stiffness is all about marketing. I don't but others clearly believe it is all marketing. I sure don't. For example, back in the day of the Vitus, the industry could have concluded that the whippiest frame is the fastest. Why not embrace this paradigm? I mean, why would stiffer be faster if whippy is fastest. And yet, the industry has continually morphed toward stiffer. Decouple lateral stiffness with vertical stiffness to make the bike ridable but make it a vault in torsion.

I would say, most that ride the modern stiffest bikes believe them to be faster. I do. I believe the modern bikes to be the most efficient.

All the top companies...all of them which employ hundreds of engineers and many PhD's...all believe stiffer is faster. They strive toward absolutely the stiffest bike laterally they can create. I don't think this is conjecture, or myth, or marketing, I believe this is what all the top bike makers believe.

The lead designer that created the new Cervelo R3, a bike I just bought...this particular frame and the R5, this bike is purported to be the 'stiffest bike Cervelo has ever produced'. Quite a statement. The lead designer is also a successful bike racer and Cervelo has its founding and history steeped in design and engineering. This bike has log for a downtube and a BB the size of a softball. How long has Cervelo been making race bikes? And what do they believe about stiffness mattering based upon all their testing and bikes that have won races on the world stage? Stiffer is faster.
I believe it as well. I also believe it is a complex analysis and hard to prove but the entire industry believes it.

Nessism

When standing out of the saddle and sprinting on a bike that's super stiff the rear wheel almost hops during peak pedal loads and the noise from the tire sounds like rubber is being scrubbed off. Is this a faster bike compared to one that has a little give which smooths out power delivery? Energy is conserved anyway; a little sideways deflection in the bottom bracket during peak loads will spring back and return that energy off peak.

Regarding a frame test, different frames can be made in a wide range of stiffness regardless of the material. Simply testing material alone won't tell you much if anything.

Campag4life

So what you saying is, you believe the entire industry is wrong. Sure it isn't you?
One would think there would be one bike company out there that believe softer is faster or at least more comfortable and comfort and fatigue have to account for something right? . Oh we do have the boutique guys who build steel and Ti bikes that never went to college. But no top companies. Curious isn't it?

Aubergine 

Nessism

You're stretching with that "entire industry" business. Yes, there are some companies in the industry that make marketing claims that stiffness = fast, but that in itself isn't evidence that all companies believe this nor is it evidence that the basic premise is even true.

Campag4life

The evidence is in the bikes. All top tier races bikes from all the leading bike mfr's are brick $h!thouse stiff. That's a fact. They could make them soft. They could make them medium. They make them brick $h!thouse stiff. After all the years Cervelo has been making world class race bikes, they made the new R-series 13% stiffer in downtube and 8% stiffer in BB. After all these years, they believe today, stiffer is faster.

Campag4life

Your opinion matters a lot compared to the industry...lol.

asgelle

The debate is whether manufacturers are making bikes stiffer because that makes them faster or because it is a marketing device to increase sales. Nothing you've written provides evidence one way or the other.

Campag4life

I can't definitively win the argument either way. Legions have tried here and elsewhere. I am stating immutable fact. Bike companies are making frames stiffer and not softer.

To believe that marketing is the pure impetus over about a ten year run of making frames stiffer to me at least is a very ignorant view. For one thing, people want a superior bike. In fact, press fit and monster BB's is predicated in fact as well. At least the industry believes it. Why else tolerate the reliability hit if lighter and stiffer wasn't faster.

Stiffness generally portends lighter too...or can...permissible flex and yield point. And stiffness I believe many feel is directionally correct to control and handling. Most strong riders prefer a stiff handlebar out of the saddle.

So, I can only believe and even as a mechanical engineer, I believe stiffer is faster. I agree with the industry and in fact defer to the collective consciousness on the subject.
There isn't a single bike manufacturer that makes a $10K race bike deliberately softer. Not one. If there was a tangible advantage to make a bike softer, even for ride comfort or removing more frame material to make the bike lighter and/or weaker, this metric would be 'used as marketing' to set itself apart from the competition.

My opinion and of course you are welcome to state yours if you have one.

Nessism

I disagree that ultimate BB stiffness is the priority of the "bike industry", even for race bikes. If that were the case down tubes would flare to huge proportions since stiffness increases exponentially with diameter. Rather, frames are designed to be stiff through the BB to a point and a balance with other attributes come into play such as rider comfort, aerodynamics, and handling. In the motorcycle world some time back, when perimeter frames were still evolving, the lateral stiffness of the frame on some bikes was overbuilt to a point where handling suffered; wheel chatter among other issues cropped up. The designers reversed course and engineered back in some lateral flex (to a point) to improve this situation. My point is that one can not simply look at these situations with a singular point of view because the real world isn't that simple.

Hermes

OP, which head unit are you going to use? 1 second recording time, IMHO, will not be fast enough since peak pedal force is produced over a relatively small portion of the pedal stroke. IMO, only the SRM scientific power meter with a PC7 head unit may be up to the task to measure crank power. There is no PM hub unit fast enough. This does not invalidate your premise per se for the experiment. It means that you will not capture all time segments.

Is the issue of lost power about lost power or lost speed due to frame dynamics? Every time one changes direction on the bike when the goal is to ride in a straight line, one scrubs speed due to going a farther distance and increasing frictional forces in the tires. Did you ever follow a rider who snaked along a straight line? Is the movement due to the rider or frame flex? So no movement in the bike along a straight line is the fastest for any given power.

Hence frames that spring may cause riders to have to compensate to stay on line.

At the track, riding in a straight line and following the radius of a turn without deviation is very important. In the turn, if one goes up track one is going uphill and for a longer distance.

Most any modern bike does well on the straight away at the track. Now let's consider a rider out of the saddle generating greater than 1000 watts on a 250 meter track on 45 degree banking. A frame that is whippy or not stiff enough will cause the rider to be all over the pole lane. Even if zero power is lost between the pedal and the tires, it is hard to control the bike. Riders often back off on power to control the bike or are just slower through the turn due to changes in direction.

Anyway, congrats on an interesting premise for an experiment and good luck. One never knows what the result will be.

Spoonrobot 

To a word, none of this is true. This post I have quoted from is merely an appeal to authority with no quantitative argument to back it up.

No single bicycle company has hundreds of engineers on their payroll. The entire industry of dozens of companies may employ around a thousand but I suspect it is substantially less.

Stiffer is not faster. This is plainly obvious by comparing different types of bikes, to make no mention of the fact that there is no comprehensive independent frame stiffness testing information available. All whitepapers and data is either released directly from the manufacturer or carefully curated by third parties (Tour is the most obvious) to only be applicable to a small subset of frames. It would be impossible for a consumer to compare frame stiffness of a Trek frame from 2016 to a Specialized frame from 2018. There is no real data and thus there is no ability to actually confirm the statement. It's nonsense.

Now, back to my first point. Track bikes are significantly stiffer than road bikes - if stiffer is faster this would never be the case. Time trial bikes would be the stiffest bikes ever created and that's also not true. It's a small point but is positive affirmation that stiffness is only one of the components to a fast frame and that stiffer to infinity does not create a faster frame for most riders. There is a point of diminishing returns and it's well known by manufacturers and designers but the end user has no idea and merely regurgitates internalized marketing (see below) to reinforce their world view. Pro Tour riders routinely complain about frames being too stiff as do cross racers and mountain bike racers.

Where did this data come from? Can you provide the stiffness measurements Cervelo has used over time? Their testing methodology?

This is correct.

Campag4life

Pretty much how the debate goes. You speak with such conviction. What a joke.

To me to deny that frame stiffness has not increased among the top race bikes laterally including BB which is the whole provocation between wide shell 30mm dia cranks tying into broader chainstays is to put one's head in the sand. Cervelo would certainly not divulge its methodology. If you go out on the web and read reviews, they quote the increase in stiffness nos. Seen it both in video reviews from GCN and Bikeradar. If you compare previous R-series, the frames are night and day certainly visually.

I think you have a real windfall opportunity. You can create the first medium flex high end race bike and make millions. Think of the marketing opportunity. You can call it a kinder and gentler race bike that your body will thank after a long hard ride.
You are on your way.

Campag4life

I too have lived in the motorcycle world. Really? You are making the extrapolation? A 170hp superbike to a .5 hp human powered road bike?
Do you think the priority of energy transfer and economies of scale are the same?

Nessism

I'm talking about lateral flex, not "economies of scale". And the flex was to augment handling, not power delivery of the engine. Some lateral flex was good for handling. I don't know why a bicycle would be any different.

Campag4life

In bold, I know you don't. Its pretty clear. Because the priority is more power delivery with a bicycle compared to handling due to reduced frame loading based upon much smaller tire contact patch compared to motorcycle that pulls much higher g's laterally. No comparison. With a bicycle, the priority is stiffness for control and power delivery. Ride any top rung race bike. Any made by the big 5. They all are super stiff laterally. Not one has moderate flex. Not one. All manufacturers are aligned in this priority...well until of course Spoonrobot, rewrites physics and comes to market with his softer flex race bike to capture the market. We wait with great anticipation.

Dean V

No question that a stiff frame is preferred by many, especially at the high performance end of the spectrum.
But that doesn't mean it is actually faster. It doesn't mean you will climb a mountain or cover a TT in less time.
It means they prefer the feel and handling of a laterally stiff bike and even better now that it is achieved without a bone jarring ride.