I have a bad wing, so Ive been on my rollers with a fork stand. It's Fredilicious, but better than no riding. Anyway....

With the fork held solid the frame flexes under you as you pedal, as we all know. The deflection appears to be primarily in the front triangle. As the frame flexes, it tilts the rear wheel slightly side to side. As this happens, the rear steers about an inch or 2 side to side. So when riding normally down the road this must be happening to some extent, which would contribute to tire scrub.

I've always said that frame stiffness doesn't affect efficiency. I'll still say that the frame itself is not 'absorbing' (I don't know what that means, honestly) power. I will say that this deflection may very well effect tire scrub which would affect overall efficiency.

Any thoughts?

I'm almost sure I agree...
Actually I read this first in a now-dead magazine called Bicycle Guide maybe as long as 25 years ago. The author's premise was that the frame acted as a spring, flexing as it absorbed power, but then returning it as it flexed back. There would be a tiny loss as energy was transformed to heat (or however that works,,,been awhile since I took that class), but it was insignificant, especially compared to the increased comfort and possibly improved handling (from a small suspension effect, keeping the tires on the ground) due to the flex. Sort of made sense to me then, and it sort of makes sense to me now.

everybody knows carbon bikes are faster.

on the road your fork isnt clamped down stationary so the effect is not the same.

Well,

there would also be loss due to more friction of the tire against the road. The less tire against the road, the more efficient you will be.

As someone who weights 152 points, I just cannot flex a frame much to notice on any of the frames I have ridden. I think a lot of these individuals who discuss how certain bikes take off like rockets are ignoring the different components (wheel set etc..) between their lower-end bikes and these rockets they ride.

Maybe there is the feeling of immediate transfer from the pedals to the wheels, and my CAAD frame does feel different from my steel Fuji. But how much of this is weight, wheels and the crank-set being used? Unfortunately there are a lot of anecdotal descriptions of stiff vs. noodly frames, but I still have not seen a lot of evidence statistically as to how much flex really impacts riding efficiency.

when the rear axle is clamped down into a trainer, the BB gives a noticeable amount of flex.

Try a smoother pedal stoke. See if you can make the tire not wiggle. Then you can see if this affects your overall speed, since you won't be wasting your watt production laterally (in the wiggle/scrub).

The pro's know that but the Fred's are another story.

True. I thought about that as well. If the front and rear essentially move the same amount at the same time, then there is effectively no net steering effect. This would argue that frame stiffness alone is not important, but frame stiffness distribution is. A stiff rear must have a stiff front to work with it. That makes sense to me.

From what I know, most road frames are similar enough that stiffness doesn't have much of an effect on efficiency. The main exception is at the BB, where more flex could result in less power making it to the pedals. (There might be reasons other than frame flex that is shifting the tires, by the way.)

Not sure what you mean by "tire scrub," unless you mean "rolling resistance." I don't think increasing lateral stiffness will reduce rolling resistance; increasing vertical compliance might drop it a tad, but probably not by an appreciable amount in most circumstances.

And keep in mind that for many reasons, a stiff frame may merely feel like it's faster. A stiff frame will have a more lively feel; also, many geometry aspects that make a frame stiff will also make it handle more responsively, which again will feel faster (without actually being faster).

Unfortunately, it seems to me the only way you could really determine the performance effect of frame stiffness is to take two bikes with identical geometries, identical tires, identical rider position, identical power meters, and frames with different stiffness (but the same weight), and compare the efficiency of the two.

Until someone does that, I wouldn't worry too much about it unless you're in a highly competitive situation. Go with how the bike feels, rather than a parameter whose effect is difficult to quantify in the first place.

It's not a point of smooth or not. You're applying a moment in one direction and then the other. It's gonna deflect. There's no way around it.

The frame flex producing heat thing has never worked for me logically. Someone did the calcs on this in one of these threads. It was tiny, tiny....hundredths of a watt.

However, putting a slip angle in a tire makes a lot more sense. It would certainly explain wearing a flat spot on the rear tire after X number of miles. The front spends the same amount of time rolling in a straight line, albeit with less load, and you don't get that wear pattern. I'm not trying to come up with some grand unified bike theory, but it is food for thought.

Motorcycle and bike tires are made to steer when cambered (angled from front view) one way or the other. This frame deflection due to pedaling tilts them slightly back and forth which creates a slight amount of steering back and forth. I'm calling that tire scrub.

For the record, this observation was on my old steel bike as the newer C'dale is in pieces (closely linked to bad wing and me being on rollers in the summer). I've never felt that there was a hill of beans difference between the 2 bikes in terms of performance. I race the C'dale because it's expendable.

But I have always felt that the flat spot that the rear tire wears is more due to my increased weight back there (pun intended) rather than on the front. When I ride my rollers and I don't have the fork mount, the entire bike moves - front and rear - as described.

What you're talking about, then, is the tire sliding side to side on the trainer's roller as you pedal.

On the road, theoretically, the rear tire would not only be propelling the bike forward, but it's also getting pushed side to side, which is the scrubbing you're describing. Then, the idea is that the sideways scrubbing contributes to the rear tire wearing down faster.

Am I reading you right so far?

The complication here is that the stationary trainer creates a pivot point at the rear axle that may not exist in regular riding. I'd suspect that, on the road, the whole bike pivots about the bottom of the wheels, at least as long as it's not getting flung around by heavy pedaling.

I also don't think that frame stiffness would affect sideways tire scrub enough to matter, either. Any frame should be stiff enough already.

We haven't started talking about wheel stiffness yet, either.

The front wheel tracks when you turn, the back wheel does not so as you make a turn the portion of the rear tire contacting the road has to pivot on the contact patch. That is more likely the cause of the flat spot and why you see it on the rear and not the front. Also more of your weight is typically on your rear wheel so that tire is going to wear more. Anyway that’s two bit theory on it.

When you're riding rollers, the fork is held tight and you get flex in the frame. But then you compare this to "riding normally down the road" where the fork is not held tight. You say the flex must be happening "to some extent" and I agree, but no where near the extent if the fork was held rigidly (which it isn't). So there must be some flex in the fork, and some in the frame. I guess there would also be some flex in the handlebars, and even some in the wheels. I dont think you can draw too many conclusions about flex during a normal ride on the road, compared to a rigidly held fork on a roller.

just because it springs back does not mean the energy is ultimately applied in any useful manner.

Exactly.

What is the point of the frame acting as a perfect spring that returns the energy ...back up your leg. There does not seem to be any benefit to this.

Stiff frame that deflects minimally during pedaling is faster. Energy spent on flexing the frame is in practice wasted energy as it does not result in forward propulsion.

V.

During a turn, though, the tires are tilted, and won't wear a flat spot -- they should wear evenly across the tread instead. You could probably say that the front is creating more turning force because it's changing the heading of the bike, so it should scrub a little more.

The rear does take more weight, but it's also handling acceleration and braking duties (and, because it's so easy to lock up, it sees more skidding than the front ever does). I'd think that that's what's creating the flattened tread.

I initiate a turn with my handlebars, but the lion's share of the actual turning is done on the rear. This is the same on a bike or motorcycle. It's damn rare for a motorcycle racer to consistently slide the front tire (some of the best do it). They all slide the rear. Hell, some of them finish the last 1/2 of the corner with the front in the air.

I can hit a pretty good bump mid-corner, skip the front and not dramatically change my line. The whole bike leans. The front and rear should both wear the sides of the tire similarly if it the wear was an issue of cornering. The weight difference might effect the rate of wear front to rear, but by itself it shouldn't effect the wear pattern.

My observations are saying to me that the wear is coming from straight line riding, which should be where the tires wear the least (although they spend 95+% of their time rolling straight). Keep in mind, the side to side motion I noticed was small. Maybe 1" either way. It was enough that over many miles it would cause wear issues, though.

I also ride with more pressure in the rear, which causes flatter wear.

We've kinda covered this one. Ultimately, most of us agree on this point. There is the issue of the magnitude of energy involved on the frame side, though. It's tiny and on any frame I've ever heard of or seen, not a legitimate player in terms of wasted energy. It might be a player in bike handling or overall feel, but in terms of getting power from the cranks to the rear wheel, it's essentially nothing.

The new thing I'm coming up with is how frame deflection can effect rear wheel tracking. I'm thinking that this could indeed be meaningful. I don't know if I'm right, it's just an observation I made last night. It made me rethink my previous stance on the subject, so I decided to kick it around the forum.

elaborate?

Yes, this does seem to be a contentious issue for some reason.

The deflection of a very stiff frame achievable by a fit rider is about 1mm to either side, so a full stroke makes it 2mm.

Normal carbon fiber frames deflect 2mm per side, "comfort" frames do about 3mm per side.

So by abusing the pharmaceutical industry "relative benefit" concept that makes them billions, stiff carbon fiber frame is 100% more efficient than a normal carbon fiber frame and 300% more efficient than a budget carbon fiber frame

...in seriousness, I'll try to look for a suitable elastic modulus versus energy formula and get some actual power numbers attached to this.

Cheers,

V.