carbon seatstays + aluminum frame - Page 5

At this rate, your butt will be detecting neutrinos next... Seriously. You claim you can detect microns of frame flex and grams of frame weight. I've offered some double-blind tests of your theories and get nothing but silence in return. Please stop making ridiculous claims that contradict a wide body of science and engineering principles. Just stop. Or put your butt to the test and prove you're right.

John Swanson
www.bikephysics.com

I think you read me wrong. I don't think that the rear triangle will deflect by millimeters. I think it will be less than that, maybe even a lot less than that. Bill Holland gave some numbers, but I can't remember them.

To be honest, I don't think a rider could reliably say that they were feeling rear triangle deflection. There is way too much other stuff going on. Hell, the rear triangle is likely much stiffer in plane than wheels are in plane, and wheels have frontal stiffness such that they only deflect parts of a millimeter in response to in plane impulses.

Also, it has to be remembered that the human body is wildly inaccurate as a detector. This has been shown over and over again.

I am not suggesting that people don't buy or do things because of how they "feel," but I am saying that "feel" likely bears no relation to actual dynamics in a bike. Moreover, I'd say the human sensor cannot distinguish one displacement from another when multiple deflections are buried in the signal a rider's ass is receiving.

Mavic's studies showed that riders could not accurately judge whether they were riding laterally stiff wheels or laterally noodly wheels. Again, lateral wheel deflection is much greater than rear triangle in plane deflection, so if riders can't tell whether their wheels are stiff laterally or not, how will they pick up rear triangle deflection. Hell, I'll be the Mavic studies data is generous in favor of riders thinking they know what's goin' on, given that the riders had a 50/50 chance of guessing right if their choice was just a guess.

It's funny you mention that. Yesterday, as I was sitting at my desk in my office, I thought I felt an electron drop to a lower energy state in my chair.

It'd be nice to see all these "science don't work" folks put their words to the test. The science is already there, as you've said. I guess if we weren't so blinded by that whole Scientific Method thingee and all that stuff about needing to be objective and stuff....heck, I'll bet we could really advance things. Heck, we could be like them Intelligent Design folks and just make stuff up and be certain that it was right because we felt it was so.

All this talk reminds me of the mechanical engineer in Youngstown, Ohio that got written up in the Youngstown paper. See, he had designed a perpetual energy machine, and had his theory to back it up. Of course, his machine was only 10% efficient, but heck, he was still developin', ya know. He wasn't gonna let that Second Law of Thermodynamics keep him from making real scientific progress.

Now that everyone here has sussed out science and seen it for the sham that is, maybe they can go over to the SERPO page and join up for the Zeta Reticuli Exchange Program.

Sounds like we need to convince velonews or some German mag to set up a decent blind test on frames with the same components, but different design concepts for the rear triangle.

We are all argueing to different points of engineering judgement, but nobody will win without proof.

For mine I just can't see how an early 90s Klien with straight large diameter seat stays could have the same level of rear end complaince as a modern bike designed for verticle flex and have it not make a difference.

__________________
Precision Sports

At this rate, your butt will be detecting neutrinos next... Seriously.

Again, 'what' evidence...seriously.

To be honest, I don't think a rider could reliably say that they were feeling rear triangle deflection. There is way too much other stuff going on.

No matter what 'other stuff' is going on, the rear triangle is an integral part of the equation and one that is a component of ride that makes up the end characteristic we describe as stiff or soft.

If this were not true we would all be riding gas pipe and saying, "Man, my new Sewerpipe Shredder rides great over dem bumps!". Sewerpipe? Hell, let's just go with powder metal tech in a ceramic...the marketing boyz will be all over it! Nah, there's more profit margin in sewerpipe. We can still spin it right over their heads with a hip flash website. Them dumb cyclists are too stoopid to know the difference.

Hell, the rear triangle is likely much stiffer in plane than wheels are in plane, and wheels have frontal stiffness such that they only deflect parts of a millimeter in response to in plane impulses.

A main triangle, with it's larger shaped tubes and much great contact areas at the junctions/nodes also deflects.

It's all marketing, homes. Them thar pencil stays on that Cervelo are as stiff and un-moving as a 5/8" steel single-taper socketed 2" below the top tube. Why any scientist can see that! A Schwinn Varsity should ride just as well an R3! The extra mass, according to science, will even help damp out those second order harmonics!

Once again science proves there's no difference in bicycles and that cyclists do not know what is under them. Thank god for a rational universe...no wait...thank those godz among us mere mortals, the scientists!

Here's something I've learned this week: psychometric models are based on the fact that there is only a 50% chance of detecting a signal if it (the signal difference from baseline) is 2-3 times greater than the noise. The probability distribution is gaussian. The psychometric model has been proven valid for touch, sight, and sound. I can provide all kinds of scholarly references on that one.

What does that mean for riding a bike? Let's say you have ~40 kg of force between your butt and saddle. Now let's suppose that going over a normal road surface causes up and down vibrations (total system flex of tires, frame,seat post, saddle, shorts, etc) of about 1-2 mm or ~0.1 g.

That means you would need to have a signal that was anywhere from 2-6 mm or 0.2-0.3 g (12 kg) before you could distinguish it from regular road buzz. You could also separate out each element of a bike to determine whether or not you can feel it. For instance, the bike tires flex about an order of magnitude more than anything else. You can feel differences in tire flex, absolutely. Seat post? Not likely since the tires dominate the overall noise. Frame? Not a hope in hell. You can not distinguish frame flex.

This is fact from proven psychometric models used by everything from psychologists to LCD monitor manufacturers (setting limits on visible defects).

Sources abound. I can point people to really great primers on the subject if they're interested. I'm also willing to be proven wrong (i.e., correct deficiencies in the model due to frequency spectrum separation for example) with a couple of easy to perform experiments.

John Swanson
www.bikephysics.com

I've thought of a better example to illustrate the pyschometric model: Imagine a flat slab of granite. The surface is kind of smoothe, but has a texture you can feel. If you measured it, the average roughness is let's say 0.1 mm. That is, there's peaks and valleys across the surface and if you took the average size of peak and valley it would be 0.1 mm. You can feel this roughness.

Now lets say I put a scratch into this surface. How big would the scratch have to be before you could feel it? The model says that in order for you to feel it 50% of the time, the scratch would have to be 0.2-0.3 mm deep. If it was only, say 0.15 mm deep you might only have a 10% chance of feeling it. And so on.

Interestingly, if the scratch was only 0.1 mm deep, you would still have a 0.0001% chance of "feeling" the scratch. This is called a false alarm and is due to neruological noise. Actual noise created by your brain firing neurons. In fact, to reduce the the level of false alarms to this level, the brain needs to operate on the basis that signals that are two to three times the noise are recognized. Any more sensitive than that and you'd have way too many false alarms. You'd be reacting to too many things that weren't actually there.

John Swanson
www.bikephysics.com

I've got a very simple test. It involves no instrumentation. No deflection calculations. No recorder charts. And who knows? It may even be fun!

Can you ride a 57-58cm frame? If so, i can offer the same wheels (32h-3x), sew-ups set at 105 psi on all bikes, same saddle (Selle San Marco Concor Profile) and seat pin (Campy Record alloy) over several rear triangles in 3 materials (steel, carbon and Ti).

Orders of magnitude be damned and to hell with spurious noise in the 'system', I'll guarantee you that even your heinnie will be agreeing there's more to it than marketers sitting around a table dreaming up ways to get our cash.

I'll agree that many folks can't tell a Ford from a Chevy despite looking at the trunklid badges side-by-side, but for god's sake...if an expirienced cyclist can't tell the compliance difference between a few similarly setup bicycles over a few miles of bad road it's high time to hang up the cleats...because you're already dead.

Not that it supports or defeats your main thesis, but a seatpost can only flex away that component of a bump or jolt which is perpendicular to its axis unless it compresses.

That said, it is my belief that if a frame rides soft, it has more to do with its geometry than its material. Also, the rails of one's seat probably flex a great deal.