Translation of this article (earlier version):
To stay young requires unceasing cultivation
of the ability to unlearn old falsehoods.
--Robert A. Heinlein
There is an amazing amount of folkloric "conventional wisdom" about bicycle frames and materials that is widely disseminated, but has no basis in fact.
The reality is that you can make a good bike frame out of any of these metals, with any desired riding qualities, by selecting appropriate tubing diameters, wall thicknesses and frame geometry.
Imagine you clamp one end of a metal bar in a vise, and you hang a weight on the free end, causing the bar to flex temporarily. When you remove the weight, the bar snaps back to its original shape.
Different materials will flex different amounts for the same amount of force applied. This is stiffness.
Now imagine hanging a heavier weight on the bar, so heavy that it becomes permanently deformed. When you remove this weight, the bar does not snap back all the way to its original shape, but remains bent to some extent. When the metal changes shape permanently, it is said to "yield."
Different materials can withstand different amounts of force before yielding. This property is strength.
Stiffness is determined by a property of the material called "elastic modulus" Elastic modulus is essentially independent of the quality or alloying elements in a given metal. All kinds of steel, for instance have basically the same elastic modulus.
Strength is determined by a property of the material called "yield strength."
Yield strength is very much affected by the quality, heat treatment and alloying elements used in a particular brand/model of tubing.
Like stiffness, the density of a given metal is not significantly affected by the addition of different alloying elements. Although your bike may have a sticker saying "Lite Steel (TM)," in fact, all steel is equally heavy.
Here are some properties of the three common frame metals:
[Sheldon was a strong proponent of the Standard International system of measurement over the English system, but inconsistent in the table below. It doesn't matter though, as the table only makes comparisons -- John Allen]
|Material||Modulus (psi)||Yield Point||Density (lb/ft3)|
|Aluminum||10 to 11 x 106||11 to 59 x 103psi (4-22 annealed.)||168.5|
|Steel||30 x 106||46 to 162 x 103 psi||490|
|Titanium||15 to 16.5 x 106||40 to 120 x 103 psi||280|
Note that the modulus (stiffness) and density (weight) are pretty much independent of the quality, heat treatment, or alloying agents of the materials. For instance, all steels, from the "gas-pipe" used in department-store bikes to the exotic alloys used in multi-thousand dollar bikes have a modulus of 30, and a specific gravity of 490.
Anybody that tells you that a particular brand of steel (or aluminum, or titanium) is "lighter" or "stiffer" than another brand or model is blowing smoke.
There are, however, real differences in yield strength among different qualities of tubing.
This modulus value shows that if you were to build identical frames from the 3 materials, using the same tubing diameters and wall thicknesses, the aluminum frame would be only 1/3 as stiff as a steel one, and the titanium frame only half as stiff.
The yield values show that the aluminum frame would be very much weaker, in the sense of being more easily damaged than either the steel or titanium frames.
The density values show that the aluminum frame would only weigh 1/3 what the steel frame weighs, while the titanium frame would be roughly half the weight of the steel one.
These generalities, however, are basically meaningless, because you wouldn't build frames out of the three different metals to the same tubing dimensions!
Real bicycles take the nature of the material into account in selecting the diameter and wall thickness of each piece of tubing that goes to make up the frame. Stiffness is mainly related to the tubing diameter. Strength is mainly related to the wall thickness, though diameter also enters into it. Weight is affected both by diameter and wall thickness.
A frame manufacturer can make trade-offs by selecting different tube diameters/wall thicknesses, allowing a frame to be made stiffer, or stronger, or lighter.
Such a frame would likely have a whippy feel due to the reduced stiffness, especially in loaded touring applications. To compensate, builders of titanium frames use somewhat larger diameter tubes to bring the stiffness more into line with what riders like. This tends to increase the weight a bit, but by making the walls of the larger tubes a bit thinner, they can compensate to some extent, and come up with a frame that is still lighter than a normal steel frame.
Why don't manufacturers do this? Two reasons.
Lateral stiffness can be an issue particularly when there's a touring load on the rear rack. A frame that is too flexy in the top tube, head tube and seatstays will feel "whippy" and may be prone to dangerous oscillations at high speeds. Some of this flex is in the luggage rack itself, but there can be enough flex in the frame to aggravate this condition.
In geometry, there's nothing as strong, or stiff, as a triangle. Diamond-frame bikes consist basically of two triangles. The elegance and simplicity of this design is very hard to improve upon. Billions of diamond-frame bikes have been made from tubing for over a century, and during that time, hundreds of thousands of very smart people have spent billions of hours riding along and thinking about ways to fine-tune the performance of their bikes. The tubular diamond frame has been fine-tuned by an evolutionary process to the point where it is very close to perfection, given the basic design and materials. I often commute on a Mead Ranger frame built in 1916. It's a tad heavier than a more modern frame, but its general riding qualities are as nice as any bike I own.
"Step-through" or "ladies" frames, where the top tube -- if there is one -- meets the seat tube a few inches above the bottom bracket, have far less torsional stiffness than diamond frames, all other things being equal. However, robust construction can compensate. This is especially practical with carbon fiber, keeping the weight reasonable. Damon Rinard's design for Betty's Bike, described in more detail elsewhere on this site, succeeded in being adequately stiff.
"Mixte" frames are triangulated, but the top tube and the seatstays lie in a single line. There may be double top tubes continuous with the seatstays, as shown. This construction is less stiff and strong than a diamond frame but more so than a step-through frame.
If there is to be any major improvement in frame design, it must come either from a completely different type of construction process, such as carbon fiber, or cast magnesium; or a completely different type of design, such as a recumbent.
Much of the commonplace B.S. that is talked about different frame materials relates to imagined differences in vertical stiffness. It will be said that one frame has a comfy ride and absorbs road shocks, while another is alleged to be harsh and make you feel every crack in the pavement. Virtually all of these "differences" are either the imaginary result of the placebo effect, or are caused by something other than the frame material choice.
Bumps are transmitted from the rear tire patch, through the tire, the wheel, the seatstays, the seatpost, the saddle frame, and the saddle top. All these parts deflect to a greater or lesser extent when you hit a bump, but not to an equal extent.
The greatest degree of flex is in the tire; probably the second greatest is the saddle itself. If you have a lot of seatpost sticking out of a small frame, there's noticeable flex in the seatpost. The shock-absorbing qualities of good-quality wheels are negligible...and now we get to the seat stays. The seat stays (the only part of this system that is actually part of the frame) are loaded in pure, in-line compression. In this direction, they are so stiff, even the lightest and thinnest ones, that they can contribute nothing worth mentioning to shock absorbency.
The only place that frame flex can be reasonably supposed to contribute anything at all to "suspension" is that, if you have a long exposed seatpost that doesn't run too deep into the seat tube, the bottom end of the seatpost may cause the top of the seat tube to bow very slightly. Even this compliance is only a fraction of the flex of the exposed length of the seatpost.
The frame feature that does have some effect on road shock at the rump is the design of the rear triangle. This is one of the reasons that touring bikes tend to have long chainstays -- they put the rider forward of the rear wheel. Short chainstays give a harsh ride for the same reason that you bounce more in the back of a bus than in the middle...if you're right on top of the wheel, all of the jolt goes straight up. Some newer carbon-fiber frames have stiffness reducers in the chainstays, but they mainly serve as dampers (see next section).
[This section added by John Allen]
All bicycles have resonant frequencies at which they ring (oscillate) in different ways. You can hear a metal tube ring if you tap it with a fingernail, but resonances below the audible range most affect the feel of a bicycle. For example, "speed wobble" is the frame's ringing in torsion as the head tube tilts relative to the rear triangle, excited by the front wheel's turning slightly to the right and left. This is generally worse with a taller frame, because the front triangle is less stiff in torsion, and the resonant frequency is lower. Resonances also can affect the feel when going over bumps. Two frames which are equally stiff may feel different if they resonate at different frequencies.
Damping is the tendency of ringing to die out. All metal frames have very low damping -- they ring long enough to produce a clear tone. A carbon-fiber frame will give a dull sound if tapped, because carbon fiber has more damping than metal. This may affect the feel to some degree, though much less at the low frequencies which affect frame feel.
Rubber, leather and flesh are highly damped -- and so the greatest damping in a bicycle/rider system by far is in the tires, the saddle and the rider's body, unless the bicycle has suspension.
A suspension fork or frame is a highly-damped resonant system -- if it weren't damped, it would bounce up and down repeatedly after every bump. Suspension, obviously, has a major effect on the feel of a bicycle. Suspension also adds weight, affecting the feel.
Modern bicycle suspensions are mostly rather stiff, intended to protect the rider and bicycle against hard impacts, while minimizing "pogo sticking" due to pedaling out of the saddle. Interestingly, recent research published in Bicycle Quarterly magazine showed that tire choice and tire pressure achieved a much greater difference in comfort than a suspension front fork in a test ride on a bumpy surface! The main reason is that: the unsprung weight of the small part of the tire that flexes is tiny, while that of the wheel and fork is substantial. The rider is not rigidly connected to the bicycle, and so the sprung weight is largely that of the frame and attached components.
Also, don't be obsessed with weight. The pound or so of difference between a cutting-edge ultralight frame and a well-constructed, heavier one (generally a steel frame) is important to a racer climbing a mountain pass but makes little difference to a bicycle tourist -- especially not when carrying a touring load. The heavier frame may actually be more comfortable, because it increases the ratio of sprung to unsprung weight, and because it also is stiffer. A steel frame with plain-gauge tubing may actually be preferable for touring, compared with an equally-strong frame with butted tubing, because of the greater torsional stiffness and reduced tendency toward speed wobble when carrying a load.
Unfortunately, in bicycle applications, carbon fiber is not a fully mature technology, as tubular-construction metal frames are. Bicycles are subjected to a very wide range of different stresses from many different directions. Even with computer modeling, the loads can't be entirely predicted. Carbon fiber has great potential, but contemporary carbon fiber frames have not demonstrated the level of reliability and durability that are desired for heavy-duty touring use. In particular, a weak point tends to be the areas where metal fitments, such as fork ends, bottom bracket shells, headsets, etc connect to the carbon frame. These areas can be weakened by corrosion over time, and lead to failure. Raoul Luescher has excellent videos on YouTube exploring issues with carbon-fiber frames and components.
For extended travel in less-developed areas, steel is probably still the best choice, because in the event of damage, repairs can be made by anybody with a torch and brazing/welding know-how.
For further reading on this topic, see also Damon Rinard's frame tests at:
Scott Nicol has a very thorough discussion of frame materials at: