jyl

There have been a few threads lately about pulling back damaged frames. Sometimes the damage includes the downtube. Usually people say the downtube isn't that important, it is only under tension, so the bike will be rideable with the straightened downtube.

This got me to thinking. In a modern carbon fiber bike, the high performance sort, the downtube is usually the most massive part of the frame, meaning it has the largest cross section and is visually thickest. The top tube is often quite thin. Also the seat stays are usually very slender, much thinner than the chain stays.

Doesn't the cross section of the frame tube indicate the forces on it, because the designers have sized the tubes to have just enough strength without excess weight? If so, doesn't that mean the downtube takes more force than any other tube?

hueyhoolihan

i think that the modern downtube's large cross section is a consequence of the designer's desire to create rigidity in the bottom bracket, among other things.

the framebuilder's forum will no doubt have a few people that have opinions too. better than mine i would imagine.

FBinNY 

It's more complicated than simple tension or compression since a bicycle operates in a 3 dimensional world. The downtube is the largest because, besides it's role as part of the frame triangle, it also functions to keep the bike in a plane against the twisting forces of the front, and bottom bracket. It's those torsional loads that call for so much beefiness, not the flatland tension/compression loads.

There's plenty f literature on frame design, and any depth is going to take more than I'm willing to type, but when we speak of whether it's OK to ride a damaged frame, we'e (or at least I) talking about safety implications, not performance.

Andrew R Stewart 

And not just the tubes are seeing complex forces but also their junctions. For such a simple design the bicycle is a very complex structure/machine. this is why so many engineering classes use a bicycle as their example of study. A very good start in learning more then you can understand is the book "Bicycling Science" third edition by David Gordon Wilson (with contributions by Jim Papadpopoulos). I just finished a quick read through,for the second time as I also have the first edition. Another cool book, although not one I've read fully but have skimmed, is Archibald Sharp's "Bicycles and Tricycles: A Classic Treatise on Their Design and Construction" from 1896. The take away I got was that bicycle design hasn't really changed in over a century and our understanding of it too. Some materials and manufacturing methods have but the basic stuff like tension spoked wheels, balled bearings, tubular frames, self centering steering and other elements are all the same these days as they were over 119 years ago.

To rant a bit- So many riders and posts on this forum have such a self centered focus in their view of cycling and such. A little historical perspective could go a long way to understand that one's place and experiences have been had by thousands before you. Andy.

dabac

Don't forget that while you do see the outer diameter(for lack of a better word), you don't see the wall thickness.

They're quite similar to the early years of aluminium bikes. Big diameters for loads of torsion strength, but really thin and flimsy walls.

wphamilton

Maybe not so much with self-stability. Not to snipe at you, I just think it's interesting how the common sense explanations became common knowledge and since discredited.

Andrew R Stewart 

wphamilton- Actually steering stability and it's cause/effect is exactly why one should bone up on the latest theories. Jim P has done some of the best/deepest research of this field. Note he is a contributor to "Bicycling Science". Also my point of design elements being the same is more to how things actually work and are configured, not how we understand them. Today's bikes employ much the same steering design that was used in the 1800s. An angled steering axis, wheel axle offset from this axis producing a trail or castor effect. I agree that our understanding has evolved but the design in practice remains the same. Further it is your point that I am talking about WRT our understandings. Today our common sense says that we have invented fiber reinforced materials, anatomical saddles, indexed shifting, clippless pedals... and the list goes on. A brief study of bicycle design history shows this current view to be wrong. Another very interesting book I've recently read is "Bicycle Design an Illustrated History" by T Hadland and H-E Lessing. While not going into the math it does show how the bicycle has evolved, or not so much. Andy.

gsa103

Most of modern bicycle tube sizing is about managing stiffness and has very little to do with load support. Carbon fiber is EXTREMELY flexible and can undergo absurd amounts of flex (see Boeing 787 wind flex testing) without any issues at all. You could make a bike out of much thinner tubing that would easily support a rider but would have the rigidity of a pool noodle.

Seat stays are thin because you want the wheel to flex upward for bump compliance (see Cervelo R-series or Cannondale Synapse). The downtube you want to large because you want a stiff BB so that pedal force is transferred into the chain rather than flexing the BB. So the seat and down tubes are sized to maximize the stiffness of the BB without giving up too much weight or aerodynamics. Stiffer BB is only good up to a certain point, once a you get stiff enough it doesn't matter anymore, since the rider is not able to distinguish any difference (either in feel or efficiency).

The downtube size is a trade-off between weight, stiffness, aerodynamics, and BB width (Q-Factor).

1saxman

And of course compression is a major factor on the down tube. Imagine running the bike into a wall; the head tube tries to deflect into the down tube which puts the down tube into compression. But the likely result of that will be bending, because of the angle between the head tube and the down tube, and a trashed frame. Control of the head tube and bottom bracket are the principle reasons for the larger cross section of the down tube.

hueyhoolihan

surprised nobody has posted a pic of one of these yet. blows a hole in all the downtube rigidity mumbo jumbo, but in defense of those that have mentioned it, me being one of them, you don't see these too often anymore, if one ever did.

at least the bike won't take much damage to the what serves as a downtube when run into a brick wall at speed, but oh , the toptube will suffer...

Jeff Wills

For some mathematical models of bicycle vehicles and stability, consider Bill Patterson's "Lords of the Chainring": https://www.calpoly.edu/~wpatters/ (Yeah, I know that web page hasn't been edited in 14 years. It still applies.)

FBinNY 

The Slingshot doesn't in any way challenge the downtube compression considerations.

Forces lie not only along tubes, but within them too. Consider single tube bikes like folding bikes or the Trimble. What might be tension or compression elements in a truss all happen as bending moments within the single tube. Likewise the slingshot. The head is rigidly attached to a high aspect top tube, and if the front axle is pushed forward or back, it would create bending moments in that tube, the same way it would to a toptube/downtube combination. That the slingshot, uses a pivot at the seat lug, and a cable stabilizer doesn't change that. Like any bike, a front end collision would buckle the tube at the head.

For those interested, the classic downtube operates in both tension and compression. As you ride there are external forces on the front wheel, which can be taken as passing directly through the axle. So sitting on level ground, the force is vertical passing forward of the head joint, and would tension the downtube. Bumps are more interesting. and the effect depends on the height. The key is the line from the point of impact through the axle. If the bump is small, that line will be near vertical,and as expected tension the downtube. But a slightly taller bump whould create force with a shallower line of action passing behind the head, and creating tension in the top tube, and compression in the down tube.

The extreme case is a wall which we all know will compress and buckle a downtube, but there's a whole spectrum from tiny to bigger bumps, and the forces vary accordingly.

Reynolds 

No downtube:

noglider 

I'm just guessing, but I'd say The reason we don't see this is that it doesn't work well precisely because it doesn't withstand twisting forces. It's the same reason down tubes have grown in recent years.

tcs

Or not, depending on where you come down on planing.

gsa103

Planing only works when your force and cadence match the spring rate of the bike. It's impossible to get a good spring rate match at 60, 90 and 120 rpm, since the BB is acting like a driven oscillator. Stiffer results in equal or better power transfer regardless of cadence and force. So if you always pedal at the same 80 rpm, planing can be a large benefit. If you want to mash up a climb at 60 rpm then sprint away at 110 rpm you're better off with a stiffer bike.