Originally Posted by eubi

It depends on how it is expected to be loaded.

The bending stress in a beam is given by:

Stress = M*y/Ixx

Where M is the bending Moment
y is the distance from the neutral axis
Ixx is the second moment of inertia of the beam along the xx bending axis

The resistance to bending stress in a beam is determined by what engineers call the second moment of inertia of the beam. When you bend a beam, the stress is highest as you move from the bending axis of the section. The reason for an "I" beam to be shaped the way it is is that it has a lot of material at the top and bottom to better resist the bending stresses. This way it is strong but light.

Square tubing can be made to simulate the stress distribution you would see in an "I" beam. Thicker material at the top and bottom, with thinner material on the sides. You have light weight, and material where you need it.

The down side to this wonderful I beam is if it is subjected to the same bending load in another direction, it will be highly stressed. Now the second moment of inertia will be Iyy. From the equation, if Iyy is less than Ixx, you can see the stress will be higher.

On the other hand, you can put a bending load on a tube from any direction and the stress will be the same. Ixx = Iyy.

I know this is probably clear as mud. We really need some diagrams and manipulatives to illustrate it better. Try this:

Take a yard stick and bend it across its long side. Hard, huh? Now bend it across its short side. Much easier, right? The amount of material is the same, but depending on you you orient it, it can be made to resist bending better.

Now take a piece of PVC pipe. It will bend the same no matter how you bend it.

Let's not even talk about when happens when you twist a square or rectangular tube!