Stress at a Point Depends on the Angle You View It
So far you always split stress into σ perpendicular to a face and τ along it. But which way you orient that face is, in fact, up to you. At the very same point, tilt the face to a different angle and the normal and shear stresses on it change completely. It sounds strange but it is obvious: stress is only defined relative to a face, and you just changed the face's direction. So the stress state at a point is not a single number but a whole family that varies with angle. Handling this is stress transformation. Rotate the angle and σ and τ swing as sine and cosine, and among them comes a special angle where the shear is zero. There the normal stress is largest, and that is the direction in which the material actually splits. This lesson shows you that swing directly.
The standard way to draw the stress state at a point is the stress element. Picture a tiny square around that point and mark the stresses on its four faces with arrows. The left and right faces carry the horizontal normal stress σx, the top and bottom faces the vertical normal stress σy, and along the faces runs the shear stress τxy. Drag τ. You see the shear arrows appear on all four faces at once. Shear never acts on just one face, since that would break equilibrium; it always comes in a set across the four faces. This is complementary shear. This one little element holds the entire stress state at that point. Three numbers, σx, σy, τxy, fully determine plane stress.
Now the key move. Looking at the same point, tilt only the element. Drag θ and the square rotates, and the normal stress σ′ and shear stress τ′ on the new faces change together. The point is unchanged, only the faces' direction changed, yet the stresses differ. This is confusing at first, but on reflection it is obvious: the same bundle of force splits into different ratios of perpendicular and parallel components depending on which way you cut it. The more slanted the face, the more the force skims across it, so the shear grows. So σ′ and τ′ are functions of the angle θ. A point does not carry one stress value but a whole set, a different one for each viewing angle. Then you naturally ask: which angle is the most dangerous?
Writing that swing as formulas gives the transformation equations: σ′ = (σx+σy)2 + (σx−σy)2·cos2θ + τxy·sin2θ, and τ′ = −(σx−σy)2·sin2θ + τxy·cos2θ. They look long, but the heart is simple: both are sine and cosine of 2θ, smooth waves. Drag θ and the two points on the σ′ and τ′ curves move together. Two things to notice. First, it is 2θ, not θ, so the period is 180 degrees, which makes sense since rotating the element 180 degrees returns it to itself. Second, the average of σ′, (σx+σy)2, does not change no matter what θ is; this sum is a quantity preserved under rotation. Once you hold these two curves, finding the dangerous angle becomes reading off the peaks and the zeros of the curves.
As you rotate the element, an angle appears where the shear stress τ′ is exactly zero. That face is called a principal plane, and the normal stress on it is a principal stress. Drag θ to bring τ′ to zero. At that instant the shear arrows on the element vanish, and what remains is pure push and pull. And it is exactly at this angle that the normal stress is largest. The greatest principal stress σ1 and the smallest σ2 come out on these faces. Why it matters: a material usually splits perpendicular to this maximum principal stress. When you twist a piece of chalk to break it, the cross-section tears along a slanted spiral, and that spiral follows the principal planes. So to predict failure, the first task is to find σ1 and its direction, the principal stress and the principal angle.
Now the opposite: find the angle where the shear stress is largest. Drag θ and feel out where τ′ peaks, and it sits exactly 45 degrees from the principal plane. Always. The maximum-shear plane is 45 degrees from the principal plane. And the size of that maximum shear is half the difference of the two principal stresses, τmax = (σ1−σ2)2. Why it matters: a ductile material like steel yields to shear before normal stress. So when you pull a tensile specimen, it slips and stretches along faces tilted 45 degrees, not along a clean transverse section. Those slanted bands are the trace left along the maximum-shear planes. To sum up: on the principal planes the shear is zero and the normal stress is extreme, and 45 degrees from there the shear is extreme. These two special angles are the conclusion of stress transformation.