Bond, cracking and what changes at first crack

A bar cast into concrete is gripped by it. Three mechanisms contribute, in ascending order of importance:

• Adhesion — a chemical bond at the interface, which breaks at very small slip.
• Friction — resistance to the bar sliding once adhesion has gone.
• Mechanical interlock — the ribs rolled onto a deformed bar bear directly on the concrete between them.

For a modern ribbed bar the third dominates so completely that the first two are effectively a bonus. The ribs push against the concrete, which pushes back; that bearing generates a ring of tension around the bar, which is why bond is improved by cover and by links, and why a bar too close to a face can split the concrete off rather than develop its force.

The consequence of bond is strain compatibility: at any section, the strain in a bar equals the strain in the concrete immediately around it. Not the stress — the two materials have very different moduli, so their stresses differ by that ratio. The strain is what they share.

That single statement is what allows a section to be analysed at all, and it appears in every derivation in this course.

Now follow a beam from zero load and watch what happens.

Uncracked. At low load the whole section acts, including the concrete in tension. The neutral axis sits close to mid-depth. The section is stiff, and the steel carries very little — it has almost the same strain as the concrete next to it, and concrete at that strain carries very little stress, so the steel does too.

First crack. The tensile stress at the extreme fibre reaches the concrete's tensile strength. A crack opens. At that section the concrete below the neutral axis stops carrying tension — instantly and completely.

What changes. The tension that concrete was carrying has to go somewhere, and there is only one place left: the bar. Steel stress jumps sharply at the cracked section. Because the tension zone has been lost, the neutral axis rises and the section becomes markedly less stiff.

What does not change. Equilibrium. The compression resultant still equals the tension resultant, and their couple still equals the applied moment. Cracking changes the section, not the physics.

Beyond first crack. More cracks form, spaced along the tension zone. Between cracks the concrete still carries some tension, picked up from the bar through bond — this is tension stiffening, and it is why a real cracked beam is a little stiffer than a fully cracked calculation predicts.

One more idea completes the picture, and it explains a lot of detailing practice.

A crack is not a failure. What matters is how wide it is, because width controls whether moisture and chlorides can reach the steel.

Given a fixed amount of extension to accommodate, you can have it as one wide crack or many fine ones. Many fine ones are far better, and what produces them is well-distributed reinforcement: several smaller bars rather than a few large ones, at a modest spacing.

The reason is bond. Each bar can only transfer force back into the concrete over a certain length. Closely spaced bars, with more total surface area for the same steel area, re-anchor the concrete more often, so a new crack forms sooner and the extension is shared between more cracks.

This is why crack-control rules limit bar spacing and bar diameter rather than simply demanding more steel. Two 32 mm bars and four 25 mm bars supply almost the same area, but the four smaller bars will give visibly better crack control.