Bond, Anchorage & Development Length

1. What is Bond? — Concept and Mechanisms

For an RCC beam to work as a composite structure, the steel bars and surrounding concrete must deform together — they must have the same strain at every point along their interface. This strain compatibility is maintained through bond — the resistance to relative sliding between the bar and the concrete.

Bond is not a single mechanism but a combination of three effects:

1. Chemical adhesion: When concrete hardens, the cement paste bonds chemically to the steel surface. This is the weakest component and is destroyed as soon as the bar slides even slightly relative to the concrete.
2. Friction: Mechanical friction between the bar surface and the surrounding concrete. Smooth (plain) bars rely mainly on friction and adhesion, which is why they develop relatively low bond stresses.
3. Mechanical interlock (bearing of ribs): The dominant mechanism for deformed (ribbed) bars. The transverse ribs on the bar surface bear against the surrounding concrete, creating a mechanical lock. This is why deformed bars have 60% higher bond stress than plain bars — the ribs dramatically increase bond capacity. Under load, the concrete in front of each rib is in compression (bearing), and small diagonal cracks form around the bar (splitting cracks). These splitting cracks are what ultimately limit bond strength and cause bond failure.

Bond failure can occur in two ways. Pull-out failure: The bar slides out of the concrete, overcoming the bond resistance over the embedded length. Splitting failure: Transverse splitting cracks propagate along the bar and reach the surface, effectively splitting the concrete cover and destroying the confinement needed for bond. Splitting failure is more common in practice and is why IS 456 specifies minimum cover and bar spacing requirements alongside development length.

2. Types of Bond Stress

IS 456:2000 in the LSM framework primarily uses the development length concept — the anchorage bond approach — rather than checking flexural bond stress explicitly. The development length requirement ensures that every bar is embedded long enough to develop its full design force, which implicitly satisfies the flexural bond requirements as well.

3. Design Bond Stress τ_bd — IS 456 Table 26

4. Development Length L_d — Formula and Derivation

The development length is derived from the equilibrium between the force in the bar and the total bond resistance along the embedded length. Consider a bar of diameter φ embedded for length L_d in concrete. The bar is at full design stress σ_s at one end and zero stress at the other end (just developed).

5. L_d Values — Pre-computed Table ⭐ GATE

6. Hooks and Bends — Anchorage Value

When there is insufficient straight bar length available to develop the full force in a bar (for example, near the ends of a simply supported beam), hooks and bends provide additional anchorage. A standard hook or bend contributes an equivalent straight embedment length, effectively shortening the required straight development length.

7. Curtailment of Bars

In practice, not all tension bars need to run the full length of a beam. As the bending moment decreases toward the supports (for a simply supported beam), fewer bars are needed. Bars that are no longer needed can be curtailed (stopped) at an appropriate point. However, IS 456 imposes strict rules on where curtailment is permitted to ensure that the development length requirement is still satisfied.

8. Checking Development Length at Supports and Curtailment Points

9. Lap Splices

When bars are not long enough to run the full span (due to available bar lengths, transportation constraints, or design requirements), they are spliced — overlapped for a specified length so that the force transfers from one bar to the next through the concrete.

10. Worked Example 1 — Development Length Calculation

Problem: Calculate the development length for: (a) Fe 415 deformed bars of 20 mm diameter in M20 concrete in tension, (b) Fe 500 deformed bars of 25 mm diameter in M25 concrete in tension, (c) Fe 415 deformed bars of 16 mm diameter in M20 concrete in compression.

11. Worked Example 2 — Checking L_d at Simply Supported End

Problem: A simply supported beam has span 5 m, b = 250 mm, d = 450 mm. It carries a factored UDL of 40 kN/m. Main steel: 3 bars of 20 mm dia Fe 415 (A_st = 942 mm²). M20 concrete. All 3 bars continue to the supports. The bearing width at each support = 230 mm. Check if the development length requirement is satisfied at the supports. No hooks provided.

12. Worked Example 3 — GATE-Style Problem

Problem (GATE-style): A simply supported RCC beam uses 4 bars of 16 mm diameter (Fe 415) in M25 concrete. If 2 bars are curtailed at a point where the factored shear force is 80 kN, find the minimum length of extension required beyond the theoretical cut-off point for the curtailed bars.

13. Common Mistakes Students Make

• Forgetting the 1.6 multiplier for deformed bars: The IS 456 Table 26 values are for plain bars. For deformed (HYSD) bars — Fe 415, Fe 500 — the design bond stress must be multiplied by 1.6. This is the most important adjustment in development length calculation. Using the plain bar τ_bd directly for Fe 415 bars overstates the required development length by 60%, giving a very conservative (wasteful) result. Always check: is the bar plain or deformed? For Fe 415 and Fe 500, always multiply τ_bd by 1.6.
• Using σ_s = f_y instead of 0.87f_y: The stress to be developed is the design yield stress = 0.87f_y (which includes the partial safety factor of 1.15). Using the characteristic yield strength f_y instead of 0.87f_y overestimates the required development length by a factor of 1/0.87 = 1.15 (about 15%). This is a common sign of confusing the characteristic and design values.
• Applying hooks to compression bars: Hooks are only effective for bars in tension. For compression bars (column longitudinal bars at splice points, compression steel in doubly reinforced beams), hooks do not provide additional anchorage — the bar is being pushed into the concrete, and the end bearing of the bar provides compression resistance, but standard hooks designed for tension don’t help. Using hooks to reduce the development length of compression bars is unconservative and incorrect.
• Using the development length for a different concrete grade or bar type: L_d = 47φ applies specifically to Fe 415 deformed bars in M20 concrete in tension. For Fe 500 bars or higher-grade concrete, L_d is different. Always re-derive L_d from the formula for the actual grades specified in the problem, or use the table of L_d/φ ratios with the correct combination.
• Confusing development length with lap splice length: Development length L_d is the embedment required to develop full bar stress at one end. Lap splice length is the overlap between two bars for force transfer and equals L_d as a minimum for up to 25% of bars lapped at a section. They happen to be numerically equal under IS 456 for the standard case, which causes confusion — but they are conceptually different. Lap length increases when more than 25% of bars are lapped at one section (multiplier up to 2.0 × L_d).

14. Frequently Asked Questions

Why does development length increase with bar diameter?

The development length formula L_d = 0.87f_yφ/(4τ_bd) shows that L_d is directly proportional to the bar diameter φ. Physically, this makes sense: a larger bar has more cross-sectional area and therefore carries more force (proportional to φ²). But the bar perimeter — which provides the bond surface — only increases proportionally to φ (not φ²). So the bond force per unit length increases as φ (perimeter × τ_bd × 1), but the total force to be developed increases as φ². To equate the two, the length must increase proportionally to φ. In simpler terms: doubling the bar diameter requires doubling the development length for the same bond stress and design stress.

What happens if the available length for anchorage is less than L_d?

If the available straight embedment length is less than L_d, the bar cannot develop its full design stress and the beam section’s moment capacity is reduced. The engineer has several options. First, provide standard hooks to gain extra anchorage equivalent to 16φ per hook. Second, reduce the bar diameter — smaller bars have shorter L_d for the same total steel area, because L_d ∝ φ while area ∝ φ². For example, replacing one 25 mm bar with four 12.5 mm bars provides the same area with much shorter L_d. Third, extend the beam into a wider bearing at the support to provide more embedment length. Fourth, use mechanical anchorage devices (end plates, couplers) that can develop the bar force over a very short length — these are permitted by IS 456 Cl. 26.2.2.4 when end anchorage is needed.

Why is the development length for compression bars shorter than for tension bars?

Two factors combine to reduce L_d for compression bars. First, the design bond stress τ_bd for compression bars is 25% higher than for tension bars — IS 456 multiplies the tension τ_bd by 1.25 for compression. This is because under compression, the concrete around the bar is laterally stressed (Poisson effect), which increases confinement and improves bond. Second, the end of a compression bar bears directly on the concrete, providing additional compression transfer through end bearing — an effect that does not exist for tension bars being pulled out. These two effects together reduce L_d in compression to about 0.8 × L_d in tension.

Why must bars be extended beyond the theoretical cut-off point?

At the theoretical cut-off point, the section’s moment capacity exactly equals the applied moment — there is no reserve. But cutting off bars also affects the shear behaviour. The presence of longitudinal bars suppresses diagonal cracking through dowel action; when a bar ends, its contribution to shear resistance suddenly disappears, creating a stress concentration and a potential for diagonal tension failure at the cut-off point. The extension requirements of IS 456 (greater of d and 12φ) provide a safety margin — the bar extends far enough past the cut-off point that the stress in the bar at the actual cut is well below 0.87f_y, reducing the risk of premature failure at that location.