Tests for compressive and tensile strength

Compressive strength test (IS 516)

1. Sampling and casting: take a representative sample from the mix (IS 456, Cl. 15). Fill 150 mm cube moulds in 3 layers, each layer compacted by a table vibrator or tamped with a 16 mm bar (35 strokes per layer for 150 mm cubes). Level and finish the top.
2. Curing: keep the mould covered for 24 ± 0.5 h at 27 ± 2 °C, demould, and cure in water at 27 ± 2 °C until testing (normally 28 days; also 3 and 7 days).
3. Testing: take the cube out of water, wipe the surface, and place it in the compression testing machine so that the load acts on two opposite faces of the cube as cast (not on its top and bottom as cast). Apply the load without shock at 14 N/mm$^2$ per minute (about 140 kg/cm$^2$/min), until the specimen fails. Record the maximum load $P$.
4. Result: $f_{c} = P/A$, where $A = 150\times150 = 22500$ mm$^2$. The test strength is the average of three specimens; if any one differs by more than ±15% from the average, that result is discarded (IS 456, Cl. 15.4).
5. For cylinders (150 mm dia × 300 mm high) the same procedure is followed, with capped ends; $f_{cyl} \approx 0.8\,f_{cube}$.

Tests in tension

Tensile strength of concrete is only about 7 to 12% of its compressive strength, but it controls cracking, shear, bond and the design of pavements, water tanks and prestressed members. Concrete is not designed to carry tension, but the tensile strength decides when the first crack appears.

Direct tension test

A dog-bone or cylindrical specimen with embedded end grips is pulled axially. It is rarely used, because it is hard to avoid eccentricity and secondary stresses from the grips, and the result scatters. Direct tensile strength is about $0.35\sqrt{f_{ck}}$ to $0.45\sqrt{f_{ck}}$ (MPa).

Splitting tension test (Brazilian test, IS 5816)

A standard cylinder (150 mm dia × 300 mm long) is laid horizontally between the platens of the testing machine, with thin plywood or fibre strips (about 3 mm thick and 15 to 25 mm wide) along the contact lines. The compressive line load is increased at a rate of 1.2 to 2.4 N/mm$^2$/min (splitting tensile stress) until the cylinder splits along the vertical diameter. The load applied gives a nearly uniform tensile stress across the loaded plane:

$$f_{sp} = \frac{2P}{\pi L d}$$

where $P$ = failure load, $L$ = length and $d$ = diameter of cylinder. Splitting strength is usually a little higher than the direct tensile strength and is about $0.5\sqrt{f_{ck}}$ to $0.6\sqrt{f_{ck}}$ (MPa).

Advantages: simple, uses the same cylinders and compression machine as the compression test, gives a more uniform and reproducible result than direct tension, with no grip problem.

Flexure test (modulus of rupture, IS 516)

A plain concrete beam 150 × 150 × 700 mm is supported on a 600 mm span and loaded at the third points (two-point loading) at 0.7 N/mm$^2$/min extreme-fibre stress rate. If the crack occurs in the middle third:

$$f_{cr} = \frac{P\,l}{b\,d^{2}}$$

If the crack is outside the middle third but within 5% of the span from the load point, $f_{cr} = 3Pa/(bd^2)$ ($a$ = distance of crack from the nearer support); otherwise the result is rejected. For single (centre) point loading, $f_{cr} = 3Pl/(2bd^2)$. IS 456 (Cl. 6.2.2) gives the flexural strength $f_{cr} = 0.7\sqrt{f_{ck}}$ N/mm$^2$. The modulus of rupture is greater than direct tension (by about 1.5 to 2 times) since the stress distribution is assumed linear while the actual stress block is plastic.