Limitations of plate load test

The plate load test (PLT) loads a rigid steel plate (30 to 75 cm) placed at foundation level and records load against settlement. Its results are not safely extrapolated to the real footing in the following circumstances.

1. Depth of influence (scale effect). The stressed zone below a plate is only about $1.5$ to $2B_p$ deep, while below the real footing it is $1.5$ to $2B$. A weak or compressible layer below the plate's influence zone is not tested at all, yet it may control the settlement of the full-size footing.
2. Clay: time-dependent settlement. The test lasts a few hours or days, so only immediate settlement is recorded. The consolidation settlement of clay, which takes months or years, is not shown. Settlement in clay also depends on the width of the footing, not in simple proportion.
3. Size effect on bearing capacity. In sand, $q_u$ increases with width (the $N_\gamma$ term), so a plate result underestimates the footing's capacity and is corrected by $q_u \propto B$. In clay $q_u$ is independent of size. A single rule cannot be applied to all soils.
4. Non-homogeneous or layered soil. Properties at the test point may not represent the whole site, and one test represents only a small area.
5. Water table. If the water table lies below the plate but within the depth of influence of the footing, the test does not reflect its effect.
6. Test conditions. Moisture and disturbance at the test pit differ from field conditions, and the pit must be dug to foundation level, so deep foundations cannot be tested.
7. Non-uniform loading and rigidity. A rigid plate and a flexible footing do not distribute pressure in the same way.
8. Cost and time. The test needs heavy reaction loading and is expensive for large footings.

Because of these limits, the PLT is used as a check and is supported by borehole data, SPT/CPT and laboratory tests.