Flow Net Applications in Seepage Analysis A properly constructed flow net for a homogeneous, isotropic soil can be used to determine which of the following quantities in seepage problems?

Difficulty: Easy

Correct Answer: All the above

Explanation:


Introduction / Context:
Flow nets are graphical solutions to Laplace’s equation for two-dimensional steady seepage in saturated soils. They are widely used beneath sheet pile walls, around cutoffs, and through earth dams to evaluate safety against boiling, piping, and uplift, and to estimate seepage quantities for drainage design.


Given Data / Assumptions:

  • Soil domain is treated as homogeneous and isotropic.
  • Steady-state conditions with well-defined boundary conditions.
  • The constructed net forms approximate curvilinear squares with orthogonal flow lines and equipotentials.


Concept / Approach:

From a flow net, discharge per unit thickness is computed by q = k * (N_f / N_d) * Δh, where k is coefficient of permeability, N_f is the number of flow channels, and N_d is the number of potential drops. The pore-water pressure at any point is u = γ_w * h, where h is the hydraulic head read from equipotential lines. Seepage pressure equals γ_w * i along a flow line, while exit gradient at the downstream toe is obtained from the head loss across the last potential drop divided by the flow path length at the exit element.


Step-by-Step Solution:

Count N_f and N_d → compute q.Read head at a node → compute u = γ_w * h.Estimate local hydraulic gradient i from potential spacing → get seepage pressure γ_w * i.At the exit, compute i_exit and compare to critical gradient to check piping risk.


Verification / Alternative check:

Finite element seepage analysis reproduces similar results when boundary conditions and anisotropy treatments are equivalent.


Why Other Options Are Wrong:

Each listed quantity legitimately comes from a correct flow net; omitting any would be incomplete. Hence 'All the above' is the comprehensive choice.


Common Pitfalls:

Using distorted grids that are not curvilinear squares; miscounting N_f or N_d; applying the method without transforming axes for anisotropic soils.


Final Answer:

All the above

More Questions from Soil Mechanics and Foundation Engineering

Discussion & Comments

No comments yet. Be the first to comment!
Join Discussion