Detailed ExplanationCELE · Geotechnical EngineeringReal content

CELE Geotechnical Engineering — Permeability and SeepageDetailed Explanation

Summary Concepts Notes Revision Deep Dive Slides Flashcards Mistakes Cheat Mnemonics Templates

Detailed explanations for CELE Geotechnical Engineering — Permeability and Seepage. This page treats you like a serious reviewer: we unpack the concepts thoroughly, show worked examples of how Professional Regulation Commission (PRC) — Board of Civil Engineering frames Permeability and Seepage questions, and explain the underlying reasoning that gets you to the right answer every time.

Exam context

Professional Regulation Commission (PRC) — Board of Civil Engineering runs the Civil Engineer Licensure Examination on May and November 2026. Its Geotechnical Engineering section sits under a "Core" weighting, and Permeability and Seepage is the 3rd chapter in the 11-chapter CELE Geotechnical Engineering rotation. The CELE passing mark is 70% weighted average, no sub-test below 50%, and the most recent 2026 paper drew about a meaningful share of questions from Geotechnical Engineering.

Permeability and Seepage - Detailed Explanation

Permeability and seepage are among the most frequently tested topics in the PRC Civil Engineer Licensure Examination under Geotechnical Engineering. These concepts govern the movement of water through soil pores and directly affect the design and safety of dams, retaining walls, sheet piles, foundations, and slopes throughout the Philippines. From checking whether a proposed irrigation embankment in Pampanga will experience piping failure to computing dewatering requirements for a deep basement excavation in Metro Manila, the engineer must be fluent with Darcy's Law, laboratory permeability tests, layered-soil equivalents, flow-net analysis, and the critical gradient (quick condition). This chapter presents every concept at board-exam depth, with worked numerical examples in SI units, proper formula derivation, and targeted advice on the pitfalls that repeatedly trap examinees.

Concepts

Darcy's Law — Discharge Velocity and Seepage Velocity

Henri Darcy (1856) established experimentally that for laminar flow through a saturated porous medium, the discharge velocity v (also called Darcy velocity or superficial velocity) is directly proportional to the hydraulic gradient i: v = k · i where k is the coefficient of permeability (m/s) and i = Δh / L is the ratio of total head loss Δh to the length of flow path L (dimensionless). The volumetric flow rate (discharge) through a gross cross-sectional area A is: Q = v · A = k · i · A Critical distinction — Discharge velocity vs. Seepage velocity: The discharge velocity v is fictitious; it assumes water flows through the entire gross area A, including solid particles. The actual velocity through pore spaces is the seepage velocity vs: vs = v / n where n is the soil porosity (0 < n < 1). Because n < 1, vs > v. Examinees frequently confuse these two velocities; remember that seepage velocity is always LARGER than discharge velocity. Validity of Darcy's Law: Darcy's law holds only for laminar (low-Reynolds-number) flow, which covers virtually all practical cases in silts, clays, and sands. In coarse gravels or rockfill, turbulent flow can occur and Darcy's law may underestimate head loss. The Reynolds number criterion Re = v·D10/ν < 1 is the usual check.

Examples

Note that vs = 1.14×10⁻⁴ m/s is about 2.86 times larger than v = 4×10⁻⁵ m/s, confirming that seepage velocity exceeds discharge velocity. In board problems that ask 'how fast does a contaminant travel through the soil?', you must use seepage velocity, not discharge velocity.

Scenario

A sandy soil layer with k = 1×10⁻⁴ m/s has a total head loss of 2 m over a flow path of 5 m. The gross cross-sectional area is 0.5 m². The soil has a porosity of 0.35. Compute: (a) hydraulic gradient, (b) discharge, and (c) seepage velocity.

Solution

(a) i = Δh/L = 2/5 = 0.40 (b) Q = k·i·A = (1×10⁻⁴ m/s)(0.40)(0.5 m²) Q = 2×10⁻⁵ m³/s ✓ (c) v = k·i = (1×10⁻⁴)(0.40) = 4×10⁻⁵ m/s vs = v/n = (4×10⁻⁵)/0.35 = 1.14×10⁻⁴ m/s

Applications

Computing seepage through earth dam cores and blanket drains
Estimating dewatering pump capacity for basement excavations
Contaminant transport time estimation in groundwater studies
Checking adequacy of filter layers and drainage blankets under pavement
Input to consolidation analysis (Terzaghi theory uses k)

Misconceptions

Thinking vs = v — seepage velocity is always LARGER than discharge velocity by factor 1/n
Using diameter instead of length for the flow path in i = h/L
Forgetting that k is a property of BOTH soil and fluid (temperature affects viscosity, hence k changes slightly with temperature)
Applying Darcy's law to coarse gravels without checking laminar flow condition

Related Concepts

Hydraulic gradient
Porosity and void ratio
Bernoulli's equation (total head)
Terzaghi consolidation theory
Critical gradient / quick condition

Common Exam Questions

Example

Given k = 5×10⁻⁵ m/s, i = 0.3, A = 0.8 m², find Q. Answer: Q = (5×10⁻⁵)(0.3)(0.8) = 1.2×10⁻⁵ m³/s

Approach

Identify Δh, L, A, and k from the problem. Compute i = Δh/L, then Q = kiA. Watch units — convert mm to m consistently.

Question Type

Direct computation of Q or v

Example

If e = 0.65 and v = 3×10⁻⁵ m/s, then n = 0.65/1.65 = 0.394, vs = 3×10⁻⁵/0.394 = 7.6×10⁻⁵ m/s

Approach

First find v = ki, then divide by porosity n. Porosity may be given or computed from e: n = e/(1+e).

Question Type

Seepage velocity computation

Key Points To Remember

v = k·i (Darcy velocity) — water velocity through gross area
Q = k·i·A — volumetric discharge
i = Δh / L — hydraulic gradient (dimensionless)
vs = v / n — seepage (pore) velocity, always greater than v
k has units of velocity (m/s or cm/s)
Darcy's law is valid for laminar flow only (Re < 1)
k for clays: 10⁻⁹ to 10⁻⁷ m/s; silts: 10⁻⁷ to 10⁻⁵; sands: 10⁻⁵ to 10⁻³; gravels: >10⁻³

Laboratory Permeability Tests

Two standard laboratory methods are used depending on soil type: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ CONSTANT-HEAD TEST (for coarse-grained soils: sands, gravels) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ The head difference h is maintained constant throughout the test. The volume V of water collected in time t gives: k = (V · L) / (A · h · t) where: V = volume of water collected (m³ or mm³) L = length of specimen (m or mm) A = cross-sectional area of specimen (m² or mm²) h = constant head difference (m or mm) t = duration (s) Alternative form using flow rate Q = V/t: k = Q·L / (A·h) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ FALLING-HEAD TEST (for fine-grained soils: silts, clays) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ The head drops from h₁ to h₂ in time t through a standpipe of cross-sectional area a. The specimen has cross-section A and length L: k = (2.303 · a · L) / (A · t) · log₁₀(h₁/h₂) Note: The factor 2.303 converts natural log to log base-10 (since ln x = 2.303 log₁₀ x). You may also write this as: k = (a · L) / (A · t) · ln(h₁/h₂) Both forms are equivalent and accepted on the board exam. Key rule for test selection: • Constant-head → coarse soils (sand, gravel) — high k, large Q, easy to measure volume • Falling-head → fine soils (silt, clay) — low k, very small Q, impractical to measure directly

Examples

This value (≈10⁻⁴ m/s) is typical for medium sand. The answer is consistent with the test being performed on a coarse-grained soil using the constant-head method.

Scenario

CONSTANT-HEAD TEST: A sand specimen 150 mm long with A = 2000 mm² is tested under a constant head of 300 mm. In 120 seconds, 180 cm³ of water is collected. Determine k.

Solution

Convert volumes to consistent units (mm³): V = 180 cm³ = 180,000 mm³ L = 150 mm, A = 2000 mm², h = 300 mm, t = 120 s k = VL / (Aht) = (180,000 × 150) / (2000 × 300 × 120) = 27,000,000 / 72,000,000 = 0.375 mm/s = 3.75×10⁻⁴ m/s ✓

This is within the silt range (10⁻⁷ to 10⁻⁵ m/s boundary region), appropriate for a fine-grained soil tested using the falling-head method. Always double-check by asking: Is the answer in the expected range for the soil type described?

Scenario

FALLING-HEAD TEST: A clay specimen with L = 150 mm and A = 2000 mm² is tested using a standpipe with a = 100 mm². The head drops from h₁ = 500 mm to h₂ = 200 mm in 90 seconds. Find k.

Solution

k = (2.303 × a × L) / (A × t) × log₁₀(h₁/h₂) log₁₀(500/200) = log₁₀(2.5) = 0.3979 k = (2.303 × 100 × 150) / (2000 × 90) × 0.3979 Numerator constant: 2.303 × 100 × 150 = 34,545 Denominator: 2000 × 90 = 180,000 k = (34,545 / 180,000) × 0.3979 = 0.19192 × 0.3979 = 0.07636 mm/s = 7.64×10⁻⁵ m/s

Applications

Quality control for compacted fill and embankment construction
Drainage layer design for road pavements and airport runways
Determining pumping test parameters for well design
Input data for seepage and consolidation analyses
Compliance testing for liner materials in sanitary landfills

Misconceptions

Using constant-head formula for a fine-grained soil problem (or vice versa)
Confusing 'a' (standpipe area) with 'A' (specimen area) in the falling-head formula
Using ln instead of log₁₀ with the 2.303 factor — both forms give the same answer only if used correctly
Forgetting to convert volume units (cm³ to mm³ or m³) before substituting

Related Concepts

Darcy's Law
Soil classification (USCS)
Void ratio and porosity
Grain size distribution
Hazen's empirical formula: k = C·D₁₀²

Common Exam Questions

Example

A CH clay is to be tested for permeability. Which test is appropriate? Answer: Falling-head test.

Approach

Identify grain size / soil classification. Sand or gravel → constant-head. Silt or clay → falling-head.

Question Type

Which test to use for a given soil

Example

Board exam often mixes mm and cm units intentionally — always convert to one system first.

Approach

Plug into the correct formula. For constant-head, identify V (or Q), L, A, h, t. For falling-head, identify a, A, L, t, h₁, h₂.

Question Type

Solve for k from test data

Example

Given k, a, A, L, h₁, h₂, find time t for head to drop. Rearrange: t = (2.303·a·L)/(A·k) × log(h₁/h₂)

Approach

Rearrange the permeability formula to solve for t. This tests algebraic manipulation under pressure.

Question Type

Solve for missing variable (e.g., required time)

Key Points To Remember

Constant-head: k = VL / (Aht) — use for sands and gravels
Falling-head: k = (2.303·a·L / A·t)·log(h₁/h₂) — use for silts and clays
In falling-head, 'a' is standpipe area; 'A' is specimen area
Keep units consistent throughout — either all in mm or all in m
log means log base 10 (common logarithm) in the standard formula
The ratio a/A controls the rate of head drop — smaller a/A = slower drop
ASTM D2434 covers constant-head; ASTM D5084 covers falling-head

Layered Soils — Equivalent Permeability

Natural soil deposits are rarely homogeneous. Engineers must compute an equivalent (average) permeability keq for a stratified soil profile. The result depends critically on the DIRECTION of flow relative to the layering. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ CASE 1: FLOW PARALLEL TO LAYERS (Horizontal flow) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ All layers experience the same hydraulic gradient i. The total discharge is the sum of discharges through each layer. For n layers: keq(H) = (k₁H₁ + k₂H₂ + ... + knHn) / (H₁ + H₂ + ... + Hn) keq(H) = Σ(kiHi) / ΣHi This is a WEIGHTED ARITHMETIC MEAN — the layer with the highest k dominates. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ CASE 2: FLOW PERPENDICULAR TO LAYERS (Vertical flow) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ All layers carry the same discharge Q (continuity). The total head loss is the sum of head losses in each layer: keq(V) = ΣHi / Σ(Hi/ki) This is a WEIGHTED HARMONIC MEAN — the layer with the LOWEST k dominates. Physical meaning: When flow is parallel, the high-k layer acts like a shortcut highway — it carries most of the flow. When flow is perpendicular, every layer is in series, and the most restrictive (lowest k) layer controls — like the narrowest section of a pipe. For a two-layer system: keq(H) = (k₁H₁ + k₂H₂)/(H₁ + H₂) keq(V) = (H₁ + H₂)/[(H₁/k₁) + (H₂/k₂)] IMPORTANT INEQUALITY (always true): keq(H) ≥ keq(V) Parallel permeability is always ≥ perpendicular permeability.

Examples

Notice that keq(H) is dominated by the high-k layer (k₁ = 10⁻³), while keq(V) is dominated by the low-k layer (k₂ = 10⁻⁵). The ratio keq(H)/keq(V) ≈ 24.5, which is typical for natural alluvial deposits with clay lenses. This anisotropy is critical for dam seepage calculations.

Scenario

A soil deposit has two layers: Layer 1 — k₁ = 1×10⁻³ m/s, H₁ = 2 m; Layer 2 — k₂ = 1×10⁻⁵ m/s, H₂ = 3 m. Find keq for (a) flow parallel to layers and (b) flow perpendicular to layers.

Solution

(a) PARALLEL (horizontal): keq(H) = (k₁H₁ + k₂H₂)/(H₁ + H₂) = [(1×10⁻³)(2) + (1×10⁻⁵)(3)] / (2 + 3) = [2×10⁻³ + 3×10⁻⁵] / 5 = [0.002000 + 0.000030] / 5 = 0.002030 / 5 = 4.06×10⁻⁴ m/s (b) PERPENDICULAR (vertical): keq(V) = (H₁ + H₂) / (H₁/k₁ + H₂/k₂) = (2 + 3) / [(2/1×10⁻³) + (3/1×10⁻⁵)] = 5 / [2000 + 300,000] = 5 / 302,000 = 1.656×10⁻⁵ m/s Check: keq(H) = 4.06×10⁻⁴ >> keq(V) = 1.66×10⁻⁵ ✓ (parallel always larger)

Applications

Horizontal drainage layer design (parallel flow analysis)
Seepage through embankment dams with clay cores and sand shells
Vertical infiltration through layered road subgrades
Analyzing anisotropic flow conditions for flow net construction
Dewatering system design for excavations in stratified soils

Misconceptions

Swapping the parallel and perpendicular formulas — the most common error on this topic
Using simple arithmetic average (k₁+k₂)/2 instead of the weighted formulas
Not checking that keq(parallel) > keq(perpendicular) as a sanity check
Forgetting to weight by thickness — using equal weights when layers have different heights

Related Concepts

Darcy's Law
Anisotropic permeability
Seepage through earth dams
Flow net construction
Terzaghi's consolidation (1D vertical flow)

Common Exam Questions

Example

Three layers: k₁=10⁻³, H₁=1m; k₂=5×10⁻⁴, H₂=2m; k₃=10⁻⁴, H₃=1m. keq = [(10⁻³)(1)+(5×10⁻⁴)(2)+(10⁻⁴)(1)]/4 = 5.25×10⁻⁴ m/s

Approach

Use weighted arithmetic mean. Multiply each k by its thickness, sum the products, divide by total thickness.

Question Type

Compute keq for parallel flow

Example

Same three layers: keq = 4 / [(1/10⁻³)+(2/5×10⁻⁴)+(1/10⁻⁴)] = 4/[1000+4000+10000] = 2.67×10⁻⁴ m/s

Approach

Use weighted harmonic mean. Sum the thicknesses for numerator; sum H/k for each layer for denominator.

Question Type

Compute keq for perpendicular flow

Example

Vertical consolidation drainage → perpendicular; horizontal seepage through an embankment → parallel

Approach

Read the problem carefully. Water flowing horizontally along layers = parallel. Water flowing down through layers = perpendicular.

Question Type

Identify which direction of flow applies

Key Points To Remember

Parallel flow (horizontal): keq = Σ(kiHi)/ΣHi — weighted arithmetic mean, HIGH k dominates
Perpendicular flow (vertical): keq = ΣHi/Σ(Hi/ki) — weighted harmonic mean, LOW k dominates
keq(parallel) ≥ keq(perpendicular) always
For flow nets under dams: horizontal flow uses keq(H), vertical recharge uses keq(V)
Anisotropy ratio keq(H)/keq(V) can be 10 to 1000 in stratified deposits
In anisotropic analysis, effective k = √(kH × kV)

Seepage and Flow Nets

A flow net is a graphical solution to Laplace's equation (∂²h/∂x² + ∂²h/∂y² = 0) for two-dimensional seepage. It consists of two families of curves that are always perpendicular to each other: 1. FLOW LINES (streamlines) — paths along which individual water particles travel. Water never crosses a flow line. 2. EQUIPOTENTIAL LINES — lines of equal total head. Flow is always perpendicular to these lines. The region between two adjacent flow lines is a FLOW CHANNEL. The region between two adjacent equipotential lines is a POTENTIAL DROP. FLOW NET PARAMETERS: Nf = number of flow channels (count the spaces, not the lines) Nd = number of equipotential drops (count the spaces, not the lines) H = total head difference (upstream head minus downstream head) SEEPAGE PER UNIT WIDTH OF STRUCTURE: Q = k · H · (Nf / Nd) [m³/s per metre of width] For anisotropic soils (kH ≠ kV), use: k_effective = √(kH · kV) PROPERTIES OF A VALID FLOW NET: • Flow lines and equipotential lines intersect at 90° • Each element (called a flow element or curvilinear square) should be approximately square (a/b ≈ 1) • The head drop between any two adjacent equipotential lines is H/Nd (equal drop) • The flow in each channel is Q/Nf (equal flow per channel) PRACTICAL NOTE on counting: Board problems always give Nf and Nd. Remember: Nf/Nd is the shape factor of the flow net. A typical flow net under a dam might have Nf = 4 and Nd = 12, giving Nf/Nd = 1/3.

Examples

For a dam 60 m long, total seepage = (4×10⁻⁵)(60) = 2.4×10⁻³ m³/s = 2.4 L/s. This is acceptable seepage for most concrete gravity dams. If the upstream were raised by 2 m (H = 8 m), seepage would increase proportionally to 5.33×10⁻⁵ m³/s per metre — note Q is directly proportional to H.

Scenario

A concrete dam rests on a permeable sand foundation with k = 2×10⁻⁵ m/s. A flow net drawn under the dam yields Nf = 4 flow channels and Nd = 12 equipotential drops. The upstream water depth is 9 m and downstream is 3 m. Compute the seepage per metre length of dam.

Solution

Total head difference: H = 9 - 3 = 6 m Nf/Nd = 4/12 = 1/3 Q = k·H·(Nf/Nd) = (2×10⁻⁵ m/s)(6 m)(4/12) = (2×10⁻⁵)(6)(0.3333) = 4.0×10⁻⁵ m³/s per metre = 4.0×10⁻⁵ m²/s (per unit width)

Applications

Seepage analysis under concrete gravity dams and weirs
Designing filter and drainage systems at the downstream toe of earth embankments
Estimating uplift pressure under dam foundations
Assessing piping risk at downstream exits of levees and dikes
Seepage under sheet pile walls and cofferdam cells

Misconceptions

Counting flow LINES instead of flow CHANNELS for Nf (if there are 5 flow lines bounding the flow region, Nf = 4)
Counting equipotential LINES instead of DROPS for Nd
Using upstream depth instead of head DIFFERENCE for H
Forgetting the 'per unit width' qualifier — Q from flow net is 2D (per metre)

Related Concepts

Laplace's equation (2D seepage)
Phreatic surface (free surface seepage)
Uplift pressure on dam foundations
Exit gradient and piping
Anisotropic permeability

Common Exam Questions

Example

k=3×10⁻⁶ m/s, H=8m, Nf=5, Nd=14. Q=(3×10⁻⁶)(8)(5/14)=8.57×10⁻⁶ m³/s per m

Approach

Directly apply Q = kH(Nf/Nd). Identify each parameter carefully. Note that H is total head DIFFERENCE, not upstream depth.

Question Type

Compute seepage Q from flow net data

Example

H=6m, Nd=12. Head per drop = 0.5 m. At 4th drop from upstream: remaining head = 6 - 4(0.5) = 4 m

Approach

Head per drop = H/Nd. Count equipotential drops from upstream to the point of interest. Subtract from upstream head to get total head at that point.

Question Type

Head loss per drop and pressure head at a point

Example

If total head at midpoint under dam = 5 m and datum is at foundation level, pressure head = 5 m, uplift pressure = 9.81×5 = 49.05 kPa

Approach

Use the flow net to find total head at each point under the base, then compute uplift = γw × pressure head at each point and integrate.

Question Type

Uplift pressure under a dam

Key Points To Remember

Q = k·H·(Nf/Nd) per unit width — the master equation for flow nets
Nf = number of flow CHANNELS (spaces between flow lines), not the lines themselves
Nd = number of equipotential DROPS (spaces), not the equipotential lines themselves
Flow lines and equipotential lines are always perpendicular
Head drop per potential drop = H/Nd
Flow per channel = k·(H/Nd)·(1) per unit width
For anisotropic media: use k_eff = √(kH·kV)
Boundary conditions: upstream face and downstream face are equipotential lines; impermeable boundaries and the phreatic surface are flow lines

Quick Condition (Boiling / Quicksand)

When water seeps UPWARD through a soil, it exerts an upward seepage force on the soil particles. If this upward force becomes large enough, it exactly cancels the effective weight of the soil — the effective stress becomes zero and the soil loses all shear strength. This state is called the QUICK CONDITION, often called 'boiling' or 'quicksand.' DERIVATION OF CRITICAL GRADIENT: For a soil element of unit area and height H with upward seepage: Downward force (submerged weight) = γ' · H = (Gs - 1)/(1 + e) · γw · H Upward seepage force = iw · γw · H At the quick condition, upward seepage force = submerged weight: icr · γw · H = γ' · H Critical hydraulic gradient: icr = γ'/γw = (Gs - 1)/(1 + e) For typical soil (Gs ≈ 2.65, e ≈ 0.65): icr = (2.65 - 1)/(1 + 0.65) = 1.65/1.65 = 1.0 This is why the critical gradient is approximately 1.0 for most natural soils — a useful memory aid! FACTOR OF SAFETY AGAINST PIPING: FS = icr / iexit where iexit is the hydraulic gradient at the exit point (downstream toe of a dam or levee). A minimum FS = 4 to 6 is typically required for important structures. SEEPAGE FORCE CONCEPT: Seepage force per unit volume: j = i · γw (N/m³ or kN/m³) Total seepage force on volume V: J = i · γw · V

Examples

Notice icr ≈ 1.0 (within rounding). If iexit ≥ icr, the soil boils. Here iexit = 0.40 < 0.98, so the soil is not currently boiling, but the margin is insufficient for long-term safety. Increasing the downstream filter thickness or adding a weighted blanket would reduce iexit.

Scenario

A soil has Gs = 2.67 and void ratio e = 0.70. (a) Find the critical hydraulic gradient. (b) If the exit gradient computed from a flow net is iexit = 0.40, find the factor of safety against piping.

Solution

(a) icr = (Gs - 1)/(1 + e) = (2.67 - 1)/(1 + 0.70) = 1.67/1.70 = 0.982 ≈ 0.98 (b) FS = icr / iexit = 0.982 / 0.40 = 2.46 This FS of 2.46 is below the recommended minimum of 4 for most earthwork structures, indicating the need for a drainage filter at the downstream exit to reduce the exit gradient.

FS = 1.16 means the levee foundation is almost at the point of boiling. Any slight increase in upstream head (e.g., during a typhoon flood peak) could push iexit above icr. This scenario is very relevant in Philippine contexts where typhoon-induced flooding regularly overtops levee safety margins.

Scenario

A levee foundation soil has γsat = 19.5 kN/m³ and the upward seepage gradient is 0.85. Check whether quick condition occurs.

Solution

γ' = γsat - γw = 19.5 - 9.81 = 9.69 kN/m³ icr = γ'/γw = 9.69/9.81 = 0.988 Since iexit = 0.85 < icr = 0.988, the soil is NOT yet in a quick condition. FS = 0.988/0.85 = 1.16 The factor of safety is critically low — this levee is near failure!

Applications

Safety assessment of earth dams and levees against piping failure
Sheet pile design for deep excavations to prevent bottom heave
Analysis of cofferdam stability during construction
Downstream filter design to reduce exit gradients
Forensic analysis of embankment failures during typhoon events (Philippines context)

Misconceptions

Thinking quicksand is a special type of soil — it is a CONDITION, not a soil type; any cohesionless soil can become quick
Using total unit weight instead of submerged unit weight: icr = γ'/γw, NOT γsat/γw
Confusing icr ≈ 1.0 as exactly 1.0 — it depends on Gs and e; always compute it
Thinking FS > 1 means safe — minimum FS for important structures is 4 to 6, not just 1

Related Concepts

Effective stress principle (Terzaghi)
Seepage force
Uplift pressure
Filter design criteria (Terzaghi filter rules)
Internal erosion and piping

Common Exam Questions

Example

Gs=2.70, e=0.60: icr = (2.70-1)/(1.60) = 1.70/1.60 = 1.0625

Approach

Use icr = (Gs-1)/(1+e). Alternatively, compute γ' = γsat - γw, then icr = γ'/γw.

Question Type

Calculate icr from soil properties

Example

icr = 0.98, iexit = 0.35. FS = 0.98/0.35 = 2.80

Approach

Find icr from soil properties, get iexit from flow net or given data, then FS = icr/iexit.

Question Type

Compute FS against piping

Example

Soil depth L=3m, icr=1.0. Required head for boiling: h = 1.0 × 3 = 3m above outlet

Approach

At quick condition, i = icr = h/L. Solve for h = icr × L.

Question Type

Find upward head required to cause quick condition

Key Points To Remember

icr = γ'/γw = (Gs-1)/(1+e) ≈ 1.0 for typical soils
Quick condition: upward gradient equals icr → effective stress = 0 → soil liquefies
FS = icr / iexit — factor of safety against piping/boiling
Seepage force per unit volume: j = i·γw (always acts in the direction of flow)
Quick condition can occur at the downstream toe of dams, downstream of sheet piles
Increasing exit gradient (by raising upstream head) is MORE dangerous than people think
Coarse aggregates and filter blankets at downstream toe increase FS by shortening the exit gradient

Practice Problems

Part (c) tests unit conversion — a common board exam twist. Daily flow per unit area = 1.73 m/day, which represents about 1.73 metres of water per day flowing through each square metre of the cross-section. This is substantial for a silty sand and would necessitate active dewatering during excavation.

Problem

PROBLEM 1 (Darcy's Law + Seepage Velocity): A silty sand layer 4 m thick underlies a construction site. The hydraulic gradient across the layer is 0.25. The permeability k = 8×10⁻⁵ m/s and void ratio e = 0.55. Compute: (a) discharge velocity, (b) seepage velocity, (c) volumetric flow per unit area per day.

Solution

(a) v = k·i = (8×10⁻⁵)(0.25) = 2.0×10⁻⁵ m/s (b) n = e/(1+e) = 0.55/1.55 = 0.3548 vs = v/n = (2.0×10⁻⁵)/0.3548 = 5.64×10⁻⁵ m/s (c) Q/A = v = 2.0×10⁻⁵ m/s × 86,400 s/day = 1.728 m/day per m² of area ≈ 1.73 m³/day per m²

This value falls in the silt range (10⁻⁷ to 10⁻⁵ boundary), which is consistent with a fine-grained soil requiring the falling-head method. Key technique: compute the areas from diameters using π/4 × D², and use consistent units throughout (all in mm for area and length).

Problem

PROBLEM 2 (Falling-Head Test): In a falling-head permeability test, the standpipe has a diameter of 12 mm and the specimen is 80 mm diameter and 200 mm long. The water level in the standpipe drops from 1200 mm to 400 mm above the outlet in 185 seconds. Determine k.

Solution

a = π(12)²/4 = 113.1 mm² A = π(80)²/4 = 5026.5 mm² L = 200 mm, h₁ = 1200 mm, h₂ = 400 mm, t = 185 s log₁₀(h₁/h₂) = log₁₀(1200/400) = log₁₀(3.0) = 0.4771 k = (2.303 × a × L)/(A × t) × log₁₀(h₁/h₂) = (2.303 × 113.1 × 200)/(5026.5 × 185) × 0.4771 Numerator: 2.303 × 113.1 × 200 = 52,119.06 Denominator: 5026.5 × 185 = 929,902.5 k = (52,119.06 / 929,902.5) × 0.4771 = 0.05604 × 0.4771 = 0.02674 mm/s = 2.67×10⁻⁵ m/s

The thin clay layer (Layer 2, k₂ = 2×10⁻⁶) completely dominates the vertical permeability — it contributes 98.5% of the total flow resistance in perpendicular flow! Meanwhile, horizontal flow is dominated by the high-k Layer 1. This is extremely common in alluvial deposits in the Philippine lowlands (Pampanga, Cagayan Valley) and explains why artesian pressures persist below confining clay layers.

Problem

PROBLEM 3 (Layered Soil): A river bank has three horizontal layers: Layer 1 (top) — k₁ = 5×10⁻⁴ m/s, H₁ = 1.5 m; Layer 2 (middle) — k₂ = 2×10⁻⁶ m/s, H₂ = 2.0 m; Layer 3 (bottom) — k₃ = 8×10⁻⁵ m/s, H₃ = 1.0 m. Find (a) keq for horizontal flow and (b) keq for vertical flow.

Solution

Total H = 1.5 + 2.0 + 1.0 = 4.5 m (a) PARALLEL (horizontal flow): keq(H) = [k₁H₁ + k₂H₂ + k₃H₃] / (H₁+H₂+H₃) = [(5×10⁻⁴)(1.5) + (2×10⁻⁶)(2.0) + (8×10⁻⁵)(1.0)] / 4.5 = [7.500×10⁻⁴ + 4.000×10⁻⁶ + 8.000×10⁻⁵] / 4.5 = [0.000750 + 0.000004 + 0.000080] / 4.5 = 0.000834 / 4.5 = 1.853×10⁻⁴ m/s (b) PERPENDICULAR (vertical flow): H₁/k₁ = 1.5/(5×10⁻⁴) = 3,000 s H₂/k₂ = 2.0/(2×10⁻⁶) = 1,000,000 s H₃/k₃ = 1.0/(8×10⁻⁵) = 12,500 s Σ(Hi/ki) = 3,000 + 1,000,000 + 12,500 = 1,015,500 s keq(V) = ΣHi / Σ(Hi/ki) = 4.5 / 1,015,500 = 4.43×10⁻⁶ m/s Check: keq(H) >> keq(V) ✓ (by factor ≈ 42)

The total seepage of 0.30 L/s over a 40-m weir is manageable. However, the head drop of 0.375 m per potential drop is needed for computing pore pressures at specific points under the foundation — essential for uplift pressure calculations that govern the structural stability of the weir against overturning.

Problem

PROBLEM 4 (Flow Net Seepage + Uplift): Under a concrete weir, a flow net gives Nf = 5 flow channels and Nd = 16 potential drops. The weir impounds 7.5 m of water; the tailwater depth is 1.5 m. The foundation soil has k = 4×10⁻⁶ m/s and the weir is 40 m long. Find: (a) seepage per unit length, (b) total seepage, (c) head loss per potential drop.

Solution

(a) H = 7.5 - 1.5 = 6.0 m (head difference) Q per m = k·H·(Nf/Nd) = (4×10⁻⁶)(6.0)(5/16) = (4×10⁻⁶)(6.0)(0.3125) = 7.5×10⁻⁶ m³/s per m = 7.5×10⁻⁶ m²/s (b) Total seepage = (7.5×10⁻⁶)(40) = 3.0×10⁻⁴ m³/s = 0.30 L/s (c) Head drop per potential drop = H/Nd = 6.0/16 = 0.375 m per drop

Part (c) tests advanced application — converting a safety factor requirement into a design load. The concept is that a surcharge on the downstream exit zone increases the downward effective stress, effectively raising the resistance to upward seepage forces. This is the principle behind downstream weighted berms used to stabilize levees during flood events — highly relevant engineering practice in flood-prone Philippines.

Problem

PROBLEM 5 (Quick Condition): A clean sand has Gs = 2.65 and e = 0.60. A flow net shows the exit gradient at the downstream toe of a levee is iexit = 0.72. (a) Compute icr. (b) Compute FS against piping. (c) If FS < 4.0, what minimum surcharge load (in kPa) placed on the exit zone would bring FS to 4.0?

Solution

(a) icr = (Gs-1)/(1+e) = (2.65-1)/(1+0.60) = 1.65/1.60 = 1.031 (b) FS = icr/iexit = 1.031/0.72 = 1.43 → Far below 4.0, unsafe! (c) To achieve FS = 4.0, we need the effective icr to be: Required icr_eff = FS × iexit = 4.0 × 0.72 = 2.88 A surcharge load q (kPa) increases the downward effective stress: The upward seepage force per unit volume = iexit × γw = 0.72 × 9.81 = 7.063 kN/m³ For equilibrium with FS = 4.0, the effective vertical stress per unit depth must be: σ'v_required = (seepage force per unit volume) × FS = 7.063 × 4.0 = 28.25 kN/m³ Natural submerged weight alone = γ' = (Gs-1)γw/(1+e) = 1.65×9.81/1.60 = 10.11 kN/m³ The surcharge must provide the remaining resistance per unit depth. For a 1-m exit zone: Required surcharge = (28.25 - 10.11) × 1.0 = 18.14 kPa ≈ 18.1 kPa NOTE: In practice, a weighted filter blanket of gravel (γ ≈ 20 kN/m³) about 0.9 m thick would provide this surcharge while also lengthening the seepage path.

Exam Preparation Tips

MEMORIZE THE FOUR MASTER FORMULAS: (1) Q = kiA, (2) k_const = VL/(Aht), (3) k_fall = (2.303·a·L/A·t)·log(h₁/h₂), (4) icr = (Gs-1)/(1+e). Every board problem on permeability traces back to one of these.
DISTINGUISH DISCHARGE vs. SEEPAGE VELOCITY: Discharge velocity v = ki uses the gross area. Seepage velocity vs = v/n uses the pore space. If asked about contaminant travel time or actual pore-water velocity, always use vs. Writing 'vs > v always' in your scratch pad prevents the switch-up error.
PARALLEL vs. PERPENDICULAR LAYERED FLOW: Draw a quick sketch showing flow direction and layer orientation. Flow along layers = parallel = arithmetic weighted mean (high k wins). Flow through layers = perpendicular = harmonic weighted mean (low k wins). Double-check with the sanity rule: keq(parallel) ≥ keq(perpendicular) always.
FLOW NET COUNTING — CHANNELS NOT LINES: The most common error is counting the number of flow lines instead of the spaces (channels) between them, and similarly for equipotential lines. If you see 5 flow lines bounding 4 channels, Nf = 4. Practice with published flow net sketches before the exam.
CRITICAL GRADIENT APPROXIMATION: icr ≈ 1.0 for most natural soils (Gs ≈ 2.65, e ≈ 0.65). Use this to quickly check your computed answer. If you get icr = 3.5 or 0.2, recheck your substitution.
UNIT CONSISTENCY IS MANDATORY: In constant-head and falling-head formulas, the result for k has units of [length/time]. Before computing, convert ALL lengths to the same unit (all mm or all m). A single cm vs. mm mismatch gives an answer off by a factor of 10.
FACTOR OF SAFETY AGAINST PIPING: FS = icr/iexit. Know that a FS of 1.43 is NOT acceptable for permanent structures — minimum is 4 to 6. If the board problem says 'determine if the structure is safe' with FS = 2.5, the answer is NO.
TEMPERATURE EFFECT ON k: Permeability k is proportional to 1/μ (dynamic viscosity). Water at 20°C has lower viscosity than at 4°C, so k is slightly higher at warmer temperatures. The board exam may give a temperature correction factor — apply it directly to k.
HAZEN'S FORMULA FOR QUICK ESTIMATES: For clean sands, k ≈ C·D₁₀² where D₁₀ is in cm and k in cm/s, C ≈ 100 to 150. This is useful for checking order of magnitude in sand problems.
FLOW NET SEEPAGE IS 'PER UNIT WIDTH': Always note that Q = kH(Nf/Nd) gives m³/s per metre of structure width. For a dam 50 m long, multiply by 50 to get total Q. Many examinees forget this and leave the answer as per-unit-width when total is asked.
QUICK CONDITION vs. PIPING: Quick condition (boiling) occurs when i ≥ icr over a large area. Piping (internal erosion) can occur at lower gradients if fine particles are progressively washed out. The board typically tests the quick condition (icr formula); piping design requires filter criteria (Terzaghi's filter rules).
USE GIVEN POROSITY OR COMPUTE FROM e: If e is given, compute n = e/(1+e). Do NOT assume n = e. For e = 0.65, n = 0.65/1.65 = 0.394, NOT 0.65. This mistake leads to a 65% error in seepage velocity calculations.

Loading diagram…

In summary

Permeability and seepage form a cornerstone of geotechnical engineering practice and consistently appear in the PRC Civil Engineer Licensure Examination. The discipline integrates fundamental soil mechanics (void ratio, unit weights, effective stress) with fluid mechanics (hydraulic gradient, Darcy's law, Bernoulli's equation) to produce practically vital tools: laboratory k measurements, layered-system equivalents, graphical flow-net analysis, and stability checks against the quick condition. For the board exam, your success in this topic rests on three habits: 1. FORMULA ACCURACY: Memorize the four master equations (Q = kiA; constant-head formula; falling-head formula; icr = (Gs-1)/(1+e)) and know the physical meaning behind each variable. Never swap the 'a' (standpipe) and 'A' (specimen) in the falling-head formula. 2. CONCEPTUAL CLARITY: Understand WHY parallel flow yields the arithmetic mean (high k layer acts as shortcut) and WHY perpendicular flow yields the harmonic mean (low k layer is the bottleneck). Understanding prevents formula confusion under exam pressure. 3. PHYSICAL REASONABLENESS: Always check your k answer against typical ranges — clay should give 10⁻⁹ to 10⁻⁷ m/s, sand 10⁻⁵ to 10⁻³ m/s. Check that seepage velocity exceeds discharge velocity. Verify that keq(parallel) ≥ keq(perpendicular). And remember that icr ≈ 1.0 for most natural soils. In the Philippine context, these concepts are not merely academic — proper application prevents dam failures, slope collapses, and flooding of communities during typhoon season. The engineer who masters permeability and seepage analysis contributes directly to national resilience and public safety, fulfilling the mandate of Republic Act 544 (Civil Engineering Law of the Philippines) to protect life and property through competent engineering practice. Practice the exercises systematically, internalize the flow-net counting rules, and approach each problem by first identifying soil type, flow direction, and the specific formula that applies. With thorough preparation, this topic will be a reliable source of correct answers — not a source of lost points — on your board examination.

Previous chapter

Soil Classification

Next chapter

Stresses in Soil Mass

Ready to practise for the CELE 2026?

Super Tutor's AI review plan adapts to your weak areas and builds a weekly practice schedule around your target CELE exam date.

Start free See the full CELE guide CELE preparation guide

← Back to Geotechnical Engineering hub