Measuring Sand Filtration Coefficient With Falling Head Tests

The falling head permeability test measures how readily water moves through a sand layer under a controlled hydraulic gradient. It is commonly used to determine the coefficient of permeability, also called hydraulic conductivity or the filtration coefficient, particularly when the soil is fine enough for water flow to be measured over a practical period.

For road bases, drainage trenches, embankments, bridge approaches and landscaped pedestrian areas, the result helps engineers predict seepage, pore-pressure response and drainage performance. A carefully performed test is especially valuable when design assumptions must be supported by laboratory evidence, rather than by grain size alone.

What The Filtration Coefficient Represents

The filtration coefficient, usually written as k, describes the rate at which water can pass through a soil under a known hydraulic gradient. A high value indicates that water travels relatively easily through connected voids, while a low value indicates finer pores, poorer connectivity or greater resistance to flow. Clean, uniformly graded sand generally has a higher k than silty sand or compacted sand containing a substantial fine fraction.

Darcy’s law provides the basic relationship:

q = k i

Here, q is the discharge velocity and i is the hydraulic gradient. In a laboratory specimen, the gradient is calculated by dividing the difference in water head by the length of the sample. The measured flow and specimen dimensions then allow the coefficient to be calculated.

The term “filtration coefficient” can be confusing because it does not describe a filter rating or a percentage of particles removed. In geotechnical work, it normally refers to water permeability. The result must therefore be interpreted alongside particle-size distribution, density, saturation condition, degree of compaction and soil structure.

For Australian projects, engineers may compare laboratory procedures with relevant Australian Standards, project specifications and NATA-accredited laboratory requirements. Local practice also needs to account for materials that behave differently across regions, such as clean alluvial sands around Brisbane, crushed sand blends used in Western Australian works, or fine coastal deposits near Sydney and Melbourne.

How A Falling Head Test Works

A representative sand specimen is placed inside a rigid permeameter between porous stones or screens. The sample is confined to a known cross-sectional area and length. Water is introduced through the specimen, and trapped air is removed as far as practical. Saturation is essential because air in the pores can interrupt flow and make the apparent permeability lower or less stable than the true value.

A narrow standpipe is connected to the inlet or outlet of the permeameter. Water initially stands at a high level in the tube. As water flows through the sand, the level falls. The operator records the initial head, final head and elapsed time. Unlike a constant head test, the driving head changes continuously, so the calculation uses the logarithm of the ratio between the two measured heads.

For a conventional arrangement, the coefficient is calculated using:

k = (aL / At) ln(h₁ / h₂)

In this expression, a is the internal area of the standpipe, L is the specimen length, A is the specimen cross-sectional area, t is elapsed time, and h₁ and h₂ are the initial and final hydraulic heads. The heads must be measured from the same reference level and expressed in compatible units.

The operator usually repeats the run until successive readings show reasonable agreement. Water temperature should be recorded because viscosity changes with temperature. Results are often corrected to a reference temperature, commonly 20°C, when the project method requires it. A laboratory report should state the testing temperature, sample condition, density, saturation method, head range and any correction applied.

Useful background on laboratory quality systems and measurement practice can be found through engineering resources when developing a broader investigation or verification programme.

Preparing And Running The Specimen

Sample preparation has a direct effect on the measured coefficient. If sand is poured loosely into the mould, the result may represent a high-void-ratio condition. If it is compacted in layers, the flow path becomes narrower and k generally decreases. The report must identify the target density or relative density, the placement method and the amount of compaction applied.

Disturbed samples can be suitable when the purpose is to assess a specified reconstructed material, such as a pavement drainage layer. An undisturbed sample is more appropriate when the natural fabric, bedding or cementation is important. In a layered deposit, the sample orientation should be recorded because permeability parallel to bedding can differ greatly from permeability across bedding.

The specimen must be sealed against side leakage. Even a small bypass between the soil and the permeameter wall can produce an unrealistically high result. Porous discs should be clean and compatible with the soil grading, while filters should prevent particle loss without creating a significant additional resistance.

The following comparison helps determine whether a falling head arrangement is appropriate:

Test feature Falling head permeability Constant head permeability
Best suited soil Fine sand, silty sand and soils with relatively low permeability Clean, coarse sand and gravel
Driving head Decreases during the test Remains approximately constant
Main measurement Time for water level to fall between two heads Volume of water collected over time
Calculation Uses standpipe area and logarithmic head ratio Uses steady discharge and hydraulic gradient
Practical advantage Sensitive to small flow rates Efficient when discharge is high
Main limitation Sensitive to air, leaks and reading errors Can be impractical when flow is very slow

A falling head test is not automatically suitable for every sand layer. Very coarse sand may drain through the apparatus too rapidly for accurate timing, while a soil with considerable clay may require a different test duration and saturation approach. The selected method should match the expected permeability and the intended field condition.

Checking Results Against Field Conditions

A laboratory coefficient is a controlled measurement, not a complete description of a site. Field layers may contain lenses, construction joints, roots, gravel pockets or discontinuities that are absent from a small specimen. For a road shoulder in regional Victoria or a drainage corridor in Western Sydney, the design team may need several samples to capture changes across the alignment.

Compaction is another important source of variation. A sand layer installed beneath a shared path or pavement may be placed at a specified moisture condition and compacted with plant that produces a different structure from the laboratory preparation. The design value should reflect the expected construction density rather than simply adopting the largest measured coefficient.

Hydraulic conductivity also changes with water viscosity. If a test is conducted in a warm laboratory and the result is compared directly with field conditions during a cooler winter period, the difference may matter for sensitive seepage or drainage calculations. Temperature correction improves comparison, but it does not remove uncertainty caused by soil fabric or field heterogeneity.

The result should be considered with infiltration demands, groundwater levels and drainage geometry. In parts of Queensland, intense summer rainfall can create short-duration flows that exceed the rate suggested by a simple average permeability value. In Perth, highly permeable sands may allow rapid infiltration while still presenting problems if fines migrate into the layer. Sydney projects may need to consider variable fill and weathered residual soils near the same alignment.

Reporting And Applying The Data

A reliable report states exactly what was tested and how. At a minimum, it should identify the project, sample location and depth, soil description, specimen dimensions, density, water temperature, saturation method, head values, elapsed time, calculated coefficient and units. Photographs or sample logs can help connect the laboratory result with the field material.

Quality control should include equipment calibration, leak checks, verification of standpipe dimensions and review of raw observations. A result should not be discarded merely because it differs from another specimen; variation may reveal changes in grading or compaction. However, unexplained scatter should trigger a review of sample preparation, air removal, seals and timing records.

Two short checklists can support consistent technical review.

Before testing

Before issuing the result

Permeability is one part of a broader infrastructure assurance process. A sand layer beneath concrete drainage structures, for example, may need permeability assessment alongside concrete strength and durability checks. Where flexural performance is also relevant, flexural testing guidance can help place that separate measurement within a coordinated quality programme.

For government and private-sector work in Jakarta, a laboratory and field investigation unit operating under a public works department can apply similar principles while working to SNI ISO/IEC 17025 and ISO 45001 requirements. Traceable records, controlled procedures and safe field practice make the results more useful during planning, construction, maintenance and defect assessment.

When the measured coefficient is used in a design, engineers should distinguish between a laboratory value, a representative design value and a conservative allowable value. That decision may involve statistical review, safety factors, anticipated clogging and the consequences of poor drainage. The number is most valuable when its test conditions and limitations remain visible throughout the project documentation.

Arrange a controlled falling head permeability investigation for the sand layers in your road, bridge, drainage or pedestrian infrastructure project. A qualified testing unit can help define the sampling plan, prepare representative specimens, verify the calculation and report a defensible filtration coefficient for design and construction decisions.