Why Sand Drains Far Faster Than Silt: The Permeameter Test

Grain-size formulas give you an estimate of hydraulic conductivity. A permeameter gives you a measurement. This video is the whole laboratory session, built so it can stand in for the bench if you cannot be there. Every piece of equipment is laid out and named first — the cell, the porous end plates, the constant-head tank and its overflow, the standpipe, the measuring cylinder, the stopwatch, the thermometer and the balance — and then the inside of the cell, because two of the three quantities you measure live in there. Then the procedure, twice. The constant-head test runs end to end as an animation before it is taken apart step by step: pack in lifts and record the porosity, saturate slowly from the base so no air is trapped, wait for the overflow to run steadily, and measure h across the sample rather than from the bench. The falling-head test then runs end to end as its own animation, for material too fine for the first method. Two worked examples: 14.4 m/day for a clean sand and 0.118 m/day for a silty sample — a factor of about 120 from the same rig on the same afternoon. Plus the temperature correction to 20 °C, which nobody remembers: water at 10 °C is 1.30 times as viscous as at 20 °C, so an uncorrected cold reading comes out about 23 % low. Ends with the three ways this goes wrong — trapped air, a head that is not constant, and leakage down the cylinder wall — and the cheapest quality check in the lab. CHAPTERS 0:00 Same sand, two groups, a factor of two 0:56 The equipment, and the cell that holds L and A 2:11 Preparing the sample: packing and saturation 3:24 Constant head: the method for coarse samples 5:40 Falling head: the method for fine samples 7:22 One correction: reporting K at 20 °C 8:01 The comparison: a factor of 120 from one rig 8:36 Trapped air, a drifting head, sidewall leakage 10:13 Measure it, then say how SUBTITLES English, Español, हिन्दी, 日本語, 한국어 and 中文（简体） — choose one from the CC button. THE NUMBERS ON SCREEN *Constant head.* L = 15 cm, A = 25 cm², h = 5.0 cm, V = 100 mL in t = 720 s. K = VL/(Ath) = 1.67 × 10⁻² cm/s = 14.4 m/day. *Falling head.* d_tube = 2.0 cm, d_cell = 5.64 cm, L = 15 cm, head 5.0 → 0.50 cm in 528 min. K = (d_t²L)/(d_c²t) · ln(h₀/h), with ln 10 = 2.303, giving 1.37 × 10⁻⁴ cm/s = 0.118 m/day. Ratio between the two samples: about 120 (121.7). *Temperature correction.* K₂₀ = K_measured × μ_T/μ₂₀, using the standard dynamic viscosity of pure water (1.307 mPa·s at 10 °C, 1.002 at 20 °C, 0.798 at 30 °C). The quoted factors are ×1.30 at 10 °C and ×0.80 at 30 °C. The K range chart follows Table 3.2 of the textbook. Both worked examples were computed for this video and checked line by line; they are not taken from a published dataset. WHERE IT SITS IN THE SERIES Video **06**. It follows 05 (sieve analysis), which estimates K from grain size for one sample and refuses the estimate for the other as out of Hazen's range. This video measures instead of estimating, and the gap between the two is worth its own section. It is followed by measurement, and it sets up 07 (pumping test), which measures the same property at field scale and compares the two numbers explicitly. #hydrogeology #geotechnical #permeability