Reading a Soil Report

Four Numbers, and Only One of Them Is a Unit Weight

Borehole BH-3, sample at 4.2 m, and the laboratory has sent back one sheet. A grading curve, liquid and plastic limits, a classification symbol, water content 18.4 per cent, dry density 1.66 Mg/m³, specific gravity of solids 2.68. On the other side of the desk is a retaining wall check that wants the moist unit weight of the retained soil above the water table and the buoyant unit weight of everything below it. Neither figure appears anywhere on the sheet, and neither is missing. They are derived, and the derivation is four lines long.

The laboratory stopped where it did on purpose. Dry density is something a technician measured on a specimen of known volume; saturated unit weight is a claim about a condition that specimen was never in. The moment a lab writes γsat it has assumed the voids are completely full of water — a statement about your site rather than about the material in the mould — and a buoyant weight assumes a water table it has not seen. That division is correct, and it is why the conversion belongs to the engineer holding the drawing.

All of it runs on one idea. Take any volume of soil and it is three things — mineral solids, water, and air — and every quantity on the sheet is a ratio between two of them. Specific gravity fixes how heavy the solids are compared with water. Dry density fixes how much solid there is in a given volume. Between them they fix everything else on the page, and the quantity that carries the information from one to the other is the void ratio.

Density, Unit Weight, and the Number That Is the Same in Both

A metric laboratory reports density, in Mg/m³ or g/cm³, and the calculators here take unit weight. Density is mass per unit volume and unit weight is force per unit volume, so the conversion is gravity: a dry density of 1.66 Mg/m³ — identical to 1.66 g/cm³, the sheet may use either — becomes 1.66 × 9.81 = 16.28 kN/m³. That single multiplication is the most common thing to leave out, and leaving it out understates the weight of the ground by an order of magnitude, which is large enough that it usually gets caught. The dangerous errors are smaller.

Imperial practice hides the distinction, and hides it in a way that never fails locally. The same soil is 103.6 lb/ft³ as a mass density and 103.6 lbf/ft³ as a unit weight, because the pound-mass and the pound-force were defined to make that true at standard gravity. A sheet reading 103.6 pcf is therefore correct whichever quantity the reader believes it to be, and nothing goes wrong until somebody converts to SI and has to pick. One lbf/ft³ is 0.1571 kN/m³, so 103.6 × 0.1571 = 16.28 kN/m³ — the same soil by the other road. Water is the reference at either end, 9.81 kN/m³ or a shade over 62.4 lbf/ft³, and a conversion that produces a soil lighter than water has gone in upside down.

The dry unit weight field follows the metric and imperial switch at the top of the page, reading kN/m³ or pcf, and a unit weight the calculators return comes back in the same unit; specific gravity and void ratios are pure ratios and read the same either way. The switch cannot turn a density into a unit weight: a metric sheet's Mg/m³ is multiplied by gravity before it goes in, as above, while a pcf reading goes in as it stands. The unit weight of water is fixed internally at 9.81 kN/m³, which is fresh groundwater; a pore fluid that is brine, a tailings liquor or a contaminated leachate has a different density and the phase relationships have to be worked longhand with the right value in them.

Specific Gravity Is the Number You Can Borrow, Within Limits

The specific gravity of soil solids is measured with a water pycnometer under ASTM D854, with AASHTO T 100 as the highway equivalent, BS EN ISO 17892-3 as the European particle density method and the corresponding parts of BS 1377 and AS 1289 elsewhere. It is reported against a reference temperature with a correction applied, because water's own density moves with temperature, and a sheet that gives a specific gravity without saying at what temperature has left out part of the result. For ordinary quartz and feldspar soils the answer lands between about 2.60 and 2.75 with dull regularity, which is exactly why it is the number people stop measuring.

There is a limit on the borrowing that catches gravelly fills. ASTM D854 is written for the material passing the 4.75 mm sieve; the coarse fraction is tested separately as a coarse aggregate under ASTM C127 and the two results are combined by mass into a composite value. A single specific gravity quoted for a sandy gravel has, in most laboratories, described only the matrix, and if the gravel is a different rock from the sand — a limestone gravel in a quartz sand, a slag or a recycled concrete — the composite is genuinely different from either. Ask what was tested before assuming the number covers the whole sample.

It helps to know what a borrowed value costs. On the BH-3 sample, a Gs of 2.68 with a dry unit weight of 16.28 kN/m³ gives a void ratio of 0.615. Push the specific gravity to 2.73 — roughly the width of the plausible mineral range — and it becomes 0.645, five per cent higher. Leave it alone and let the dry density be three per cent low at 15.79 kN/m³ instead, and the void ratio goes to 0.665, eight per cent out. The measurement least trusted moves the answer less than a modest error in the one most trusted. That collapses for organic and micaceous soils, iron-rich residual soils and manufactured fills such as ash or slag, where the specific gravity is not 2.65 and no convention makes it so.

Percent Compaction and Relative Density Answer Different Questions

Percent compaction measures a field dry density against a laboratory maximum obtained by impact compaction at an optimum water content, under ASTM D698 or ASTM D1557. Relative density measures a field void ratio against the loosest and densest states obtainable by the index methods. Those are different reference states produced by different energies, and there is no general conversion between them — a percentage on one scale cannot be translated to the other for an arbitrary soil. For clean, free-draining granular materials the impact compaction curve is often poorly defined in the first place, with no clear peak to call a maximum, and that is precisely the situation the index density tests were written for.

The backfill above makes the point concretely. Its standard effort maximum dry density is 1.815 Mg/m³, so the field value of 1.723 Mg/m³ is 94.9 per cent compaction — a fail against a 95 per cent specification, by a tenth of a per cent, on a soil the relative density test calls dense at 72.6 per cent. Neither is wrong; they are two rulers with different zero points laid against the same sand. The argument on site comes from a specification naming both with limits chosen independently, so a load can pass one clause and fail the other. Decide which governs before the compaction plant arrives, and let the other be information.

Where the Fields Stop, and What That Is Telling You

Every field on the calculators here carries a range, and the field states it. Those are not decoration. A value typed outside them is clamped to the nearest end and the field says out loud what it did and what you entered, which is the right behaviour — the alternative is an answer that looks the same as any other answer and was computed from a number you did not supply. If a clamp notice appears, read it rather than working around it.

The edges of those ranges are informative. Soft organic clays and peats sit outside them twice over: their dry unit weight can fall below the field's floor, and their specific gravity below its floor too, organic matter being far lighter than mineral grain, which is what ASTM D2974 exists to quantify. The phase relationships stay valid at any void ratio — they are definitions, not correlations — so run those soils by hand and note on the sheet that the calculators were outside their stated range.

What the Sheet Does Not Say

A specific gravity is frequently run once, on a composite made from several samples, and then applied to every density in the report. That is defensible where the deposit is uniform and quietly wrong where it is not — a made-ground fill with pockets of brick rubble, ash and clay has no single particle density, and the void ratios computed for it inherit whichever sample dominated the composite. The report will not tell you this unless you ask which specimens the pycnometer test was run on, and that question is usually answered by return of email.

Then there is what happened to the sample between the ground and the mould. A thin-walled tube specimen taken under ASTM D1587 can swell, drain or lose its structure in transit, and the dry density that comes off it belongs to the tube rather than to the ground. In-place methods avoid that — the sand cone of ASTM D1556 and the nuclear gauge of ASTM D6938 measure the fill where it lies — but they measure only the top of a lift and they measure it where the technician chose to stand. Neither limitation is a reason to distrust the number; both are reasons to know how many tests produced it and how far apart they were.

Finally, the water table on the report is a reading with a date on it rather than a property of the site. A standpipe dipped at the end of a dry summer describes a different wall from the same standpipe in late winter, and perched water over a clay layer or a river-linked groundwater regime moves it further still. Where the design is sensitive — and a wall retaining three metres below the water table is very sensitive — the honest input is the highest credible level with a justification attached, which is how EN 1997-2 frames the water regime: something established, not assumed.

The Line You Write Down

What comes out of an hour of this is four unit weights, a void ratio, a relative density where the soil is granular, and — the part that gets skipped — a note of where each one came from. Write the converted figure next to the original so the multiplication can be checked without opening the laboratory report again. Name the test and the sample against each derived value. State the water table depth and the date it was dipped beside the buoyant weight, because that weight is only meaningful with the level attached.

The reason is not tidiness. These four numbers will be read by a structural engineer sizing the wall, by whoever runs the bearing check on the footing, by the person correcting blow counts for overburden, and eventually by somebody investigating a movement three years after handover. All four will assume the values were derived rather than measured, and only one of them will be in a position to ask what they were derived from. The sheet you write is the answer to that question.

1. Convert every reported density to kN/m³ before touching anything else, and keep the original figure and its units written alongside.
2. Establish which specimens the specific gravity was run on, and whether a gravel fraction was tested separately and combined.
3. Compute the void ratio from specific gravity and dry unit weight, and read the porosity beside it as a plausibility check against the soil description.
4. Put that void ratio back through the unit weight relation and confirm it returns the dry unit weight you began with.
5. Apply the water content to get the moist unit weight, then compute the degree of saturation and reconcile it with the reported water table.
6. Subtract the unit weight of water from the saturated value for the buoyant weight, and use it below the water table only, with hydrostatic pressure carried on its own line.
7. For granular fill, compute relative density from index void ratios measured on that same material, and record the percentage rather than the descriptive band.
8. Write each figure down with the test standard, the sample reference and the date it came from, on the same page as the calculation that consumes it.