Lesson 3: Gravel, sand, silt or clay?
Particle size and grading curves · about 40 minutesYou will learn
What size are the grains?
Pick up a handful of soil and the first question an engineer asks is simple: how big are the grains? The answer tells you a surprising amount about how the soil will behave. In this lesson you will learn how engineers sort soils by grain size, and how to read the graph they use to show it.
- Name the main soil types by particle size, and tell them apart by look and feel.
- Turn sieve test results into a grading curve, and read it to describe the soil.
You only need a calculator. If you can, have a pinch of soil from a garden or park nearby for the hands-on step.
Why it matters
Soil comes as it is
Steel and concrete are made to order. Soil is not. Engineers have to work with whatever nature left on the site, a mix of grains that came from weathered rock and was moved around by water, wind, ice and gravity. No two soils are exactly alike.
To make sense of this variety, engineers sort soils into groups that behave in similar ways. The most useful thing to sort by is grain size, because it controls so much:
- How easily water flows through it. Water rushes through gravel but barely moves through clay.
- How strong it is, and where that strength comes from.
- How much it squashes under a building, and how quickly.
Two words you will meet
Classification sorts a soil into a group with similar behaviour, using tests such as the ones in this lesson.
Description is a written statement of what the soil is and what state it is in, for example “firm brown sandy CLAY”. A classification is usually part of a description.
The idea, in plain words
Four main soil types
From biggest to smallest grains, the four main soil types are gravel, sand, silt and clay. Gravel and sand are coarse soils: you can see the grains. Silt and clay are fine soils: the grains are too small to see.
Silt, sand and gravel are each split again into fine, medium and coarse:
| Soil | Fine | Medium | Coarse |
|---|---|---|---|
| Gravel | 2 to 6.3 mm | 6.3 to 20 mm | 20 to 63 mm |
| Sand | 0.063 to 0.2 mm | 0.2 to 0.63 mm | 0.63 to 2 mm |
| Silt | 0.002 to 0.0063 mm | 0.0063 to 0.02 mm | 0.02 to 0.063 mm |
These are the boundaries used in the UK, from BS EN ISO 14688-1. Notice the pattern: each boundary is about three times the one before (0.2, 0.63, 2, 6.3, 20, 63).
Wherever you are
Not every system draws the lines in the same place. The Unified Soil Classification System, used widely around the world, puts the sand to silt boundary at 0.075 mm and the gravel to sand boundary at 4.75 mm. Agricultural systems use different lines again. The idea is always the same. Before you use a number, check which system it comes from, and say so in your work.
The idea, in plain words
Telling soils apart with your hands
Before any lab test, engineers describe soil by eye and by hand. With practice, you can get surprisingly close.
Gravel and sand
You can see the separate grains. Rub a little between your fingers and it feels gritty, even fine sand. Dry sand runs through your fingers.
Silt and clay
You cannot see the grains. Wet, they feel smooth. To tell silt from clay, try the three checks below.
| Check | Silt | Clay |
|---|---|---|
| Shake test. Put a small wet pat in your palm and shake it. | Water rises to the surface and it looks shiny. Squeeze it and the water disappears. This is called dilatancy. | Little or no change. |
| Feel when wet | Slightly gritty or floury, not very sticky. | Smooth and very sticky. |
| When dry | Powdery. Brushes off your hands. | Hard lumps. Needs washing off. |
Try it (5 minutes)
If you have a pinch of soil, wet it slightly and try the shake test, then rub it between your fingers. Is it gritty, smooth, sticky or floury? Write down what you think it is. Real soils are usually mixtures, so “mostly sand with some silt” is a perfectly good answer.
Hand tests tell you whether a soil is coarse or fine quite reliably. For mixtures, engineers measure the sizes in the lab. That is the rest of this lesson.
Measuring it
The sieve test
For grains bigger than 0.063 mm, the size is measured by sieving. A stack of sieves sits with the largest openings at the top and the smallest at the bottom, with a pan underneath.
In the UK, BS EN ISO 17892-4 asks for sieves at the boundary sizes you met earlier: 63, 20, 6.3, 2.0, 0.63, 0.2 and 0.063 mm, plus enough others to show the shape of the curve properly.
- Dry sieving works for clean sands and gravels.
- Wet sieving is used when the soil contains silt or clay. The soil is washed through the sieves with water, so the fine grains do not stay stuck to the bigger ones. Each portion is then dried and weighed.
- The check: add up all the retained masses, including the pan. If the total differs from the starting mass by more than 1%, the test is repeated.
Dry sieving clayey soil hides the fines
If a soil with clay in it is dry sieved, the clay clings to the sand and gravel and gets counted as coarse grains. The soil then looks cleaner and coarser than it really is, and its drainage and strength can be badly overestimated. When I check lab results, I look for which method was used before I trust a grading curve.
Worked example
From masses to “percentage passing”
A sample of soil weighing 500.0 g (oven dry) was wet sieved. Here are the results, and how each column is worked out:
| Sieve (mm) | Mass retained (g) | % retained | Cumulative % retained | % passing |
|---|---|---|---|---|
| 6.3 | 0.0 | 0 | 0 | 100 |
| 2.0 | 15.0 | 3 | 3 | 97 |
| 0.63 | 85.0 | 17 | 20 | 80 |
| 0.2 | 160.0 | 32 | 52 | 48 |
| 0.063 | 150.0 | 30 | 82 | 18 |
| Pan | 90.0 | 18 | 100 | – |
| Total | 500.0 | 100 |
Check: the retained masses add up to 500.0 g, the same as the starting mass, so the test passes the 1% check.
Each row gives one point for the graph: (sieve size, % passing). For example, 80% of the soil is finer than 0.63 mm.
Notice the pan
18% of this soil passed the smallest sieve, so 18% is silt and clay. Sieving cannot tell us how that 18% splits between silt and clay. For that we need the next step.
Measuring it
Below 0.063 mm: let the grains settle
Sieves with openings finer than 0.063 mm are not practical, so the fine grains are measured a different way. The fine soil is mixed with water (and a chemical that stops the grains clumping together) and left to settle.
Big grains sink faster than small ones. This is described by Stokes’ law: a grain’s settling speed rises with the square of its size, so a grain twice as big settles four times as fast. By measuring how much soil is still in suspension at set times, from 30 seconds to a day or more, the lab works out the size of grains still settling.
Hydrometer method
A floating instrument measures the density of the liquid as the grains settle out.
Pipette method
Small samples of liquid are drawn off at a set depth and dried to weigh the soil in them.
This is usually only worth doing when the soil has a significant amount of fines, more than about 10%.
One step people forget
The settling test only uses the fine part of the soil, so its results are percentages of the fines, not of the whole sample. Scale them before plotting. In our example, fines are 18% of the sample. If the settling test says 22% of the fines are finer than 0.002 mm, then:
22% × 0.18 = 4% of the whole sample is clay-sized.
Our sample’s settling results, already scaled, are: 12% finer than 0.02 mm, 7% finer than 0.006 mm and 4% finer than 0.002 mm.
Reading the graph
The grading curve and its shape
All the points are plotted on a grading curve (also called a particle size distribution, or PSD). Particle size goes along the bottom on a log scale, because the sizes range from microns to centimetres. Percentage passing goes up the side. The shape of the curve tells you a lot at a glance.
| Shape | What it means |
|---|---|
| A: Well graded | A smooth, gently sloping curve across a wide range of sizes. Small grains fill the gaps between big ones, so the soil packs tightly. |
| B: Uniform (uniformly graded) | A steep, almost vertical curve. Most grains are about the same size, like beach sand. |
| C: Gap graded | A flat section where a range of sizes is missing. Here the soil has fine and coarse grains, but very little between about 0.6 and 6 mm. |
Avoid “poorly graded”
Uniform and gap-graded soils are sometimes both called “poorly graded”. The term is vague, because it lumps two very different shapes together. Say uniform or gap graded instead, so the reader knows exactly what you mean.
Reading the graph
Reading off how much of each soil type
Here is our worked example plotted as a grading curve. Any point on the curve means “this percentage of the soil is finer than this size”.
Each soil type is the difference between two readings:
| Soil type | Working | Amount |
|---|---|---|
| Gravel (bigger than 2 mm) | 100 − 97 | 3% |
| Sand (0.063 to 2 mm) | 97 − 18 | 79% |
| Silt (0.002 to 0.063 mm) | 18 − 4 | 14% |
| Clay (smaller than 0.002 mm) | 4 − 0 | 4% |
The biggest share is the primary soil type, written in capitals. The next biggest is the secondary type, written as an adjective in front. Add the grading word from the shape of the curve, and you have a short description:
Well graded silty SAND
A full engineering description adds more, such as colour, density or strength, and uses set percentages to decide words like “slightly silty”. You will learn those rules in Lesson 5.
Reading the graph
D10, D30, D60 and the grading coefficients
“Well graded” can also be measured. Read three sizes off the curve:
- D10: the size that 10% of the soil is finer than. It is called the effective size, because the smallest grains control how easily water can flow through the soil.
- D30 and D60: the sizes that 30% and 60% of the soil are finer than.
From our curve: D10 = 0.012 mm, D30 = 0.10 mm, D60 = 0.31 mm. They give two coefficients:
| Soil | Well graded if |
|---|---|
| Gravel | Cu of at least 4, and Cc between 1 and 3 |
| Sand | Cu of at least 6, and Cc between 1 and 3 |
Our sand has Cu ≈ 26 and Cc ≈ 2.7, so it is well graded, matching the smooth shape of its curve. A perfectly single-sized soil would have Cu = 1 and Cc = 1.
Wherever you are
The limits in the table come from the Unified Soil Classification System, which is used for clean sands and gravels in many countries. Other standards, including the UK soil description standard BS EN ISO 14688-2, set their own bands for words like “well graded”. The coefficients are calculated the same way everywhere. When you quote a grading term, say which rule you used.
Your turn
Describe this soil
Task (15 minutes)
A 500.0 g oven-dry sample was sieved. The masses retained were: 2.0 mm: 0 g; 0.63 mm: 10 g; 0.2 mm: 380 g; 0.063 mm: 100 g; pan: 10 g.
- Check the masses against the starting mass.
- Work out the % passing each sieve.
- Work out the % of gravel, sand and fines.
- The curve gives D10 = 0.10 mm, D30 = 0.23 mm and D60 = 0.36 mm. Work out Cu and Cc. Is the sand well graded?
- Write a short description.
Show the answers
1. 0 + 10 + 380 + 100 + 10 = 500 g, the same as the starting mass. The check passes.
2. % retained: 0, 2, 76, 20 and 2 (pan). Cumulative: 0, 2, 78, 98. % passing: 2.0 mm: 100; 0.63 mm: 98; 0.2 mm: 22; 0.063 mm: 2.
3. Gravel 100 − 100 = 0%. Sand 100 − 2 = 98%. Fines 2%.
4. Cu = 0.36 ÷ 0.10 = 3.6. Cc = 0.23² ÷ (0.10 × 0.36) = 0.0529 ÷ 0.036 ≈ 1.5. Cc is between 1 and 3, but Cu is well below 6, so the sand is not well graded. Most of it (76%) sits between 0.2 and 0.63 mm, so its curve is steep.
5. Uniform SAND (a medium sand). There are too few fines to name a secondary type.
Quick check
Five quick questions
This is not a test. It simply shows you whether you are ready for Lesson 4. Choose an answer to see the explanation.
Key points
What to take away
- From biggest to smallest: gravel, sand, silt and clay. In the UK the boundaries are 63, 2, 0.063 and 0.002 mm.
- Coarse soils feel gritty. To tell silt from clay, use the shake test, stickiness and how it behaves when dry.
- Sieving measures grains down to 0.063 mm. Use wet sieving if there is any silt or clay, and check the masses add up to within 1%.
- Finer grains are measured by letting them settle. Scale the results by the fines percentage before plotting.
- The grading curve’s shape shows whether a soil is well graded, uniform or gap graded.
- Read the amount of each soil type from the curve, then describe it as grading word + secondary type + PRIMARY TYPE.
- Cu = D60 ÷ D10 and Cc = D30² ÷ (D10 × D60) put a number on the grading. Always say which rule you used.
Up next: Lesson 4
Why does clay behave like plastic? Grain size is not the whole story for fine soils. You will learn how water changes clay from hard to plastic to liquid, and how the Atterberg limits measure it.
Assessment (optional)
Test yourself properly
Unlike the practice questions, this assessment has no answers on the page. Attempt it on your own, as if it were an exam.
Instructions
- Total: 40 marks. Suggested time: 60 minutes.
- Answer all questions. The marks for each question are shown in brackets.
- You may refer to the lesson notes, but write your answers in your own words.
- Show your working in calculations. Where a question asks you to justify or explain, marks go to your reasoning, not just the final answer.
Section A: Concepts (12 marks)
- Name the four main soil types from coarsest to finest, and give the UK particle size range for each. [4 marks]
- Explain why wet sieving is used when a soil contains silt or clay, and what goes wrong if such a soil is dry sieved. [3 marks]
- Describe the shake (dilatancy) test, and how a silt and a clay respond to it. [2 marks]
- Describe the shape of a well graded, a uniform and a gap graded grading curve. [3 marks]
Section B: Calculations and interpretation (16 marks)
An oven-dry soil sample of mass 800.0 g was wet sieved, with these results:
Sieve (mm) Mass retained (g) 20 0.0 6.3 40.0 2.0 120.0 0.63 200.0 0.2 240.0 0.063 120.0 Pan 80.0 (a) Work out the % retained, cumulative % retained and % passing for each sieve. [4 marks]
(b) State the percentages of gravel, sand and fines in the sample. [3 marks]
(c) Does the test pass the mass check? Explain. [1 mark]
(d) Estimate D60 for this soil. Explain how you read it. [2 marks]
A different sand has D10 = 0.12 mm, D30 = 0.35 mm and D60 = 1.1 mm. It has 10% fines.
(a) Calculate Cu and Cc. [2 marks]
(b) Is the sand well graded under the rule used in this lesson? Explain. [2 marks]
(c) A settling test finds that 25% of the fines are finer than 0.002 mm. What percentage of the whole sample is clay-sized? [2 marks]
Section C: Applied case (12 marks)
A contractor wants to reuse sand dug from a site as a drainage layer under a floor slab. The laboratory report shows a smooth grading curve with 3% fines, from a dry sieve test. The field description of the same soil says it is “slightly clayey” and that it sticks to the hands when wet.
Write a short technical note to the contractor (maximum 300 words) that:
- explains why the 3% fines result may not be reliable [3 marks]
- states what testing should be done instead, and when a settling test would be needed [3 marks]
- explains how a higher fines content would affect the sand’s use as a drainage layer [3 marks]
- notes why the report should state which sieve sizes and classification system were used [1 mark]
Clear, professional communication suitable for a non-specialist. [2 marks]
How your work is judged
| Criterion | What a strong answer shows |
|---|---|
| Technical accuracy | Correct terms, methods and calculations |
| Use of evidence | Conclusions backed by specific information in the question (test results, descriptions, site details) |
| Engineering judgement | Results linked to what they mean for design and construction |
| Communication | Clear, concise and well organised, in your own words |
Want your work marked? (optional)
You’re welcome to complete this assessment for your own practice. If you’d like me to mark it personally and send written feedback within 5 days, you can send it to me for £20. The marking & feedback page explains how.
- Upload one file (PDF, Word or a clear photo of handwritten work) with your name at the top.
- I only mark assessments from this website. I do not mark university or college coursework, assignments or exam answers.
- How your submission and personal details are handled is explained in the privacy policy.
References
- ASTM International (2017) ASTM D2487-17 Standard Practice for Classification of Soils for Engineering Purposes (Unified Soil Classification System). West Conshohocken, PA: ASTM International.
- British Standards Institution (BSI) (2016) BS EN ISO 17892-4:2016 Geotechnical investigation and testing. Laboratory testing of soil. Determination of particle size distribution. London: BSI.
- British Standards Institution (BSI) (2018) BS EN ISO 14688-1:2018 Geotechnical investigation and testing. Identification and classification of soil. Identification and description. London: BSI.
- British Standards Institution (BSI) (2018) BS EN ISO 14688-2:2018 Geotechnical investigation and testing. Identification and classification of soil. Principles for a classification. London: BSI.
- Knappett, J.A. and Craig, R.F. (2019) Craig’s Soil Mechanics. 9th edn. Boca Raton, FL: CRC Press.
Further learning
- Read: Chapter 1 of Craig’s Soil Mechanics covers particle size analysis with more worked examples.
- See it done: look for videos of a sieve analysis and a hydrometer test to see the equipment in use.
- Practise: draw the curve from the “Your turn” task on log graph paper or in a spreadsheet, and read off D10, D30 and D60 yourself.