Concrete in a bridge spends most of its working life in compression, so we test it by squashing specimens and reading the number at failure. The awkward part is deciding which specimen. A cube cast at the pour, a cylinder cured in a laboratory, and a core drilled from a pier can all originate from the same load of concrete and still produce three different figures. Choosing between them is not a matter of preference. It is a matter of what you are trying to prove, and what the contract says you must do.
Three tests, three different questions
Start by being precise about purpose, because the methods are not interchangeable.
- Cubes are the standard UK presumptive test. They are made from fresh concrete, cured under controlled conditions and tested at a set age. They measure the potential strength of the mix as supplied.
- Cylinders do the same job in a different shape. They are the reference strength in Eurocode 2, which is why a concrete class is written as C32/40 — 32 N/mm² cylinder, 40 N/mm² cube.
- Cores are cut from hardened concrete in the finished element. They measure strength as placed, which is a harder and more nuanced question.
Cubes and cylinders tell you about the concrete that arrived. Cores tell you about the concrete that is now carrying load in your bridge. Those are not the same thing, and the gap between them widens the longer a structure has been in service.
Why the numbers disagree
Shape matters more than most people expect. A cube tested between the platens of a compression machine is restrained at its loaded faces by friction, and that restraint artificially stiffens the specimen. A cylinder of the same concrete, typically 150 mm diameter by 300 mm high, has more of its height free from that effect, so it fails at a lower stress. Cylinder strength is therefore the more conservative measure of the two.
The ratio between them is not a constant. It shifts with the grade of concrete, the type and size of aggregate, the rate of loading and the condition of the specimen at test. That is why blanket conversion factors passed around site should be treated with suspicion. If you need to move between cube and cylinder values for anything that matters, use the relationships given in the relevant standard or ask the concrete specialist advising the project.
Cubes in UK bridge practice
For most bridge works in the UK, cubes remain the default. They are familiar, cheap to produce and directly tied to the compliance rules in BS EN 206 and its complementary British Standard, BS 8500. If you are casting cubes, a few habits separate a useful result from a contentious one:
- Sample from the middle of the discharge, not the first or last part of the load.
- Make and cure specimens to BS EN 12390-2, keeping them at the specified temperature and protected from frost, sun and drying.
- Compact properly in layers, and never leave cubes sitting on the back of a pickup overnight.
- Get them to the laboratory promptly, and test to BS EN 12390-3 within the permitted age window.
- Keep seven-day results for what they are good for — striking formwork, early loading decisions — and never for compliance.
Record the cube reference, the load it came from, the location it represents and the time of sampling. Most disputes are won or lost on paperwork, not on the testing machine.
Cylinders: a smaller role, an important reference
In UK site practice, cylinders appear less often than cubes, but they are not optional trivia. They dominate precast factories, sprayed concrete and specialist work, and they are the natural specimen where design has been carried out to Eurocode 2. If a specification calls for cylinders, check the size, the curing regime, the end preparation and the age at test before you start, because a poorly prepared cylinder end can knock a surprising amount off the result.
Cores: when the cube is not the whole story
Cores come into play when the cube result is in doubt, when an existing structure is being assessed for a heavier load, or when there is no reliable record of what was placed. Taking and testing them is covered by BS EN 12504-1, with in-situ strength assessment addressed by BS EN 13791.
Core work rewards planning. Avoid reinforcement, keep clear of construction joints, choose a diameter that gives a sensible length-to-diameter ratio, and tell the laboratory exactly where each core came from and how it was drilled. Drilling disturbs the concrete, and the moisture condition at test changes the answer, so core results are not a straight swap for cube results. Interpretation should be done by someone used to in-situ assessment, not by reading the failure load off the certificate.
What specifications require, and what happens when a result is challenged
Compliance under BS EN 206 is assessed on groups of results using statistical rules, not on a single cube. The Specification for Highway Works and individual project specifications may add their own requirements on specimen type, sampling rate, age at test and where the results must originate. Read those clauses before the pour, because arguments afterwards rarely change them.
When a cube result comes back low, work through it in a sensible order:
- Verify the basics — sampling, cube making, curing, transport, labelling and the laboratory's calibration records.
- Look at the element itself for evidence of poor compaction, honeycombing or curing problems.
- Decide whether the open question is about the mix or about the structure. They need different answers.
- If the structure is the question, take cores under a proper plan and test them.
- Interpret cube, core and visual evidence together, with specialist input.
- Treat load testing as a last resort, properly specified and justified.
Practical closing: decide the regime before the pour
The cheapest testing decision you will ever make is the one taken at the pre-pour meeting. Agree the specimen type, the number of samples, the age at test, the laboratory and the acceptance criteria, and write it down. Match the test to the question: cubes for conformity of the mix, cylinders where design demands them, cores when you need to know about the concrete already in the bridge. Do that, and a difficult result becomes a manageable engineering conversation rather than a contractual one.
Photo: analogicus / Pixabay
