A strength class is a performance target you can argue about. Aggregate size is geometry, and geometry does not negotiate: either the stone fits between the bars and the form, or it bridges across them and leaves a void behind the steel that nobody will see until a core comes out.
That is the whole reason the governing limits on nominal maximum size are written as fractions of member dimensions rather than as a recommended size. They are placement limits, not durability limits, and on most Alberta submittals they are the clauses that get copied forward from the last job without anybody checking the section they now apply to.
What the two terms mean, and why a submittal should say which one it uses
Two similar phrases circulate, and they do not mean the same thing. Under the long-standing material definitions, maximum size is the smallest sieve through which 100 per cent of the sample passes, while nominal maximum size is the largest sieve that retains some particles but generally not more than 10 per cent by weight.
So a nominal maximum size of 20 mm describes a blend in which a little material is larger than 20 mm. The two terms differ by roughly one sieve, and asphalt work uses a different convention again.
The practical consequence is narrow but real. If a specification sets a limit of 20 mm and the submittal reports a nominal maximum size of 20 mm, there is permitted material above the limit. On a congested member that is the difference between a clean pour and a honeycomb. Say which term the number is, every time.
Why the limits are fractions of the member, not a recommended size
The governing requirement in North American practice sets nominal maximum size at the least of three values. ACI 318-19 puts it in §26.4.2.1(a)(5), and the three are:
| Limit | What it is a fraction of |
|---|---|
| One-fifth | the narrowest dimension between sides of forms |
| One-third | the depth of slabs |
| Three-fourths | the minimum specified clear spacing between individual bars or wires |
The three-quarters figure applies not only to individual bars but to bundles of bars, prestressed reinforcement, tendons and ducts. The Code commentary is explicit about why they exist: the limits are there to “facilitate placement of concrete around the reinforcement without honeycombing.” They are about getting the mix past the steel, not about how strong it ends up.
The same commentary then adds the instruction most specifications ignore. Because maximum size affects both shrinkage and cost, “the largest aggregate size consistent with the requirements of 26.4.2.1 should be permitted.” Read together, the clause is a ceiling and the commentary is a nudge upward toward it. A spec that calls 14 mm into a member that could take 28 mm is leaving paste, water and money on the table, and the slab-guide limits for flatwork are the same two fractions — three-quarters of the clear bar spacing, or one-third of the section depth.
Which of the three actually governs
The useful exercise takes two minutes and almost nobody does it. Run all three and take the smallest.
A 150 mm interior slab-on-grade with mesh:
- One-third of the depth: 50 mm.
- Form width: not controlling on a slab.
- Three-quarters of clear spacing: if the mesh is at 150 mm centres with a small wire, clear spacing is roughly 140 mm, so about 105 mm.
Depth governs, at 50 mm, which means a 20 mm or even a 28 mm aggregate is comfortably inside the limit and the choice reverts to finishability and economy.
Now a 200 mm structural wall with two curtains of 15M at 200 mm centres: one-fifth of the narrowest form dimension is 200 ÷ 5 = 40 mm, while three-quarters of the roughly 184 mm bar-to-bar clear spacing gives 138 mm. Form width governs, at 40 mm, which still clears 20 mm comfortably — and is why walls rarely cause trouble.
The members that do cause trouble are thin and reinforced at once. A 50 mm bonded topping is the classic case: one-third of 50 mm is 16.7 mm, so 20 mm is already outside the limit before the steel is considered. Any pile cap or deep beam with bundled bars and a duct in the middle is the other case, where the three-quarters rule on the duct, not the member size, is what bites.
The rule of thumb worth carrying: on flatwork, depth governs; in walls, the form governs; in anything congested or thin, the reinforcement and duct spacing governs — and that is a number you read off the drawing rather than assume.
The exception that is not a loophole
There is a documented exception, and it is narrower than people hope. The three limits do not apply where the licensed design professional judges that the workability and methods of consolidation available are such that concrete can be placed without honeycombs or voids.
That is a signed engineering judgment about a specific placement method, not a general allowance. In practice it is what makes a self-consolidating mix in a congested element legitimate, and it should arrive with the method attached: the mix, the consolidation plan and the name of the person who accepted it.
What a Canadian owner specification does instead
Canadian public specifications tend to skip the fractions and pin the aggregate size directly to the class and the application. Alberta Transportation’s bridge specification is a clean example — its cast-in-place concrete section sets a size band per class:
| Class | Minimum specified 28-day strength | Coarse aggregate size | Maximum w/cm |
|---|---|---|---|
| S | 20 MPa | 28 to 5 mm | 0.50 |
| B | 25 MPa | 28 to 5 mm | 0.45 |
| Pile | 25 MPa | 28 to 5 mm | 0.45 |
| D | 30 MPa | 14 to 5 mm | 0.42 |
| C | 35 MPa | 20 to 5 mm | 0.42 |
| HPC | 45 MPa | 20 to 5 mm | 0.38 |
Two notes attached to that table are more instructive than the table itself. For mass pours such as piers and abutments, the coarse aggregate for the 35 MPa class is moved up to the 28 to 5 mm band — bigger stone, less paste, less heat to get rid of. And where a deck overlay is 50 mm or thinner, the nominal maximum top size is dropped to 14 mm, which is the one-third-of-depth rule arriving by a different route.
Notice also what the specification does not do. It sets no aggregate size for pumped concrete; the pumping clause asks only that mixes intended for a pump be designed for pumping, which pushes the question back to the supplier and the mix. That is the correct allocation of the problem, and it is the reason the conversation about line size belongs on the order rather than in the specification.
Why bigger is cheaper and smaller is stronger
Both halves of that sentence are documented, and they pull in opposite directions.
At a fixed water-to-cementing-materials ratio, a larger maximum size lowers the cement content needed to reach a given consistency — or, held the other way, lowers the water requirement at a given cement content. Less paste per cubic metre is less shrinkage and less cost, which is the economic argument the ACI commentary is making when it tells you to permit the largest size the limits allow.
Against that: where strength requirements are high, a smaller maximum size delivers the highest strength at a given water-to-cementing-materials ratio. The surface area goes up, the bond improves, and the stress concentrations around large particles come down. It is why high-strength mixes trend toward 14 mm and 10 mm rather than 28 mm, and why Alberta Transportation’s 45 MPa class sits at 20 to 5 mm while its 20 MPa class sits at 28 to 5 mm.
So the choice is not “what size is good.” It is which of the two pressures your member is under:
| If the controlling risk is… | Move the nominal maximum size… |
|---|---|
| Shrinkage, heat of hydration, cost per cubic metre | Up, to the largest the three limits allow |
| High specified strength, bond, thin cover | Down |
| Placement through congested steel or ducts | Down, to three-quarters of the tightest clear spacing |
| Finishability of exterior flatwork | Toward the middle, with the blend checked |
That last row is where aggregate size and aggregate grading get confused with each other. They are separate questions: size is the ceiling set by geometry, while the distribution of everything beneath that ceiling is what governs water demand and finishability. We have treated the second question separately in what the sand-to-stone split actually does to your pour. Choosing 20 mm does not tell you whether the blend under it is any good.
Where the pump changes the answer
A pump line imposes its own ceiling, and it is independent of the three structural limits. The constraint is the smallest inside diameter in the system — not the pump, and not the first metre of boom, but the narrowest reducer or the hose at the end of the string. A size that satisfies §26.4.2.1 comfortably can still bridge in a reducer, and the blockage will be blamed on the mix.
Two practical rules follow. The line size, the reach and the smallest reducer in the string belong on the order alongside the aggregate size, because the mix has to be designed around the whole system rather than around the member. And where the geometry forces a size down for pumping, it is the paste and the sand fractions that have to carry the extra workability — which is a mix design decision rather than a site one, as set out in our note on pumpable concrete mix design.
What to put on the order
Bring four things and the size question answers itself:
- The controlling member geometry. Thinnest section, narrowest form dimension, and the tightest clear spacing between bars, bundles or ducts — the three numbers the limits are fractions of.
- Which term you mean. Nominal maximum size or maximum size, stated.
- The exposure class. Durability requirements set the water-to-cementing-materials ratio and the air content, and those interact with the size you choose; see how the exposure classes map onto Alberta work.
- The placement method, including line diameter and the smallest reducer if it is being pumped.
Where a specification has called a size that the member cannot take, say so before the pour rather than at the chute. The fix is almost always a one-line change to the mix. Discovering it when the first truck will not move through the line is a considerably more expensive version of the same conversation.
FAQ
What is the maximum aggregate size allowed in a concrete member? The least of one-fifth the narrowest dimension between sides of forms, one-third the depth of slabs, and three-fourths the minimum specified clear spacing between individual bars, bundles, prestressed reinforcement, tendons or ducts.
What is the difference between maximum size and nominal maximum size? Maximum size is the smallest sieve through which all of the sample passes. Nominal maximum size is the largest sieve that retains some material, generally not more than about 10 per cent by weight. They differ by roughly one sieve, so a specification should state which it means.
What aggregate size should a 50 mm topping use? One-third of 50 mm is about 16.7 mm, so 20 mm is outside the limit and 14 mm is the usual call. A Canadian bridge specification reaches the same answer directly, dropping the nominal maximum top size to 14 mm for overlays 50 mm or thinner.
Does larger aggregate make concrete stronger? Not where strength is the controlling requirement. At a given water-to-cementing-materials ratio, a smaller maximum size yields the highest strength. Larger sizes reduce the paste and water needed for a given consistency, which is an economy and shrinkage benefit rather than a strength one.
Can the limits be exceeded? Only where the licensed design professional judges that the workability and consolidation methods available allow placement without honeycombs or voids. That judgment should arrive with the placement method named.
Send the geometry, not just the strength class
The size question is settled by three dimensions off the drawing and one line diameter off the pump truck. With those in hand it is a ten-minute decision; without them it is a guess that shows up in a core.
Send us your project details — pour date, mix type, PSI, volume — and our team will get back to you within 1 business hour with a quote and delivery slot. If the member is thin, congested or being pumped a long way, include the thinnest section and the tightest bar spacing and we will tell you which of the three limits governs before anything is batched. See custom mixed concrete in Calgary for how the mixes are built around these constraints.
Sources
- American Concrete Institute, What nominal maximum sizes of aggregate are recommended? (ACI 318-19 §26.4.2.1(a)(5) and commentary): https://www.concrete.org/frequentlyaskedquestions.aspx?faqid=754
- Alberta Transportation, Specifications for Bridge Construction, Section 4 — Cast-in-Place Concrete: https://www.transportation.alberta.ca/content/doctype246/production/07bcs04.pdf
- Pavement Interactive, Gradation and Size (maximum size and nominal maximum size definitions): https://pavementinteractive.org/reference-desk/materials/aggregate/gradation-and-size/
