Designing a Concrete Mix and Reading the Submittal

One page, three claims, and a return-by date

The submittal arrives as a single page: a plant printout of batch weights in kilograms per cubic metre or pounds per cubic yard, one aggregate gradation dated some weeks back, admixtures listed by trade name and dose, and a strength history compressed into a mean and a standard deviation. Every line on it is defensible on its own, because no quality control department publishes a figure it cannot support. The review is not about whether the lines are true individually. It is about whether they are true together, and about which of them the structural model on your side of the transaction has already spent.

Three lines carry the weight. The water-cement ratio settles durability and most of the strength, and it is the figure a plant can quietly miss by adding water at the truck. The fine aggregate's fineness modulus is what the proportions were computed against, and the number most likely to have drifted since the sieve analysis was run. The modulus of elasticity is not on the sheet at all - it is inferred from unit weight and strength - yet the deflection checks, the camber and the distribution of lateral load between elements were all done with a value somebody assumed months ago. Each is governed by a different document, and nobody in the chain has a duty to make the three agree.

Be blunt about what the review is not. Under ACI 301, Specifications for Structural Concrete, proportioning the mixture and demonstrating it will reach the specified strength belongs to the supplier, and a stamp does not move that duty. What the stamp asserts is that the proposal as written satisfies the contract documents - which is why a line the documents do not support has to come back marked, in the record, rather than raised in an email nobody will find in two years.

Durability sets the ceiling before strength gets a vote

Read the exposure assignment before the strength line, because in a great many mixes the strength on the drawing is not the number that governs. ACI 318 sorts exposure into four categories - F for freezing and thawing, S for sulfate attack, W for contact with water where permeability matters, and C for corrosion protection of reinforcement - and assigns each element a class within them. Every class carries its own maximum water-cementitious ratio and its own minimum compressive strength, and the mixture has to satisfy the most restrictive requirement it inherits from any of the four.

The arithmetic of that is where reviews go wrong. A slab specified at 4,000 psi (28 MPa) but assigned to freeze-thaw class F3 picks up a maximum water-cementitious ratio of 0.40 and a minimum strength of 5,000 psi (35 MPa) from the durability table, so the mixture that satisfies the drawing does not satisfy the code. If the submittal quotes strength and says nothing about exposure, the first question back is which class was assigned and by whom - not because the supplier should have known, but because the exposure assignment is the engineer's to make and it belongs in the specification, not in the reviewer's head.

Outside North America the same logic runs through different letters. EN 206, with BS 8500-1 as its United Kingdom complement, classifies exposure as XC for carbonation, XD and XS for chlorides from de-icing salt and seawater, XF for freeze-thaw and XA for chemical attack, and sets limiting values for maximum water-cement ratio, minimum cement content and minimum strength class against each. BS 8500-1 ties those limits to an intended working life and a nominal cover as well, so a change of cover on the drawing can change the mix allowed underneath it. Either way, durability is a ceiling on the ratio and a floor on the strength at once.

Air entrainment rides along with the freeze-thaw classes, and it is the requirement most often applied where it does harm. ACI 318 sets a target air content by nominal maximum aggregate size for each F class, and ASTM C231 is the pressure method that verifies it in the field, with ASTM C173 substituted where lightweight or otherwise porous aggregate makes the pressure method unreliable. But ACI 302.1R, Guide to Concrete Floor and Slab Construction, warns against hard-trowelling air-entrained concrete, because the entrained air under a densified surface is what produces blisters and delamination. An interior slab that will be power-trowelled and an exterior slab that will freeze wet want opposite things, and a mix submitted for both is submitted for neither.

The strength column is a statistic, not a promise

No supplier proportions a mix to the specified strength. They proportion it to a required average strength high enough that the specified value is met with a defined probability, and the gap between the two is the plant's own variability. ACI 318 Section 26.4.3 sets out the route: a sample standard deviation from a record of at least thirty consecutive strength tests, or two records totalling forty-five, produced with similar materials under similar control, and representing a specified strength within 1,000 psi (7 MPa) of the one now being supplied. Records shorter than thirty tests are scaled up by a modification factor - about 1.16 at fifteen tests, easing to 1.00 at thirty - which is the code's way of charging for a thin history.

The required average is then the larger of two expressions, one governing the average of three consecutive tests and one the single low test. For specified strengths up to 5,000 psi they are the specified strength plus 1.34 standard deviations, and the specified strength plus 2.33 standard deviations less 500 psi - less 3.5 MPa in SI; above 5,000 psi the second changes shape. With no record at all, ACI 318 substitutes flat over-design increments, and the supplier may instead demonstrate the mixture by trial batches at three different ratios.

ACI 214R, Guide to Evaluation of Strength Test Results of Concrete, is where the standard deviation on the page acquires meaning. Genuinely tight control earns a lower required average and buys a leaner, cheaper, less shrinkage-prone mix - that is the reward the method pays. A low standard deviation quoted with a short or borrowed record proposes to sail nearer the acceptance criteria on somebody else's evidence, and those criteria are unforgiving: every average of three consecutive tests must reach the specified strength, and no individual test may fall below it by more than 500 psi (3.5 MPa) at or under 5,000 psi. A strength test is itself an average, of at least two 150 by 300 mm cylinders or three 100 by 200 mm cylinders, cast to ASTM C31 and broken to ASTM C39.

What each line has to reconcile with

A mix submittal is a set of cross-references pretending to be a list. The value in reviewing it line by line is that each claim has exactly one other claim it must agree with, and exactly one published test method that settles the argument if it does not. Where a line has no counterpart on the page - a modulus assumed in the analysis, an exposure class assigned in the specification - the counterpart is in your file, not theirs. Run them in the order below rather than the order the page prints them in: durability caps the ratio, the ratio caps the paste, and the stiffness at the bottom is whatever everything above it has quietly decided.

Yield: the summation nobody runs

Every mixture proportioned by ACI 211.1 is built on absolute volume, and absolute volume is checkable in about four minutes with the numbers already on the page. Divide each ingredient's mass by its relative density times the density of water to get the volume it actually occupies, add the design air as a volume, and the total must come to exactly one cubic metre or 27 cubic feet. Nothing else is acceptable, because the sheet claims to describe a unit volume of concrete.

When the sum misses, it misses for a reason worth knowing. Over unity usually means an aggregate relative density is a bulk oven-dry value where a saturated surface-dry one was needed, or a specific gravity carried over from a source that no longer supplies the plant. Under unity often means the air was left out of the summation, a sign the mix was scaled from an older non-air-entrained design rather than proportioned fresh. Either way the batch weights are not the weights for a cubic metre.

The field version is ASTM C138, which measures the fresh density of a known volume and turns it into yield and relative yield: the volume produced against the volume designed. A relative yield of 0.97 means every load is three per cent short, and that gets discovered by a finisher running out of concrete two metres from a construction joint, not by anyone holding the submittal. So demanding the relative densities and absorptions to ASTM C127 and C128, and the dry-rodded unit weight to ASTM C29 that the coarse aggregate table is entered with, is not pedantry - without them nobody can perform the summation, including the person who wrote the sheet.

Where the three numbers pull against each other

The reason a mix review cannot be done one line at a time is that the three headline numbers are coupled, and tightening any of them loosens another. Drop the water-cementitious ratio to satisfy a freeze-thaw class and, at constant cementitious content, the mix loses water and slump; the supplier restores workability either with a water reducer to ASTM C494 or by adding cementitious material, and the second route increases paste volume, drying shrinkage and heat of hydration all at once. The mix that best satisfies the durability table is not automatically the mix that cracks least.

The sand sits in the middle of that trade. A low fineness modulus raises water demand at the moment the ratio has none to spare, so a fine sand and an aggressive durability class are a hard pairing that usually ends in higher admixture doses; a coarser sand eases water demand but loses cohesion, which is felt at the pump line and on a steeply cambered deck. Hence the fineness modulus is a supply term, not a description: the ratio, the admixture dose and the placement method were all chosen for one grading, and none of them for the grading that arrives if the pit face moves.

Supplementary cementitious materials shift the picture in time as well as in magnitude. Replacing part of the portland cement with fly ash or slag cement lowers early strength and heat, stretches the finishing and curing window, improves late-age permeability, and is a standard mitigation for alkali-silica reaction and sulfate attack. ACI 318 sets the cementitious material options for each sulfate class, with expansion testing to ASTM C1012 as the alternative route for a combination that is not listed. The reviewer's question is not whether the replacement level is high, but whether the strength was demonstrated at the age the structure needs it and whether the specified curing supports it.

Where the three do not reconcile, say which one moves. That is a judgement the mix designer cannot make, because it depends on what the structure needs and not on what the plant can batch: relax the specified strength and accept a lower modulus, hold the strength and re-check deflection at the stiffness that actually results, or hold both and pay for it in admixture and aggregate control. All three are legitimate answers. Sending the page back with all three questions open is not.

What goes back with the stamp

A review that concludes with a mark and no words is a review that will be relitigated at the pour. Write the basis down: which exposure classes were assigned, which ratio governs, what fineness modulus the proportions assumed and therefore what the supply is now tied to, and what modulus the analysis used. Those four sentences turn a stamp into a record, and they are what the next person to look at this project - a testing agency, a delay claim, a repair engineer with cores - will actually need.

The items below are worth demanding before the stamp, in the order they stop being obtainable. Gradation history vanishes the moment the plant moves source, strength records only mean something while they still describe the materials proposed, and the analysis assumption is cheap to confirm now and expensive once formwork is on order.

1. The exposure classes assigned to each element, in writing, with the governing maximum water-cementitious ratio and minimum strength each one produces.
2. Relative density and absorption for every aggregate to ASTM C127 and C128, the dry-rodded unit weight to ASTM C29, and enough of both to sum the absolute volumes to unity.
3. The sieve analysis with its sample date, plus the running range of fineness modulus over the last quarter rather than the single result.
4. The strength record behind the quoted standard deviation: how many consecutive tests, at what specified strength, with which cement and aggregate sources.
5. Admixture and supplementary cementitious material data sheets by product name, with the ASTM C494, C618, C989 or C1240 designation each is supplied against.
6. A statement of the maximum water addition permitted at the site, so it can be printed on the delivery ticket instead of argued about at the chute.