Site-mixed versus ready-mixed concrete, and what actually fails

The argument against site-mixed concrete is almost never that a drum standing on a site cannot mix as well as a drum standing at a plant. It is that nobody can say afterwards what went into it. Materials on a site are proportioned by volume — buckets, barrows, a gauge box — and the FAO's construction textbook is blunt about where that breaks: completely dry and very wet sand occupy the same volume, while damp sand in between occupies more. So a stockpile that has dried on the surface but is still damp through delivers less sand per bucket than the same stockpile bone dry, at the same nominal mix. The effect peaks in the damp middle, which is where site stockpiles usually sit. Nothing on the site looks different. The proportions have changed anyway, and so has the water already riding in with the aggregate.

That is why the failure is not one bad batch but drift between batches, and why it is invisible until cubes come back from a laboratory, if cubes are taken at all. The same source recommends batching by weight for grade C30 and above and notes that weight batching is in practice confined to large construction sites, which is a fair description of the problem rather than a solution to it: the accurate method was the one a small site could not have. It also offers a site remedy — checking the bulk density of the aggregate, which it says allows greater accuracy when grade C20 or above is batched by volume — but that remedy is itself a measurement, retaken as the stockpile changes. The rest of the objection, that hand mixing or a small drum mixer gives an uneven mix, is real but downstream. Fix the mix and the measuring is still in front of you.

What a ready-mix plant sells is not mixing. It is a measuring regime with a paper trail. A US Federal Highway Administration plant inspection checklist sets out the items: all scales checked and calibrated, sensitivity of scales checked, water meter checked for accuracy, cement weighed as dry weight, aggregate weighed as dry weight plus total moisture, and haul tickets filled out. Moisture has a section of its own — percent total moisture for each aggregate, effective moisture per batch, water to be added at the mixer — plus a standing daily duty to keep testing it. That section is aimed at the failure described above. The scale lines are there because a drifting scale hides the same error, and the ticket is what lets an engineer reconstruct the batch a month later. A buyer comparing a plant with a site is comparing that regime against a bucket, and on the measuring question the plant wins on documents alone.

The return the plant asks for is a clock. ASTM C94/C94M carried a one-and-a-half-hour limit to end of discharge from its publication in 1935 until the current edition removed it: the time limit and the drum-revolution limit are now something the purchaser states, or the producer sets and communicates before delivery. At least one authority has put the old figure back by its own rule. California's health-care construction regulator requires discharge completed within one and a half hours of the mixing water meeting the cement and aggregates, and a maximum of 300 drum revolutions, of which no more than 100 may be mixing. The same notice makes those limits non-waivable by producer or purchaser, but lets an unplanned delay run to 120 minutes if temperature and slump are recorded at set intervals, and allows more only where the agency approves an alternative compliance method. Whatever number applies to you, it is a number, and the truck spends it driving.

The clock is the radius. Because the limit runs from the moment water meets cement to the end of discharge, every minute of driving, queuing and waiting at the gate is spent out of the same allowance — the subject of a separate page on this site about how long concrete can stay in a truck. We have not sourced a delivery radius in kilometres for any market and do not state one, because it is a different number for every plant, road and traffic condition. The structural point stands without it: a site beyond the practical reach of a plant is not a site that pays more for ready-mix, it is a site that cannot buy ready-mix at all. That is the ordinary condition in rural districts, on road work where the pour front moves all day, and anywhere the nearest batching plant has not been built.

The second constraint is physical access, where we can only speak for the machines we publish. Ten of the sixteen self-loading mixers we hold specifications for carry real overall dimensions: widths from 1.60 m to 2.30 m, lengths from 4.35 m to 7.06 m, heights from 2.45 m to 2.90 m. Width is the figure to plan against, because it barely moves with the posture a machine was measured in. Height does move with it, and for nine of those ten the sheet the figures came from does not say which posture was measured, so we do not offer those heights as travelling heights. The exception is the SLM-18, measured at 7.06 × 2.21 × 2.58 m on a drawing that separates carrying height from bucket-raised height. We have not sourced dimensions for ready-mixed delivery trucks in any market, so no comparison appears here. Measure your worst gateway first, and ask for a drawing where height decides the answer.

The case for a self-loading mixer is narrower than the marketing for the category, and it is worth stating exactly. One of the six mixers published on this site carries an electronic hydraulic weighing system in its standard configuration. Four more can have it as a listed factory option. The sixth has no option list from its plant at all, so we do not know whether the weighing system can be had on it, and we will not guess by reading across from a similar machine. This is not a range-wide feature and should not be bought as one. Where it is fitted, it moves the measuring step from a bucket to a load cell and puts it at the pour instead of at a plant, which is the only honest description of what the machine changes about the argument above.

Two things we do not hold belong in the same paragraph. There is no accuracy tolerance and no calibration certificate for that weighing system anywhere in our data, so we make no claim about how close it holds; ask for the tolerance in writing before it matters to a specification, and we will ask the plant for it too. The water side is better documented: seven of the sixteen mixers carry an automatic water-fill controller and an external suction pump, and eight more carry a high-capacity water tank with its own filling system. That matters because water is the other half of the batching failure, and a machine that draws its own water is not carrying a hose from wherever the site's water happens to be. Whether it measures that water is a separate question, and we cannot answer it.

Described plainly, the cycle is: the machine drives to the aggregate, loads it with its own front bucket, draws its own water, mixes on the way to the pour front, and discharges through a drum that lifts and rotates through 270°. Loading and mixing in transit are what the category is, not items a plant has itemised for us. What is itemised: the lifting, 270°-rotating drum on seven of the sixteen mixers, the water-fill controller and suction pump on the same seven, a reversing and drum-monitoring camera on those seven, and a high-definition driving display on eight of the others. Six of the seven are articulated and the seventh is a rigid chassis with front-wheel steering. We hold no turning-circle or turning-radius figure for any mixer in the record, and we do not state one.

What none of that proves is any figure about output or labour. There is no cubic-metres-per-day figure, no loads-per-shift figure, no fuel consumption and no crew size anywhere in our product record, so no such number appears on this site and none should be inferred from the cycle above. Fourteen of the sixteen mixers carry a real discharge figure, running from 1.0 to 4.5 m³ of discharged concrete; what that is worth on your job depends on your haul distance, your aggregate, your crew and your pour, and the arithmetic is yours rather than ours. Two inputs you would want for it are thin: rated power is real on seven of the sixteen, at 55 to 91 kW, and drum geometric volume on four, at 2,900 to 6,500 litres. A supplier quoting you a daily output has measured something on somebody else's site, or nothing at all.

We have not checked whether site-mixed concrete is accepted for any given class of work in any market we sell into, and we are not going to imply otherwise by staying quiet about it. Acceptance is decided by the project specification, the building control authority and, on funded work, the client's own engineer — and those three can disagree inside one country. Some specifications require concrete from a certified plant for structural elements and permit site mixing for blinding, kerbs, drainage and surrounds; others set a strength grade above which plant supply is mandatory; others say nothing and leave it to the supervising engineer. Which of those applies to you is a document you already have or can ask for, and it is a better first question than any question about a machine. If the specification rules site mixing out for the element you are pouring, no equipment argument reopens it.