How long concrete can stay in a mixer truck, and what that does to a site an hour from the plant

The honest answer to how long concrete can stay in a mixer truck is that the number is a contract term, not a constant. ASTM C94/C94M-21, balloted over six years and issued in April 2021, deleted the default limit of one and a half hours that the specification had carried since its original publication. Section 12.9 now reads that discharge shall be completed within the time limit established in 6.1.10, and 6.1.10 puts the duty on the purchaser: he states a time limit from the start of mixing to the completion of discharge, and if he does not, the producer establishes one and communicates it to him before delivery. The limit has to appear on the delivery ticket. The current edition adds a parallel clause, 6.1.9, which does the same for the drum revolution limit.

The change was not a relaxation. The subcommittee's stated intent was to remove a default, not to lengthen or shorten it, and the ticket requirement makes the number visible on every load instead of assumed. Why a default existed at all is worth keeping in view: under placement conditions between 40°F and 95°F, an ordinary mix of cement, fine and coarse aggregate and water begins to set somewhere between sixty and a hundred and twenty minutes, depending on cement chemistry, water-cement ratio, cement content and the temperature of the mixed concrete. That band is what the original hour and a half was drawn around. Admixtures move it, which is the argument for letting the purchaser set the figure — and also why a figure copied from another project is worth nothing.

A standards body declining to set a default does not mean nobody sets one. California's Department of Health Care Access and Information issued Policy Intent Notice 82, effective 9 October 2025, because ASTM had stopped stating a figure. For the hospital work it regulates, discharge must be completed within one and a half hours of the mixing water meeting the cement and aggregates, and within 300 drum revolutions, of which no more than 100 may be mixing revolutions and the rest agitating. Neither may be waived by producer or purchaser except under an alternative compliance method submitted to and approved by the department. Unplanned delay has its own written procedure: at ninety minutes the operator records time, slump and temperature; temperature is rechecked every ten minutes, or every five between 85°F and 88°F; slump is measured at 110 minutes or at the end of placement, whichever comes first; placement ends at 120.

Two things follow for a buyer anywhere. The number that governs a pour is the one in the project specification or the national code that applies to it, and a supplier quoting ninety minutes from memory is quoting a figure the issuing body itself withdrew as a default. And even where a longer limit is agreed, the verification points may not move with it: ACI 301, incorporated into ASTM C94 by reference, requires that air content, slump and temperature be verified at one and a half hours regardless of the discharge time specified. The hour and a half does not leave the job when the discharge limit is extended — it turns into a test point. Ask which document governs, and ask to see the delivery-ticket field where the limit is printed.

The usual reason given for a discharge limit is strength, and the largest body of testing we have found does not support it. A Washington State research report, WA-RD 831.1, evaluated extended mixing time and drum-revolution counts against coarse and fine aggregate, supplementary cementitious content and type, chemical admixtures and temperature. Its abstract reports that mixing time and revolution counts had no significant effects on the majority of the mechanical properties and durability characteristics measured — the report's key words list compressive and tensile strength, modulus of elasticity, modulus of rupture, freeze-thaw resistance and diffusivity — and that what limits an aged load instead is loss of workability. On that basis it concludes that the existing specification limits may be appropriate, for that reason, while also being very conservative. A load turned away at the chute is, on this evidence, one that can no longer be placed rather than one that would be weak.

The regulator cited in the section above takes the opposite view, and we have not reconciled the two. The background to California's notice states that concrete cylinder breaks show a strength decrease the longer the concrete remains in the mixer truck beyond the hour-and-a-half limit before being discharged, and that there may also be an adverse relationship between strength and an excessive number of drum revolutions. That is its stated reason for refusing to let the limit be waived. One document reports a test programme and the other states a regulatory position; neither is ours, and a buyer is entitled to know that the technical case behind the number he is given is contested. The specification itself takes the workability line: the limit may be waived where the slump afterwards is still such that the concrete can be placed, without the addition of water.

The clock does not start when the truck leaves the plant. It starts when the mixing water meets the cement and aggregates, so everything up to the last of the concrete leaving the chute comes out of one allowance: batching, the outbound run, the queue at the site, and the discharge itself. Placement is last in that sequence, so a minute lost upstream is paid for out of the placement window and nothing else. Take a ninety-minute limit — the figure California still imposes on the hospital work described above, and the one the specification carried until 2021, not necessarily yours. Assume ten minutes from water to the plant gate, thirty-five out, fifteen waiting: sixty gone, thirty left for discharge and placement. Move the pour front ten kilometres on at an assumed thirty kilometres an hour and the outbound leg grows twenty minutes: the window falls from thirty to ten.

None of those four durations is measured. They are illustrative: substitute your own and the shape survives, because the allowance is fixed and the haul is not. A second clock runs in parallel on drum revolutions. Our specifications record a drum rotation speed of 16 revolutions per minute on four of the six self-loading mixer models published on this site, the SLM-18 among them, and no figure at all on the other two. Divide a 300-revolution budget by that speed and you get 18.75 minutes of continuous rotation; divide the 100-revolution mixing allowance and you get 6.25. Both divisions are performed here and neither is a holding time, because a transit mixer's budget is written around a slow agitating speed for which we hold no figure. Where a revolution limit applies it is read off a drum counter, and we hold no record of one being fitted to these models.

A building site is a fixed destination. Measure the cycle once — batch, run, discharge, return — and the same figure holds for the whole pour, so the plant can schedule against it and the limit is either met every time or missed every time. A road is not a destination; it is a sequence of them. The pour point advances with the work, so the outbound leg grows over the job and the cycle time measured on the first morning is wrong by an amount that increases daily. Nothing warns you: each load is only slightly worse than the one before, and a schedule that worked yesterday keeps very nearly working until it does not. The load that fails is the one at the far end of the day's advance, furthest from any help.

Two further things compress the window exactly where it is already shortest. The queue at the front is unavoidable, because a pour is one-directional and a truck cannot discharge past the one ahead of it — waiting there is a property of the geometry, not a planning failure. And the access at the head of a road job is by definition the worst on the site: no hardstand, a surface still under construction, and none of the manoeuvring room the finished road will eventually provide. The available answers are to move the batching plant, to dose a retarder, or to shorten the haul. The third is the only one that acts on both clocks at once, and the way to shorten a haul to almost nothing is to stop hauling.

A self-loading mixer does not repeal the discharge limit; it stops the limit binding, because the interval between water and discharge collapses to the machine's own cycle. It loads its own aggregate with its bucket, mixes in its drum, drives to the pour point and discharges there, so there is no plant-to-site leg to subtract and no queue behind another truck. On the six self-loading mixer models published on this site the water is carried on board in one of two forms, both of them standard equipment rather than options. One model carries a water tank with a self-priming system, and its capacity is not on record. The other five, the SLM-18 among them, have an automatic water-fill controller with an external suction pump. That is what makes a batch possible at the work front rather than at a plant: the water arrives with the machine.

The proportions are the harder half, and here our own specifications say something less convenient than a brochure would. Electronic hydraulic weighing is standard equipment on exactly one of the six models. On four others, the SLM-18 among them, it appears in the optional list, and on the sixth it appears in neither list. Without it the bucket is loaded by eye and the mix is an estimate — a real difference on work tested against an approved design mix, none at all on a rural slab. Specify it deliberately rather than discover it on delivery. Five of the six carry a reversing and drum-monitoring camera and a drum that lifts and rotates through 270° as standard, and four of the six are articulated, one of the remaining two being a rigid chassis with front-wheel steering.

What we will not do is turn any of that into an output figure. No hourly output and no cycle time is confirmed for any of the six, so we publish neither, and we will not convert a discharge capacity into a day's production by assuming how many cycles fit into one. Five of the six carry a published discharge figure and one does not; across the range those figures run from 1.0 to 4.5 cubic metres, and the SLM-18 sits at 1.6 to 1.8. How many of those batches your crew places in a shift is arithmetic only you can do, because it turns on your access and your labour rather than on the machine. A comparison table covering all six sits on the catalogue page of this site, blank wherever a figure has not been supplied; the SLM-18 is the one model with a specification table of its own, and the complete sheet for any of the others is sent in writing on request.

The three documents behind this page are American — one state regulatory notice, one state-funded research report and one commercial testing laboratory's summary of the standard — and none of them governs a pour in Vietnam, Indonesia, the Philippines, Nigeria or Saudi Arabia. We have not read the ready-mixed concrete code of any destination market, so we state no national discharge limit for any of them, and make no claim that ninety minutes is or is not the local rule. Nor do we know how far the 2021 ASTM change has propagated: a code that cites ASTM C94 by designation rather than by edition now points at a clause stating no number, which is a live question for a specifier and one we have not answered. The figure that governs your pour is in your project specification or your national standard. Ask for it in writing before the mix design is approved, not at the chute.