Mass Concrete Thermal Gradient vs Surface Curing

Frequently asked questions

Why does insulating a hot surface help?

Because the crack risk is the DIFFERENCE between the core and the surface, not the surface's temperature. The core is hot and expanded; if the surface is allowed to cool to the air, it contracts relative to the core, is restrained by it, and goes into tension — which young concrete cannot take. Insulating the surface keeps it closer to the core's temperature, so the differential stays within the limit and the tension never develops. It feels wrong because the instinct is to cool something that is too hot, but cooling the surface is the one intervention that makes the differential worse. The core cannot be cooled by anything applied outside it; it is cooled from within by embedded pipework on very large pours, or its peak is reduced in the first place by the mix and the placing temperature.

Why can stripping formwork crack a mass pour?

Because formwork is insulation, and removing it exposes a warm surface to cold air suddenly. While the forms are on, the surface is held close to the core's temperature and the differential is small. Strip them on a cold morning and the surface drops rapidly, the differential spikes, and the surface cracks — potentially within hours, on an element that had been behaving perfectly. That is why stripping times on mass concrete are a thermal decision informed by the monitored differential rather than a programme decision based on strength, and why in cold weather the forms are sometimes left on longer than strength alone would require, or replaced immediately with insulation blankets. It is also why the stripping sequence is planned rather than left to whoever is available.

How is the peak temperature reduced?

By putting less heat in and generating it more slowly. The placing temperature sets the baseline the rise builds on, so cooling the mix — chilled water, ice substitution, cooled aggregate — lowers the peak directly. The cement content and type determine how much heat is generated and how fast: replacing a substantial proportion of the Portland cement with supplementary cementitious materials such as ground granulated blast-furnace slag or fly ash reduces both the total heat and the rate at which it is released, which is why mass concrete mixes routinely use high replacement levels. It also slows early strength gain, which on a mass pour is usually acceptable because nothing is waiting on it. On very large elements, embedded cooling pipes circulating water remove heat from the core directly.

What limits apply to the differential?

A maximum differential between the core and the surface, and usually a maximum core temperature as well, both set by the specification for that element and that mix. The differential limit exists to keep the surface tension below what the young concrete can resist, and it is commonly expressed as a temperature difference to be maintained throughout the period the core stays hot. The core temperature cap exists for a different reason: above a certain temperature, delayed ettringite formation becomes a risk, which is an expansive reaction occurring much later in the concrete's life and capable of damaging it years afterwards. Both limits are monitored rather than assumed, and the specification also states the response when a limit is approached — usually more insulation rather than less.

How is it monitored?

With thermocouples cast into the pour, at the core and near the surface, logged continuously — because the differential is the control variable and it changes over days. Sensors are placed at the thickest point, where the peak will be highest, and close to the surface at a stated depth, and the data is read at intervals or logged automatically so that the differential can be tracked and acted on. The action is usually to add insulation rather than to remove it, and to delay stripping. On significant pours the monitoring plan is agreed before the concrete arrives, with the sensor positions, the reading frequency, the limits and the response defined — because a differential exceedance discovered after the fact is a crack rather than a decision.

How thick does a section have to be to count as mass concrete?

There is no single dimension, and the definition is behavioural rather than geometric: a section is mass concrete when its dimensions are such that measures are required to deal with the heat generated and the resulting volume change. Guidance gives indicative thicknesses, commonly in the region of a metre or more, but the threshold depends on the mix as much as the geometry — a high-cement-content, high-strength mix generates far more heat than a low-heat mix, so a thinner section of the first can behave as mass concrete while a thicker section of the second does not. The practical test is to estimate the temperature rise for the mix and the section and see whether it approaches the limits, which is what the calculation is for, rather than to apply a rule of thumb about thickness.

Does the surface still need normal curing?

Yes, and the two requirements coexist rather than one replacing the other. The thermal measures address the differential; the concrete surface still has to retain its water so hydration can proceed and the surface does not end up weak, dusty and permeable. In practice the same measure often does both — an insulated blanket over a covered surface retains heat and moisture together — which is convenient but should be a deliberate choice rather than an assumption. Where a curing compound is used on a mass pour, it addresses the water and does nothing about the heat, so insulation is still required over it. The sequence and the responsibilities are worth stating in the pour plan, since the thermal protection is usually managed by someone thinking about cracking and the curing by someone thinking about the surface.

What does mass concrete cracking look like?

Through-surface cracks in a pattern set by the element's restraint, appearing days after the pour on concrete whose test results are entirely satisfactory. Where the surface is restrained by the hot core alone, the cracking is superficial to moderate depth and distributed. Where the element is also restrained externally — a wall cast against an older base, a slab restrained by piles or by adjacent construction — the cracking can go through the section as the whole element cools and contracts against that restraint, which is a different and more serious mechanism and is why external restraint is considered alongside the internal differential. The distinguishing feature in both cases is the timing: these cracks appear over days rather than in the first hours, which separates them from plastic shrinkage cracking.