In general, surface-mounted underfloor heating systems in new buildings are more expensive than embedded underfloor heating systems. However, the building has an operating period (60-year economic service life), where operation is important, both economically and in terms of environmental impact. In operation, surface-mounted systems are more energy-efficient than embedded systems. To make embedded systems more energy-efficient, double casting is sometimes carried out where, after the first casting operation, insulation, heating pipes and then a thinner layer with concrete are laid. This method is more expensive than single casting and takes much longer.

In the renovation and reconstruction sector, surface-mounted systems are the obvious option.

• In a renovation project, it is easier to install surface-mounted systems (no need to embed pipes in concrete). Having to cast over the top significantly increases the floor level.
• For intermediate joists, surface-mounted systems are the only and obvious choice, not least because of the joists’ bearing capacity and weight.
• Surface-mounted systems are more energy-efficient. They are easier to control and regulate: the system emits heat when required and will stop emitting heat when the indoor temperature rises. Embedded systems do not have time to react to rapid fluctuations in outdoor temperatures.
• The systems normally require lower forward temperatures than embedded systems. The reason for this is that heat is supplied next to the floor surface. Therefore, it is easier for the heat to rise than in cases where the coils are located far down in the structure. This also means that the downward losses are less.
• The risk of leaks during the installation procedure is greater with embedding. Dealing with leaks is obviously easier with surface-mounted systems.

Make your own judgement – which arrangement means less risk of accidents?

In a series of simulations performed by the Building Technology Division at KTH Royal Institute of Technology, which were presented at a seminar at the Nordbyggmässan fair in 1996, the thermal properties of the various designs of underfloor heating were demonstrated. In particular, the inertia of the system was highlighted. As a condition, the water temperature was increased in a single step by 1°C. The objective of the study was to see how long it took for the change to take place, i.e. how long it took for the heat to fully permeate the overlying floor structure.

The requirement was for a floor structure equivalent to a slab on the ground where the slab’s thickness was no more than 150 mm with 150 mm of underlying insulation. The pipes’ center-to-center distance was 300 mm. 8 mm tiles with adhesive (4 mm) were used as a surface layer.

In the simulations for the embedded underfloor heating system, pipes were placed so that there was

1. 110 mm of concrete above the top of the pipe;
2. 55 mm of concrete above the top of the pipe;
3. 30 mm of concrete above the top of the pipe.

In a fourth simulation, a surface-mounted system similar to Flooré was installed instead, but this had a center-to-center distance of 300 mm and aluminum foil 0.5 mm thick.

Temp = Temperature
Timmar = Hours
Effekt = Power [W7m3]
Tidskonstant = Time constant

One way to describe the inertia is by using a time constant which, in this context, is the time it took until the process had “settled down”. In this case, the time constant was defined as the time it took for 63% of the process to take place before all the heat passed through the structure’s surface.

The results are presented below.

Ingjutet = Embedded
Cellplast och folie = Panels with aluminium foil
Timmar = Hours
Effekt = Power [W7m3]

The following conclusions can be drawn from the calculation example:

• The closer the coil to the floor surface, the more efficient the heat transfer (you get more power upwards per degree Celsius). The concrete also has some thermal insulation property in relation to underfloor heating, particularly if the pipes are embedded deep down in the structure.
• Because of the concrete’s heat-storing capacity, the structure will be sluggish. It takes time for the heat to migrate through the concrete mass. Figure on the top left shows that no heat is emitted from the floor surface for the first hour, as the heat needs to permeate the concrete layer.
• Surface-mounted systems react quickly to meet the heating requirement, with a lower water temperature than for the embedded systems. What is not illustrated in figure on the bottom left is that the floor surface temperature becomes completely irregular: there is a risk of being able to feel where the pipes are located. The “bend” shown in figure on the bottom right is due to the heat being routed quickly upwards in the initial stage. Then the concrete slab underneath the system gradually heats up, after which the slab gets warmer and more heat is emitted upwards.