Hygrothermal simulation of a ground slab

The lengths L1 to L4 specified in the drawing are still to be determined. These lengths can be defined, for example, in accordance with the specifications of ISO 10211 (thermal bridges). The length L2 should correspond to half the width of the building. L1 can be set to three times the wall thickness in accordance with the guidelines for thermal bridge calculation. For L3 and L4, there are now two different approaches in ISO 10211.

In our case, temperatures are to be calculated so that L3 could be set at 3 m and L4 at 3 times the wall thickness. However, it should also be noted that ISO 10211 is primarily intended for steady state calculations. The following graphics illustrate the influence of different distances on the temperature.

The two diagrams above show the temperature curves for different ground dimensions. L3 and L4 were set to 3 m, 7 m, and 15 m respectively. As can be clearly seen, there are greater deviations, especially in the temperature in the center of the building. The thicker the ground layers, the lower the temperatures. It is also interesting to note that the differences become smaller again over time. For safety reasons, however, you should still select the data from the variant with the greatest ground thickness.

By default, DELPHIN sets 20°C and 80% relative humidity as initial conditions. At greater depths, the annual average temperature can be expected in the ground. Therefore, it would be appropriate to set this temperature as the initial value for the ground. An initial moisture content should also be set for the ground in this simulation. The ground in Central Europe is normally quite moist. Since the moisture content of the ground has a significant influence on its thermal conductivity, a high initial moisture content should be set. In ground science, a capillary pressure of -40 kPa is recommended. This can also be set in DELPHIN. To do this, first create a new initial condition for capillary pressure by clicking on the green plus button in the initial conditions.

This initial moisture must then be assigned to the entire ground.

In order for the moisture content of the ground to be used for thermal conductivity in a purely thermal calculation, this must be adjusted in the solver settings.

A suitable location is selected from the climate database as the climate. Any model can be used for the indoor climate. Note that the indoor climate can have a major influence on the temperature under the floor slab, especially if the floor slab is poorly insulated. For the outdoor climate, we need two surfaces: one for the wall (inclination 90°, any orientation) and one for the surrounding ground (inclination 0°). Since only thermal calculations are performed, no moisture-related boundary conditions need to be applied.

The image above shows the surface for the ground. The type selected here is Standard outdoor climate with user-defined settings. This automatically accesses the local climate. Only heat conduction and short- and long-wave radiation were selected for the boundary currents. The surface for the wall can be set in exactly the same way, only with a different inclination. Both surfaces still need to be assigned to the corresponding boundaries.

The following boundaries remain adiabatic (no surface assigned):

For the lower ground boundary, you can also set the annual average temperature as a fixed boundary condition or a fixed heat flow. However, this would require additional information that must be justified. If it is known that groundwater flows in the ground area under consideration, this can be set as a fixed temperature boundary condition.

The most important outputs are the temperatures required as boundary conditions for the 1D calculations. This requires outputs with the following formatting:

These outputs must then be arranged at selected positions directly below the floor slab. You can assign as many outputs as you like. However, at least two are necessary:

Additional temperature-related outputs can also be assigned. For example, the temperature field of the entire structure would be quite useful for evaluation and documentation.

Since the structure has a high heat retention capacity due to the large mass of the ground, it is recommended to perform the simulation over a longer period of time (6 to 10 years).

In DELPHIN, a climate condition defines the type and value of a climate component. This can then be used in boundary conditions and field conditions. The next image shows the settings dialog for a climate condition for temperature.

The following settings are important for this example:

The kind of data indicates its source. Here you can specify constant values, parameterize sine waves or, as in this case, import an external data source. When Data points is selected, you can choose the file. You can choose from different formats:

The first two formats are described in the Tutorial 6. The last format should be used here. This format allows the output of the preliminary calculation to be used directly as climate data. However, there is one point to note. In DELPHIN, a distinction is made between two different types of climate data handling:

Cyclical data is mostly used for assessment calculations. It covers exactly one year and can be used by DELPHIN for any length of time. Non-cyclical data usually covers a specific period and has a specific start date. It can then only be used for this specific period. Such data is usually generated from measurements. Since the preliminary calculation ran over several years, its outputs can only be used non-cyclically. The period of the hygrothermal calculation must therefore correspond to the period of the preliminary calculation (or be shorter). The outputs can be prepared for cyclical use. However, this is quite complex and will therefore not be described here. In the following, we will use the data non-cyclically. The procedure for this is as follows:

If everything is correct, you will see the graphical representation of the data as shown in the image above. You should then give this climate condition a meaningful name.

Now create a new boundary condition in a similar way (green plus sign for boundary conditions). The following settings must be made here:

Finally, give the new boundary condition a new name and it is ready.

Since we now want to perform a hygrothermal simulation, we need boundary conditions for moisture transport. Since the ground is directly adjacent at the bottom, there is the possibility of vapor transport and liquid water transport. We therefore need two additional boundary conditions:

We do not have separate climate conditions for either boundary condition, which is why the climate is assumed to be constant. A capillary pressure of -40 kPa was assumed as the initial condition for the ground. This can be converted into relative humidity using the Kelvin equation, resulting in 99.97% relative humidity as the climate boundary condition for vapor diffusion. For a constant climate, there is no need to set a separate climate condition; the value can be entered directly into the input field for the climate. A pressure height must be specified for water contact. Since 100 kPa corresponds to a 10 m water column, a pressure height of -4 m must now be set for the selected -40 kPa. Below you can see the two completed dialogs and the climate condition for the pressure height.

As with the other two conditions, you first create the new surface. Select ‘Detailed surface defined by several boundary conditions’ as the type. Once the type has been selected, a list of existing boundary conditions appears at the bottom. Then check the box for the condition you just created. Finally, give it a good name and assign this surface to the construction at the bottom.

Alignment and inclination are irrelevant here and can be ignored. The finished surface must then be assigned to the bottom element.

Now the outputs must be added. First, you have to ask what is to be evaluated (see also Tutorial 4). In our case, it is probably sufficient to check for moisture accumulation. To be on the safe side, you can also add a test for mold on the interior surface. For a better overview, it would be good to have the profiles for relative humidity and moisture content. The temperature profile should still be there. So we need the following outputs with the corresponding settings:

The first three outputs are assigned to the entire structure and the last two to the upper edge. The evaluation can then be performed with PostProc 2.