Authors: David Edwards and Harald Konrad Malzer

The present generation of planners is very familiar with 3D design tools, but unfortunately the issue of energy efficiency plays a minor role in the initial design process and is often taken into consideration too late. This leads to suboptimal details and complex solutions with high end costs. In order efficiently plan Passive Houses, the PHPP was developed as a detailed high quality tool for thermal calculations. The input of the building’s geometrical data, such as surfaces, windows and shading elements, is often challenging and can lead to errors. Therefore, a visual feedback makes the tool more user-friendly and facilitates the control of mistakes. designPH combines the well-known 3D designing tool SketchUp and the high quality calculation of PHPP. Thus, the issue of energy efficiency comes to the fore, right in the designing stage.

The designPH plugin has been developed by PHI to provide a 3D model interface for entering building geometry into PHPP. The benefits of the tool are twofold; firstly it will simplify the process of entering data into PHPP and secondly it can provide preliminary feedback on the performance of the design within SketchUp.

The model geometry is marked up with thermal properties, with the aid of some automatic analysis functions. The tool uses an heuristic algorithm to infer element types, temperature zones and area groups, in order to save input time, but these can be overridden by the user if required. The external heat loss areas and treated floor area are collected and formatted for export to PHPP. Each window is analysed to identify external shading objects and these are exported as input parameters for each of the three key shading types in PHPP (reveal, overhang and horizontal object).

The export function saves model data to a PPP file, the interchange format developed by PHI, which can then be imported into a PHPP model. Through this process, a 3D digital model can easily be converted to a PHPP workbook. After importing a model, the primary data on the Areas, Windows & Shading sheets will be mostly complete, enabling a result for specific space heat demand to be calculated quickly, without the need for much direct data entry. The advantages of this approach over entering the geometry directly into PHPP are, firstly that it should save time on data entry and secondly that it is possible to visually verify in the 3D model that all the heat loss surfaces have been correctly taken into account.

One main feature is that a simplified energy balance is provided within the 3D modelling environment. This facilitates a more effective iterative design process, allowing the user to get an immediate idea of the building’s energy performance, and rule out poorly performing design options, before exporting to PHPP to fine-tune the design and make the verification.

Figure 1: Possible workflows in designPH, demonstrating iterative design capabilities

A simplified energy balance is calculated by designPH within the SketchUp environment. This is based on the same conventions as the Annual Method in PHPP, with some additional simplifications.

Surface areas – The input of surface areas is provided graphically by the user drawing a model in 3D. The designPH automatic analysis algorithm assigns a thermal area group to each surface (using the same nomenclature as in PHPP) and a corresponding default U-value. These values can also be overridden by user assignment. Windows are inserted using a predefined parametric component, which provides links to lookup the thermal properties from the internal components library.

Climate data – Annual climate data is used in designPH; this is of sufficient accuracy to give a preliminary indication of the energy balance. A number of representative climate data sets are supplied with designPH, but (as of Jan 2014) these are not as comprehensive as in PHPP. Additional climate data sets can be generated by selecting the desired data set in PHPP (or inputting user data), then copying the data from cells N6-N8, N26-N29 labelled “Transfer to Annual Method”. Future versions of designPH will allow for the input of additional climate data sets. Components library – Window frame, glazing units and opaque assembly types can be selected from the built-in components library and the corresponding U-values, g-values, psi-values and window frame widths are transferred to the internal calculation. The current set of Certified Components (as of Jan 2014) is provided with designPH and user-defined components may also be entered, including via a built-in U-value calculator.

Transmission heat losses – The total surface area for each area group is automatically summed from the model. U-values are then taken from either automatic or user-assigned assemblies for each surface. Temperature reduction factors are looked up from the assigned area group. Currently designPH uses a default value of 0.6 for the ground reduction factor. More detailed configuration of this may be provided in future. Temperature reduction factors to account for the sheltering effect of adjacent structures (such as garages, porches or unheated basements) are not supported, again this will be a subject for future development. Thermal bridges can be assigned an area group for export from designPH but are not yet included in the heat balance calculation.

Ventilation heat loss – is calculated under the assumption that MVHR is being used and the Passive House minimum level of airtightness is achievable. An allowance is also made for duct losses. See fig. 2 opposite for full details. Future versions of designPH are likely to allow further configuration of these values, either through optional “expert settings” or by selection of predefined configurations suitable for existing stock and EnerPHit projects.

Winter solar heat gains are calculated from the annual perpendicular radiation values provided in the climate data. A reduction factor for frame factor, dirt and non perpendicular radiation is applied as in PHPP. Version 1.0 uses a default value of 75 % for the shading reduction factor, however it is anticipated that the shading reduction algorithm from PHPP will be implemented in designPH Version 2.0. Summer solar heat gains – The current version of designPH makes no calculation of summer solar heat gains, therefore an indication of overheating is not provided. This will be investigated for a future version.

Internal heat gain – is simply calculated on the basis of 2.1 W/m2 of TFA, multiplied by the heating period provided in the annual climate data.

Figure 2: designPH calculation methodology; Version 1.0 (01/2014)

A simple raytracing algorithm is used to perform automatic detection of shading objects, for compatibility with PHPP, these are resolved into three types: horizontal objects, reveal shading and overhang shading. The origin point for these tests is the centre of the cill for horizontal and overhang objects and the centre of the opening for the left and right reveal shading. A reference point on the edge of the opening is calculated and the frame width and reveal depth are added as appropriate to give the offset from the shading edge to the glazing edge, for export to PHPP. For visual confirmation of the process, a dotted line (b in figs. 3 & 4) is plotted in the model from the reference point to the detected point on the shading edge.

A prescan is made in both the horizontal and vertical axes, perpendicular to the plane of the window. Rays are traced at approximately 6 degrees intervals. This will usually pick up the dominant shading objects, but objects smaller than this angular resolution may not be correctly detected. The results of the prescan are analysed to find the largest gap between detected objects; this is assumed to be the sky. A binary search is then performed to refine the result and find the shading edge for each shading component (horizontal, overhang, left and right reveal).

Figure 3: Reveal shading detection (plan view)

a … Prescan lines
b … Reference lines shown in model
c … Shading edge refined by binary search
d … Largest gap (sky)

Figure 4: Horizontal and overhang shading detection (section view)

The method works reliably when obstructions can easily be resolved into the three distinct categories of horizontal, overhang or reveal shading. However, overhangs and horizontal obstructions will only be detected if they intersect the vertical plane passing through the centre of the window and perpendicular to the plane of the window. Therefore intermittent obstructions such as trees, and continuous obstructions at non perpendicular angles from the window may not be detected. In these cases, it may be necessary to model obstructions in a simplified manner in order for them to be detected as the appropriate form of shading.

Objects having an angular resolution of less than 6 degrees may not be correctly detected.

In densely shaded situations - where there is not one dominant gap which gives a sky view or where there is no view of the sky - it can be difficult for the algorithm to resolve the detected objects into the correct components.

Performing additional scans at a higher resolution and non perpendicular angles may help to improve the accuracy of the process, however there will still be situations where it is difficult to resolve the objects into the three components without the need for user intervention.

A more robust solution would be to create a ‘shading mask’ for the sky visible to each window, and export this as a precalculated shading reduction factor, rather than attempting to resolve into the simplified model of horizontal, overhang and reveal shading objects.

The recent development of designPH shows that the use of a 3D model as a basis for the development and planning of passive houses is beneficial and helpful for the user. However, the development potential of designPH, in those areas where further development is needed, as well as in the implementation of many additional features (e.g. integration of building services, urban efficiency analysis, and much more) is promising. The excellent visual feedback to the user (even with complex shapes), as well as the active fault avoidance by means of a rigorous automated analysis of the model, can help to accelerate the whole design process in the field of energy efficiency.

[PHI 2014] designPH version 1.0, (Jan 2014), available at www.designPH.org

[PHI 2013] Feist et al, Passivhaus Planning Package, version 8.

www.designph.org the main website for information on designPH plugin

Visit the “Passive House Tools” section on Passipedia

Read more about the Passive House Planning Package (PHPP)

List of all released conference proceedings of the 18th International Passive House Conference 2014 in Aachen

Conference Proceedings of the 18th International Passive House Conference 2014 in Aachen