Greater dependency on natural forces creates uncertainties about ventilation performance, particularly the effects of changing wind conditions. The location and size of ventilation openings in maintaining the desired rates of air extract are also important variables.
In many cases, natural ventilation systems are designed around the stack effect, with the assumption that wind will increase ventilation. However, building features such as chimneys and atria, intended to help ventilation may actually counteract the system if incorrectly designed.
As a result, designers are using various forms of modelling to determine the effects of wind on buildings and their ventilation systems. Town planners are also requesting that modelling be carried out in the initial stages of projects in order to determine the effects of a proposed building on the existing environment.
Essentially, external wind forces can affect a building's internal environment in two main ways: by creating points of high pressure and through high windspeeds. By looking at the effects of these variables prior to construction, potential problems can be identified and the wind put to good use in the ventilation system.
The information required from testing should include:
- the pressure coefficients over the building (in particular at inlet and outlet points);
- a visualisation of air movement around the building.
Methods of modelling
The two main methods used for modelling wind are computational fluid dynamics (cfd) and wind tunnels.
Several well established cfd computer packages are available to model the effectiveness of natural ventilation chimneys. The common approach is to create a two-dimensional slice of the building in question, which is then subject to various computer simulations.
By altering the boundary conditions of each variable in turn, it is possible to identify performance thresholds. Often, in order to get a full picture of possible problems, the design must be viewed from several angles and in a series of slices.
The use of cfd packages can provide accurate results. However, it can be difficult to set up very complex geometries and an accurate layout of the local environment. Some prior knowledge of possible problem areas may also be required to optimise the cfd model.
These problems can be overcome by using physical three-dimensional models of a building in a wind tunnel.
Scale models of buildings and the surrounding site can be set up on a rotating platform and then subjected to air movement over 360 degrees. Various substances, including smoke and sand, are used to show a three-dimensional result of the changes in windflow caused by the building and any possible problems which may arise.
Video cameras can capture wind flow patterns on computer for subsequent investigation and analysis. The shape and positioning of ventilation chimneys can then be examined, with alterations being quick and easy to make.
Wind pressures on the envelope of the building are also measured with a hollow version of the model being wired to a meter. The difference in pressure gradient from the inlet to the outlet can be used to ensure that a chimney intended to vent air will do so successfully.
One such tunnel is currently being operated in the Welsh School of Architecture at Cardiff University. The tunnel is being used to test the effectiveness of a wide range of ventilation chimney designs.
Case study 1: atrium ventilation
One typical area where wind tunnel modelling can be used to fine-tune building design is in the natural ventilation of building atriums. The usual technique is to use buoyancy forces to draw cool air into the area at low level, and expel warm air at high level.
A common problem with this approach is that, depending on the degree of exposure of the atrium vent and the direction of the prevailing wind, air may be pushed back into the building. This conflict can prevent full ventilation taking place and may result in the building overheating.
Wind tunnel tests have been carried out on a scale model of such a system proposed for the Saga Group headquarters1. For certain wind directions and speeds, the buoyancy and wind forces were found to cancel each other out.
The effects of varying the building design and adding wind devices were investigated in order to solve these problems. One approach was to move the outlet point to the terraced area. However, this did not suit architectural design and produced the possibility of problem wind in the terrace.
In this case, the addition of a wing-shaped deflector in front of the outlet point was found to maximise the ventilation flow. The plate shielded the outlet from direct wind pressure and created wind flow across the vent, reducing positive pressures and generating a strong suction. This in turn supported the buoyancy flow for the predominant wind directions.
While this solved the problem, it would be sensitive to disruption from the addition of solid solar shades.
Case study 2: H-pot chimneys
This case concerned the use of H-pot chimneys for the Short Ford Associates designed Contact Theatre in Manchester. The building was designed to be fully naturally ventilated and passively cooled. It is sited beside a large building and low level inlets were included in the building design2.
A combination of wind tunnel, cfd and building energy model analysis was used to test the validity of the design. In addition to the standard requirements of natural ventilation, the potential conflict of acoustic design had to be considered.
Air enters through banks of acoustic isolators and is extracted via large stacks at roof level. Although the stack effect naturally promotes air flow due to the higher internal temperatures, wind effects could have caused problems when combined with the effects of the surrounding environment. The aim was to ensure a positive pressure gradient between the inlet and outlet under all wind conditions.
It was found that the H-pots maintained a negative suction pressure relatively independent of wind direction. The location of the low level inlets, however, did affect the functioning of the system and were the main variable of the design. Positioning of the inlets and pots was changed on the model and re-tested until the optimum combination was found.
Case study 3: balanced flues
With an inlet/outlet balanced flue the inlet faces the prevailing wind and the outlet is in the wake.
Tests were carried out to examine the changes caused to inlet and outlet flow by wind direction, chimney location and height and the proximity of surrounding buildings. In addition to smoke visualisation, laser doppler was used for speed measurement1.
It was found that the chimney is wind direction dependent and the location of the inlet and outlet positions must be carefully planned to avoid reverse-wind flows.
With duo-pitched roofs, for example, if the device is positioned at low level on the downward side, wind induced ventilation would be minimal as the mean pressure would be equal around the opening. Positioning the chimney exhaust at height (typically 1 m) above the ridge was found to produce the best results.
It was also concluded that the use of stack-only conditions would have minimal benefit with these chimneys, instead a wind enhancing or opposing situation should be used.
Limitations
The scaling of buildings in a wind tunnel will provide an accurate picture of what would happen under real circumstances. 100% exact results are unlikely to be obtained due to the lack of thermal effects.
An alternative is to combine cfd with wind tunnel modelling, a method being used at Cardiff. External airflow patterns and wind pressures on the building around the openings are examined in the wind tunnel, while a cfd model looks at the internal airflows. With this combination, it should be possible to get accurate results.
Source
Building Sustainable Design
