By continuing to use the site you agree to our Privacy & Cookies policy

Wind in the bath

In the third article on building modelling, we look at the benefits of water-bath tests to assess natural ventilation performance

A fundamental requirement of the ventilation of school buildings is that adequate airflow must be provided to all areas of a classroom. And with the budgetary demands for both capital expenditure and operating costs in the education sector, low-energy systems have become increasingly attractive.

Naturally ventilated buildings often include large stack structures to provide additional buoyancy to assist the flow of air through the space. The benefit of such stacks lies in the additional buoyancy force that they produce to drive the flow in still conditions. They can also provide additional draw in windy conditions if the dampers are designed appropriately. In some situations, for example with a sloping roof, the geometry of the building restricts the location of the stack to lying below the uppermost point of the building.

This can reduce the efficiency of the stack and, depending on the detailed flow paths, it may even lead to flow reversal in the stack.

Such complicated flow patterns are often not picked up in computational fluid dynamics or in physical wind-tunnel tests, but water-bath laboratory modelling has been used to help inform the designs of a number of new school buildings by exposing these ventilation issues. The application of water-bath modelling has identified that such designs can exhibit a multiple range of flow regimes for exactly the same operating conditions.

This has led to changes in the design of the natural ventilation system.

Playing draughts The 12-classroom extension at Hagley Primary School by Associated Architects has been designed for natural ventilation. Here an opening issuing from an intermediate level in the classroom feeds into a stack, while high up on the outer facade there is a window for outflowing air. A stack combined with high-level windows can enable cross-room upward displacement ventilation, under appropriate conditions.

But the Hagley School project team used insights from detailed water-bath modelling, which exposed some of the flow discrepancies that had not been picked up by other flow assessments, and used these to modify the design and improve the ventilation scheme.

For much of the year, adequate air supply can be provided by operating a ventilation scheme based on a simple upward displacement, cross-flow ventilation mode. Cool outside air enters through a lower vent in the external wall of the classroom, is heated in the classroom and exits into a stack or a corridor area, which has high-level roof vents. On very hot, still days, it can be desirable to increase the ventilation and make use of additional vent area on the external facade. Upperlevel windows on the external wall of a classroom can be included to provide an additional outflow, although the draw from the two outflow openings is likely to be rather different. The design can lead to improved ventilation throughout the classroom if both vents act as an outflow, but depending on the relative sizes and the heights of the upper openings, the stack may not operate in the most effective mode.

Stacked odds The first-floor classrooms at Hagley have a sloping roof. In this case, the stack draws air from the lower side of the slope, which is furthest from the external wall, while a window allows air to vent from the upper part of the classroom. It is not clear that the flow will always be upwards within the stack.

Furthermore, if the flow in the stack is downwards, then the benefit of building a tall stack may be compromised.

The difference between the modes of behaviour can be understood by considering the inflow through the lower window and the outflow through the upper window. The inflow requires a pressure at the base of the room smaller than the external pressure, while the outflow at the upper window requires a pressure in excess of the exterior pressure. As a result, there is some horizontal surface in the room at which the pressure in the interior matches the pressure in the exterior (the neutralpressure level). Above this point there will be outflow and below this point there will be inflow. (It is worth noting that exactly at this point there will be negligible flow. ) The height of the neutral-pressure level depends on the ratio of the size of the inflow opening (lower window) to the outflow opening (upper window). With a large lower opening, the neutral-pressure level is low in the room, while a large upper opening pulls the neutral-pressure level closer to the upper opening.

If there is a third opening, then the flow through this opening may be inflow if the neutral-pressure level lies above the height of the base of the stack. However, there may also be outflow through this stack if the top of the stack lies above the neutral-pressure height, in which case there may be a net outflow from the stack.

Control of the inflow regime, through modification of the area of the lower vent, or through changing the design of the stack, ensures that the ventilation flow through the stack is effective. The occurrence of a cold draught through the stack may be highly undesirable, especially if it leads to cold spots on the base of the room.

In the initial design of the Hagley extension, with an upper opening window and a stack for outflow ventilation, the stack was designed to vent from the first-floor classroom through the wall into a central atrium space and then into a stack. This led to the possibility of inflow ventilation through the stack.

So the access point from the classroom to the stack was moved upwards; this was achieved by venting to the stack directly through the classroom roof.

The entrance to the stack was then above the neutral-pressure level and outflow ventilation developed.

The revelation of multiple-flow regimes in a building is of fundamental importance to the overall design and management strategy. Water-bath laboratory modelling technique is uniquely able to expose the existence of multiple-flow regimes. The results of laboratory modelling can then be used to develop the appropriate mathematical models for sizing the airflow vents, etc, and formulate the appropriate operation strategy for the building.

Shaun Fitzgerald is a research fellow and Andrew Woods is Professor of Petroleum Science at the BP Institute, University of Cambridge

Have your say

You must sign in to make a comment.

The searchable digital buildings archive with drawings from more than 1,500 projects

AJ newsletters