Natural History Museum - Darwin Centre Phase II

Introduction

Phase 2 of the Darwin Centre at the Natural History Museum will link directly onto Waterhouse's historic Cromwell Road Building. It features a 'cocoon' situated within a 78m wide by 32m high atrium. The conditions within the atrium and cocoon are critical since the latter will house an entomological and botanical collection of over 28 million insects and 6 million plants.

At the early design stage there was immediate concern over the difficulties involved in controlling the environment in such a large atrium. Flow Analysis were commissioned by Fulcrum Consulting to work alongside and enhance their Building Physics Department's modelling and CFD capability. Flow Analysis studied specific aspects of the project that were pretty labour intensive and involved focussed CFD planning and resolution. Close collaboration between Fulcrum Consulting and Flow Analysis was crucial for the success of the project.

Modelling Approach

The geometry was built from the architect's drawings and plans. The cocoon was imported directly into the CFD software from AUTOCAD software.

Some features of the design are an architectural mesh hanging over the steel framework and a series of deep mullions on the glazing system which act like horizontal fins and thus break-up and reduce downdraughts. Some careful modelling was required for these small-scale geometric features.

Detailed surface temperatures were provided from dynamic thermal simulation for summer, winter and mid-season conditions. These were input into the model along with casual heat gains from people, lighting and equipment.

Winter Simulation

The results show that although the surface temperature differences appear to be fairly small, the atrium is sufficiently large for there to be significant buoyancy-driven air movement.

The figure above illustrates the complex, time-dependent flowfield at two lateral slices through the atrium. In places, the air velocity is predicted to exceed 0.5m/s - sufficient to cause discomfort.

Results indicate the use of underfloor heating alone is insufficient to heat the space and counter the severe air movement. Therefore convector heating has been added to the model, and the effects of varying heat inputs have been quickly analysed. Resultant temperatures within the atrium have been raised to more comfortable levels, but the convector is only partly successful in reducing the downdraught.

Work is ongoing to establish the most satisfactory heating strategy to optimise comfort conditions in the occupied zones of the atrium.

Summer Simulation

The design and size of the atrium make it an ideal candidate for stack-driven natural ventilation. A high level slit vent is proposed for exhaust and one or more lower-level slit vents are proposed for inlets.

Initial simulations with a single lower-level inlet vent indicate a high level of ventilation - around 6.5 ach. However, on the downside, velocities through the inlet vent are high - over 1m/s (see the figure above).

Work is ongoing to establish the ideal number and positions for the vents - the neutral pressure level is critical in this respect. The ventilation which is achievable through natural means is likely to result in comfortable resultant temperatures (see the figure to the right).

BACK TO NATURAL VENTILATION PROJECTS

HOME