Thermal Network Analysis Research Articles

Thermal performance of adobe structures with domed roofs and moist internal surfaces

The inside air and the mean radiant temperatures of two buildings, one built of brick having a flat roof and the other built of lightweight adobe and having a domed roof, were estimated through a thermal network analysis. The analysis was repeated for both buildings when their ceilings and inside wall surfaces were kept moist and evaporatively cooled. A design day for a hot, arid region was considered for the analysis. It is shown that the temperature of the moist surfaces is reduced appreciably and the floor and the air temperatures are also reduced by their heat transfer to the moist surfaces. When natural ventilation rate is high, the room air does not become uncomfortably humid. With low mean radiant temperature in the moist buildings, thermal comfort can be maintained for the occupants. The total area and the duration when the surfaces are kept moist, along with the natural ventilation rate, can be controlled by the occupants to provide thermal comfort when it is otherwise uncomfortable. The use of a domed roof with a hole in its crown can ensure high ventilation rates at low wind velocities and in buildings which cannot otherwise be sufficiently vented.

Weekly storage of coolness in heavy brick and adobe walls

Weekly storage of coolness in heavy walls (walls with large thermal inertia or large characteristic time constants or low Fourier numbers) was investigated numerically by considering one-dimensional heat conduction through the walls. The study consisted of first analyzing the heat flow through a single wall and considering various boundary conditions on the inside. The boundary conditions were: constant inside air tempeture throughout the day, variable inside air temperature on a 24-hour cycle, and variable inside air temperature on a 48-hour cycle. Next, the heat flow through walls was studied through a thermal network analysis of a simple building. In this case it was assumed that the ambient air temperature (following a periodic distribution) was increased suddenly and followed this new distribution for many days thereafter. It was concluded that walls with high thermal inertia or time constant can store coolness for several days. The larger the time constant of the wall (for example adobe as compared with brick) the longer it takes for the wall temperature to reach a steady periodic distribution after the change has occurred. However, because of low thermal conductivity of adobe, the retrieval of the stored coolness in these walls is slow, and the mean daily temperature of the room air in the adobe building does not change appreciably beyond seven days after the change. Increase of the wall thickness beyond 50 cm does not improve the thermal performance of the building significantly.

Thermal network analysis of a conditioned building: The effect of location of insulation

Thermal network analysis has been applied to an air-conditioned enclosure to predict the daily variation of heat flux. The effect of putting the insulation on the concrete roof/wall has been analysed in detail. It is found that, when only the ceiling is insulated, the difference in concrete thickness, either on the top side or bottom side with a fixed thickness of insulation sandwiched in between, total concrete thickness remaining the same, does not affect the total heat flux appreciably. The contributions to the heat flux from the south and west walls are also found to be quite significant. The effect of putting the insulation on the south, as well as on the west, walls is also significant.

Thermocirculation characteristics of a Trombe wall passive test cell

Passive heating, using Trombe wall elements for solar energy collection, can be readily integrated into new building structures and can be retrofitted to existing buildings. Although the basic physical principles underlying the natural operation of such air thermosiphon devices have been well known for many years now, there is still insufficient design information available to allow sensible sizing and installation decisions to be undertaken by architects and thermal system designers. In this experimental investigation, flow visualisation studies have given a deeper insight into the fundamental flow mechanisms; whilst air velocity and temperature measurements have been used to explore the natural convection heat transfer processes involved in the thermocirculation flow. Particular attention has been paid to the effects on operational performance of the wall parameters such as wall/glazing distance and vent size. Adequate data correlation of the experimental results has been achieved by using expressions derived for natural (free) convection processes occuring between vertical paralled plates; in this case the vertical heated wall surface and the cooler vertical glazing panel. It is felt that this type of fundamental heat transfer information is vital so that the overall performance of Trombe wall systems can be adequately modelled using the large range of simulation techniques currently available for thermal network analysis.

Theta — a desktop computer solution for nonlinear time dependent heat flow problems

THETA is a thermal network analysis program having an in-core solution, written in Hewlett Packard extended basic and implemented on the Hewlett Packard 9845 desktop computer. Both steady state and transient solutions may be obtained by an iterative solution technique in which all material properties may have non-linear temperature dependence, independent of each other. Material properties are accessed from permanent archives which may be progressively extended. Transient solutions are governed by a flexible system of sequence control instructions, which allow repetitive cycles to be specified as sub-sequences in the main sequence. The modelling facilities allow structure and background nodes to be connected by radiation, convection and conduction links, the latter including non-homogenous/interface effects. Nodal temperatures may be either constrained at particular values, with certain bounds, or used to trigger temperature dependent heat sources/sinks. The software includes data checks leading to warning messages and extensive editing routines to facilitate the modification, extension and rearrangement of model data as a design progresses. An output option allows the user to specify a mix of results printouts, giving either full details of nodal temperatures, link heat fluxes and heat balances, or an abbreviated output listing only temperatures and heat balances. Post processor facilities allow results to be sorted to indicate significant areas, and complementary programs permit the calculation of radiation view factors between arbitrary surfaces.

Cooling of Distribution Transformers in Vented Underground Vaults

The thermal performance of an instrumented distribution transformer in a vault was measured and compared for still air ambients and low-speed wind with various covers (grates and caps) and flow-control barriers and splitters. Correlations for free convection from the tank and vault as well as correlations of the thermal impedances due to air recirculation for various cover and flow-control designs were obtained using the data and thermal network analysis of the radiant and convective transfers. Data show distributions of oil, tank wall and cover temperatures, heat fluxes, and thermal conductances.