Study of the average heat transfer coefficient at different distances between wind tunnel models

The paper presents investigations of physical and climatic factors with regard to design and process variables having effect on heat transfer in the building model system at different distances between them in the airflow direction. The aim of this work is to improve energy efficiency of exterior walls of buildings. A method of physical simulation was used in experiments. Experimental results on the average values of the heat transfer coefficient in the building model system are presented herein. A series of experiments was carried out on a specific aerodynamic test bench including a subsonic wind tunnel, heat models and devices for giving thermal boundary conditions, transducers, and the record system equipment. The paper contains diagrams of the average heat transfer distribution at fixed Reynolds number and the airflow angle of attack; the average values of the heat transfer coefficient for each face and wind tunnel models as a whole at maximum, medium, and large distances between them. Intensification of the average heat transfer was observed on the downstream model faces depending on the distance between models.


Introduction
Researchers of the Construction Engineering Technology Department of Tomsk State University of Architecture and Building (Tomsk, Russia) are involved in a variety of problems dealing with the airflow structure [1], and the local and the average values of the heat transfer coefficient [2] with the aim to improve the design technique and energy efficiency of exterior walls.
All experiments on the average value of heat transfer coefficient were carried out using the subsonic open-return wind tunnel with air sucked through a duct. The experimental methodology and processing of wind tunnel test results are described in the work of Mokshin et al. [16].
In this study, the key parameter is calibration of wind tunnel models expressed by the distance ratio L1/a between models in the light (L1) and the cross-sectional dimension of the wind tunnel model (a) equal to 50 mm. Figure 1 a, b shows schematic layouts of the test Model 2 relative to the test Model 1 at the airflow angle of attack  = 0 о .     At a distance between the wind tunnel models ranging from 0,5 to 2,0 a sharp increase of the average value of the heat transfer coefficient on (С -D) face of Model 2 is observed, the value of which is lower than on the same face of Model 1. At distance ratio L1 / a = 0,5 the heat transfer coefficient of (С -D) face of Model 2 is 9,9 % lower than on the same face of Model 1. At a maximum short distance, the arch-type vortex formed by Model 2 is less intensive than that one formed by Model 1. In increasing the distance ratio up to 2,0 the arch-type vortex formed by Model 2 promotes a stronger effect on Model 2. At that, average values of the heat transfer coefficient on (С -D) face of Model 2 are practically the same (0,46 % difference) as on (С -D) face of Model 1.

Experimental
At a distance between the wind tunnel models ranging from 2,0 to 4,5 average values of the heat transfer coefficient on (С -D) face of Model 2 are being sharply increased, and become higher than on the same face of Model 1. At distance ratio L1 / a = 4,5 the heat transfer coefficient of (С -D) face of Model 2 is 5,7 % higher than on the same face of Model 1. At this distance, the intensity of the arch-type vortex formed by Model 2 exceeds the similar arch-type vortex formed by Model 1.
At a distance between the wind tunnel models ranging from 4,5 to 12,0 the intensity of the archtype vortex formed by Model 2 on its (С -D) face becomes lower. The average value of the heat transfer coefficient on (С -D) face of Model 2 is reduced, thereby approximating to that of (С -D) face of Model 1. At a distance between the wind tunnel models ranging from 0,5 to 5,5 the average value of the heat transfer coefficient increases on the entire Model 2. At a maximum short distance L1 / a = 0,5 the heat transfer coefficient on Model 2 is 9,7 % higher than on Model 1. This is because Model 2 is affected by arch-type vortices formed by Model 1 and the strong flow separation occurred at the top of Model 2. The ultimate effect from these phenomena is recorded at a distance ratio L1 / a = 5,5 when the average value of the heat transfer coefficient achieves its maximum, i.e. 24,9 % higher than on Model 1.
At a distance between the wind tunnel models ranging from 5,5 to 27,0 these forces become lower, and the average value of the heat transfer coefficient on Model 2 is reduced, thereby approximating to that of Model 1.
The As shown in Figure 6, at the increase of distance ratio L1 / a from 0,5 to 30 the airflow behavior is changed between the wind tunnel models and on lateral faces of Model 2. This leads to the intensification of the average heat transfer on Model 2 faces, while the distance between wind tunnel models increases. At the same time, in varying the distance ratio L1/a not more than by 10%, heat transfer on (C -D) face of Model 2 changes insignificantly.