Outline
- Abstract
- Keywords
- Nomenclature
- 1. Introduction
- 2. Formulation of the Problem
- 2.1. Experimental Setup
- 2.2. Numerical Setup
- 3. Results
- 3.1. Comparison Between Numerics and Experimental Data
- 3.2. Pressure Drop
- 3.3. Throw and Drop of the Airflow
- 4. Conclusions
- Acknowledgment
- References
رئوس مطالب
- چکیده
- کلید واژه ها
- 1. مقدمه
- 2. فرمولاسیون مسئله
- 2.1. روال آزمایشگاهی
- 2.2. روال عددی
- 3. نتایج
- 3.1. مقایسه بین داده های عددی و آزمایشگاهی
- 3.2. افت فشار
- 3.3. پرتاب و سقوط جریان هوا
- 4. نتیجه گیری
Abstract
In this paper we have developed a Micro/Macro Level Approach (MMLA) method to model a conventional diffuser. This procedure reduces the number of points in order to solve the airflow in a room. This can be seen as a numerical box method, and it provides a number of variables at the inlet of the room (outlet of the diffuser) which are difficult to obtain experimentally, e.g. turbulent properties. Flow visualizations have been carried out to determine the shape of the plume. These qualitative experimental data are also compared with numerical temperature fields for sufficiently high Reynolds numbers and give accurate results. In addition, we report the throw and drop of the diffuser for different flow rates. Finally, the pressure drop of the terminal device has been obtained experimentally, and it is also shown that numerical results can predict it accurately. A discussion of two different methods to compute the pressure drop is given, showing the differences in relation to the airflow characteristics near the outlet of the diffuser.
Keywords: Air - CFD - Diffuser - HVAC - MMLAConclusions
In this paper we have developed a method of studying the characteristics of an HVAC diffuser with CFD and experiments. We have used experiments to visualize the fluid flow by introducing oil droplets, which allows us to check the accuracy of the numerical simulations. This method is less precise than measuring the real flow using PIV or hot-wire anemometry, but easier to setup and analyze since the only experimental devices required are a laser and a digital camera. It has been stated that only one measurement of the velocity field inside the duct is required to obtain the airflow rate and some image processing. Reasonable good agreement was found between the temperature field obtained by numerical simulations and the experimental data of the plume shape. A method of numerically solving the fluid flow was developed. This procedure separates the fluid flow into two linked problems.
Firstly, the numerical problem inside the diffuser was solved with a very fine mesh due to the presence of boundary layers near the vanes, and secondly the airflow was determined on a coarser mesh in the room. The connection between the two problems lies in the fact that all the variables at the outflow of the diffuser are set as the inlet variables of the room problem. This speeds up the calculations and good agreement can be found between experimental results and numerical data, the source of small errors being the limitations of the experimental technique and the number of pixels of the camera. Besides, this method allows us to map turbulence properties that are usually estimated on the box method or the PVmethod. It has been shown that the equation that governs the shape of the plume is different from those presented in literature [31]. We have shown that other authors predicted a quadratic flow with other conditions, as the results presented in the paper. In addition, from the numerical results we computed the throw and drop of the diffuser, which are significant parameters for designing. The drop shows a saturation in its maximum value, the explanation of which is the movement of the airflow in the transverse direction.
Finally, we have also shown that we can reproduce the results for the pressure drop in the diffuser, another important parameter for designing airflow distribution systems. We have distinguished the effect of the air loss of kinetic energy discharging in a room with the effect of the vanes in the diffuser, comparing numerical results and experimental data. The integral method allows us to obtain a mean pressure drop in the air terminal device by taking into account the vane geometry which gives a better estimation than the line method for higher flow rates.