H. Koskela, P. Saarinen, Finnish Institute of Occupational Health, Turku, Finland; J. Heikkinen, VTT Technical Research Centre of Finland, Espoo, Finland; E. Sandberg, Satakunta Polytechnic, Pori, Finland.
CFD-modeling provides a method for predicting the airflow pattern, thermal comfort, and contaminant distribution in a space. The application of the method to practical ventilation problems has constantly increased, together with the development of the modeling software and the processing power of computers. An important issue for the applicability of CFD-modeling is the availability of proper boundary conditions for the objects in the room, especially for air diffusers. For an individual modeler, it is usually not possible to determine reliable boundary conditions for each diffuser type in a modeled space. Therefore, validated boundary conditions should be made available for modelers. This study presents simplified boundary conditions for a square multinozzle ceiling air diffuser. The model is a so-called slot model, where the correct momentum of the supply airflow is adjusted by changing the dimensions of the inlet opening. The boundary conditions were determined based on jet flow measurements in the near zone of the diffuser. The measurements were carried out in a laboratory test room with an automatic traversing system and an ultrasonic velocity sensor which is able to detect all three velocity components in the jet. The slot model was created by dividing each side of the diffuser to six parts and giving different velocity boundary conditions for each part. The model was validated in the laboratory room in nonisothermal test cases with four different Archimedes numbers for the supply airflow. Heat was generated in the room with a floor heating system. The model correctly predicted the flow pattern in all test cases. The selection of turbulence model was found to have significant effect on the spreading of the supply jet. Detailed flow measurements in the near zone were found to be necessary for determining the boundary conditions for diffusers with a complex airflow pattern.
A. Kolesnikov, CD-adapco, Melville, NY; J. Baker, M. Grubert, S. Ericson, The University of Tennessee, Knoxville, TN; J. Bennett, NIOSH, Cincinnati, OH.
Many common laboratory handling practices such as weighing, pipetting, container transfer, autoclaving, and incubating can introduce aerosols, gases, and vapors into the inhabited environment. In addition, laboratory use of high-throughput technology with hazardous solvents (e.g., DMSO, methanol), biological agents (e.g., HIV, TB, hepatitis), and novel compounds of unknown potency (e.g., drugs), is rapidly expanding. Increased safety awareness leads to steady refinement of regulations emphasizing the need to accurately perform risk assessment studies of human exposure to airborne contaminants in cases of their accidental or intentional release into an indoor occupied space. Experimental tests are costly and provide only limited data. The goal of this study is to explore, hence validate the use of computational fluid dynamics (CFD) for generating reliable predictions of laboratory airflow patterns hence contaminant mass transport distributions. Laboratory contaminant mass transport is dictated by room airflow patterns and contaminant source momentum. Mechanical ventilation (including the buoyancy effects of heating/cooling), location and characteristics of contaminant sources, occupancy, and lab equipment determine the contaminant concentration field. Quantitative inhalation exposure assessment therefore requires detailed assessment of the indoor ventilation velocity vector distribution. The presentation is inaugurated via CFD prediction for established experimental room airflow benchmarks describing supply, forced and mixed ventilation scenarios. The availability of quality experimental data for each case enables detailed validation studies to assess the impact of different turbulence closure models, numerical dissipation and domain discretization density. Following, airflow and contaminant mass transport CFD experiments are detailed for a current NIOSH experimental laboratory environment. Pollutant source, ventilation supply/exhaust and equipment locations are all shown to play important roles in the resultant contaminant distributions within the floor plan, The impact of approximations made in forming the CFD models is specifically addressed, to quantitatively assess the associated error mechanisms in prediction fidelity.
Q. Chen, T. Zhang, Purdue University, West Lafayette, IN.
Computational fluid dynamics (CFD) has been widely used to study how contaminants are transported from one building location to another together with airflow. In practice it is also important to know where the contaminant sources are if we know the airflow pattern and contaminant concentration distributions in a building. This presentation shows an inversed CFD method for the identification of the contaminant sources. By inversing the CFD calculations, the numerical procedure becomes unstable and the convergence is a major problem. This presentation imposes a remedy method to solve the numerical instability. With the new method, this presentation will show, through a few demonstration cases, that it is possible to identify contaminant source locations if the corresponding airflow patterns and the contaminant distributions are available.
P. Mustakallio, R. Kosonen, Halton Oy, Kausala, Finland.
The main design target for the power plant hall ventilation is to keep the difference in the engine level temperature and the ambient air below 10 Kelvin, and to keep the temperature of hall air cooling the generator as low as possible. Nowadays, the systems are based on mixing ventilation and relatively high airflow rate. The studied system uses big fans supplying the air straight to the engine hall and mixing it well. The design targets are met but energy consumption for the ventilation is quite high, which is directly from the generated electricity. Novel ventilation strategy using displacement ventilation was studied in two-diesel engine power plant cases. New ventilation strategy was developed for power plant module. Diesel engine power plant with existing and new ventilation systems was modelled by using computational fluid dynamics (CFD) tool. Periodic boundary conditions were used on the both sides of the module to model the multiengine power plant. In the most workable configuration, air is supplied through the low-velocity units from both ends of the engine hall, and cooling air circulated through the generator is directed straight to the exhaust opening in the ceiling. This makes it possible to reduce the supply airflow rate 30% from the current setup and distributes the supply air more uniformly.
S. Kotha, R. Ryan, Flowsciences, Leland, NC; D. Walters, KCP, Inc, Raleigh, NC.
Fume hoods, also known as laboratory chemical hoods, are one of the most important and widely used engineering controls in laboratories. Fume hoods were introduced about 100 years ago to protect personnel working with hazardous materials. While many changes and improvements have been made, the basic concept and design remains the same. The majority of the laboratory hoods are constant air volume (CAV) hoods, which draw a constant amount of air at all times. Rising energy costs have made these hoods exceptionally expensive to operate. In addition, CAV hoods do not react rapidly to airflow disturbances within the hood or within the laboratory and, hence, their sole purpose of containment and protection can be seriously compromised. This presentation describes a new low-flow; variable air volume (LF-VAV) hood, which was developed, in part, to help overcome the high operating costs of CAV hoods. Because LF-VAV hoods operate at considerable lower exhaust volumes than CAV hoods they are energy efficient and result in considerable savings. The control system designed for use in this LF-VAV hood responds rapidly to airflow disturbances within the hood and the laboratory and automatically adjusts the face velocity so that containment is not lost. Computational fluid dynamics (CFD) was used to optimize the airflow and ensure there is minimum turbulence so there are no containment losses from the hood. By using CFD, the performance of the hood can be seen even before a prototype is built and, hence, serves as a very cost-effective and timesaving tool, which also significantly improves containment. The new low-flow, variable air volume hood, significantly improves containment, and because it is coupled with an efficient, rapid-response control system helps ensure the safety of laboratory personnel and provides substantial energy savings.
D. Gubler, A. Schaelin, AFC Airflow Consulting AG, Zurich, Switzerland.
A given configuration of working places with a high contamination rate of CO has to be renovated. The actual situation is a mixing ventilation system based on dilution of the CO in the working area. The paper describes how CFD was applied to investigate various basic ventilation concepts such as local exhaust hood, dilution, push-pull ventilation and others. Based on first CFD simulations, a decision was made in favor of a local exhaust ventilation hood. Further CFD simulations were made to optimize the design and the operating parameters of the exhaust hoods. After the installation was realized, measurements and smoke tests were performed under real working conditions to verify if the system works correctly and if the performance is sufficient. This project showed how CFD simulations can be used in an early project phase to make the basic concept decisions. In a second phase, CFD can be used for systematic optimization of operating and design parameters. Also, the comparison between the measurements and the simulations showed excellent agreement. The measurements confirm the predicted contaminant removal efficiency completely and prove the functionality of the system.
D. Gubler, A. Schaelin, AFC Airflow Consulting AG, Zurich, Switzerland.
In a production plant for pharmaceuticals Switzerland, a new packaging (blister-)line was installed. The air quality of the environment in this room is in general zone 3 (no limitation on the number of particles), whereas the required air quality for the product is “zone 2” (less than 100,000 particles per square foot). A ventilation solution was developed using CFD simulations. Various concepts are known to solve this problem. Typically, the whole blister-line with the loading part is put below a laminar flow. This solution guarantees the required air quality but a huge amount of airflow is needed. The costs for installation and operation of this solution are enormous. CFD was used to develop a local ventilation device which creates a local zone of “zone 2”-air around the loading zone. The solution consists of a push (supply air) and a pull (exhaust air) device with integrated filters (H13). The CFD simulation were used to optimize the design (size and position) of the apparatus and the operating parameters for the whole loading process including movement of parts. Based on the results of the CFD simulations, the design team was enabled to decide for a very energy efficient ventilation systems, which (a) guarantees the required air quality, (b) reduces the expected airflow rates and hence the energy consumption, and low maintenance costs.
J. Fontaine, J. Blaise, INRS, Vandoeuvre, France; P. Hynynen, Lappeenranta Regional Institute of Occupational Health, Lappeenranta, Finland; R. Devienne, LEMTA CNRS UMR 7563, Vandoeuvre, France.
Many industrial processes bring into play heat sources (e.g., furnaces, surface treatment, welding, metal cutting). Through natural convection, these thermal sources create plumes likely to convey various pollutants in the occupied zone. Naturally, this situation creates occupational risks and must be controlled through knowledge and control of the structure of implicated flows. Characterization of thermal plumes generated by industrial heat sources is necessary for the design and dimensioning of the appropriate ventilation systems used to achieve a safe workplace environment. An experimental method to characterize in real-scale the plumes of three-dimensional thermal sources of finite dimensions was developed at INRS. A test bench was designed and instrumented for this purpose. It consists of an aeraulic cell whose dimensions are 4.2 m × 4.8 m × 5.6 m equipped with an air treatment unit and a ventilation system allowing a vertical airflow inside the cell. The bench is instrumented with a three-dimensional displacement robot allowing the positioning of speed and temperature sensors in any point of volume. An experimental study of the thermal plume developed above a cylindrical source of finite dimensions was conducted with the aim of verifying the point source model. The heat source was a 1 m diameter, 1 m high cylinder featuring five temperature controllable zones (four 0.25 m high cylindrical surfaces and a top disk) of identical surface area. For the same total convective power, various distributions of source surface temperature were considered. The velocity and temperature fields of the plume were measured simultaneously by anemometric probes and thermocouples. Two-dimensional elliptical Gaussian distribution was introduced to model these fields in the developing zone of the plume. These measurements allowed the plume thermal and dynamic radii to be deduced. These radii vary linearly with height and enable the plume virtual origin to be located.
A. Prits, Stantec Consulting Ltd., Mississauga, ON, Canada.
Stantec Consulting Ltd. was retained by a client to evaluate the potential risk to the life safety of operators from an accidental release of a large volume of nitrogen being stored under high pressure. The objectives of the assignment were to (1) Determine the expected distribution of the large scale nitrogen leak and specifically consider the potential to displace the life sustaining oxygen atmosphere in the pit area where operators carry out their duties; and (2) if a potential for displacing the life sustaining oxygen atmosphere in the pit area exists, determine recommended modifications to prevent such an occurrence. To meet these objectives, Stantec chose to use CFD modeling validated with field gathered data and observations of the operation. This presentation will summarize the analysis carried out and the methods used in the application of CFD techniques. The major conclusion from the investigation was that a leak could result in an atmosphere with a reduced oxygen content. Another interesting conclusion was that the high pressure release could result in windstorm like conditions that would impact on any attempts by operators to leave the area via the available stairwells. The proposed action plan to ensure operator safety developed as a result of this investigation will be reviewed.
R. Braconnier, INRS, Vandoeuvre-les-Nancy, France.
Paint booths incorporating a pit are used to apply paint by spraying on the lower surface of substrates where mechanical turning over is impossible. This study used CFD techniques to simulate the air movements and the dispersion of pollutants in a booth during the painting of the underside of the chassis of a trailer of a heavy goods vehicle. Two substrate models were considered: open-work substrate (without platform body) and solid substrate (with platform body). For each substrate model, the performance of two types of pit ventilation were compared: (1) ascending vertical ventilation obtained by the introduction of new air through a grating at the bottom of the pit, and (2) descending vertical ventilation obtained by extracting the air through this grating. The structure of the flows in the movement zone of the painter was studied for different booth ventilation operating conditions: the air velocity at the ceiling varying from 0.15 to 0.40 m/sec and the velocity at the bottom of the pit from 0.30 to 1.0 m/sec. For an open-work substrate and descending pit ventilation, the painter was located in a descending flow and found to be in the path of the pollutant, between the source and the extraction. For an open-work substrate and ascending pit ventilation, the painter was located in a descending or ascending velocity region depending on the ratio of the blowing velocity at the bottom of the pit to the blowing velocity at the ceiling. For a substrate with platform body, whatever the longitudinal position in the booth, the painter could be located, depending on the local direction of horizontal air current and on the direction of spraying, downstream of the source and therefore in the path of the pollutant, whether the ventilation of the pit was descending or ascending.
R. Braconnier, INRS, Vandoeuvre-les-Nancy, France.
During spray painting operations, the initial emission of pollutants occurs inside an air jet coming from the head of the paint gun. The aim of this study was to improve knowledge of the characteristics of this type of jet (values of the air velocities produced, dimensions and appearance ratio of the jet) in order to facilitate the design of ventilation systems. Anemometric measurements were made, in the absence of paint, using a very fine Pitot static tube in front of a conventional pneumatic gun set to operate with a flat jet. The relative pressure at the handle was 4.4 bars and the consumed airflow rate 17.5 m3/hour. The air velocities were measured along the axis of the jet as well as in three planes orthogonal to this axis encompassing the usual spraying distance and located 100, 200, and 300 mm from the gun. The air velocity along this axis drops sharply with the distance from the gun: from 98 m/sec at 25 mm to 7.6 m/sec at 600 mm, according to a profile that can take the form of a power function of distance. The three-dimensional character of the jet is very marked. The velocity contours in the orthogonal planes have an elongated ovoid shape along the perpendicular direction of air injection by the two spraying head lugs. The lateral velocity profiles in the orthogonal planes can be adjusted to gaussians. The jet appearance ratio varies from 3.6 at 100 mm from the gun to 2.1 at 300 mm.
Posted May 30, 2006