Airflow inside school building office compartments with moderate environments.

INTRODUCTION

In general the air temperature levels in the South of Portugal (the Algarve region) are not very low during winter, are very high during summer, and are moderate during spring and autumn (there are also moderate winter and summer days), the time of year that represents the wider period of school activities. This research analyzed and developed a forced ventilation system, to be used in spring and autumn (and also on moderate winter and summer days), being of easy implementation in all school building offices, with low initial investment and low consumption level. This ventilation system, which worked in general in spaces not subjected to direct solar radiation or subjected to low direct solar radiation, with acceptable thermal comfort levels, presents acceptable local thermal discomfort and air quality levels.

In order to obtain good thermal comfort and air quality in indoor environments used as educational spaces for young people, the inlet and outlet air, through strategically placed points, are necessary. When entering the spaces, the clean air should create a pleasant micro-climate around the occupants, assure good air quality in the breathing area, and extract contaminants released by the occupants.

In order to evaluate the indoor air and thermal quality, micro-models and macro-models can be used (see, for example, Heinsohn [1989]). In the micro-models, using space discretizing grids, environmental variables are calculated in all grid nodes using computational fluid dynamics (CFD), while in macro-models these variables are calculated inside all spaces and bodies.

In order to climatize the indoor space, the heating, ventilation, air-conditioning, and radiation systems are usually considered. These systems can, in general, operate in accordance with basic piston flow, perfect mixing, short circuiting, hybrid, displacement, natural, or cross-flow ventilation topologies or in combinations of these. To study these systems with these topologies, experimental chambers or adapted full-scale chambers can be used. Koskela et al. (1998), for example, presented an experimental study in a test room to evaluate the displacement airflow efficiency, and Hashimoto et al. (2000) measured airflow profiles to evaluate the energy consumption and thermal comfort for a new heating, ventilation, and air-conditioning system using low air temperature. In other studies, when the full-scale compartment is not available, the scale model is also used (see, for example, Sandberg and Elvsen [2004]).

A combination of CFD numerical models and full or other scale models is frequently shown. Teodosiu et al. (2000) and Gobeau and Saunders (2002), for example, showed applications of this technique.

The characterization of the expected thermal comfort conditions for a given environment is frequently made through the evaluation of the human thermal sensation using the predicted mean vote (PMV) and predicted percentage dissatisfied (PPD) indices described in Fanger (1970). This kind of evaluation is used, for example, in CR 1752, Ventilation for Buildings: Design Criteria for the Indoor Environment (CEN 1998) and international standards like ISO 7730, Moderate Thermal Environments--Determination of the PMV and PPD Indices and Specification of the Conditions for Thermal Comfort (ISO 1993). The above-mentioned CR 1752 (CEN 1998) defines three comfort categories (A, B, and C), establishing limits for PMV and PPD indexes. These indexes consider the human body as a whole and are able to make a good analysis of the thermal sensation of an occupant.

A person can express thermal comfort sensations and present local thermal discomfort problems. Local thermal discomfort sensations in localized regions of the body may occur, produced namely by incident airflows from ventilators, due to their intrinsic characteristics such as exit air velocity and airflow symmetry as well as their location in relation to the occupants. According to Fanger et al. (1988), local thermal discomfort depends on the local air temperature and velocity and turbulence intensity. This draft risk index is also presented in ISO 7730 (ISO 1993) and CR 1752 (CEN 1998), defined in the latter standard as three local thermal discomfort categories (A, B, and C). According to Fanger and Pedersen (1977), local discomfort sensations associated with air velocity frequency fluctuations, are verified in frequencies between 0.3 and 0.5 Hz and according to Zhou and Melikov (2002) and Zhou et al. (2002) are verified to equivalent frequencies between 0.2 and 0.6 Hz.

The airflow rate inside an occupied compartment can be calculated using different recommendations and methodologies presented in national and international standards (see Olesen [1997]). According to ANSI/ASHRAE Standard 62.1, Ventilation for Acceptable Indoor Air Quality (ASHRAE 2004), for offices in buildings the recommended values are 2.5 L/s (5.3 [ft.sup.3]/min) per person (for outdoor air rate) and 8.5 L/s (18 [ft.sup.3]/min) per person (for combined outdoor air rate). The Portuguese normalization presented in D.-L. n[degrees] 79.2006 of April 4th, Regulamento dos Sistemas Energetico de Climatizacao em Edificios (DR 2006) defines an airflow rate to each occupant and kind of space of 35 [m.sup.3]/h (9.7 L/s [20.6 [ft.sup.3]/min]) per occupant. CR 1752 (CEN 1998), where the airflow rate is based on the occupants' comfort level, was also analyzed. In this methodology, which defines three quality levels (Category A, with 10% of dissatisfied persons; Category B, with 20% of dissatisfied persons; and Category C, with 30% of dissatisfied persons), occupants and materials existing inside the compartment are considered pollution sources. This perspective, based in olf and decipol units, is presented in detail in Fanger (1988). Categories A, B, and C are recommended, respectively, for 16 L/s (33.9 [ft.sup.3]/min), 7 L/s (14.8 [ft.sup.3]/min), and 4 L/s (8.5 [ft.sup.3]/min) per olf. In all categories it is also necessary to consider the pollution load caused by the building itself (including furnishings, carpet, and ventilation systems). For school building office compartments, a value 0.3 olf per floor area is recommended.

Carbon dioxide can be used as a tracer gas. The advantage in using this, when compared to other gases, is due to its cost (generically negligible) because it is generated by the occupants themselves (see Levine et al. [1993]). The tracer gas method, allowing the evaluation of air quality in indoor environments, can be used to determine the air renovation rate and the airflow rate, as well as the air age and ventilation system efficiency, among other values. Several methods can be used to calculate these values, namely the decay, crescent, and constant concentration, and constant emission and pulsed injection (see Sandberg [1981], Sandberg and Sjoberg [1983], Niemela et al. [1989], Roulet [1991], Grieve [1991], and Roulet and Cretton [1992] as examples). The option for each method is associated with the specificities of the studied compartment, the existing ventilation system, and the values to be evaluated, among other issues.

The mean value of air renovation rate is defined as the ratio between the volume of outdoor air entering an enclosure per hour and the effective volume of the compartment. In other words, it is the number of times, during one hour, that the indoor air is replaced by outdoor air. Three methods are frequently used to determine the air renovation rate, or airflow rate, with the tracer gas method: the concentration decay method, the constant emission method, and the constant concentration method. In the first method, which is used in this work to measure the air renovation rate, a small quantity of tracer gas is mixed with the room air at the beginning of the test, and afterwards the tracer gas concentration decay is measured. The air renovation rate is, in this work, calculated by the slope of the concentration logarithmic decay obtained through the regression of the experimental data. A compartment air renovation rate can be calculated either locally or in mean terms. In the local air renovation rate of a compartment, the tracer gas concentration is obtained at a specific point for a determined system of ventilation. In the mean air renovation situation, this evolution can be determined either at an exhaustion point or at an indoor point, being conditions for a guaranteed perfect mix.

The airflow inside school building compartments, used as work offices in the South of Portugal (the Algarve region), in general present problems to conciliate the occupants' thermal comfort, local thermal discomfort, and air quality levels with low energy consumption levels. In general, the airflow inlet and outlet in these kind of spaces, when existing, are located in parallel walls above the door and window level, using small windows with opening control, and work through natural principles based on wind forces. This kind of ventilation system is easy to implement with low investment cost and significantly reduces draft risks to which occupants are subjected. Nevertheless, this kind of system can promote airflow short-circuit.

Thus, an experimental chamber with similar dimensions to the school building office compartments and a CFD numerical model were developed to analyze in detail a new airflow topology inside this kind of school building compartment. The idea is to develop a forced-air system with inlet and outlet airflow located above head level in nonparallel walls, divided into two areas: the non-occupied upper area, which contains the airflow inlet and outlet, with high air velocity levels, used to make contaminant extraction easier, and the occupied lower area with low air velocity levels, used to ensure good thermal comfort, reduced local thermal discomfort, and acceptable air quality levels to the occupants. This kind of forced-air system, with low energy consumption and reduced initial investment, is easy to implement in the Mediterranean climate for moderate air temperature levels.

In this study, in steady-state conditions, where the airflow inside a school building office compartment with a particular configuration is experimentally and numerically analyzed, tools and criteria for indoor environments are tested and presented; these will be used in more complex studies on school compartments. The omni-directional air velocity fluctuation is experimentally measured and the air velocity root mean square, the air turbulence intensity, the draft risk, the air velocity fluctuation frequencies, and the air velocity fluctuation equivalent frequencies are calculated. The mean and local air renovation rates, using the decay tracer gas method with a carbon dioxide concentration, are also calculated. The three-dimensional air velocity and the air renovation rate are numerically calculated. Both these experimental and numerical results are used to gain a better understanding of the airflow topology inside this particular kind of compartment and evaluate the comfort level, namely the thermal comfort, local thermal discomfort, and air quality levels, to which occupants are subjected.

CFD NUMERICAL MODEL

In the CFD numerical model, which simulates the three-dimensional turbulent airflow at steady-state conditions inside an occupied space in cartesian coordinates, one mass continuity, three moments, one energy, one turbulent kinetic energy, and one turbulent energy dissipation rate equation are used. The developed numerical software calculates the air velocity, temperature, turbulent kinetic energy, and turbulent energy dissipation rate inside occupied spaces. In the partial differential equations discretization, the finite volume method is used; in the convective/ diffusive fluxes, the hybrid scheme is used; in the velocity and pressure equations, the SIMPLE (Semi-Implicit Method for Pressure-Linked Equations) algorithm is used; in the grid generation, a non-uniform methodology is used; in the equations system resolution, an iterative TDMA (Tri-Diagonal Matrix Algorithm) method is used; in turbulence simulation, the k-[EPSILON] model for high Reynolds number is used; and in the surfaces proximity, the wall boundary is used (see Patankar [1980]).

EXPERIMENTAL SETUP

In this chamber, developed using an easy building philosophy, built by modules, it is possible to easily change the chamber dimensions and the position and dimensions of the door and windows and the different panel constitutions. The main purpose of this chamber is to simulate in a laboratory a real school building office compartment and to make experimental tests in the project phase.

This chamber, enveloped by different antechambers that allow simulation of different surrounding temperatures and are isolated from the floor, is equipped with an air-conditioning system installed on a moving platform (the system controls the air temperature, velocity, and relative humidity), a radiant floor with heated water in solar collectors located inside the laboratory, a ventilated window with internal/external and opaque/transparent shading devices, an auxiliary window used in demonstration studies, solar radiation simulators in the window and collectors (with several rotation angles possible) using lamps with spectra similar to solar radiation, a traversing system with manual positioning, a measuring and controlling system, and more.

The experimental chamber was built in a wooden structure with modules of 0.45 x 0.48 [m.sup.2] (1.48 x 1.57 [ft.sup.2]) in the lateral panels and 0.45 x 0.45 [m.sup.2] (1.48 x 1.48 [ft.sup.2]) in the ceiling. Each module, built with a thickness of 10 cm (0.33 ft), is composed of three layers: an internal layer of wood, a central layer of isolating thermal material, and an external layer also made of wood. This philosophy allows for easily changing the chamber dimensions and the position and dimensions of the windows (replacing the panels with glass) and of the door.

A heating, ventilation, and air-conditioning system that controls the air temperature, velocity, and relative humidity, installed on a moving platform with two articulated conduits, allowing for easily changing the chamber air inlet and outlet, is set up in the chamber. The philosophy used in the building allows total recirculation, total renovation, or partial renovation/recirculation. This climatization system is important in the simulation of winter, summer, spring, and autumn conditions in a laboratory situation.

The experimental chamber is made of panels orientated toward the south, north, west, and east; a ceiling; and an isolated floor. The idea is to use different antechambers in each lateral panel and the ceiling, in order to simulate different environmental temperatures. In the present situation, where a school office with dimensions of 2.7 x 2.4 x 2.4 [m.sup.3] (8.86 x 7.87 x 7.87 [ft.sup.3]) was simulated, three antechambers were developed: the south one (which simulates the external environment and is equipped with solar radiation simulators), the east one (which simulates one internal space, equipped with monitoring and controlling systems), and the west one (which simulates one internal space, equipped with an air climatization system). The ceiling is in contact with a small space, while the north panel is in contact with the laboratory.

A three-dimensional positioning system allowing the positioning of probes inside the full-scale experimental chamber was projected and developed. This system is made of a light aluminium structure.

The circular inlet and outlet airflow, with an area of around 0.05 [m.sup.2] (0.54 [ft.sup.2]), are located in the YZ plane and in the XZ plane, respectively (see Figure 1).

[FIGURE 1 OMITTED]

The idea of the present forced ventilation system, implemented in the full-scale experimental chamber, is to reduce the occurrences of the short-circuit phenomena and to promote a thermally comfortable micro-climate with good air quality levels around the occupants, with low local thermal discomfort levels and a low cost fan and low consumption level. The airflow topology, with the forced-air inlet and outlet located above the occupants' head level and located in a non-parallel wall, is divided into two areas: the non-occupied upper area, which contains the air inlet and outlet, with high air velocity levels, used to make contaminant extraction easier, and the occupied lower area, with low air velocity levels, used to ensure better occupant contaminant transport to the upper area and a low draft risk level area.

Two indoor climate analyzers were used in the experimental tests: the air velocity fluctuations are measured with three omni-directional probes from Sensor, and the mean values of other environmental variables are measured using a Babuc-A of LSI.

In the experimental tests, the air velocity fluctuation values were measured during ten minutes at a rate of six points per second. The measured values were performed with 6 x 7 x 6 grids in the occupied area, 36.4 cm (1.19 ft) spaced in the X direction, 37.8 cm (1.24 ft) spaced in the Y direction, and 38.4 cm (1.24 ft) spaced in the Z direction. The mean air temperature and relative humidity are 21.6[degrees]C (70.88[degrees]F) and 35%, respectively. The surface wall temperature is similar to the air temperature value. The inlet mean air velocity and turbulent intensity, measured at several outlet points, are 4.5 m/s (14.76 ft/s) and 11%, respectively.

In order to evaluate the air renovation rate using the decreasing tracer gas method, the carbon dioxide concentrations are measured 1.2 m (3.94 ft) above the floor level, in the respiration level, performed in a horizontal plane with 3 x 3 grids, in the occupied area, 65 cm (2.13 ft) spaced in the X direction and 87 cm (2.85 ft) spaced in the Y direction (see Figure 7b).

For the numerical simulation, the calculated values were performed with 24 x 28 x 24 grids, 10.9 cm (0.36 ft) spaced in the X direction, 10.4 cm (0.34 ft) spaced in the Y direction, and 10.9 cm (0.36 ft) spaced in the Z direction. This numerical grid presented a good compromise between the final results quality and the calculation time. In order to make the comparison between the experimental and numerical results easier, it was decided not to consider a refined grid near the wall surface and in the air inlet and outlet.

The airflow developed in this study, to be used in school building office compartments during not only autumn and spring days but also in moderate winter and summer days in the south of Portugal (the Algarve region), promotes (in general) good thermal comfort conditions, without local thermal discomfort conditions and with low cost. The idea is to use an air ventilator located above the head level, as discussed at the beginning of this section, to inject fresh air inside the office directly from the external environment or from the corridor.

The inlet air mean velocity in the experimental chamber, promoted by the air ventilator, is 4.5 m/s (14.76 ft/s). This air velocity, in accord with previous tests and confirmed by the present results, promotes an air mean velocity in the occupied spaces of 0.49 m/s (1.6 ft/s). Using standard people with an activity of 1.2 met and with an air mean relative humidity of 35% (measured in the experimental test) to guarantee a maximum 10% of people thermally uncomfortable (Category B of CR 1752 [CEN 1998]), in the winter (with a clothing level of 1 clo) the operative temperature can change between 21.6[degrees]C (70.88[degrees]F) and 25.6[degrees]C (78.08[degrees]F), while in the summer (with a clothing level of 0.5 clo) the operative temperature can change between 25.5[degrees]C (78.08[degrees]F) and 28.2[degrees]C (82.76[degrees]F). Thus, in spaces with low direct solar radiation, considering that the air mean temperature is equal to the mean radiant temperature, with the air mean temperature changing between 21.6[degrees]C (70.88[degrees]F) and 28.2[degrees]C (82.76[degrees]F)--which in general are the environmental conditions verified for this region's school days--this system guarantees global thermal comfort conditions.

In the experimental tests, the airflow topology, the airflow fluctuations, and the carbon dioxide concentration evolution are analyzed in order to evaluate the local thermal discomfort and the air quality to which the occupants are subjected in this kind of ventilation system. In the experimental test, in steady-state conditions, the mean air temperature, equal to the mean radiant temperature, was 21.6[degrees]C (70.88[degrees]F). These conditions are associated with the worst conditions of local thermal discomfort levels verified in the global thermal comfort interval defined between 21.6[degrees]C (70.88[degrees]F) and 28.2[degrees]C (82.76[degrees]F).

This space, with an area of around 6.5 [m.sup.2] (69.96 [ft.sup.2]), is used, in general, by one seated teacher. Thus, in the Fanger philosophy, to calculate the airflow rate (CR 1752 [CEN 1998]), 1 olf for the pollution load caused by the nonsmoking occupant is considered and around 2 olfs for the pollution load caused by the building (including the furnishings, carpet, and ventilation system) are considered. With these 3 olfs, according to CR 1752 (CEN 1998), in the respiration area, for Category A an airflow rate of 47.1 L/s (99.8 [ft.sup.3]/min) is necessary, for Category B an airflow rate of 20.6 L/s (43.6 [ft.sup.3]/min) is necessary, while for Category C an airflow rate of 11.8 L/s (24.9 [ft.sup.3]/s) is necessary. Nevertheless, if the ASHRAE Standard 62.1 (ASHRAE 2004) criteria are used, 8.5 L/s (18 [ft.sup.3]/min) for combined outdoor air rate are applied, while if the Portuguese criteria in D.-L. n.[degrees] 79.2006 of April 4th (DR 2006) are used, 9.7 L/s (20.6 [ft.sup.3]/min) are applied.

In the present system, with an inlet air mean velocity of 4.5 m/s (14.76 ft/s), the airflow rate is around 795 [m.sup.3]/h (221 L/s or 468.3 [ft.sup.3]/min) and the mean air renovation time for a volume of around 15.5 [m.sup.3] (547.32 [ft.sup.3]) is around 51 L/h (108.1 [ft.sup.3]/min) if the ventilation system creates a perfect mix.

RESULTS AND DISCUSSION

In this work a CFD numerical model that simulates turbulent airflow with heat and mass transfer inside school building office compartments and a full-scale experimental chamber that simulates a school building office compartment are presented and used. The numerical model, after being compared to the experimental data, is used to gain a better understanding of the airflow topology in this kind of situation and, in a later phase, will be used in different school building office compartments. In this study, in isothermal steady-state conditions, the local thermal discomfort level and the air mean local renovation rate for this particular airflow topology, developed in this work, are analysed.

The air mean velocity inside the experimental chamber was numerically determined and is represented by small arrows in Figure 1. In this figure, the arrows have a direction similar to the local air velocity and a dimension proportional to the local air velocity values. This figure shows that the inlet airflow incident in the parallel wall placed in front is divided into two parts: one at the outlet direction, in the upper area, and the other at the lower, occupied area. In the outlet are concentrated the previous airflow, which is associated to a little short-circuit phenomena, and another ascendant airflow came to the lower area.

Figure 2 shows the experimental and numerical grid plans represented in Figures 3, 4, and 5, the comparison between the air mean velocity values (in m/s) obtained (a) experimentally and (b) numerically for the XY (Z = 215 cm [7.05 ft]), XZ (Y = 244 cm [8 ft]), and YZ (X = 205 cm [6.73 ft]), respectively, are represented. The experimental areas represented in Figures 3a, 4a, and 5a are indicated in the numerical images (Figures 3b, 4b, and 5b) by a black rectangle.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

The numerical values presented in Figures 3, 4, and 5 were subjected to a correction presented by Popiolek and Melikov (2004). In this correction, done by an empirical model, the numerically calculated air mean velocity is converted to the air mean speed, to be compared with the experimental air mean speed obtained by the omni-directional sensors. In this conversion, the numerical air mean speed is a function of the numerical air mean velocity and the kinetic turbulence energy.

The comparison between the experimental values and numerical results, in general, show a good correlation in the measured area. After being compared with success, the first values of the numerical and experimental results are used to gain a better understanding of the airflow philosophy verified in the occupied area. This analysis is important for evaluating the descendent or ascendant airflow in the occupied area.

It is verified that the air velocity Z component value decreases when the distance to the floor decreases. The decreasing air velocity values are verified in the opposite side of the air inlet, being the increasing values verified in the remaining area, with the principal importance under the air inlet area.

An occupant can express thermal comfort, in general, for the whole body area while simultaneously feeling local thermal discomforts due to the draft risk or air velocity fluctuations. Thus, in this section, these phenomena are analyzed.

The measured air velocity fluctuations are used to calculate the mean air velocity, used in the previous figures, the root mean square values, and, consequently, the turbulence intensity. This information, in combination with the local air temperature value, is used to calculate the draft risk (Fanger et al. [1998]). In Figure 6a, the percentage of dissatisfied people due to the draft risk in the seated occupied area (101 cm [3.31 ft] above the floor level) for the air temperature of 21.6[degrees]C (70.88[degrees]F) is shown.

[FIGURE 6 OMITTED]

The percentage of dissatisfied people due to draft risk for 21.6[degrees]C (70.88[degrees]F), changes, in general, between 12% and 53%, the higher value levels verified, in general, near the wall located on the opposite side of the air inlet. In the occupied area, for 25 cm (0.82 ft) above the floor level, the percentage of dissatisfied people due to draft risk changes between 28% and 52%, while for 63 cm (2.07 ft), 101 cm (3.31 ft), and 140 cm (4.59 ft) above the floor level, this draft risk changes between 16% and 32%. In conclusion, it is verified that at 25 cm (0.82 ft) above the floor level, the local thermal discomfort associated with the draft risk is not in accordance with the CR 1752 (CEN 1998) recommendation, and above 63 cm (2.07 ft) in the central large area it is in accordance with Category C (< 25% of uncomfortable people). Nevertheless, if lower body parts are protected by clothing, as in general is verified in winter conditions, this local thermal discomfort due to the draft risk is not verified. For 101 cm (3.31 ft) above the floor level in the central large area, when the air temperature increases to 24.9[degrees]C (76.82[degrees]F) Category B is verified, and when it increases to 28.2[degrees]C (82.76[degrees]F) Category A is verified.

For the seated occupied area, the air velocity fluctuation frequency and the air velocity fluctuation equivalent frequency, which can cause local discomfort sensations associated with air velocity fluctuations, were also calculated.

The air velocity fluctuation frequencies presented a maximum discomfort level at frequencies around 0.3 and 0.5 Hz, as was presented previously. These frequencies are located, in general, at the feet level; nevertheless, near the walls located in front of the air inlet and in the air outlet area, uncomfortable air fluctuation frequencies in the occupied area were also verified.

The more uncomfortable air velocity fluctuation equivalent frequencies (between 0.2 and 0.6 Hz, as was presented previously) do not present uncomfortable frequencies in the horizontal plane located 25 cm (0.82 ft) above the floor level; nevertheless, in the horizontal plane located 63 cm (2.07 ft) above the floor level, the local discomfort sensations are verified in the air inlet frontal area. For the other horizontal planes, located 101 cm (3.31 ft) (see Figure 6b) and 140 cm (4.59 ft) above the floor level, these frequency areas increase. At 140 cm (4.59 ft) above the floor level, these more uncomfortable frequencies are verified in the occupied central larger area.

Finally, the air renovation rate in the respiration area, experimentally and numerically obtained, is shown in Figure 7. In the experimental results, using the tracer gas decreasing method, the tracer gas is measured around 0.2 hours, in the respiration area, at nine points. The air renovation rate, experimentally measured, changes between 1.18 [h.sup.-1] below the air outlet area and 3.77 [h.sup.-1] in the wall located in the air inlet area. In these tests, the correlation coefficient changes between 0.67 and 0.93.

[FIGURE 7 OMITTED]

The comparison between the experimental and numerical values, obtained in the analyzed nine points, can be seen in Figure 7a. This figure shows good agreement between the experimental and numerical values, the maximum difference, verified above the air inlet area, being 0.5 [h.sup.-1]. Finally, Figure 7b presents the air renovation rate numerically calculated in the respiration area.

The mean air renovation rate in the respiration area, calculated using the mean value of the numerical values in the respiration area, is 2.65 [h.sup.-1], and the airflow rate is 11.4 L/s (24.16 [ft.sup.3]/min). This value can promote the specifications for one person recommended by ASHRAE Standard 62.1 (ASHRAE 2004), by D.-L. n[degrees] 79.2006 of April 4th (DR 2006), and by Category C of CR 1752 (CEN 1998).

CONCLUSIONS

In this work, a numerical model to simulate a turbulent airflow, and a full-scale experimental chamber to simulate a school building office compartment, were developed and presented. The air velocity fluctuations, a carbon dioxide concentration evolution in the respiration area, and other environmental mean values in the occupied area were measured. The air velocity root mean square, the air turbulence intensity, the draft risk, the air velocity fluctuation frequencies, the air velocity fluctuation equivalent frequencies, and the mean and local air renovation rates in the respiration area were calculated.

The analyzed forced-air ventilation system, to be installed in school building office compartments and to be used in the spring and autumn as well as on moderate summer and winter days, promotes thermal comfort conditions, in general, when the mean air temperature, equal to the mean radiant temperature, changes between 21.6[degrees]C (70.88[degrees]F) and 28.2[degrees]C (82.76[degrees]F).

The airflow topology is divided into an upper non-occupied area and a lower occupied area. The inlet airflow incident in the parallel wall placed in front, divided in two parts, came out to the outlet direction, in the upper area, and came out to the lower occupied area. In the occupied area, the descendent airflow is verified on the opposite side of the air inlet, and the ascendant airflow is verified in the remaining area, with the principal importance under the air inlet area. In the outlet is concentrated airflow, which is associated with a short-circuit phenomena, and one ascendant airflow came out to the lower area.

The local thermal discomfort in the larger central occupied area, due to the draft risk, for 21.6[degrees]C (70.88[degrees]F) is in accordance with Category C, for 24.9[degrees]C (76.82[degrees]F) is in accordance with Category B, and for 28.2[degrees]C (82.76[degrees]F) is in accordance with Category A of CR 1752 (CEN 1998).

The more uncomfortable air velocity fluctuation frequencies were not verified in the large central occupied area; nevertheless, the more uncomfortable air velocity fluctuation equivalent frequencies were verified in the air inlet frontal area from low distances to the floor level and in the large central area from high distances to the floor level.

The air renovation rate, obtained in the respiration area by the tracer gas decreasing concentration, is the highest in front to the airflow outlet area. The measured airflow rate mean value of around 11.4 L/s (24.16 [ft.sup.3]/min) can promote the specifications per person recommended by ASHRAE Standard 62.1 (ASHRAE 2004), D.-L. [n.sup.[degrees]] 79.2006 of April 4th (DR 2006), and Category C of CR 1752 (CEN 1998).

Thus, in accordance with the main purpose of this paper, the occupant in terms of local thermal discomfort due to the draft risk can be seated in the central large occupied area. Therefore, in order to reduce local discomfort levels, the protection of lower members with clothing is suggested. The central large occupied area is not subjected to more uncomfortable air velocity fluctuation frequencies; nevertheless, some more uncomfortable air velocity fluctuation equivalent frequencies can occur, mainly in the upper body parts. Finally, the occupant in terms of the air quality level can be seated around the central large occupied area.

ACKNOWLEDGMENTS

The authors acknowledge the support from the 2005-2006 ASHRAE Undergraduate Senior Project Grant Program, financed by the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

This research activity is also being developed in a project approved and financed by the Portuguese Foundation for Science and Technology and POCI 2010, sponsored by the European Comunitary Fund FEDER.

The authors acknowledge the support from NAMB/FCMA/University of Algarve.

REFERENCES

ASHRAE. 2004. ANSI/ASHRAE Standard 62.1-2004, Ventilation for Acceptable Indoor Air Quality. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

CEN. 1998. CR 1752, Ventilation for Buildings: Design Criteria for the Indoor Environment. Brussels: Comite Europeen de Normalisation (CEN).

DR. 2006. Decreto-Lei [n.sup.[degrees]] 79.2006, Regulamento dos Sistemas Energeticos de Climatizacao em Edificios (RSECE). I Serie-A, Nr 67, April 4th. Lisbon: Diario da Republica.

Fanger, P.O. 1970. Thermal Comfort: Analysis and Applications in Environmental Engineering. New York: McGraw-Hill Book Company.

Fanger, P.O. 1988. Introduction of the olf and the decipol units to quantify air pollution perceived by humans indoors and outdoors. Energy and Buildings 12:1-6.

Fanger, P., and C. Pedersen. 1977. Discomfort Due to Air Velocities In Spaces. Paris: Instituit International du Froid, pp. 289-96.

Fanger, P., A. Melikov, H. Hazawa, and J. Ring. 1988. Air turbulence and sensation of draught. Energy and Buildings 12:21-39.

Gobeau, N., and C.J. Saunders. 2002. Computational fluid dynamics (CFD) validation of the prediction of pesticide dispersion in a naturally-ventilated building. Proceedings of ROOMVENT'2002, 8th International Conference on Air Distributions in Rooms, Copenhagen, Denmark, September 8-11, pp. 81-84.

Grieve, P.W. 1991. Measuring ventilation using tracer-gases. Bruel & Kjaer, Technical Review, Naerum, Denmark, February.

Hashimoto, Y., Y. Tsubota, Y. Nakano, W. Urabe, T. Oka, Y. Kaneko, and T. Imai. 2000. Development of a ductless air supply system using low temperature air. Proceedings of ROOMVENT'2000, 7th International Conference on Air Distributions in Rooms, Reading, UK, July 9-12, pp. 553-58.

Heinsohn, R.J. 1989. Macro-modelling using sequential box models. Building Systems, Room Air and Air Contaminant Distribution, pp. 200-5. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

ISO. 1993. International Standard 7730, Moderate Thermal Environments--Determination of the PMV and PPD Indices and Specification of the Conditions for Thermal Comfort. Geneva: International Organization for Standardization.

Koskela, H., T. Hautalampi, E. Sandberg, and M. Lindgren. 1998. Test room and measurement system for active displacement air distribution. Proceedings of ROOMVENT'98, 6th International Conference on air Distributions in rooms, Stockholm, Sweden, June 14-17, pp. 225-32.

Levine, K.B., E.M. Sterling, and C.W. Collett. 1993. Estimation of outdoor air ventilation rates using CO2 concentrations. ASHRAE Transactions 1:1554-59.

Niemela, R., A. Saamanen, and H. Koskela. 1989. A comparison between the step-up, step-down and pulse injection technique for the measurements of the mean age of air. Proceedings of the 10th AIVC Conference, Dipoli, Finland, September 25-28, pp. 17-30.

Olesen, B.W. 1997. International development of standards for ventilation of buildings. ASHRAE Journal 39(4):31-39.

Patankar, S.V. 1980. Numerical Heat Transfer and Fluid Flow. Washington, DC: Hemisphere Publishing Corporation.

Popiolek, Z., and A. Melikov. 2004. Improved interpretation and validation of CFD predictions. Room- Vent 2004, Coimbra, Portugal.

Roulet, C.A. 1991. Use of tracer gases to assess indoor air quality. ASHRAE Symposium on Indoor Air Quality and Energy Consumption, Montreal, September.

Roulet, C.A., and P. Cretton. 1992. Field comparison of age of air measurement techniques. Proceedings of ROOMVENT'1992, Air Distribution in Rooms, Third International Conference, Aalborg, Denmark, 3:213-29.

Sandberg, M. 1981. What is ventilation efficiency? Building and Environment 16(2):123-35.

Sandberg, M., and M. Sjoberg. 1983. The use of moments for assessing air quality in ventilated rooms. Building and Environment 18(4):181-97.

Sandberg, M., and P.A. Elvsen. 2004. Rapid time varying ventilation flow as a mean increasing the ventilations efficiency. ROOMVENT'2004, 8th International Conference on Air Distributions in Rooms, Coimbra, Portugal, September 5-8.

Teodosiu, C., S. Laporthe, G. Rusaouen, and J. Virgone 2000. Experimental and numerical predictions of indoor air quality. Proceedings of ROOMVENT'2000, 7th International Conference on Air Distributions in Rooms, Reading, UK, July 9-12, pp. 95-100.

Zhou, G., and A. Melikov. 2002. Equivalent frequency--A new parameter for description of frequency characteristics of airflow fluctuations. ROOMVENT'2002, 8th International Conference on Air Distributions in Rooms, Copenhagen, Denmark.

Zhou, G., A. Melikov, and P. Fanger. 2002. Impact of equivalent frequency on the sensation of draught. ROOMVENT'2002, 8th International Conference on Air Distributions in Rooms, Copenhagen, Denmark.

Received September 13, 2006; accepted August 29, 2007

Eusebio Z.E. Conceicao, PhD ASHRAE Member Vitor D.S.R. Vicente ASHRAE Student Member [M.sup.a] Manuela J.R. Lucio

Eusebio Z.E. Conceicao is an assistant professor and Vitor D.S.R. Vicente is a researcher at FCMA-University of Algarve, Campus de Gambelas, Faro, Portugal. [M.sup.a] Manuela J.R. Lucio is a teacher at Vertical Grouping of Schools Professor Paula Nogueira/Regional Directory of Education of Algarve, Olhao, Portugal.