Contaminant Distribution Around Persons in Rooms Ventilated by Displacement Ventilation Brohus, Henrik; Nielsen, Peter Vilhelm

Aalborg Universitet Containant Distribution Around Persons in Roos Ventilated by Displaceent Ventilation Brohus, Henrik; Nielsen, Peter Vilhel Publication date: 1994 Docuent Version Publisher's PDF, also known as Version of record Link to publication fro Aalborg University Citation for published version (APA): Brohus, H., & Nielsen, P. V. (1994). Containant Distribution Around Persons in Roos Ventilated by Displaceent Ventilation. Aalborg: Dept. of Building Technology and Structural Engineering. Indoor Environental Technology, No. 40, Vol.. R9415 General rights Copyright and oral rights for the publications ade accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requireents associated with these rights.? Users ay download and print one copy of any publication fro the public portal for the purpose of private study or research.? You ay not further distribute the aterial or use it for any profit-aking activity or coercial gain? You ay freely distribute the URL identifying the publication in the public portal? Take down policy If you believe that this docuent breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will reove access to the work iediately and investigate your clai. Downloaded fro vbn.aau.dk on: January 11, 2019

NSTJITUTTET F DEPT. OF' BUILDING T ECHNOLOGY AND AALBORG UNIVERSIT ET AUC INDOOR ENVIRONMENTAL TECHNOLOG Y PAPER NO. 40 Present ed at R OOMVENT '94, Four th International Confer ence on Air Dist ribut ion in Roos, June 15-17, 1994, C r acow, Pola nd H. B R O H US, P. V. N IELSEN CONTAMINANT D ISTRIBUTION AROUND PERSONS IN ROOMS V E N T I LAT E D BY DISPLACEMENT VENTILATION J U N E 1994 ISSN 0902-7'513 R9415

The papers on INDOOR ENVIRONMENTAL TECHNOLOGY are issued for early disseination of research results fro the Indoor Environental Technology Group at the University of A a~lborg. These papers are generally subitted to scientific eetings, conferences or journals and should therefore not be widely distributed. \i\thenever possible reference should be given to the final publications (proceedings, journals, etc.) and not to the paper in this series.

INSTITUTTET FOR BYGNINGSTEKNIK DEPT. OF BUILDING TECHNOLOGY AND STRUCTURAL ENGINEERING AALBORG UNIVERSITET AUC AALBORG DANMARK INDOOR ENVIRONMENTAL TECHNOLOGY PAPER NO. 40 Presented at ROOMVENT '94, Fourth International Conference on Air Distribution in Roos, June 15-17, 1994, Cracow, Poland H. BROHUS, P. V. NIELSEN CONTAMINANT DISTRIBUTION AROUND PERSONS IN ROOMS VEN TILATED BY DISPLACEMENT VENTILATION JUNE 1994 ISSN 0902-7513 R9415

CONTAMINANT DISTRIBUTION AROUND PERSONS IN ROOMS VENTILATED BY DISPLACEMENT VENTILATION Henrik Brohus and Peter V. Nielsen Departent of Building Technology and Structural Engineering Aalborg University Aalborg, Denark SUMMARY An optial design of the ventilation syste needs a proper prediction of the velocity, teperature and containant distribution in the roo. Traditionally this is done either by the use of siplified odels or by a soewhat ore coprehensive CFD-siulation. Coon to both ethods is usually the lack of consideration for the persons present in the roo. This paper deals with soe of the effects of persons present in a displaceent ventilated roo, especially the effect on the containant distribution. It is deonstrated that although the containant distribution is affected the stratification in the flow is stable when people are oving around in the roo. The exposure of a sitting and a standing person in proportion to the stratification height is exained. It is found that the flow in the boundary layer along a person to a great extent is able to entrain air fro below the breathing zone. Measureents also show the possible disadvantage when containant sources are located in the lower part of the roo. Two new quantities, applied in connection with personal exposure in ventilated roos, are defined.

CONTAMINANT DISTRffiUTION AROUND PERSONS IN ROOMS VENTILATED BY DISPLACEMENT VENTILATION Henrik Brohus and Peter V. Nielsen Departent of Building Technology and Structural Engineering Aalborg University Aalborg, Denark 1. INTRODUCTION During recent years displaceent ventilation has becoe an often used way to ventilate offices and industrial buildings, especially in the Scandinavian countries. With this there is an increasing deand for energy efficient design which, at the sae tie, is able to ensure a high level of theral cofort and indoor air quality. An optial design of the ventilation syste needs a proper prediction of the velocity, teperature and containant distribution in the roo. Traditionally this is done either by the use of siplified odels or by a soewhat ore coprehensive CFD-siulation. Coon to both ethods is usually the lack of consideration for the persons present in the roo. A person produces heat due to the etabolis and the surface teperature is usually several degrees above the abient teperature level. In a state of theral cofort the teperature of the huan skin is about 33-34 oc depending on the activity level [1]. The clothing fors an insulating layer between the skin and the surrounding air and causes a drop in surface teperature. The teperature drop ay be 5-8 o C for persons standing relaxed wearing usual indoor clothing. The resulting excess teperature causes an upward air flow along the body which entrains air fro the surroundings. The boundary layer around the person is able to entrain and to transport clean air as well as containated air to the breathing zone which ay cause an exposure to pollution in the roo. Another effect of persons in ventilated roos is the disturbance caused by oveents. When displaceent ventilation is applied the influence on the stratification of the flow is especially iportant. In this paper the topics entioned above are discussed and full-scale easureents are presented showing different aspects of persons present in a displaceent ventilated roo.

2. EXPERIMENTAL SET -UP 2.1. Full-scale test roos The easureents reported are perfored in three full-scale test roos with a length, width and height, respectively: Sall test roo I Sall test roo 2 Large test roo 4.2 x 3.6 x 2.4 5.4 x 3.6 x 2.6 8x6x4 Three different wall-ounted low-velocity inlet devices are used. They are shown in figure I. The return opening in the sall test roo I is located at the corner of the roo in the wall 0.15 below the ceiling. In the other roos the return openings are located in the ceiling. Fig. I. Low-velocity inlet devices used in the three full-scale test roos. The devices are used in the sall test roo I (left), sall test roo 2 (centre) and in the large test roo (right), respectively. 2.2. Heat sources Three different heat sources are used. A point heat source consisting of heated coils ounted on an iron base surrounded by a 0.2 high tube 0 0.15. A person siulator in shape or a 1.0 high black-painted closed cylinder 0 0.4, heated by four light bulbs. Finally a theral anikin acts as a heat source.

3 2.3. Theral anikin To be able to easure the personal exposure properly the theral anikin seen on figure 2 is used. The anikin is shaped as a 1.7 high average sized woan. The tight-fitting clothes have an insulation value of0.8 clo. Fig.2. Theral anikin used to easure the personal exposure etc. The anikin is separated in 16 individually controlled parts of the body, each with the sae surface teperature and heat output as people in theral cofort. An artificial lung provides for the breathing. The anikin consists of a fibre ared polyester shell, wound with nickel wire used sequentially both for the heating of the anikin and for easuring and controlling the skin teperature. The skin teperature and the heat output correspond to people in theral cofort. An <utificial lung provides for the breathing. It is possible to adjust both the frequency of respiration (nuber of breaths per inute) and the pulonary ventilation (litres per inute). 3. RESULTS AND DISCUSSION 3.1. Theral and passive sources The location of the containant source in the roo has a great ipact on the concentration distribution and with it the personal exposure. In figure 3 the concentration profiles for two different locations of the source are shown.

4 y/h 1. 0 0 Th er al sourc e 0.8 0 Passive sourc e :E ~ QJ I 0.6 0.4 0.2 0.0 0.0 0.5 1.0 Co ncenlralia n Fig. 3. Concentration profiles easured in the sall test roo 1 with a theral (O) and a passive (D) containant source. The roo is ventilated after the displaceent principle with an air change rate of 5 h- 1 and a heat load of 400 W supplied by four person siulators. The neutral density tracer gas is C0 2 ixed with He. Figure 3 shows concentration easureents perfored in the displaceent ventilated sall test roo 1. The air change rate is 5 h- 1 and the heat load is 400 W supplied by four person siulators. The vertical concentration profile in the roo is easured by a theral and a passive containant source. The theral source is established by supplying the tracer gas in the plues above the four person siulators. The passive source is established by supplying tracer gas outside the plues 0.5 above the floor near the centre of the roo. In both cases the tracer gas is supplied by a ping-pong ball with six evenly distributed holes at neutral density. The stratification in the displaceent ventilated roo is clearly seen on the concentration profile when the theral source is applied. In this case the displaceent principle works very well. A concentration gradient is established in the roo where the air in the upper part is separated fro the less containated air in the lower part. However, when the passive containant source is located in the lower part of the roo it lacks any buoyancy or oentu to raise it above the stratification height. The vertical concentration profile reveals that the diensionless concentrations approach 1 corresponding to coplete ixing. Pollutants generated in neutral places of the roo cause considerable local concentrations which ay be entrained into the boundary layer of the present persons giving rise to exposure as pointed out by Holberg et al. (1987) and Nielsen (1993). As deonstrated it is iportant to supply the tracer properly to ensure that a containant source assued to be theral, really becoes a theral source in practice, ditto passive sources.

5 3.2. Convection around sources The containant transport is governed by diffusion and convection. Even a rather low velocity around the containant source ay result in a convection doinated dispersion. This fact has great ipact on the propagation of containation in displaceent ventilated roos because of the stratified flow where the velocity field alters in vertical direction due to the reverse flow as shown in figure 4. 2.0 AB I I I I I I ~!. :~... :~...J!. llll~...-: _!1 -...J... 3.0 1.0 /s - 0.3-0.2-0.1 0.0 0.1 Ve locity 2.0 A 2 0 l B :c CJ1 QJ I 1.5 1.5 :c 1. 0 ~ 10 I -l:?==- 1/ c/cr c/cr 0.5 05 0.0 0.0 0 8 12 16 20 24 0 8 12 16 20 24 Concenlro lio n Co ncentration Fig. 4. Velocity and concentration profiles fro the displaceent ventilated large test roo. The velocity profile is easured 3.0 fro the inlet device at the sae location as the containant source. Two concentration profiles are easured. 0.1 upstrea the source (A) and 0.1 downstrea the source (B). The neutral density tracer gas is N 2 0 ixed with He. Air change rate is 1.5 h- 1 and heat load is 400 W. As indicated in figure 4 the source is located 0.75 above the floor in the reverse flow of the displaceent ventilated large test roo. The air change rate is 1.5 h- 1 and the heat load is approxiately 400 W generated by two person siulators, a point heat source and the theral anikin. The easureents are perfored at steady state.

6 The velocity profile is easured at the sae location as the containant source. It is seen that the velocity at the height of the source is below 0.05 /s. Two concentration profiles are easured, one at the upstrea side and the other at the downstrea side of the containant source, both at a distance of 0.1. The concentration profiles deonstrates the influence of the convective ass transfer. Even at a velocity below 0.05 ls the diffusion is doinated extensively by the convective transport. Figure 4 indicates that the vertical location of the containant source in a displaceent ventilated roo ay be an iportant paraeter to consider if the source is found in the lower part of the roo. Even a vertical difference of few centietres ay give rise to soewhat deviating results. 3.3. Persons' ipact on stratification The displaceent principle is usually chosen to achieve an energy efficient ventilation syste capable to ensure a high level of theral cofort and indoor air quality. A basis for eeting the deands is the stratification of the teperature and the concentration fields in the displaceent ventilated roo. When there is no disturbance fro oveents in the roo the displaceent principle ay work well, but will oving persons or achinery be able to destroy or affect the stratification? Results fro full-scale easureents on the topic are shown below. To describe the efficiency of an air distribution syste different quantities are coonly used. The ventilation effectiveness e in the occupied zone is given by. DC (1) where er is the concentration in the return opening and coc is the ean concentration in the occupied zone. In this paper the occupied zone is defined as the area up to 1.8 above floor level. The local ventilation index E P is defined as er (2) where Cp is the concentration in a point of the roo. A new effectiveness is defined, the inhalation effectiveness E e E e (3) where ce is the concentration in the air inhaled by a person. ce is also an indication of the exposure to containation in the roo. If the pulonary ventilation is known it is possible to calculate the aount of inhaled atter.

Equations (1) to (3) assue that the supply air is uncontainated. The teperature effectiveness in the occupied zone ET is defined as 7 (4) where T R, T 0 and T oc are the teperature in the return opening, the teperature in the supply opening and the ean teperature in the occupied zone, respectively. To exaine the influence on stratification fro oving persons concentration and teperature easureents are perfored in the sall test roo 1 and the sall test roo 2. In figure 5 concentration profiles are found in three different cases where the heat sources and the containant sources are constituted by four person siulators ( o), four sitting persons (D) and four persons of who two are sitting and two are walking about in the roo (c.). y/h 1.0 0.8 +-' 0.6 _c O'l QJ I 0.4 0 4 person siulators 0 4 persons sitting 6 4 persons: 2 sitting; 2 walking 0.2 0.0 -j---,-------r---r----.---.-----, 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Concentration Fig. 5. Concentration profiles in the displaceent ventilated sall test roo 1. Air change rate is 5 h- 1 and heat load is 400 Wand 600 W (c.). Heat and containant sources are four person siulators (o), four sitting persons (D) and four persons of who two are sitting and two are walking about (c.). C0 2 ixed with He is used as tracer gas in the case of person siulator easureents. In the other two cases the persons supply the tracer by the respiration.

8 In the case of four person siulators ( o) the stratification is very distinct showing the separation of the cleaner lower part of the roo and the ore containated upper part. When the siulators are replaced by four persons the concentration profile changes and a tendency of soothing is seen. The ventilation effectiveness in the occupied zone E 0 c is 1.68 (o), 1.25 (D) and 1.38 (LI), respectively. Soe deviation between the case of siulators and the case of persons is found, while the effectiveness in the case of four sitting persons approaches the case of two persons sitting and two persons walking about. This indicates that neither oving persons nor sedentary persons are able to destroy or to disturb the stratification considerable in the previous case. The oveents were restrained to usual indoor walking and presuably, ore intense activity would anage to affect the concentration field to a greater extent as reported by Mundt (1993) where transient disturbances such as the opening and closing of a door are addressed. Disturbance of the teperature field due to oveents is exained in the two sall test roos with person siulators as well as sitting and oving persons. 1.0 y/ H 4-J. : 01 Q) I 0.8 0.6 0.4 0 4 person siulators 0 4 persons sitting 6 4 persons: 2 sitting; 2 wal ki ng 0. 2 0. 0 +----,-----,--W-I:IL.,-~9l6---r------, 0.0 0.2 0.4 0.6 0.8 1.0 Teperature (T - To) / (T R - To) Fig. 6. Teperature profiles in the displaceent ventilated sall test roo 1. Sae conditions as figure in 5. Four person siulators (o), four sitting persons (D) and four persons of who two are sitting and two are walking about (t.). Figure 6 shows the diensionless teperature profiles corresponding to the concentration profiles in figure 5. Fro the teperature profiles a rather good agreeent between the easureents with the four person siulators (o) and the four persons sitting (D) is found.

9 When the two of the four persons are walking about in the roo (t.) the profile shows a slight deviation towards the state of coplete ixing, i.e. a diensionless teperature on 1 throughout the roo. The teperature effectiveness for the occupied zone er is 1.20 in the case of four siulators as well as four sitting persons, while the effectiveness aounts to 1.13 in the case of two persons sitting and two persons in otion (t.). As expected the effectiveness becoes the sallest during the oveents. Although the profile changes, there is only a inor influence on the stratification due to the disturbance in the previous case. 1.0 y/h 0 4 person siulators 0.8 0 4 persons sitting 6 4 persons: 3 sitting; 1 walking -<-' 0.6..c 01 Q) I 0.4 0 4 persons: 2 sitting; 2 in otion 0.2 0. 0 -+---,-----1~~"'-'l-6--<7---,.---..., 0.0 0.2 0.4 0.6 0.8 1.0 Teperature (T - T 0 )/(TR - To) Fig. 7. Teperature profiles in the displaceent ventilated sall test roo 2. Air change rate is 7.6 h- 1 and heat load is approxiately 400 W. Heat sources are four person siulators (o), four sitting persons (O), three persons sitting and one walking about (t.) and two persons sitting and two in otion ( () ). Phan and Schulz (1991). In the sall test roo 2 easureents, alost siilar to the previous ones, are shown in figure 7. Also in this case four person siulators (o) and four sitting persons (O) take part, but now there are two different patterns of oveent. In the one case three person are sitting while one person is walking about (t.). In the other case two persons are sitting while two persons are walking about interrupted by heavy oveents of the ars and slight exercises ( () ). The teperature effectiveness in the occupied zone is 1.99 (o), 1.73 (O), 1.70 (t.) and 1.60 ( () ), respectively.

10 The effectiveness fro these easureents is not directly coparable to the effectiveness corresponding to figure 6 because of different inlet devices, air change rates and Archiedes nubers, but the tendencies are the sae. The axiu effectiveness is obtained when the disturbance is the slightest and the effectiveness decreases when the oveents in the roo increase. Sununarizing it ust be concluded that even intense oveents in the displaceent ventilated roo are not able to destroy the teperature stratification, but the teperature effectiveness shows a slight decrease when disturbance in the roo increases. The concentration stratification sees to be ore sensitive to persons' oveents. The disturbance created ay soothen the gradient to soe extent, which is also pointed out by Nielsen (1992). Heavy oveents ay cause a ixing of the upper containated zone and the lower cleaner zone, i.e. E - 1. oc 3.4. Personal exposure to containants As entioned previously one of the purposes of displaceent ventilation is to provide a good indoor air quality for the persons in the roo. The indoor air quality exerts influence on persons due to the breathing. Both fro a health point of view as well as a perceived air quality point of view, the influence arises fro the inspired air. Therefore, it is ore relevant to know the concentration of the containant in the inhaled air than the concentration at a neutral place of the roo easured at the sae height. Often the exposure in a ventilated roo is estiated fro the concentration easured in the height of the breathing zone in a neutral place of the roo. This ay be a good approxiation in ixing ventilated roos, but when displaceent is applied this ay lead to erroneous exposures. The reason is the cobined effect of the entrainent of roo air into the huan boundary layer and the concentration gradients in the displaceent ventilated roo. The transport of fresh air as well as containated air in the boundary layer ay cause the concentration in the inhaled air to deviate considerably fro the air outside the breathing zone easured at the sae height. Subsequently, the above-entioned topics are discussed by eans of the full-scale easureents on personal exposure in a displaceent ventilated roo at different stratification heights and pollutant source locations. 3.4.1. Exposure in proportion to stratification height In the following "exposure" is defined as the containant concentration in the inhaled air ce. The exposure easureents are perfored with the theral anikin by eans of an artificial lung able to provide the respiration either through the outh or through the nose. Since the anikin is controlled to obtain the sae heat output and the sae skin teperatures as a huan being under the sae circustances, the results are, to a great

11 extent, supposed to approach the personal exposure of a real person. Certain liitations are discussed later. Exposure easureents are carried out with respiration through the outh. In the present easureents no significant difference between respiration through the outh and respiration through the nose is found. The present easureents are perfored at steady state conditions. The real exposure to a certain atter in a roo requires detailed tie dependent knowledge of the pulonary ventilation and the pattern of oveents besides ce. In figure 9 the exposure of a sitting and a standing anikin is shown. The easureents are perfored in the large full-scale test roo at three different stratification heights Yst The corresponding teperature and velocity profiles are found in figure 8. 4 2.0.._,..c (J) QJ I 3 0 A y,, = 1.0 0 8 y,, = 1.35 2 c y,, = 2.25 "' o A y, 1 = 1,0 0 8 "' c 1.5 0 +----.------~~~~~-- 0.0 0.2 0. 4 0.6 0.8 1.0 Teperature (T - T 0)/(T. - To) -0.3-0.2-0. 1 0.0 Velocity /s 0.1 Fig. 8. Teperature and velocity profiles in the displaceent ventilated large test roo The easureents corresponds to the concentration easureents in figure 9. The velocity profiles are easured 4.0 fro the inlet device at the sae location as the theral anikin. q is the airflow rate, cl? is heat load and d T is the teperature difference between inlet and outlet. Please note the different ordinate axis heights. Case A: Yst = 1.00, q = 145 3 /h (0.8 h" 1 ), cl?= 771 W, d T = 9.8 oc Case B: Yst = 1.35, q = 290 3 /h (1.5 h- 1 ), cl?= 376 W, d T = 5.0 oc Case C: Yst = 2.25, q = 395 3 /h (2.1 h" 1 ), cl?= 781 W, d T = 7.6 C Measureents are carried out in collaboration with Christensen and Stevnhoved (1993).

12 4 c, = 0. 47 4 c, = 0.57 A Sitting Coc :::::: 1.44 Eoc = 1.54 A Standing 3, 2. 13 3, 1.75 :c 7), ~ 2 ~ 2 I I :c 7), c/cr c/ cr 0 0.0 0.5 1. 0 1.5 0 0.0 0.5 1.0 1.5 Concentration Concentration 4 c, = 0.28 B Sitting 4 c, = 0.50 B Standing Coc = 1.65 l:oc = 1.84 3, = 3.57 3, 2.00 :c ~ 2 ~2 I I :c 7), 0.78 J::t c/cr c/cr 0 0.0 0.5 1.0 1.5 0 0.0 0.5 1.0 1.5 Co ncentration Concentro tion 4 c. = 0.06 4 c, ~ 0.06 C Sitting Coc = 12.41 toe = 12.08 3, = 16.67 3, = 16.67 C Standing :c :c ~ 2 ~2 I I 0+=------,------,------~ 0+=------.------.------~ 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 Concentration Concentration Fig. 9. Personal exposure in the displaceent ventilated large test roo. Exposure (<1-) and concentration (o) are shown at the stratification heights l.o (A), 1.35 (B) and 2.25 (C) for sitting (left) and standing (right) anikin. Exposure ce, effectiveness in the occupied zone e, inhalation effectiveness e and DC e effectiveness of entrainent in the huan boundary layer 11 e. The tracer gas C0 2 is supplied above a heat source. Measureents in collaboration with [7].

13... Figure 9 shows the concentration profile (o) and the exposure (P.) at the three stratification heights 1.0 (A top of figure), 1.35 (B iddle of figure) and 2.25 (C botto of figure). In all three cases the theral anikin (approxiately 75 W), the two person siulators (2 x 100 W) and the point heat source (1 00 or 500 W) constitute the heat load in the roo. Due to the uninsulated walls there is a slight heat transfer between the roo and the surrounding laboratory, which is presuably of no particular iportance. As seen in figure 9 the effect of entrainent and transport of roo air fro the lower and cleaner zone to the breathing zone is distinct, which is also pointed out by Holberg et al. (1990). This eans that the concentration in the inhaled air ce is saller than the corresponding concentration in the sae height at a neutral place of the roo. In this case the entrainent provides a better indoor air quality in the displaceent ventilated roo than in the case of ixing ventilation. In figure 9 the exposure ce, the effectiveness in the occupied zone e oc and the inhalation effectiveness e e for the three cases are entioned. The figure shows that the effectiveness in the occupied zone is higher than in the case of ixing ventilation, where e ac approaches 1 in the ideal case. As expected the effectiveness increases with an increasing stratification height. In all cases the inhalation effectiveness e e exceeds e oc, indicating that the quality of the inhaled air exceeds the ean air quality in the occupied zone. As revealed, the exposure is uch depending on the concentration in the lower zone of the roo. This concentration ay vary (at the san1e stratification height) fro case to case depending on inlet device, heat sources, Archiedes nuber etc. Therefore a new quantity is defined, the effectiveness of entrainent in the huan boundary layer, designated '11 e 'lle (5) where Cr is the concentration at the floor which typically corresponds to the concentration in the lower, cleaner zone of the displaceent ventilated roo. The effectiveness of entrainent in the huan boundary layer '11 e, expresses the ability to supply (fresh) air fro the floor area to the breathing zone. It expresses the utilized fraction (cp- ce) of the possible concentration difference (cp- Cr). When 'lle is 1 all the inhaled air coes fro the lower zone (ce equals the concentration at the floor). When '11 e is 0 the concentration in the inhaled air ce equals the concentration at a neutral place at the sae height Cp, i.e. no particular effect because of the convective transport in the huan boundary layer. In the case of coplete ixing as well as the case of ideal displaceent ventilation with a stratification height located above the breathing zone, fie is not defined. In these cases ce equals the hoogeneous concentration in the lower zone.

14 One of the advantages of 11 e is the independence of the concentration in the lower zone which ay be specific in separate cases. With it an iproved possibility of coparing results fro different test roos and different set-ups arises. In figure 9 11 e is shown for case A and case B-standing where 11 e is defined. The figure illustrates to what extent the inhaled air is supplied fro the lower and cleaner zone. In case A-sitting alost all air coes fro the lower zone, while in case A-standing only a little ore than half of the air arises fro there. The reason why 11e is lower in the case A standing is obviously that the breathing zone is located at a higher level, and as a consequence ore containated air is entrained in the boundary layer before it reaches the breathing zone where it is inhaled. Holberg et al. (1990) have perfored easureents with four person siulators and four persons in a displaceent ventilated roo at a stratification height of 0.7. If the effectiveness of entrainent in the huan boundary layer 11 e is calculated using these easureents the following results are obtained. In the case of four person siulators 11 e is 0. 78 and in the case of four sitting otionless persons and four sitting persons in natural otion 11e aounts to 0.94 and 0.66, respectively. The case of the present easureents which is getting closest to the above-entioned conditions is the case A-sitting where 11 e is 0.91. This has to be copared with 0.94 (otionless sitting persons and Yst = 0.7 ). Despite the different roo sizes, inlet devices, heat sources and persons versus a theral anikin a fair agreeent is found. One disadvantage of the theral anikin exposure easureents is presuably the iobility of the anikin. In ventilated roos with real persons there will always be soe otion and disturbance. This point is supported by the easureents of Holberg et al. (1990), where 11e changed fro 0.94 to 0.66 when the four persons changed fro otionless to natural otion in sitting posture. This indicates what is previously entioned, that oveents decrease the ventilation effectiveness in the occupied zone resulting in increased ixing and increased exposure. In the extree case when a person oves very fast through a displaceent ventilated roo the exposure cc approaches Cp even though the stratification is very stable because of the huge disturbance of the boundary layer along the person and the considerable horizontal velocity generated around the breathing zone. Measureents concerning oveents are at the present tie perfored by eans of the theral anikin placed in a wind channel. The effect of oveents on the anikin is assued to be equivalent to the ipact fro the unifor velocity field. The easureents show a significant influence fro the velocity field on the huan boundary layer and with it an influence on the ability to transport fresh air as well as containated air upwards to the breathing zone, i.e. 11 e decreases when oveents increase.

3.4.2. Exposure in proportion to point containant source location 15 Apart fro personal exposure due to the general concentration distribution in the displaceent ventilated roo (3.4.1.) the effect of a point containant source location on the exposure ay be iportant. In figure 10 the exposure of a standing theral anikin is found in the case of a high and a low location of a point containant source. r 1. ~2. l.o _/./ /1 3.0 4 l ' 4 2..3.3 _c 0' 2 Q) I ~-- r--- --~ _c 0' 2 Q) I c,/c 0 1.74 c/cr 0+----.---,----,----,---, 0 2 4 6 8 10 Co ncentration 0 0 2 4 6 8 Co ncentra ti on Fig. 10. Personal exposure (-Q-) at concentration profiles ( o) corresponding to a high ( 1) and a low (2) location of a point containant source in the displaceent ventilated large test roo. The concentration profile is easured 0.9 fro the anikin. The easureents are perfored at conditions corresponding to case B (see figure 8). Neutral density tracer gas is N 2 0 ixed with He supplied through a porous foa rubber ball. Figure 10 shows the effect of two different concentration distributions corresponding to different point source locations. In the case of high location of the source ( 1) cc is 1.17 which ust be considered rather low copared with the high concentration peak in the height of the breathing zone. However, in the case of low location of the source (2), the exposure aounts to 1.74 while the concentration in the height of the breathing zone approaches I. This reveals the advantage and the disadvantage of entrainent and transport in the huan boundary layer. In case I the boundary layer transports fresh air to the breathing zone and only slight influence fro the very high local concentration in the

16 breathing zone height is seen. In case 2 the boundary layer obviously transports containated air to the breathing zone even though no particular high concentration is found in the breathing zone height at a distance of 0.9. The influence on personal exposure in proportion to the point source height is exained in figure 11. Two different cases are investigated, the anikin standing facing the inlet device (1) and the anikin standing reversed (2) as shown in figure 11. 1. 2. exp r 2.0 exp l +-' _c CJl Q) _c Q) u L :J 0 (.f) 0 Case 1 0 Case 2 0.5 1.0 Exposure 1.5 Fig. 11. Personal exposure ce as a function of point containant source height in the displaceent ventilated large test roo. The vertical distance between the source and the anikin is 1.5. Case 1, anikin facing inlet device (D) and case 2, anikin reversed (o). The easureents are perfored at conditions corresponding to case B (see figure 8). Neutral density tracer gas is N 2 0 ixed with He supplied through a porous foa rubber ball. Figure 11 shows how the exposure ce clearly varies with the elevation of the point source in the roo. Both iproved as well as deteriorated indoor air quality proportional to the case of coplete ixing is found. A distinct dependence on the flow field stresses the iportance of the convective ass transport when pollutant dispersion is addressed (see figure 4).

17 When the anikin, still facing the source located 1.5 away, is turned around (2) the exposure is alost opposite the previous case ( 1) as if irrored in the ordinate axis of figure 11. In the one case the vertical convective flow reoves the containation either giving rise to iproved (ce < 1) or soewhat ixed conditions (ce - 1) in the lower zone. In the other case the vertical convective flow transports the containation directly against the anikin where a fraction is entrained in the boundary layer giving rise to enhanced exposure (ce > 1). Actually, this result ay to soe extent be expected when the flow stratification is considered (ain flow at the floor, reverse flow above etc.). To illustrate the route of the containant in the boundary layer easureents are perfored on the containant concentration at the chest and the containant concentration at the back in proportion to the point source height. The conditions are otherwise the sae as in case 2 in figure 11. 2.0 Q)..c Q) u '- ::J 0 (f) 0 : Chest cone. 0 : Back cane. 0.5 1.0 Concentration Fig. 12. Concentrations easured at the chest (o) and the back (D) at a distance of0.3 fro a virtual point between the ears of the person. The concentrations are found as a function of the point source height. The conditions are otherwise the sae as case 2 in figure 11. The two concentrations (o) and (D) in figure 12 are easured at the sae height of the roo. If the person was not present they would be approxiately the sae, while in the actual case there is a significant difference. What akes the difference is the presence of a person and the entrainent and transport of the containant as well as the fresh air in the huan boundary layer.

18 If the concentration at the chest ( o) is copared with the exposure ce in case 2 figure 11 a very good correspondence is seen, due to the fact that the ain part of the inhaled air coes fro the boundary layer in front of the person, anyhow, in the case of an alost otionless person in a roo where the velocities fro the air distribution syste are sufficiently low to avoid draught and to ensure theral cofort. The easureents show the effect of persons' facing relative to the inlet device and the point containant source. If the anikin was turned around to face the inlet, the exposure ight presuably be reduced 25% in the present case. The topic of concentration distribution at different containant source locations in a displaceent ventilated roo is also discussed by Styne et al. ( 1991 ), where easureents are perfored by eans of a passive tracer gas technique. 4. CONCLUSIONS Persons present in a displaceent ventilated roo ay influence the air distribution to soe extent and vice versa the ventilation influence the persons. It is deonstrated that although the containant distribution is affected the stratification in the flow is stable when people are oving around in the roo. The teperature field is shown to be soewhat ore stable than the concentration field. When oveent and disturbance in the roo increase the ventilation effectiveness decreases and the exposure creases. The exposure of a sitting and a standing person as a function of the stratification height is exained. It is found that the flow in the boundary layer along a person, to a great extent, is able to entrain air fro below the breathing zone iproving the quality of the inhaled air. Persons' oveents causes the air quality to decrease due to the disturbance of the stratification as well as the disturbance of the huan boundary layer which prootes the high air quality transporting fresh air to the breathing zone. Entrainent of air in the huan boundary layer is usually an advantage, but easureents also show the possible disadvantage when containant sources in the lower part of the roo are present. In this case the plue around the person transports contain-ated air to the breathing zone giving rise to increased exposure. Two new quantities are defined: - The inhalation effectiveness e e which expresses the concentration of the air inhaled by the person relative to the return concentration.

19 - The effectiveness of entrainent in the huan boundary layer 11 e which expresses the ability to supply (fresh) air fro the floor area to the breathing zone. 'lle describes the utilized fraction of the possible concentration difference between the floor and a neutral point in the breathing zone height. ACKNOWLEDGEMENTS This research was supported financially by the Danish Technical Research Council (STVF) as a part of the research prograe "Healthy Buildings", 1993-1997. REFERENCES [1] Fanger, P.O. "Theral Cofort- Analyses and Applications in Environental Engineering". McGraw-Hill Book Copany, 1972. [2] Holberg, R.B., Folkesson, K., Stenberg, L.-G. and Jansson, G. "Experiental Analysis of Office Roo Cliate using Various Air Distribution Methods''. Roo vent '87, International Conference on Air Distribution in Ventilated Spaces, Stockhol, Sweden, 1987. [3] Nielsen, P.V. "Displaceent Ventilation- theory and design". ISSN 0902-8002 U9306, Departent of Building Technology and Structural Engineering, Aalborg University, 1993. [4] Mundt, E. "Containation Distribution in Displaceent Ventilation Influence of Disturbances". Proc. Indoor Air '93, Vol.5, Helsinki, Finland, 1993. [5] Phan, D.N. and Schulz, L.-M. Private counication, Aalborg University, 1991. [6] Nielsen, P.V. "Air Distribution Systes- Roo Air Moveent and Ventilation Efficiency". International Syposiu on Roo Air Convection and Ventilation Effectiveness, Tokyo, July, 1992. [7] Christensen, R. and Stevnhoved, L. Private counication, Aalborg University, 1993. [8] Holberg, R.B., Eliasson, L., Folkesson, K. and Strindehag, 0. "Inhalation Zone Air Quality Provided by Displaceent Ventilation". Roo Vent '90, International Conference on Air Distribution in Ventilated Spaces, Oslo, Norway, 1990. [9] Styne, H., Sandberg, M. and Mattsson, M. "Dispersion Pattern of Containants in a Displaceent Ventilated Roo- Iplications for Deand Control". 12th AIVC Conference, Ottawa, Canada, 24-27 Septeber, 1991.

PAPE R S ON INDOOR ENVIRONMENTAL T ECHNOLOG Y PAPER NO. 9: Per Heiselberg, Peter V. Nielsen: Flow Conditions in a Mechanically Ventilated Roo with a Convective Heat Source. ISSN 0902-7513 R8835. PAPER NO. 10: Peter V. Nielsen: Displaceent Ventilation in a Roo with Low Level Diffusers. ISSN 0902-7513 R8836. PAPER NO. 11: Peter V. Nielsen: Airflow Siulation Techniques - Progress and Trends. ISSN 0902-7513 R8926. PAPER NO. 12: M. Skovgaard, C. E. Hyldgaard & P. V. Nielsen: High and Low Reynolds Nuber Measureents in a Roo with an Ipinging Isotheral Jet. ISSN 0902-7513 R9003 PAPER NO. 13: M. Skovgaard, P. V. Nielsen: Nuerical Prediction of Air Distribution in Roos with Ventilation of the Mixing Type using the Standard I<, E Model. ISSN 0902-7513 R9042. PAPER NO. 14: P. Kofoed, P. V. Nielsen: Theral Plues in Ventilated Roos - Measureents in Stratified Surroundings and Analysis by Use of an Extrapolation Method. ISSN 0902-7513 R9043. PAPER NO. 15: P. Heiselberg, M. Sandberg: Convection fro a Slender Cylinder in a Ventilated Roo. ISSN 0902-7513 R9044. PAPER NO. 16: C. E. Hyldgaard: Water Evaporation in Swiing Baths. ISSN 0902-7513 R9045. PAPER NO. 17: H. Overby, M. Steen-Th0de: Calculation of Vertical Teperature Gradients in Heated Roos. ISSN 0902-7513 R9046. PAPER NO. 18: P. V. Nielsen, U. Madsen, D. Tveit: Experients on an Exhaust Hood for the Paint Industry. ISSN 0902-7513 R9146. PAPER NO. 19: L. Gerann Pedersen, P. V. Nielsen: Exhaust Syste Reinforced by Jet Flow. ISSN 0902-7513 R9147. PAPER NO. 20: P. V. Nielsen: Models for the Prediction of Roo Air Distribution. ISSN 0902-7513 R9148. PAPER NO. 21: M. Skovgaard, P. V. Nielsen: Modelling Coplex Inlet Geoetrie8 in CFD - Applied to Air Flow in Ventilated Roo8. ISSN 0902-7513 R9149. PAPER NO. 22: M. Skovgaard, P. V. Nielsen: Nuerical Investigation of Transitional Flow over a Backward Facing Step using a Low Reynolds Nuber k - c Model. ISSN 0902-7513 R9150. PAPER NO. 23: P. Kofoed, P. V. Nielsen: A uftriebsstroungen verschiedener Wiirequellen - Einfiuss der ugebenden Wande auf den geforderten Voluenstro. ISSN 0902-7513 R9151. PAPER NO. 24: P. Heiselberg: Concentration Distribution in a Ventilated Roo under Isotheral Conditions. ISSN 0902-7513 R9152.

PAPERS ON INDOOR ENVIRON MENTAL TECHNOLOGY PAPER NO. 25 : P. V. Nielsen: A ir Dis tribntion Systes - Roo Air Moveent and Ventilation Effec tiveness. ISS N 0902-75 13 R9250. PAPER NO. 26 : P. V. Nielsen: D escription of Snpply Openings in N n erical M odels for Roo A 1:r Distribntion ISSN 0902-7513 R9251. PAPER NO. 27: P. V. Nielsen : Ve loc d y Distribntion in the Flo w fro a W allo1tnied Di.ff v.ser in Roos with Displace ent Ventilation. ISS N 0902-7513 R.9252. PAPER NO. 28: T. V. J acobsen & P. V. Nielsen : Velocity and Tepetnre D is tribntion in Flow fro an Inlet D evice in Roos with Displaceent Ventilation. ISSN 0902-7513 R9253. PAPER NO. 29: P. Heiselberg: Dispersion of Containants in Indoor Cliate. ISSN 0902-7513 R9254. PAPER NO. 30: P. Heiselberg & N. C. Bergs0e: M eas ureents of Containant Dispers ion in Ventilated R oo.9 by a Pas.s ive Trac er Gas Techniqne. ISSN 0902-7513 R9255. PAPER NO. 31 : K. S. Christensen : Nu eriwl Predi ctio n of Airflow in a R oo with Ceiling-M o1ted Obstacles. ISSN 0902-75 13 R9256. PAPER NO. 32: S. G. Fox & P. V. Nielsen: Model E xperi ents in 1990 and On Site Validab:on in 199 2 of the Air Moveent in the Danish Pavilion in S ev1:lle. ISS N 0902-7513 R9335. PAPER NO. 33: U. Madsen, N. 0. Breu & P. V. Ni elsen: Local E xhau.9 t Ventilation - a N 1terical and E xperi ental Stndy of Capture Efficiency. ISS N 0902-7513 R9336. PAPER NO. 34: T. V. J a.cobsen, P. V. Nielsen : Nnerica.l Modelling of Ther al Environent in a D1:spla.c eent- Ventilated R oo. ISS N 0902-7513 R9337. PAPER NO. 35: P. Heiselberg: Dught R isk fr-o Cold Vertical S urfaces. ISSN 0902-75 13 R9338. PAPER NO. 36: P. V. Nielsen : Model Experients for the D et etination of Airflow in Large Spaces. ISS N 0902-75 13 R9339. PAPER NO. 37: K. Svidt: Nu etical Prediction of Bnoyant A ir Flo w in Livestock Buildings. ISSN 0902-75 13 R9351. PAPER NO. 38: K Svidt: I;estigation of Inlet B o1 dary Co n ditions Nn etical Prediction of A ir Flow in Livestock B uildings. ISS N 0902-7513 R9407. PAPER NO. 39: C. E. Hylclga.a.rd: Hnans as a Sov.rce of Heat and Air Pollv.t1:o n. ISSN 0902-7513 R9414. PAPER NO. 40: H. Brohus, P. V. Nielsen: Conta. inant DistTibv.tion a.rov.n d P etsons in Roos Venb;la.ted by Displaceent Ve ntilation. ISSN 0902-7513 R9415. Departent of Building Techn ology and Structural Engineering A a lborg Univer sity, Sohngaardshol svej 57. DK 9000 A albor g Telephone: + 45 98 15 85 22 'I'elefax: + 45 98 14 82 43