Ar Echange and Ventlaton n an Underground Tran Staton Mkael Björlng 1* 1 Unversty of Gävle, Faculty of Technology and Envronment, Department of Buldngs, Energy, and Envronment, 1 76 Gävle * Correspondng emal: Mkael.Bjorlng@hg.se SUMMARY The ndoor ar clmate of an underground tran staton was nvestgated durng two days n January 08. The underground platform was accessed from tcket halls on each sde wth ar volumes of 1 000 m 3 and 1 430 m 3, respectvely. The staton platform ar volume was appromately 14 300 m 3. Ar from the outsde could enter ether va the tcket halls or va the tran tunnels from ventlaton towers stuated on each sde of the platform area. The local mean age of ar was determned n several locatons at dfferent heghts usng pumped samplng and homogeneous emsson of PFTs. In addton, the temperature and relatve humdty was measured at selected locatons. The average ar echange rate per hour (ACH) was found to be 3.62 h -1, rangng from 4 h -1 at rush hours to slghtly more than 3 h -1 at nght and n the mddle of the day. The largest ACH (4.5 h -1 ) was found at rush hour n one of the tcket halls, correspondng to a flow rate of 75 000 m 3 /h. The lowest ACH (2.8 h -1 ) was found n the other tcket hall at nght, correspondng to a flow rate of 47 000 m 3 /h. In the mddle of the staton platform the ACH was lower than the ACH at the platform ends. KEYWORDS Tunnel, PFT tracer gas, homogeneous emsson technque, mult-zone model, relatve humdty. 1 INTRODUCTION The ventlaton of large premses s a challenge. The arflows n these systems can consst of a complcated array of nlets, outlets, and nter-zonal flows that may be dffcult to analyse. The homogeneous tracer gas emsson technque s one way to smplfy measurements of some ventlaton parameters. In ths technque, t smply suffces to measure the local concentraton of the tracer gas to evaluate the local mean age of ar. The technque has also been used n a smlar project to measure the ventlaton of road tunnels (Brng et al, 1997). By usng several dfferent tracer gases, t s also possble to nvestgate the flows nsde the premses (Etherdge and Sandberg, 1996). Ths study was performed to map the ventlaton flows n an underground tran staton as a part of the documentaton n a larger project wth the am of redesgnng the underground tran system. 2 MATERIALS/METHODS Measurements have been performed on the 2 and 3 of January 08 n four tme perods each rangng appromately one and a half hour (Perod 1: nght 22:00 23:40, 2/1; 2: mornng (8 am) 08:00 09:40, 3/1; 3: noon 11:40 13:, 3/1; and 4: evenng (5 pm) 16: 18:00, 3/1). The nfltrated ventlaton ar was measured usng the homogeneous emsson tracer gas technque descrbed n NORDTEST Standard VVS 118 (1997) and ISO 100-8 (05). Two dfferent perfluorocarbon tracers (PFTs), perfluorobenzene (tracer A) and perfluorotoluene (tracer B), were used. For the tracer gas emsson we have chosen to use passve tracer gas sources (Stymne and Boman, 1994). For the eperment 00 sngle tracer gas sources were
grouped n perforated boes and dstrbuted n the mornng of the 2 of January. The homogeneous emsson tracer gas technque requres the emsson of tracer gases to be proportonal to the volume of each zone. A schematc layout of the underground staton wth ts ar flows s shown n fgure 1. The natural dvson of the underground tran staton s nto three zones: the two tcket halls and the platform area. In each of the tcket halls only tracer gas B was emtted from boes n 5 postons. In the platform area only tracer gas A was emtted from 18 boes hangng at a heght of 4 meters over the platform level. In each measurement perod samples were obtaned n trplcates usng smultaneous pumped samplng for mnutes onto charcoal tubes wth calbrated and programmable SKC samplng pumps. There were 3 measurng ponts n each of the tcket halls, and 5 measurng ponts evenly dstrbuted on the platform. All measurng ponts were stuated at heghts close to 2.5 meters. Temperature and RH were measured n the tcket halls (close to the stars), at the platform ends and n the mddle of platform, wth a frequency of one value every 2 mnutes. Outdoor temperatures and RH were obtaned as three hour averages from the nearby SMHI weather staton at Observatorelunden. The tran frequency was also recorded durng the measurement perods. Tcket Hall X Tcket Hall Y Tunnel Tunnel Fgure 1. Schematc desgn of the underground staton. The tcket halls are n drect contact wth the outsde ar. On ether sde of tcket halls, the tunnels are n contact wth the outsde ar va ventlaton shafts. Therefore the platform may receve fresh ar ether va the tcket halls or the tunnels. 3 RESULTS Ventlaton system descrpton The ventlaton of the underground tran staton (fgure 1) s affected by the tran traffc n the tunnels and by natural ventlaton. There s no forced ventlaton system. Fresh ar may enter the system ether va the doors n the tcket hall or va ventlaton shafts connectng the tunnels to the street level. The ventlaton shafts are located n the tunnels, on ether sde of the platform, just after the pont where the tracks separate to arrve on ether sde of the platform. Trans act as pstons n the tunnels to transport ar from the tunnel and the ventlaton shafts nto the platform area. Natural drvng forces lead to transport of ar through the doors of the tcket halls and the ventlaton shafts n the tunnels. Natural ventlaton s thus affected both by the outdoor condtons and the frequency of door openngs. Tcket hall X s buser and has more doors than tcket hall Y. Temperature and relatve humdty The temperature and relatve humdty varatons for the two tcket halls are shown n Fgure 2. The correspondng varatons for the platform and outdoors are shown n Fgure 3. January 2 was moderately cloudy wth 0.7 mm precptaton and wnd speeds 1-3 m/s whereas January 3 was cloudy wth 1.1 mm precptaton and wnd speeds 1-7 m/s. Fgures 2 and 3 clearly show that tcket hall X s buser and has longer perods where doors are constantly open so
that the values approaches the outdoor condtons, whereas tcket hall Y and the platform s moderately affected. Temerature [ºC] 18 16 14 12 10 8 6 18 22 00 02 04 06 08 10 12 14 16 January 2 and 3, 08 40 30 Relatve Humdty [%RH] Fgure 2. Temperature (red) and relatve humdty (blue) n tcket hall X (left) and tcket hall Y (rght). Measurng perods are marked. Temperature [ºC] 18 16 14 12 10 8 6 18 22 00 02 04 06 08 10 12 14 16 January 2 and 3, 08 40 30 Relatve Humdty [%RH] Temperatur [ºC] 18 16 14 12 10 8 6 18 22 00 02 04 06 08 10 12 14 16 January 2 and 3, 08 40 30 Relatve Humdty [%RH] Temperature [ºC] Fgure 3. Temperature (red) and relatve humdty (blue) n the mddle of the platform (left) and outdoors (rght). The measurng perods are marked. Determnng the local mean age of ar The natural ventlaton varable measured by the method s the local mean age of ar τ [h] (Stymne and Boman, 1994). The local age of the ar, snce t entered the system depends on the local mng condtons. Older ar means less effcent removal of contamnants and supply of fresh ar. Snce we label the ar n the system wth two dfferent tracer gases, A and B, the local mean age of ar s gven by 5 4 3 2 1 0-1 -2-3 -4-5 02 00 02 12 03 00 03 12 04 00 Date Tme 100 95 85 75 65 55 Relatve Humdty [%RH] τ = τ + τ A B = c k + c A A B k B (1) where c s the concentraton of tracer gas [g m -3 ]and k s the emsson rate of tracer gas per volume [g h -1 m -3 ]. The concentraton s gven by c = M κ T (2) where M s the mass of tracer gas collected n the samplers [g], κ s the ar samplng rate [m 3 h -1 ], and T s the samplng tme [h]. The mass collected s measured usng a gas chromatography technque. The results of the measurements are presented n Fgure 4 for all the measurng perods. The relatve rush hours at 8 am and 5 pm stand out as the age of the ar s clearly less than durng noon and nght for all measurng ponts. The nflu of fresh ar s larger for the rush hour perods compared to noon and nght. Ths may partly be due to the
larger number of trans (Table 1) and partly because the doors n the tcket halls are open for longer perods durng rush hour. The ar generally s older toward the mddle and the Y sde of the platform. The mddle of the platform s least affected by the rush hours. In spte of the fact that tcket hall X s buser than Y, ther local mean ages do not dffer much. A quck analyss of the two tracer gas concentratons n the tcket halls shows that tcket hall X has a larger nflu of old ar from the platform area than tcket hall Y. 0.5 0.4 nght 8am noon 5 pm 0.3 0.2 0.1 0.0 Tcket Hall X (X-sde) (mddle) (Y-sde) Tcket Hall Y Fgure 4. The local mean age of ar, τ [h] measured at selected ponts for each of the measurng perods (see legend). Table 1. Number of trans n each drecton durng the measurng perods perod X-drecton Y-drecton nght 12 12 8 am 21 15 noon 18 17 5 pm 23 21 The local ar change rate and the total nflow of ar In steady state condtons, the local ar change rate ACH [h -1 ] can then be obtaned as the nverse of τ. In Fgure 5, the calculated results are presented wth error bars. The same local trends as for the local mean age are observed. The measured local ar change rates are nearly double the recommended ar change rates for resdental buldngs,.e. twce per hour. Ecept for the mornng rush hour, the mddle of the platforms s the least ventlated spot. The total ar flow rate Q tot [m 3 h -1 ] s gven by the total room volume dvded by the average τ for the etng ar (Etherdge and Sandberg, 1996). Snce there are many et ponts n the system measurng n all of them s dffcult. However, we may estmate the τ for the etng ar by the volume-weghted average of the local mean ages. These estmates are presented n Table 2. As epected the total nflu of ar nto the underground tran staton s qute large and s the largest durng rush hours. The average ar echange rate per hour (ACH) was found to be 3.62 h -1, rangng from 4 h -1 at rush hours to slghtly more than 3 h -1 at nght and n the mddle of
the day. The largest ACH (4.5 h -1 ) was found at rush hour n one of the tcket halls, correspondng to a flow rate of 75 000 m 3 /h. The lowest ACH (2.8 h -1 ) was found n the other tcket hall at nght, correspondng to a flow rate of 47 000 m 3 /h. 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 Tcket hall X (X-sde) nght 8 am noon 5 pm (mddle) (Y-sde) Tcket Hall Y Fgure 5. The local ar change rate, ACH (h -1 ) measured at selected ponts for all the measurng perods (see legend). The error bars are also gven. Table 2. Volume-weghted averages of the local mean ages of ar and the estmated total arflow nto the system. τ average [h -1 ] Measurng Object volume [m³] nght 8 am noon 5 pm Tcket Hall X 1000 0.27 0.22 0.31 0.23 14300 0.33 0.26 0.31 0.27 Tcket Hall Y 10 0.36 0.26 0.26 0.24 Total 160 Volume Averaged Mean Age 0.33 0.26 0.31 0.26 Estmated Total Arflow[ m 3 h -1 ] 0 64000 500 64000 4 DISCUSSION The measurng perods were chosen to compare perods of less frequent travel wth the relatve rush hours n the mornng and afternoon. However, at the dates of the measurements, one may epect rush hour to be less ntense than a normal work day snce many people may have chosen to reman on Chrstmas vacaton. The ventlaton ncreases durng rush hour but the reasons for that must be nvestgated further. Is t manly due to the ncreased frequency of door openngs or to the ncreased frequency of trans that could lead to a greater nflu of fresh ar from the tunnel ventlaton shafts? We are also able to see local patterns of mng condtons. The mddle of the platform appears to be the most statonary part of the ar. Usng
two dfferent tracer gases wll also enable a further study of the nternal flows n the underground tran staton. Such a quanttatve study requres a more lengthy analyss and wll be publshed n a subsequent paper (Björlng, 12). However, by nspecton we could conclude that a large contrbuton to the local mean age of ar n the buser tcket hall X come from the nflu of older ar from the platform area. 5 CONCLUSIONS We show that determnaton of the local mean age of ar can be used to study the mng condtons n an underground tran staton. In turn we can evaluate other mportant parameters such as the local ar change rate as well the total nflu of fresh ar. Usng dfferent tracer gases also enable us to draw some conclusons concernng the nternal nter-zonal flow patterns. However, because our mult-zone system s underdetermned snce we only use two dfferent tracer gases the analyss s necessarly ncomplete. Some addtonal quantfcaton of the nternal flows s nevertheless possble from the measurements but requres a more lengthy analyss and some etra assumptons. ACKNOWLEDGEMENT Hans Stymne and Carl-Ael Boman (Pentaq AB), and Bengt Chrstensson (IVL) are acknowledged for assstance. AB Storstockholms Lokaltrafk (SL) s gratefully acknowledged for fnancng the study. 6 REFERENCES Aley, J.W. 1989. Mult-zone dspersal analyss by element assembly, Buldng and Envronment. 1989;24: 113-130. Björlng, M. 12 To be submtted to Tunnellng and Underground Space Technology. Brng, A. Malmström, TG. and Boman, CA. 1997 Smulaton and Measurement of Road Tunnel Ventlaton, Tunnellng and Underground Space Technology, vol. 12, No. 3, pp 417-424. Etherdge, D. and Sandberg, M. 1996. Buldng Ventlaton Theory and Measurement. John Wley ISO/DIS 100-8. 05. Indoor Ar Part 8: Determnaton of local mean ages of ar n buldngs for characterzng ventlaton condtons, Internatonal Organzaton for Standardzaton. NORDTEST. 1997. Ventlaton: Local mean age of ar homogeneous emsson technque. Nordtest Method: NT VVS 118, Nordtest, Fnland. Stymne H. and Boman CA. 1994. "Measurement of ventlaton and ar dstrbuton usng the homogeneous emsson technque - a valdaton". Healthy Buldngs '94, Proc. of the 3rd Internatonal Conference, vol. 2, pp 539-544, Budapest, Hungary.