CIB3722

03-023
The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
ANALYSIS ON THE HYGROTHERMAL ENVIRONMENT OF CRAWLSPACES TO PREVENT MOISTURE DAMAGE BASED UPON COMBINED SIMULATION OF HEAT, AIR AND MOISTURE TRANSFER
Akihito OZAKI Dr.Eng.1 Soichi HASEGAWA1
1
Faculty of Environmental Engineering, University of Kitakyushu, 1-1 Hibikino, Wakamatsu, Kitakyushu, 808-0135, Japan, ozaki@env.kitakyu-u.ac.jp and a1402901@hibikino.ne.jp
Keywords: crawlspace, hygrothermal environment, ventilation, CFD, heat and moisture transfer
Summary
Hygrothermal environment in crawlspace of Japanese residential building was estimated to clarify characteristics of temperature and humidity variation and to prevent moisture damage by the complex simulation of heat and moisture transfer and airflow. The distribution of airflow velocity in the crawlspace obtained from the CFD analysis (in case of natural ventilation and mechanical ventilation) are compared for the general opening and the spacer opening set with the foundation, respectively. The analysis elucidated that the natural ventilation doesn’t sufficiently ventilate the crawlspace by either of the two openings. Then the combined heat and moisture transfer including moisture sorption and desorption of walls was incorporated with the calculation results of CFD. The variation of temperature and humidity in the crawlspace was simulated throughout the year. It was concluded that the humidity in the crawlspace, especially from around the center to leeward sides, is high during summer under natural ventilation, and that the humidity becomes low under the mechanical ventilations. Intermittent mechanical ventilation in the daytime during high temperature and low relative humidity of the outside air remarkably lowered the humidity in crawlspace less than relative humidity 80%.
1.
Introduction
One of the biggest issues on durability of the residential buildings in Japan is moisture damage such as condensation and decay in crawlspace. Natural ventilation of the crawlspace through an air vent opened at the foundations is standardized by the regulations to prevent moisture damage. The natural ventilation mainly purposes to exhaust moisture which is permeated through an earth floor from the ground. However a recent technological advance on the earth floor restraining moisture permeation diminishes an importance of the natural ventilation. On the contrary, there are some cases where the natural ventilation through the crawlspace with moisten air dampen building materials and causes the moisture damage in summer, so that the new techniques to prevent moisture damage such as intermittent mechanical ventilation in the daytime during high temperature and low relative humidity of the outside air that enhances drying of building materials is practical used, particularly in renovation of residential buildings. In this paper, the hygrothermal environment of the crawlspace and the effects of mechanical ventilation to prevent moisture damage are clarified by the computer simulation taking into consideration the complex relationship between heat and moisture transfer and airflow.
2.
2.1
CFD Analysis
Building Models and Calculation Conditions
Figure 1 illustrates the 1st floor plan and the foundation plan of the residential building. Figure 2 shows the position of respective air vents (general openings and spacer openings set with the foundation) used for the natural and mechanical ventilations in the crawlspace. Although the same foundation plan was used for both openings, the effective ventilation area of 600cm2 for the general opening and 1,000cm2 for the spacer opening were specified for every 4 meter length of the perimeter foundation based on the Housing Loan Corporation standard in Japan. Table 1 shows calculation conditions for CFD analysis. The control volume for analysis (x: 94m, y: 67m, and z: 25m) is approximately 10 times of that of the building model in the middle (refer to Figure 3.) The outdoor wind direction is hypothetically put on the y-axis positive direction with
- 1517 -

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
wind velocity of 1m/s when the airflow velocity at the windward openings (in case of natural ventilation) is supposed as 0.4m/s for the natural opening and 0.3m/s for the spacer opening. Three ventilators are used for mechanical ventilation. The ventilation amount per ventilator unit is 214m3/hour by the natural opening, and 100m3/hour by the spacer opening. Incidentally, the CFD analysis uses k ? ε turbulence model, logarithm-law for the building surfaces and the earth floor of crawlspace and 1/4 power-law for the ground outside the model as boundary conditions.
bathroom dressing room closet
WC
kitchen
closet lobby
Japanese style room hall porch
Japanese style room
a) The 1st floor plan [mm]
b) Foundation plan [mm]
Figure 1 The 1st floor plan and the foundation plan of the residential building
ventilation fan
3.5m
3.5m
ventilation fan
0.45m
7m
0.45m
7m
6.
9.4m
6.
9.4m
a) General openings
b) Spacer openings
Figure 2 Position of respective air vents (general openings and spacer openings set with the foundation) used for the natural and mechanical ventilations
94m
Figure 3 Control volume for CFD analysis and outdoor wind direction
- 1518 -
67 m
wind
25m

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
Table 1 Calculation conditions for CFD analysis Natural ventilation Calculation conditions Building Outside ground Ventilation amount Airflow velocity 0.3 m/s General opening Spacer opening Mechanical ventilation General opening Spacer opening
Boundary Supposition
Logarithm-law 1/4 power-law 0.4 m/s 642 m3/h 300 m3/h -
2.2
CFD Results
2.2.1 Natural Ventilation Figure 4 shows distribution of airflow velocity in the crawlspace obtained from the CFD analysis (in case of natural ventilation) for the general opening and the spacer opening, respectively. Most of the air that flowed into the crawlspace B1 and B2 zones flowed out from the lateral sides of the same zones in both the general openings and the spacer openings. The total air amount entering from the two air vents at B1 and B2 in the general opening is 76m3/hour, and the total air amount flowing out from the lateral openings is 54m3/hour, equivalent to 71% of the air that entered the area immediately flowing out. On the other hand, the total air amount flowing to the air vents at B1 and B2 in the spacer opening is 92m3/hour, and the total air amount flowing out from the lateral openings at B1 and B2 is 77m3/hour (equivalent to 84% of the air that entered the area flowing out). Due to the detachment zone created on the lateral sides of the building where air pressure drops, most of the air flowing into the crawlspace flows out from the lateral sides in both opening. This tendency is more remarkably observed in the spacer opening. As a result, the air velocity in the center of the crawlspace is slightly faster in case of the general opening. Incidentally, the airflow amount at the foundation corners is larger in the spacer opening compared to the general opening. However, in both openings, the airflow in the crawlspace is extremely slow (especially in the leeward zone from the center).
0. 8 0. 2 3. 7 1. 3 0. 7
6.0
(m3/h)
16.8
9.2
B3
3. 2 2. 8 1. 4 0. 6 0. 3
2.3
0. 7 B3
1. 7 0. 2 0. 8 5. 0
B4 ( m3/ h) 10. 3 11. 1 B1
B4
B4
B4
31.6
(m/s) 0.03 0.06 0.09 0.12 0.15 0.18 37.1 38.8
22.0
6. 3 8. 0 9. 9 5.4 8. 4 3. 6 17.0 11. 0
(m/s) 0.03 0.06 0.09 0.12 0.15 0.18
8. 8 B2 9. 9 5. 4 8. 7 7. 7 13. 6 8. 4 15. 7 3. 9 5. 3
a) General opening
b) Spacer opening
Figure 4 Distribution of airflow velocity in the crawlspace in case of the natural ventilation
2.2.2 Mechanical Ventilation Figure 5 shows distribution of airflow velocity in the crawlspace obtained from the CFD analysis (in case of the mechanical ventilation with three ventilators installed at B3 and B4) for the general opening and the spacer opening, respectively. The air amount flowing into the crawlspace in the general openings is 642m3/hour, and despite of the air velocity of 0.25m/s even at the center, air stagnation is still observed along the foundation corners. The air amount flowing into the crawlspace in the spacer opening is 300m3/hour with the current equally entering from all around of the foundation except from the sides where ventilators are installed. The air velocity in the center is 0.10m/s, which is slower compared to the general opening, but the space is equally ventilated all round.
- 1519 -

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
214.0 214.0 214.0 138.8 B4 B4 B3
(m3/h)
2. 7 100. 0 ( m3/ h) 4. 8 B4
2. 72. 9
2. 9 6. 8 2. 9 100. 0 B4
14. 4 13. 0 9. 2 B3 17. 9 15. 1 12. 8
B1 125.8
12. 5
(m/s) 0.25 0.20 0.15 0.10 0.05 125.8 125.8 125.8 B2
(m/s) 0.15
B1
11. 8 14. 2 B2 12. 8 10. 3 9. 9 11. 2 9. 9 9. 2 3. 9
10. 2 11. 6 9. 8 12. 9 14. 5 12. 2 10. 5 4. 5
0.12 0.09 0.06 0.03
a) General opening
b) Spacer opening
Figure 5 Distribution of airflow velocity in the crawlspace in case of the mechanical ventilation
3.
3.1
Combined Simulation of Heat and Moisture Transfer and Airflow
Simulation Software THERB
A Heat, Air and Moisture (HAM) simulation software program called THERB has been developed for the purpose of estimating the hygrothermal environment within buildings. This software has complete HAM features including principles of moisture transfer within walls. Generally simulation software to predict temperature, humidity, heating and cooling load of building spaces does not take into account moisture transfer in wall assemblies. Humidity calculation in most software is simply affected by ventilation and focuses on just the building spaces. THERB was developed to simulate humidity conditions in both building spaces and wall assemblies in detail. Thermal theories on conduction, convection, radiation and ventilation are based upon the detailed phenomena. The P-model using the water potential, which is defined as thermodynamic energy, is a progressive feature of THERB, which incorporates moisture transfer including moisture sorption and desorption of walls. The following is a brief outline of P-model. Water Potential which is derived by applying the chemical potential of thermodynamics to moisture diffusion is used as the driving force of moisture transfer. This approach is proposed to be more accurate than other models based on physical properties such as vapour pressure. The model called P-model using water potential makes it possible to combine moisture transfer with heat transfer perfectly, and take into account internal energy and external forces such as gravity. Balance equations of heat and moisture transfer in material are obtained as follows. z Heat Balance
?CρT ′ ? μw + μ f + clw jlw?T = ?λ?T + rv ?λ g ?t
(
)
(1)
μw
μw +
?μw ?x
dx
T Inflow
z
T +
?T ?x
dx
Moisture Balance
?φ ?μ ′ ? μ w + μ f + ?λl′? μ + μ f ρ lw = ?λ g ?μ ?t
(
)
(
)
A
A
Outflow
(2)
dx Cρ
?T ?t
Heat and Water Storage ,
? ρ g wψ ?t + ? ρ lwφ ?t
where C and ρ are specific heat and specific weight of material containing water. clw , ρ lw and jlw are specific heat, specific weight and flux of liquid phase water. λ is thermal ′ and λl′ are gaseous and liquid phase water conductivity. λg conductivity for μ w and μ gradients. rv is heat of sorption (latent heat of evaporation).
x
Figure 6 Numerical model of combined heat and moisture transfer
μ w is the water potential and defined from the basic thermodynamic principles as Eq.(3) to Eq.(5). The water o o potential is composed by saturated water potential μ w and unsaturated water potential μ . μ w expresses the thermodynamic energy of saturated vapour and μ expresses the difference of thermodynamic energy between saturated vapour and unsaturated vapour of moisten air.
o (T ) + μ ( p ) μ w ( p, T ) = μ w
(3)
- 1520 -

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
o μw (T ) = 6.44243 × 105 + c p ,w (T ? 273.15) ? Tc p ,w ln
kg
kg
ps T + Rwkg T ln 273.15 1.01325 × 105
(4) (5)
μ ( p ) = Rw T ln
kg
pw ps
where p w is the vapor pressure of the humid air, and p s is the saturated vapor pressure at temperature T . c p , wkg is the specific heat which is expressed in units of [J/(kg K)] and Rwkg = 461.50 [J/(kg K)] which is calculated by dividing the gas constant R = 8.31441 [J/(mol K)] by the molecular weight of water 18.016x10-3 [kg/mol].
μ f is the force water potential caused by internal energy and external forces. For instance, the force water potential which includes the influences of gravity and internal pressure is calculated by Eq.(6).
μ f = gz + pV w
(6)
where g is acceleration of gravity, z is height from reference position, Vw is the volume per unit weight of water and pVw is equal to Rwkg T . Boundary conditions of heat and moisture balance equations are expressed as follows. z
?λ
Heat Balance
?μ w ?T ′ = α c (Ta ? Ts ) + rv ? α ′(μ w,a ? μ w,s ) + qs ? rv ? λ g ?nv ?nv
(7)
z
′ ? λg
Moisture Balance
?μ w ′ (μ w, a ? μ w, s ) = αμ ?nv
(8)
where nv is normal line vector directed inward on a boundary surface, qs is quantity of radiant heat. Ta , Ts , μ w,a and μ w,s are the temperature and water potential of the air and surface, respectively. α c is convective ′ is convective moisture transfer coefficient for the water potential gradient. heat transfer coefficient and α μ ′ can be converted from general convective moisture transfer coefficient α ′p for the vapour pressure αμ gradient on the basis of Eq.(3).
? ′ = α′ αμ p? ? ?pw ? ps μ ? e = α ′p ? Rwkg T ? ?μ w ?
RWkg T
(9)
Indoor air temperature and humidity can be calculated from heat and moisture balance of a space based on convection, ventilation, internal generation of heat and moisture. Indoor humidity is interrelated with sorption and desorption of walls through the application of P-model. Thus THERB can predict the hygrothermal environment of the whole building taking into consideration the complex relationship between heat and moisture transfer and airflow.
3.2
Calculation Conditions
By using the simulation software of THERB and CFD, all the phenomena of heat and moisture transfer and airflow in the crawlspace were combined to calculate temperature and humidity of each zone. Incidentally, the crawlspace is divided into 38 zones (52 for the entire building) as illustrated by Figure 1. The sorption and desorption of moisture on the foundation, earth floor (vapor proof in the boundary between earth floor and soil level) and the 1st floor of building are also involved. The air amount flowing into the each zone of crawlspace was calculated on the basis of the results of natural and mechanical ventilations through the general openings and the spacer openings simulated by CFD. The influence of the different types of ventilation on the temperature and humidity in the crawlspace was examined in the general opening and the spacer opening (30mm urethane for insulation underneath the 1st floor of building in both techniques). Three different ventilation methods were supposed: natural ventilation; intermittent mechanical ventilation (mechanical ventilation with more than 300W/m2 of global solar radiation, and natural ventilation other than this condition); and all-day mechanical ventilation. Table 2 shows the calculation conditions. The standard weather data in Fukuoka was used for input data, where preset indoor temperature is 22 degree during December to March with constant relative humidity of 40%, 26 degree and constant 60% during June to September, and natural conditions are followed for the periods during April to May and October to November. The constant indoor ventilation is 0.5 times/hour all day.
- 1521 -

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
Table 2 Calculation conditions to estimate the hygrothermal environment in crawlspace Item Weather date Run-up Calculation time Time interval Space conditioning Heating Cooling Ventilation of building Natural ventilation Ventilation of crawlspace Intermittent mechanical ventilation All-day mechanical ventilation Insulation underneath the 1st floor Urethane 30mm CFD General opening Spacer opening
Standard weather data in Fukuoka 8 Months January - December(12 Months) 30 minutes December - March: 22degree, 40% June - September: 26degree, 60% 0.5 times/hour
3.3
Calculation Results of the Temperature and Humidity in the Crawlspace
Figure 7 and 8 illustrate monthly average of air temperature and relative humidity in the central zone of the crawlspace under respective ventilations of natural, intermittent-mechanical and all-day mechanical in the general opening and the spacer opening. The air temperature in the crawlspace shows higher value during summer by mechanical ventilation than by natural ventilation, and lower during winter. On the other hand, the relative humidity in the crawlspace is lower during summer by mechanical ventilation than by natural ventilation. The relative humidity under natural ventilation for both openings exceeds 80% between the period of July and August, from which it seems that there is high possibility of moisture damage during summer.
Outdoor General pening (natural) General opening (intermittent mechanical) General opening (all-day mechanical) Spacer opening (natural) Spacer opening (intermittent mechanical) Spacer Opening (all-day mechanical)
Outdoor General opening (natural) General opening (intermittent mechanical) General opening (all-day mechanical) Spacer opening (natural) Spacer opening (intermittent mechanical) Spacer opening (all-day mechanical)
[degree]
30 25 20 15 10 5 Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.
[%] Relative humidity
100 90 80 70 60 50 40 Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.
Temperature
Figure 7 monthly average of temperature
Outdoor General opening (natural) General opening (intermittent mechanical) General opening (all-day mechanical)
Figure 8 monthly average of humidity
Spacer opening (natural) Spacer opening (intermittent mechanical) Spacer opening (all-day mechanical)
[%] Cumulative ratio
100% 80% 60% 40% 20% 0% 50% 60% 70% 80% 90% 100% [%] Relative humidity
Figure 9 Cumulative ratio of relative humidity in August
- 1522 -

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
Figure 9 shows cumulative ratio of relative humidity in August in the central zone of the crawlspace under either natural, intermittent-mechanical and all-day mechanical ventilation by the general opening and the spacer opening, respectively. Higher cumulative ratio for the high humidity range is shown in order of the natural ventilation by the spacer and general opening, the mechanical ventilation by the spacer opening and the mechanical ventilation by the general opening. High relative humidity range more than 80% is unavoidable under natural ventilation by both openings in August, but the ratio drops to less than 40% with mechanical ventilation. Especially, the intermittent-mechanical ventilation demonstrates remarkable lowering in humidity. Figure 10 and 11 illustrate the relative humidity distribution (average in August) in the crawlspace by the general opening and the spacer opening. The humidity in the vicinity of the foundation is lower under natural ventilation by the spacer opening than by the general opening, but it becomes higher contrariwise from around the center to the leeward sides. The CFD analysis demonstrated slower airflow velocity from around the center to the leeward sides by the spacer opening than by the general opening, from which it is known that the air current affects the humidity lowering in the crawlspace. However, high humidity environment is created over wider areas by both openings under the natural ventilation. On the other hand, the humidity becomes low by either of the openings under the intermittent and all-day mechanical ventilations, generally showing relative humidity of less than 80%. Especially, lower humidity is obtained by intermittent mechanical ventilation than by all-day mechanical ventilation.
R H (% )
91 89
a) General opening (natural ventilation)
a) Spacer opening (natural ventilation)
87 85 83 81 79 77 75 73 71
b) General opening (intermittent mechanical)
b) Spacer opening (intermittent mechanical)
69 67 65 63
Average in August c) General opening (all-day mechanical) Figure 10 Relative humidity distribution in the crawlspace of the general opening c) Spacer opening (all-day mechanical) Figure 11 Relative humidity distribution in the crawlspace of the spacer opening
- 1523 -

The 2005 World Sustainable Building Conference, Tokyo, 27-29 September 2005 (SB05Tokyo)
4.
Conclusions
In this paper, CFD analysis was performed to compare the air current distribution of crawlspace under the natural and mechanical ventilations for the general opening and the spacer opening. The analysis elucidated that the natural ventilation doesn’t sufficiently ventilate the crawlspace by either of the two openings and that the air especially becomes stagnant from around the center to the leeward sides, and that the entire crawlspace is comparatively equally ventilated by mechanical ventilation. Then, by the interrelated simulation of heat and moisture and airflow which is calculated through the CFD analysis, the difference in the temperature and humidity in the crawlspace was compared under the natural ventilation, the intermittent and all-day mechanical ventilations in the general opening and the spacer openings, respectively. As a result, it was clarified that the humidity in the crawlspace is high during summer under natural ventilation by either of the general opening and spacer opening and that high humidity environment is especially created by the spacer opening from around the center to leeward sides, and that the humidity becomes low by either of the openings under the intermittent and all-day mechanical ventilations, generally showing relative humidity of less than 80%.
Acknowledgment
This paper has been supported by Grant-in-Aid for Scientific Research of Japan Society for the Promotion of Science.
References
Ozaki A., 2005. Combined Simulation of Heat and Moisture Transfer and Airflow on the Hygrothermal Environment and Heating/Cooling Load of Residential Buildings, Technical Paper of Annual Meeting of IBPSA-Japan/2005, 19-26 Ozaki A., Watanabe T. and Takase S., 2004. Simulation Software of the Hygrothermal Environment of Buildings Based on Detailed Thermodynamic Models, Proc. of eSim 2004, The Canadian Conference on Building Energy Simulation, 45-54 Ozaki A., Watanabe T., et al., 2001. Simulation Software to Describe the Thermal Environment of Residential Buildings Based on Detailed Physical Models, Proc. of eSim 2001, The Canadian Conference on Building Energy Simulation, 66-73 Ozaki A., Watanabe T., et al., 2001. Systematic Analysis on Combined Heat and Water Transfer through Porous Materials Based on Thermodynamic Energy, International Journal of Energy and Buildings, Vol.33, No.4, 341-350 Ozaki A., Watanabe T., et al., 1990. Heat and Mass Transfer at Outside Surface of Buildings – Wind Tunnel Tests of Heat and Mass Transfer on Horizontal Surfaces, Journal of Architecture, Planning and Environmental Engineering, Architectural Institute of Japan, No.407, 11-25
- 1524 -