Investigation of flow and heat transfer around internal channels of an air ventilation vest

Abstract

The effects of a fabric’s ventilation and cooling are important to human comfort in the design of clothes. Persons in a high-temperature situation can experience imbalance of thermoregulation, which in turn results in the symptoms of heat exhaustion and heat stroke. The purpose of this research was to achieve cooling enhancement by designing air ventilation vests as an effective measure to control human thermoregulation in hot environs. In this study, the Taiwan Textile Research Institute devised a new air ventilation vest characterized by seven internal flow channels. The computational fluid dynamics software ANSYS/Fluent® was applied using the three-dimensional conservation equations of mass, momentum and energy to simulate the airflow and heat transfer phenomena within the internal channels. The incompressible turbulent flow was also treated with a standard k-ɛ two-equation model for turbulence closure. We compared the predicted steady axial velocities at the centers of channel exits with the measured data for software validation. The simulated results were employed to investigate the complicated flowfield and heat transfer phenomena around internal channels of the air ventilation vest. In practice, the design with the outlet arranged along the airflow direction produces a higher flow rate due to a relatively lower flow resistance, and thereby achieves better air-cooling outcomes. To enhance the cooling performance, simulations were also conducted to examine the influences of channel outlet design, inlet temperature and velocity on the convective heat transfer coefficient distributions over the inner channel surface of the vest.

Keywords

air ventilation vest computational fluid dynamics simulation convective heat transfer coefficient

Workers in several professions frequently perform tasks in hot environments, where they can be subjected to excessive heat tension, leading to a decline in work performance and productivity.¹ Heat tension weakens the human abilities of mental awareness, concentration and coordination,^2–5 as well as causing muscle fatigue,^6,7 and it is a primary reason for accidents and injuries in many workplaces. In these types of situations, personal cooling (PC) apparatus is a practical and cost-effective approach to cool the micro climate of wearers. For instance, in the scenarios where protective clothing is necessary, PC garments can be effective in controlling the micro climate of individuals for heat-stress management.^8–10

In general, there are three categories of PC garments, including phase change material (PCM) cooling modules, evaporative cooled garments (ECGs) and fluid cooled garments (FCGs). The PCM modules utilize the PCMs having the ability to absorb a certain amount of latent heat from the change of solid to liquid state in an appropriate temperature range for PC.¹¹ Similarly, the inner fabric structure of ECGs can hold water supplied from a hydration system or an external reservoir (e.g. a water bottle) by a built-in hand pump to realize water evaporation for removal of latent heat from the wearer's body core.^12,13 Some examinations of commercial vests have shown that PCM garments are portable but provide only limited cooling capacities.^14,15 In addition, the high latent heat of water evaporation can assure the potential of making effective ECGs using a small amount of water. However, conventional ECGs have the main drawback of poor performance in highly humid environments attributable to a small evaporation flux.^16,17

PC technologies are considered to be highly efficient nowadays. For instance, FCGs utilize the coolant to attain continuous cooling at capacities ranging from 600 to 1000 W.¹⁸ Nevertheless, systems using water or non-toxic aqueous solutions are restricted by lack of transportability, because of the needs of complex refrigeration units and continual power supply.¹⁹ The air-based FCGs are a feasible alternative for microclimate control to reduce the human body temperature via convective cooling because of the advantages of supple, form-fitting, low-cost and lightweight structures.²⁰ Efforts have been made to develop various air-cooling/ventilation vests as well as to study the thermo-fluid behavior and performance of the associated systems. An early study of Muza et al.²¹ examined the effectiveness of decreasing thermal strain of the soldiers by furnishing an air-cooled vest with each of four different temperatures and airflow rate arrangements. The test results were used to determine minimum air conditioning requirements for several military vehicles. In an endeavor to improve the intermittent conditioned air-cooling effect, Chen et al.²² adopted a method of supplementing constant air cooling through ambient air supplied from a lightweight portable unit during work with conditioned air cooling delivered during rest periods to enhance personal comfort and decrease the cumulative fatigue over repeated work/rest cycles in selected military and industrial applications.

Subsequently, Weber²³ devised a simple, lightweight cooling garment for medical personnel working intensively in warm environments. The garment involved two layers made of an outer, air-impermeable fabric and an inner, air permeable fabric with the layers allowing air to flow between such layers and through the inner air, then the permeable layer to the body of wearers. Likewise, Uglene²⁴ proposed a sophisticated PC garment having inner and outer layers that created a confined space to contain liquid water. Both layers were impermeable to air and liquid water yet permeable to water vapor, whereas the innermost layer of the garment was in direct contact with the wearer’s skin. The liquid water between the layers diffused as vapor through the outer layer; it took away latent heat required for evaporation and hence had a cooling effect on the human body. Next, McCarter and Hayden²⁵ developed a protective vest with multiple ridge-and-channel ribs to form air channels on the inner surface next to the wearer for cooling. The cooling garment consisted of a detachable air exhaust fan and power supply with stretchable sheet material used to conform to body shapes for facilitating airflow through the channels.

Meanwhile, Horn²⁶ designed a fan-operated cooling vest with a structure that created a plenum between a garment and the body surface to facilitate the airflow. This scheme could deliver forced air by means of a fan or similar device to effect evaporation of moisture from a person and thus cool wearers in heavy protective equipment. Farnworth and Dacey²⁷ presented a lightweight, conformable, cooling garment containing a basically gas-impermeable first substrate and a gas-permeable second substrate attached to form a cavity. Spacer elements were inserted between the first and second substrates to maintain gas flow in the regions likely to be subjected to compression caused by the human body.

Numerical studies were also performed to determine the flow and heat transfer characteristics of PC garments. Li et al.²⁸ developed a model of simultaneous transport in hygroscopic porous materials to simulate the moisture diffusion process in wool fabrics. The numerical predictions showed the contours of temperature, moisture concentration, liquid water saturation and atmospheric pressure in the void space between fibers, indicating the important influence of atmospheric pressure gradient on the associated heat and mass transport processes. Ghaddar et al.²⁹ formulated a three-dimensional dynamic model to analyze the ventilation radial airflow via the fabric, the angular and axial airflow induced by the oscillating motion of the inner cylinder, and the sensible and latent heat losses from the skin due to ventilation. It was found that the predicted ventilation rates within the annulus were in good agreement with the measured data at higher frequencies. To study the processes of moisture transfer in the inter-fiber void spaces and in the fibers, Li and Li³⁰ formed a dynamic model to describe the heat and moisture transfer in porous textiles with PCM microcapsules. The computations under the ambient transient conditions were performed to investigate the influences of the phase transition temperature range and the heating/cooling rate on the phase change processes of the PCM. Moisture capacitance is important to appropriately model evaporative heat loss from the body through clothing. Taking both heat and moisture capacitance of clothing into consideration, Voelker et al.³¹ proposed a numerical model by implementing a clothing node to calculate the absorption amount of moisture for a specific fabric at a given humidity. Afterwards, Barauskas and Abraitiene³² and Barauskas et al.³³ demonstrated a structural finite-element method computational model to investigate the forced ventilation performance of a three-dimensional textile layer. The analysis could be applied to predict the time point of initiation of sweating at given heat and water vapor generation rates.

Most previous studies have concentrated on the design and development of air-cooling/ventilation vests with numerical calculations and experimental measurements conducted to analyze the performance of cooling garments for personal microclimate control purposes. However, comparatively limited computational investigations were conducted to determine the detailed thermo-fluid characteristics in the cooling device. In this research, we employed the computational fluid dynamics (CFD) software ANSYS/Fluent® for calculations to simulate such complex flows.³⁴ This study aimed to investigate the flowfield and heat transfer phenomena of the internal channels of an air ventilation vest. Considering the delivered inlet airflow rate obtained from the intersecting point of the flow resistance curve and the fan pressure–flow rate (P-Q) curve in balanced operations, the predicted steady axial velocities at the centers of seven channel exits were compared with the measured data for software validation. In response to the design needs, CFD simulations were also conducted to explore the influences of channel outlet design, inlet temperature and velocity on the convective heat transfer coefficient distributions for improving and accelerating the development process of air ventilation vests.

Methods

Design of the vest

The present study aimed to demonstrate the effectiveness of a simple, low-cost, reliable and lightweight air ventilation vest for reducing the thermal tension of wearers during exposure to a hot, dry environment. Figure 1 shows a schematic diagram of air ventilation vests with two different outlet designs. Figure 2 presents pictures of the air ventilation vest (in front, back and side views) worn by a man 170 cm tall. The dimensions of the air ventilation vest were 38 cm wide, 70 cm long and 15 cm thick. Two vest prototypes were proposed, with the major difference being the outlet layouts of internal channels. Design A (as illustrated in Figure 1), considered as the baseline design, has an exit along the flow path of internal channels with a rectangular port of 4 cm wide and 0.5 cm high, while Design B adopts a 90^o-turn at the flow exit with a circular outflow port of 2 cm in diameter. Both the inner and outer surfaces of internal channels of air ventilation vests were made of air-impermeable urethane-coated blend fabric of nylon and cotton to avoid airflow leakage. A spacer fabric structure was employed to maintain the shape of channels in the vests for ensuring a smooth low-resistance airflow pathway. For the baseline case, an axial fan (Y.S. TECH FD126010HB) was utilized to deliver airflow via a 6-cm diameter inflow port.

Figure 1.

Schematic diagram of air ventilation vests with two different outlet designs.

Figure 2.

Pictures of the air ventilation vest in front, back and side views.

Model theory

Numerical calculations were performed using the commercial CFD software ANSYS/Fluent® to predict the velocity and temperature fields in the air ventilation vest.³⁴ For the incompressible turbulent airflow, the theoretical analysis was based on the steady, three-dimensional, conservation equations of mass, momentum and energy equations with a k-ɛ two-equation model for turbulence closure. The governing equations are as follows:

\frac{\partial u_{i}}{\partial x_{i}} = 0

(1)

\frac{\partial ρ u_{i} u_{j}}{\partial x_{j}} = ρ g_{i} - \frac{\partial p}{\partial x_{i}} + \frac{\partial}{\partial x_{j}} [μ_{eff} (\frac{\partial u_{i}}{\partial x_{j}} + \frac{\partial u_{j}}{\partial x_{i}})]

(2)

In the foregoing equations, the symbol u_i designates the velocity component in the i axis; moreover, p, ρ, μ_eff and ρg_i represent the pressure, density, effective viscosity (defined as the sum of laminar viscosity μ and turbulent viscosity μ_t) and gravitational force, respectively. The energy equation can be written as

\frac{\partial ρ u_{j} h}{\partial x_{j}} - u_{j} \frac{\partial p}{\partial x_{j}} = \frac{\partial}{\partial x_{j}} [(λ + λ_{t}) (\frac{\partial T}{\partial x_{j}})] + μ_{eff} ϕ

(3)

The signs h, λ/λ_t and μ_effψ are the sensible enthalpy, molecular/turbulent conductivity and viscous dissipation terms, respectively. Considered as the most popular, well-established and broadly tested turbulence model, a standard k-ɛ two-equation turbulent model was adopted for turbulence closure, as follows:

\frac{\partial ρ u_{i} k}{\partial x_{i}} = ρ P - ρ ɛ + \frac{\partial}{\partial x_{i}} [(μ + \frac{μ_{t}}{σ_{k}}) \frac{\partial k}{\partial x_{i}}]

(4)

\frac{\partial ρ u_{i} ɛ}{\partial x_{i}} = C_{ɛ 1} \frac{ρ P ɛ}{k} - C_{ɛ 2} \frac{ρ ɛ^{2}}{k} + \frac{\partial}{\partial x_{i}} [(μ + \frac{μ_{t}}{σ_{ɛ}}) \frac{\partial ɛ}{\partial x_{i}}]

(5)

The production term is defined as

P = \frac{μ_{t}}{ρ} (\frac{\partial u_{j}}{\partial x_{i}} + \frac{\partial u_{i}}{\partial x_{j}}) \frac{\partial u_{j}}{\partial x_{i}}

(6)

where μ_t is C_μ ρk²/ɛ, k is the turbulent kinetic energy and ɛ is the turbulent energy dissipation rate. As a common practice, the constants C_μ, C_ɛ1, C_ɛ2, σ_k and σ_ɛ were set as 0.09, 1.44, 1.92, 1.0 and 1.3, respectively. In this study, the ambient pressure outside the internal channels was 1 atm. The zero normal gradients of p, k and m were imposed on the solid surface, and the log-law wall functions were applied for the next-to-surface grids.³⁵ As shown in the measurements, the walls were roughly isothermal over the surfaces of internal channels. For the baseline conditions, the inner channel surface temperature corresponding to the human body side was 37.5℃ in the summertime, while the temperatures of the vest inlet and outer channel surface were specified to be 33℃ in accordance with the environmental conditions. In calculations, the density, viscosity and conductivity were 1.15 kg/m³, 1.87×10^–5kg/m-s and 0.026 W/m-K at the temperature of 33℃, respectively. Following the test conditions, airflow in the channels was driven by the axial fan with the specified mean velocity of 1.86 m/s (equivalent to the delivered flow rate of 5.3 L/s). To determine the proper inlet turbulence values of the airflow into the internal channels, the turbulence kinetic energy and dissipation rate were computed by the following equations, respectively,

k = (3 / 2) (IU)^{2}

and

ɛ = C_{μ}^{0.75} k^{1.5} / κ L

. Here U is the inlet velocity; I and L are the average turbulence intensity and a reasonable length scale at the inlet,³⁴ respectively. In this work, the turbulent intensity (I) was specified to be 5.0%. The constant κ is equal to 0.4 and the typical value of L is suggested as the radius of the inflow port, which is 0.03 m. The outlet boundaries of the computational domain were arranged at the locations beyond the openings and treated as a fixed pressure of 1 atm. The converged solution was used to check if the exhaust mass flow rate equaled the supply mass flow rate from the fan for mass conservation. The above mathematical equations were discretized by the finite control volume approach. A second-order accurate central difference scheme was utilized to model the convective terms through the adaptive damping to deal with the non-physical oscillations. The diffusion terms were also estimated by a second-order accurate central difference scheme in numerical solutions. An iterative semi-implicit method for pressure-linked equations consistent (SIMPLEC) numerical scheme was employed for velocity–pressure coupling.^36,37

Experimental data collection

Since the airflow is essentially unidirectional with respect to the internal channels, the axial velocity was measured by a hot-wire anemometer (TSI 8495) operating in a constant-temperature mode. The single straight hot-wire probe was mounted on a steel extended-contact bearing stage with a minimum resolution of 10 µm along the moving direction. Because the probe was not sensitive to flow direction, the measurements were limited to the region of small transverse velocities with the measuring locations at a 0.5-cm distance out of the center of the exit plane for each channel. The measuring time period was at least 1-min duration for each point to circumvent the effect of relocation of the sensing device for causing airflow disturbance. An airflow velocity calibrator (TSI8392 CERTIFIER®) was utilized to calibrate the hot wire with a measured accuracy of ±1% over the course of the trials. The measured results can be readily used to validate the accuracy of the CFD predictions.

Results and discussion

Simulations were conducted with the CFD software ANSYS/Fluent® to validate the present theoretical model by comparing the predictions with the measured results. Figure 3 illustrates the numerical grids of the air ventilation vest with enlarged views for inflow and outflow ports. The mesh system consisted of several major structured portions: the joint chamber, internal channels, entrance and exits. Finer grids were deposited in the regions near the channels, entrance and exits, as well as near the wall boundaries. The average cell length in the channels was about 1.25 mm with the smallest spacing of 83.7 µm for resolving steep variations of thermo-fluid properties near the surfaces. Calculations were performed on total grids of 1,061,126, 1,273,351 and 1,578,955 points. In determining the velocity and temperature distributions, the normalized residual errors of the flow variables (u, v, w, p, T, k and ɛ) converged to 10^–6 in steady calculations with the mass conservation check under 0.5%. The calculated centerline velocity and temperature profiles across the joint chamber and internal channels at different grids and time-step values indicated that satisfactory grid independence could be achieved using a mesh setup of 1,273,351 grids. It generally needs 30 hours of central processing unit (CPU) time to attain a converged steady-state solution on an Intel-Xeon E5-2643-3.30 GHz (24GB RAM) work station for investigating the flow circulation process in internal channels of air ventilation vests.

Figure 3.

Numerical grids of the air ventilation vest with enlarged views for inflow and outflow ports.

In the analysis, the delivered flow rate from the fan is required to prescribe the velocity at the inlet. Because the intrusion of a hot-wire probe into channels could substantially change the wake flow structure of the fan and thereby deteriorate the accuracy during velocity measurements, this study incorporated a fan-curve model into the CFD software to generate the relationship of flow rates respecting pressure differences between the front and rear of the fan according to the P-Q curve. The benefit of this model is to significantly reduce the computational load due to avoidance of calculating the detailed flowfield around rotating blades. Figure 4 illustrates the fan P-Q curve and flow resistance characteristics for Designs A and B with the intersecting points indicating the delivered flow rates of balanced operating points in two air ventilation vests. In view of the fact that the flow resistance of Design A was much lower than that of Design B as a result of the abrupt flow turning at the channel exit, the predicted results show a relatively higher delivered flow rate of 5.3 L/s for Design A, as compared to the flow rate of 4.1 L/s for Design B.

Figure 4.

Fan pressure–flow rate (P-Q) curve (in red color) and flow resistance characteristics (in blue and green colors) for Designs A and B with the intersecting points indicating the corresponding balanced operations.

It is crucial to validate the accuracy of the computed thermo-fluid field before attaining proper simulations of the heat transfer phenomenon. Figure 5 shows a comparison of the predicted steady axial velocities at the centers of channel exits with the measured data in balanced operations for Designs A and B. Each data point denotes the averaged value of at least five measured records with the error bars signified by ±3σ _sd , where σ _sd is the standard deviation. The calculated axial velocities were compared with the measurements of different channels. The results clearly indicated that the exit velocities for Design A were greater than those for Design B with the corresponding ranges of 3.5–3.8 and 1.6–2.1 m/s, respectively. The layout of Design A with higher airflow velocities traversing internal channels was expected to achieve a better cooling performance. The maximum discrepancy between the predictions and the experimental results was within 3.24%, suggesting that the simulation software can predict the air flowfield with reasonable accuracy.

Figure 5.

Comparison of the predicted steady axial velocities at the centers of channel exits with the experimental data in balanced operations for Designs A and B (in blue and green colors).

In designing an air ventilation vest for maintaining human comfort, it is important to understand the airflow structure inside the vest. Figures 6 and 7 illustrate the predicted velocity vector plots, velocity magnitude and temperature iso-contours in a cross-sectional view at 2.5 mm from the inner channel surface for two different outlet designs. As shown in Table 1 listing the detailed simulation test matrix, the airflow was supplied using the axial fan at an inlet temperature of 33℃ with inlet velocities of 1.86 and 1.43 m/s (i.e. Case Nos. 1 and 2) for Designs A and B, respectively. Serially, the flow impacted on the opposite surface and then expanded outwards with a reduction of velocity speed from roughly 2.0 m/s near the inflow port to 0.7 m/s approaching the fork-shaped channels connected to the joint chamber. When the fluid passed through the junction and entered the central five channels (corresponding to the human back), the airflow was obviously accelerated up to 4.46/3.85 m/s for Design A/B because of the sudden contraction of the sectional area. The frictional effect tended to slow down the flow along the channels. Alternatively, gentle reduction of the sectional area could continuously increase the flow speed for the outer two channels (on the human front). It was also observed that some weak rolling eddies or stagnant fields appeared around the concavities of those fork regions. In general, the internal channels of Design A with streamlined exits had a relatively lower resistance to attain smoother flow discharge and thereby showed much higher airflow velocities than those of Design B.

Figure 6.

Predicted velocity vector plots and magnitude contours for (a) Design A and (b) Design B.

Figure 7.

Predicted temperature contours for (a) Design A and (b) Design B.

Table 1.

Test matrix for simulation conditions

Case	Channel outlet design	Inlet velocity (m/s)	Inlet temperature (cm)
1 ^a	A	1.86	33
2	B	1.43	33
3	A	1.86	30
4 ^a	A	1.86	33
5	A	1.86	36
6	A	0.93	33
7 ^a	A	1.86	33
8	A	3.72	33

Cases corresponding to the baseline condition.

To calculate the temperature distributions of Designs A and B, the temperatures of the vest inlet and outer channel surface were 33℃, corresponding to the environmental condition during the summertime as the baseline case. Meanwhile, the inner channel surface temperature was 37.5℃ toward the human body side. Essentially, substantial temperature variations were observed in the airflow course across the junctions. In response to the heat transfer effect, which occurred over the inner surface of the ventilation vest, the air largely remained at the vest inlet temperature of 33℃ inside the joint chamber, and steadily increased in temperature to 33.7℃ around the junctions and 34.8℃ near the exits of internal channels, respectively. In the case of Design B, the stagnation of flow could even produce temperatures up to 36℃ at the corners of channel outlet ends. In this investigation, the predicted temperature field was used to calculate the convective heat transfer coefficient distribution between the inner surface of the vest and airflow for evaluation of the cooling performance. Figure 8 illustrates the predicted convective heat transfer coefficient contours over the inner surface of the vest for (a) Design A and (b) Design B. The convective heat transfer coefficient h_conv is computed by q_s/(T_s−T_a_f), whereas the signs q_s, T_s and T_af denote the heat flux of the inner surface, the inner surface temperature and the averaged fluid temperature of the vest, respectively. It was observed that the impingement of inflow on the inner surface tended to produce high convective heat transfer coefficients in the core area of the joint chamber, while the airflow across the junction into internal channels was accelerated to augment the convective heat transfer effect over the surface. As indicated by the flowfield study, the internal channels of Design A could reach greater airflow velocities as compared to those of Design B for enhancing the heat transfer coefficient attributable to a steeper temperature gradient at the inner surface. The simulated results clearly indicated higher convective heat transfer coefficients of Design A (ranging from 17.4 to 49.5 W/m²K in the joint chamber and 20.85 to 26.51 W/m²K in the central three channels) than those of Design B (ranging from 12.2 to 39.1 W/m²K in the joint chamber and 15.4 to 19.3 W/m²K in the central three channels), suggesting a better air-cooling performance for Design A. Consequently, Design A was chosen as the baseline case with numerical simulations conducted to investigate the cooling effect of the vest at different inlet temperatures and velocities.

Figure 8.

Predicted convective heat transfer coefficient contours for (a) Design A and (b) Design B.

Numerical calculations were extended to examine the influences of the inlet temperature and velocity on the cooling performance in terms of the heat transfer coefficient distributions over the inner channel surface of the vest under pre-specified simulation conditions. Figure 9 shows the predictions of (a) temperature contours at 2.5 mm from the inner surface and (b) convective heat transfer coefficient distributions along the centerline of the inner surface of channel b at different inlet temperatures. In calculations, the temperatures of the inlet and outer channel surface were set to be 30, 33 and 36℃ (i.e. Case Nos. 3, 4 and 5, respectively) as per the variations of environmental temperature. The air stream flowed into the joint chamber with its temperature in the core area remaining at the inlet value. Subsequently, the human body tended to heat the fluid via the inner surface of the vest with the air temperature progressively increased as the flow moved downstream. Accordingly, the peak air temperature could be usually located near the exits of internal channels. The calculated results indicated that the temperatures at the channel outlets were 32.6, 34.8 and 36.6℃ for Case Nos. 3−5. For the thermal transport behavior, the convective heat transfer coefficient tended to be high in the impinged region of the joint chamber. In turn, as the airflow was away from the impingement location, the heat transfer effect rapidly declined from around 49.5 to 22.5 W/m²K due to reduction of flow velocity in the stagnation region near the junction. Essentially, the simulations revealed the minor effect of the inlet temperature on the convective heat transfer coefficient distributions along the centerline of the inner surface of channel b with the maximum discrepancy less than 4.9% at the axial distance of 0.237 m.

Figure 9.

Predicted (a) temperature contours and (b) convective heat transfer coefficient distributions along the centerline of the inner surface of channel b at different inlet temperatures.

The operational parameter of inlet velocity can appreciably affect the cooling outcome. Figure 10 shows the predicted (a) velocity vector plots and magnitude, (b) temperature and (c) convective heat transfer coefficient contours at different inlet velocities. In this investigation, simulations were conducted by presetting the inlet velocities of 0.93, 1.86 and 3.72 m/s (i.e. the flow rates of 2.7, 5.3 and 10.6 L/s at the associated Reynolds numbers of 3820, 7640 and 15,280) corresponding to Case Nos. 6, 7 and 8, respectively. In general, the airflow speed was relatively low in the joint chamber with the blue color marking the velocity of 1.5 m/s, as compared to the velocity magnitude contours that appeared in internal channels. The predictions indicated that the channel velocities ranged around 2.2−2.6 m/s, 3.5−3.8 m/s and 7.9−8.4 m/s in response to the drop of inlet velocity. At the same inlet temperature of 33℃, the computed outlet temperatures of 35.2, 34.9 and 34.7℃ for Case Nos. 6−8, respectively, indicated that a lower airflow speed essentially developed a longer flow residence time, and thereby resulted in a relatively higher temperature at the channel outlet. To investigate the cooling process, the simulated results indicated that the convective heat transfer coefficients were up to 17.7, 24.3 and 45.2 W/m²K in internal channels for the inlet velocities at 0.93, 1.86 and 3.72 m/s, respectively. This was attributable to the fact that a greater airflow velocity could generate a steeper temperature gradient at the inner surface of the vest, causing the enhanced heat transfer outcome. Generally, the thermal powers from the inner surface to the airflow through convection, defined as $(\sum \overset{\cdot}{m} C_{p} T_{af, outlet} - \overset{\cdot}{m} C_{p} T_{af, inlet})$ , were 7.1, 12.8 and 22.0 W with respect to the increasing inlet velocity from 0.93 m/s to 3.72 m/s.

Figure 10.

Predicted (a) velocity vector plots and magnitude, (b) temperature and (c) convective heat transfer coefficient contours at different inlet velocities.

Figure 11 illustrates a comparison of the predictions of the convective heat transfer coefficient distributions along the centerline of the inner surface of channel b with the calculations from the Kim^38,39 and the Gnielinski⁴⁰ equations at inlet velocities of 0.93, 1.86 and 3.72 m/s. The thermal phenomena were mainly characterized by the impingement heat transfer in the core of the joint chamber and the convective heat transfer in the channel (with the entrance located at the axial distance of ∼0.3 m). In the former case, the convective heat transfer coefficients were essentially lower than those of the Kim equation, which is an extensively used correlation in investigating the heat transfer performance of impinged flows. The differences between the numerical simulations and empirical correlations were 10.4% and 2.5% at the origin for the inlet velocities of 3.72 and 0.93 m/s, respectively. Alternatively, the Gnielinski equation, a common correlation for computing the heat transfer of turbulent flows in tubes, can reasonably estimate the heat transfer coefficients in the channel for the inlet velocities of 0.93−3.72 m/s.

Figure 11.

Comparison of the predicted convective heat transfer coefficient distributions along the centerline of the inner surface of channel b with the calculations from Kim^32,33 and Gnielinski³⁴ equations at inlet velocities of 0.93, 1.86 and 3.72 m/s.

Conclusion

Numerical simulations and experimental measurements were performed to study the airflow distribution and heat transfer in an air ventilation vest. To examine the effects of design and operation parameters on the cooling performance, we also illustrate the convective heat transfer coefficient distributions over the inner channel surface of the vest at different channel outlet designs, inlet temperatures and velocities. The major results are summarized as follows.

The comparison of calculated axial velocities with measured data at the exits of different channels indicated the associated maximum difference of less than 3.24%, suggesting that the simulation software predicted the flowfield with a reasonable accuracy. Owing to a lower flow resistance, the exit velocities for Design A (with the outlets arranged along the flow direction) were higher than those for Design B (using a 90^o-turn at the exits). A greater airflow velocity for the former resulted in the higher convective heat transfer coefficients over the inner surface of the vest, and thereby achieved a better cooling performance.

By varying inlet temperature and velocity within the ranges of 30–36℃ and 0.93–3.72 m/s, the predictions revealed that the inlet temperature effect on the convective heat transfer coefficient distributions was insignificant. Alternatively, the heat transfer coefficient was greatly enhanced from 17.7 to 45.2 W/m²K in internal channels with respect to an increase of the inlet velocities from 0.93 to 3.72 m/s. The powers from the inner surface transferred to the airflow via convection were 7.1, 12.8 and 22.0 W at the inlet velocities of 0.93, 1.86 and 3.72 m/s, respectively.

The correlation of the Kim equation for impinged flow overestimated the convective heat transfer coefficient up to 10.4% in the core region of the joint chamber for the inlet velocities of 3.72 m/s. The correlation of Gnielinski slightly underestimated the heat transfer coefficients and provided a conservative prediction for the airflow passing through internal channels over the tested cases with the inlet velocities of 0.93−3.72 m/s.

Footnotes

Funding

This works was supported by the Taiwan Textile Research Institute and National Science Council, Taiwan, ROC (Contract No. NSC101-2221-E-027-148 and NSC101-2627-E-027-001-MY3).

Nomenclature

References

Howes MJ and Nixon CA. Development of procedures for safe working in hot conditions. In: Ramani RV (eds) proceedings of the 6th international mine ventilation congress, Pittsburgh, SME, 1997, pp.191–197.

Hancock

Vasmatzidis

. Effects of heat stress on cognitive performance: the current state of knowledge. Int J Hyperthermia 2003; 19: 355–372.

McMorris

Swain

Smith

. Heat stress, plasma concentrations of adrenaline, noradrenaline, 5-hydroxytryptamine and cortisol, mood state and cognitive performance. Int J Psychophysiol 2006; 61: 204–215.

Nybo

. Exercise and heat stress: cerebral challenges and consequences. Prog Brain Res 2007; 162: 29–43.

Cheung

. Neuromuscular response to exercise heat stress. Med Sport Sci 2008; 53: 39–60.

Hunter

St Clair Gibson

Mbambo

. The effects of heat stress on neuromuscular activity during endurance exercise. Pflugers Arch 2002; 444: 738–743.

Razmjou

. Mental workload in heat: toward a framework for analyses of stress states. Aviat Space Environ Med 1996; 67: 530–538.

Rissanen

Jousela

Jeong

. Heat stress and bulkiness of chemical protective clothing impair performance of medical personnel in basic lifesaving tasks. Ergonomics 2008; 51: 1011–1022.

Bernard

Ashley

Trentacosta

. Critical heat stress evaluation of clothing ensembles with different levels of porosity. Ergonomics 2010; 53: 1048–1058.

10.

Havenith

den Hartog

Martini

. Heat stress in chemical protective clothing: porosity and vapour resistance. Ergonomics 2011; 54: 497–507.

11.

Zalba

Marin

Cabeza

. Review on thermal energy storage with phase change: materials heat transfer analysis and applications. Appl Therm Eng 2003; 23: 251–283.

12.

Bishop

Nunneley

Constable

. Comparisons of air and liquid personal cooling for intermittent heavy work in moderate temperature. Am Ind Hyg Assoc J 1991; 52: 393–397.

13.

Flouris

Cheung

. Design and control optimization of microclimate liquid cooling systems underneath protective clothing. Biomed Eng 2006; 34: 359–372.

14.

Hunter

Hopkins

Casa

. Warming up with an ice vest: core body temperature before and after cross-country racing. J Athletic Training 2006; 41: 371–374.

15.

Bendkowska W, Bogdan A, Kopias K, et al. Thermal manikin evaluation of microclimate cooling vests containing PCMs. In: ITC and DC: book of proceedings of the 4^th international textile, clothing and design conference - magic world of textiles, 2008.

16.

Perez

Tooker

Nunez

. Theoretical study of a personal cooling garment that utilizes natural evaporation. ASHRAE Trans 1994; 100: 49–53.

17.

Rothmaier

. Design and performance of personal cooling garments based on three-layer laminates. Med Biol Eng Comput 2008; 46: 825–832.

18.

Lee

Hwang

. Subminiature cool pad applying sorption cooling effect. HVAC and R Res 2006; 12: 797–806.

19.

Grzyll LR and Balderson WC. Development of a man-portable microclimate absorption cooling device. In: proceedings of the 32nd intersociety energy conversion engineering conference, 1997, pp.1646--1651.

20.

Cheung

McLellan

. Influence of hydration status and fluid replacement on heat tolerance while wearing NBC protective clothing. Eur J Appl Physiol Occup Physiol 1997; 77: 139–148.

21.

Muza

Pimental

Cosimini

. Effectiveness of an air cooled vest using selected air temperature, humidity and air flow rate combinations. U.S. Army Res Inst of Env Med, T22-87, 1987.

22.

Chen YT, Bomalaski SH and Constable SH. Physiological efficacy of a lightweight ambient cooling unit for various applications. Air Force Materiel Command Brooks Air Force Base, 1995.

23.

Weber S. Air cooling garment for medical personnel. Patent US005970519A, USA, 1999.

24.

Uglene WV. Enhanced personal cooling garment. Patent US 006134714A, USA, 2000.

25.

McCarter K and Hayden JR. Cooling garment for use with a bullet proof vest. Patent US 20050246826A1, USA, 2005.

26.

Horn ST. Support structure for forced air cooling vest. Patent US 20070199124A1, USA, 2007.

27.

Farnworth B and Dacey P. Gas distribution garment having a spacer element. Patent US 007016940B2, USA, 2010.

28.

Liu

. Numerical simulation of coupled heat and mass transfer in hygroscopic porous materials considering the influence of atmospheric pressure. Numer Heat Transfer Part B 2004; 45: 249–262.

29.

Ghaddar

Ghali

Harathani

. Ventilation rates of micro-climate air annulus of the clothing-skin system under periodic motion. Int J Heat Mass Transfer 2005; 48: 3151–3166.

30.

. A computational analysis for effects of fibre hygroscopicity on heat and moisture transfer in textiles with PCM microcapsules. Model Simul Mater Sci Eng 2007; 15: 223–235.

31.

Voelker C, Hoffmann S, Kornadt O, et al. Heat and moisture transfer through clothing. In: eleventh international IBPSA conference, 2009, pp.1360--1366.

32.

Barauskas

Abraitiene

. A model for numerical simulation of heat and water vapor exchange in multilayer textile packages with three-dimensional spacer fabric ventilation layer. Textil Res J 2010; 12: 1195–1215.

33.

Barauskas

Baltusnikaite

Abraitiene

. Experimental investigations and finite element model of heat and moisture transfer in multilayer textile packages. Fibres Textil East Eur 2012; 95: 112–118.

34.

ANSYS/FLUENT, version 13 User guide manual, ANSYS Inc., Canonsburg, PA, 2010 (Website: www.ansys.com).

35.

Launder

Spalding

. The numerical computation of turbulent flow. Comput Methods Appl Mech Eng 1974; 3: 269–289.

36.

Van Doormaal

Raithby

. Enhancements of the SIMPLE method for predicting incompressible fluid flows. Numer Heat Transfer 1984; 7: 147–163.

37.

Jang

Jetli

Acharya

. Comparison of the PISO, SIMPLER, and SIMPLEC algorithms for the treatment of the pressure-velocity coupling in steady flow problems. Numer Heat Transfer 1986; 10: 209–228.

38.

Kim

Aihara

. A numerical study of heat transfer due to an axisymmetric laminar impinging jet of supercritical carbon dioxide. Inst Fluid Sci 1992; 35: 2515–2526.

39.

Whelan

Robinson

. Nozzle geometry effects in liquid jet array impingement. Appl Therm Eng 2009; 29: 2211–2221.

40.

Gnielinski

. New equations for heat and mass transfer in turbulent pipe and channel flow. Int Chem Eng 1976; 16: 359–368.