Air Movement and Airflow Patterns of Attachment Ventilation

Abstract

This chapter presents the principle of attachment ventilation (attached ventilation) and airflow patterns. Attachment ventilation is based on the extended Coanda effect, which refers to the phenomenon that the jet flows along the wall surface and attaches to the adjacent walls, maintains continuous flow until it reaches the floor, and continues to adhere to the floor for extending flow to produce air reservoir or air lake. The airflow patterns of vertical wall attachment and column attachment ventilation are fully illustrated and compared by 2D-PIV and trajectory experiments.

The goal of ventilation is to control air movement as required in a room or given space to provide a comfortable and healthy indoor air environment. The indoor airflow is invisible, and the visualization technique makes the airflow pattern visible and observable macroscopically. The airflow pattern visualization by 2D-PIV and trajectory experiments depicts the fundamental interaction between the forced jet delivered from diffusers and the heat convection. This chapter provides the airflow pattern of vertical wall attachment and column attachment ventilation, and clarifies the significant factors affecting attachment ventilation performance.

2.1 Coanda Effect and Ventilation

2.1.1 Coanda Effect

The Coanda effect simply refers to the tendency of a fluid jet to stay attached to an adjacent solid surface. The principle of the Coanda effect is that when fluid 1 flows into another fluid 2 (or environmental fluid) from an orifice with a certain initial velocity, fluid entrainment will be formed in the surrounding environment, as shown in Fig. 2.1a. The jet will deflect towards the side with greater flow resistance when the entrainment effect of the jet on ambient air is unbalanced (Fig. 2.1b, c) (Coanda 1936). The airflow direction of a jet can be altered by appropriately adjusting the boundary conditions of the jet. If the boundary conditions of the flow near the wall are continuously changed, the jet can be formed into a streamlined flow in any desired direction theoretically (Coanda 1964, 1938; Panitz and Wasan 1972).

Fig. 2.1

Coanda effect of jet flow. a Free jet, b wall-attached jet, c curved surface-attached jet

There is an inertial movement of fluid flowing out of an orifice; the entrainment effect of surrounding fluid will lead to a low-pressure region near the wall surface. Consequently, a wall-attached flow is formed. The process of the Coanda effect driving the jet toward the wall can be further shown in Fig. 2.2. The jet is initially flowing in a straight line from an orifice, gradually driving the surrounding air with decreasing velocity, as illustrated in Fig. 2.2a. With the further extension of the jet, the pressure distribution between the upper and lower sides of the jet is different. The jet’s pressure in the lower side is equal to the atmosphere pressure, attributed to the jet being filled with sufficient surrounding air to maintain the atmospheric pressure.

Fig. 2.2

Formation process of the Coanda effect. a Jet discharge, b jet deflection, c wall attachment

On the contrary, the supplementary airflow near the wall can only enter the low-pressure zone through the gap between the upper side of the jet and the wall. The jet deflects to the wall due to the pressure difference between the upper and lower sides, as shown in Fig. 2.2b. As the jet continues to deflect upward, the replenishment channel of airflow becomes narrower, and the pressure of the jet’s upper side is further decreased so that the jet finally adheres to the wall surface. Eventually, a jet that is fully attached to the wall surface is formed and pushed forward, as depicted in Fig. 2.2c.

Some studies have shown that if there is no Coanda effect, almost all the jet throw of supply air will be reduced by about 30% (ASHRAE 2008). The following factors have a dominant influence on the Coanda effect (Awbi 2003, 2011; Wen 1982; Zhu 1984):

• Normal distance between the slot and the attached wall surface S;

• Width of the slot inlet b, also called slot outlet (ASHRAE 2017);

• Slot jet velocity u;

• Wall obstacles and other influencing factors;

• Temperature difference between jet and ambient air △t.

In point of fact, the air jet can be affected by the Coanda effect when the distance between the supply air jet and the wall surface is less than the maximum attachment distance S_max (see Sect. 2.2), and a confined wall-attached jet is formed finally. The airflow pattern is closely related to the location of the air jet inlet (see Fig. 2.3). When the air inlet is close to a wall surface, it is more likely to form a wall-attached jet. The axis direction of the jet is inclined to the wall surface. With the jet throw extending, the entrained airflow rate increases, and the jet velocity decays, resulting in the attenuation of the Coanda effect (ASHRAE 2008).

Fig. 2.3

Airflow structure of attachment ventilation by extended Coanda effect (ECE)

It should be noted that the buoyancy effect resulting from the temperature difference between a jet and the ambient air will directly affect the flow feature of a wall-attached jet. The flow field details will be explored in the subsequent chapters. For office rooms with a temperature difference of 8–12 °C when supplying cold air, the distance between the air inlet and wall surface should be kept within 300 mm, and 0 mm is preferred. Furthermore, the obstacles may also cause jet separation, making the Coanda effect no longer valid. When obstacles’ thickness is smaller than the attached boundary layer, the jet could remain attached to the wall. This phenomenon is related to jet velocity, obstacle height, obstacle position, etc. (ASHRAE 2017; JGJ/T 177-2009 2009), which are discussed in detail in Chap. 4, Sect. 4.8.

2.1.2 Extended Coanda Effect and Air Movement

The extended Coanda effect (ECE) (Li 2019) refers to the phenomenon that the jet flows along the wall surface, forming the Coanda effect and maintains continuous flow until it impinges on the floor, and continues to adhere to the floor for extending flow. The difference between the ECE and the Coanda effect lies in the presence of an impinging region. The principle of the ECE is as follows: when the jet exits from the adjacent solid surface, the jet will deflect and be inclined to attach to the wall surface (that is, forming the Coanda effect in the traditional sense, see Region I in Fig. 2.3). Because of the influence of inertia momentum, it moves along the original direction, reaches a separation point, and causes a stagnation phenomenon after impinging the ground. With the recovered dynamic pressure, the jet moves further forward along the floor and entrains the surrounding air above the floor. As shown in Fig. 2.3, there are two key points worth noting: jet separation point and floor reattachment point. The jet separation point refers to the location where $\partial u_y {/}\left. {\partial x} \right|_{x = 0} = 0$ and the Coanda effect fails, or it resembles the jet “perceives” the occurrence of the impinging effect, and then separates from the vertical wall. On the contrary, the floor reattachment point refers to the position where the airflow re-attaches to the floor under the ECE, and here $\partial u_x {/}\left. {\partial y} \right|_{y = 0} = 0$. The pressure of the stagnation zone between the separation point and the reattachment point (Region II) is close to the ambient pressure (the pressure distribution is related to the air jet parameters). In the downstream of Region II of the stagnation point, the dynamic pressure gradually increases and reaches a maximum value. With the recovered dynamic pressure, fluid can overcome the floor resistance and move along a horizontal floor surface, forming horizontal air reservoir region (Region III), as illustrated in Fig. 3.4.

2.2 Vertical Wall-Attached Jet and Airflow Pattern

As mentioned above, mixing ventilation has the characteristic of high supply air velocity (momentum). The air openings are often arranged in a room’s upper space, and the room’s lower spaces are saved. For mixing ventilation, the entire room is conditioned so that the ventilation efficiency or temperature efficiency is relatively low (Yin and Li 2015).

The features of displacement ventilation lie in that air is supplied at floor level with a very low velocity (0.1–0.3 m/s), and the supply air momentum is so low that it does not substantially impact the dominant thermal plume. The supply air temperature is usually 2–4 °C lower than the occupied zone temperature. The cool air is denser and sinks to the floor, and spreads throughout the room, creating a piston-like flow or air reservoir in the occupied zone. The indoor air slowly rises due to the rising thermal plume’s entrainment effect and the exhaust opening’s suction effect. Indoor pollutants (including thermal pollution) are removed from the upper exhaust openings. The exhaust air temperature is higher than the indoor occupied zone air temperature. Correspondingly, the ventilation efficiency or temperature efficiency is relatively high (Li 2000). However, wall-mounted diffusers of displacement ventilation often require significant spaces and occupied zones, which greatly limits the applications of displacement ventilation.

Attachment ventilation combines the advantages of the above two types of ventilation modes. Attachment ventilation mainly consists of the following modes (Song 2005; Li et al. 2012a, b; Yin and Li 2012):

• Vertical wall attachment ventilation (VWAV);

• Rectangular column attachment ventilation (RCAV);

• Circular column attachment ventilation (CCAV);

• Adaptive attachment ventilation (AAV).

• 1. Principle of VWAV

  Essentially, attachment ventilation is to create appropriately air environment parameters for the occupied zone by designing the air supply temperature, velocity, air supply volume, air terminal types, locations, etc. For VWAV, the supply air moves along the vertical wall to close to the ground, an “downstream air reservoir” is formed due to impinging, and the dynamic pressure is transformed into static pressure, and momentum is transformed into the impulse, which consumes a small percentage of the kinetic energy.

The theory of vertical wall-attached jets and air reservoir airflow is the basis for the design of attachment ventilation. The air distribution design of VWAV aims to determine the physical parameters of the supply air jet appropriately, to deliver the maximum amount of fresh air or conditioned air to the occupied zone, and to ensure that the air temperature and velocity in the occupied zone meet the comfort requirements (Yin and Li 2014; Zhang 2005; Zheng et al. 2019).

Figure 2.4 shows a diagram of the airflow pattern of VWAV mode. After the jet is discharged from the linear slot inlet (if l/b ≥ 10:1, jets can be regarded as two-dimensional plane jets), it keeps the momentum conservation and continuously entrains the ambient air. The entrained air near the wall is far less than that on the other side, and a low-pressure zone is formed at the upper corner A of the vertical wall (see Fig. 2.4). The pressure difference between the two sides of the supply air jet drives it to deflect toward the vertical wall and attach to the surface. The airflow continues to flow downward along the wall and extends to the floor level, the adverse pressure gradient increases, and the air supply jet will separate from the vertical wall. After impinging the floor, it extends forward along the floor to form an air reservoir. Then, the load in the control zone (occupied zone) is “taken away”, forming a kind of air temperature, velocity, and humidity fields that are similar to displacement ventilation mode (Yin and Li 2015). This air reservoir is created by the attached jet flow from the upper part of the room after impinging, which is different from the airflow mechanism of traditional displacement ventilation.

Fig. 2.4

Indoor airflow pattern of VWAV

From the perspective of jet mechanics, the VWAV can be regarded essentially as a combination of a vertical wall-attached jet and a horizontal floor-attached airflow (Wang 2009).

According to the distance between the air inlet and the wall surface, the VWAV can be divided into two modes, complete attachment (i.e., tangential attachment) and deflected attachment, as shown in Fig. 2.5. For the theoretical analysis, the slot width b, which has a more significant influence on the air supply characteristics, is selected as the characteristic size. s is the normal distance between the opening center and the attached wall surface, and S = s − b/2 is defined as the distance between the slot’s inner side and the attached wall surface, called the deflection distance. When S/b > 0, i.e., s/b > 0.5, the air supply jet deflection will occur, and the full attachment ventilation will turn into partially attached or deflected attachment ventilation. The difference between S/b and s/b is noted in Fig. 2.5.

Fig. 2.5

Complete and deflected attachment ventilations. a Complete attachment, b deflected attachment

For indoor air movement of VWAV, three regions can be distinguished (see Fig. 2.3) (Wang 2009).

Region I, Vertical attachment region

For the deflected attachment ventilation, there is S > 0. The discharge jet is a free turbulent jet within a short distance from the air inlet (Cho et al. 2008). After that, the jet deflects toward the vertical wall and attaches to its adjacent surface due to Coanda Effect. It is found that the initial region zone has a crucial influence on the confined attachment jet, which will be analyzed in detail in the subsequent chapters. For deflected attachment, there is usually a vortex flow at corner A (see Fig. 2.4). The possible reason is that the adverse pressure gradient increases gradually in the jet direction, which hinders the jet flow, and the fluid micro cluster near the wall will be more strongly blocked. When their kinetic energy is depleted, they are forced to turn back, forming a vortex flow phenomenon (Cui 2010).

The air supply jet attaches to the vertical wall under the action of the pressure difference between the two sides of the supply jet, and moves downward along the wall, realizing the effective transport of the jet along the vertical wall, and the centerline velocity of the air supply jet decays very slowly. With the decrease of s, the range of the initial region becomes smaller. When s equals b/2, the deflected attachment is converted into a complete attachment.

Region II, Jet impinging region

When the discharge jet approaches the ground along the vertical wall, the air supply jet separates from the vertical wall due to the influence of the adverse pressure gradient of the ground, impinges the ground, and turns into a horizontal flow. The jet flow direction deflects by 90° (the jet deflection angle is related to the wall structure), the jet mixes with the ambient air in a large amount, and the centerline velocity decays rapidly in this region.

Region III, Horizontal air reservoir region

A long region of major engineering importance that is often called horizontal air reservoir region, i.e., fully established turbulent floor-attached flow and its distance is depends on the air supply velocity (initial velocity), the shape and area of the slot inlet, and the dimensions of the space into which the slot inlet discharges.

In region III, the air supply jet spreads along the floor, forming a forward piston flow and presenting air reservoir distribution over the floor. It is the control zone for the air distribution of VWAV. It resembles a displacement ventilation mode in the flow pattern, meets the thermal comfort of the human body, and achieves high ventilation efficiency.

The essence of the design method of the air distribution of the attachment ventilation is how to calculate and determine the appropriate air supply parameters, to ensure the velocity field, temperature field, etc., in the air reservoir region.

• 2. Airflow pattern of VWAV

  Air movement in a room is a rather complex process. Airflow pattern visualization is helpful in deeply understanding the air movement mechanism and then effectively optimizing air distribution.

Airflow pattern visualization can be performed by many methods. It is a common way to display (e.g., visually inspect, detecting and photographing, etc.) the movement and state of the airflow through the tracer gas.

The following assumptions are made about the tracer particles in the visualization tests.

• (a) The concentration of particles in the airflow is low and will not interfere with the flow field.

• (b) The particles are spherical with a small diameter. Their flow is in the low Reynolds number region, and there is no interaction between particles. The gravity and buoyancy are ignored.

• (c) The tracer particles are completely and uniformly mixed with the airflow.

• (d) The temperature of the tracer particle is the same as that of the airflow. According to our experience, the smoke generator can be connected with a 2–5 m hose to reduce the smoke generation temperature and ensure that the temperature of the tracer particle is the same as that of the airflow.

The movement of smoke tracer particles can be described by Eq. (2.1)

$$M_{\text{P}} \frac{{{\text{d}}u_{\text{P}} }}{{{\text{d}}\tau }} = C_{\text{d}} \frac{1}{2}\rho_{\text{f}} .A_{\text{p}} \left( {u_{\text{f}} - u_{\text{p}} } \right)^2$$

(2.1)

where

M _p :

tracer particle mass, kg;

C _d :

resistance coefficient;

ρ _f :

airflow density, kg/m³;

u _p :

cross-sectional area of tracer particle, m²;

u _f :

airflow velocity, m/s;

u _p :

tracer particle velocity, m/s;

$\tau$ :

time of tracer particle moving together with the airflow.

If u_p is a constant (i.e., does not change with time), C_d is related to the Re number and is only a function of u_f − u_p. When the time τ is equal to 0, the tracer particles enter the airflow, and the initial velocity u_p is equal to 0, and then Eq. (2.1) can be simplified as Eq. (2.2).

$$u_{\text{p}} = u_{\text{f}} \left[ {1 - \exp \left( { - \frac{{18\mu_{\text{f}} }}{{\rho_{\text{p}} d_{\text{p}}^2 }}\tau } \right)} \right]$$

(2.2)

where

μ _f :

dynamic viscosity of airflow, Pa⋅s;

ρ _p :

tracer particle density, kg/m³;

d _p :

tracer particle diameter, m.

The above equation indicates that when u_f, d_p, ρ_p are determined, the tracer particle velocity u_p is a function of time. Once the tracer particles enter the flow field, u_p can immediately converge to the airflow velocity u_f (i.e., u_f = u_p).

It is a common visualization method to trace the flow field with ethylene glycol (less than 1.0 μm in diameter), which has good light reflection performance and particle tracking performance, and it is easy for CCD high-speed digital camera to photograph the experimental process (Song 2005).

The test model and geometric parameters for the ethylene glycol visual test are shown in Fig. 2.6. During the test, either a CCD high-speed digital camera or an ordinary digital video system can be used to take pictures of the airflow patterns of the air supply jet. The CCD digital camera system is labeled in the figure, which is used to record the changing air jet pattern qualitatively. The lens of the camera system is kept in a horizontal fixed state to eliminate the influence of image distortion on the test data, and the horizontal direction of the lens is perpendicular to the plane where the central axis of the jet flow is located. In fact, ordinary digital video can also better qualitatively record the airflow process and change the filming position according to actual needs (Qiu 2008; Song 2005; Zhang 2005). Figures 2.7 and 2.8 show the visualization results and airflow patterns of VWAV under different scenarios.

Fig. 2.6

Visualization test system

Fig. 2.7

Visualization of air supply flow patterns at different distances from the vertical wall (u₀ = 5.15 m/s, t₀ = 22.0 °C, t_n = 24.5 °C), a vertical baffle is set on the right side of the folding ladder to form a movable vertical wall in the visualization test. a s = 0.13 m, b s = 0.60 m, c s = 0.78 m

Fig. 2.8

Airflow patterns in rooms of different distances from the vertical wall (Re = 17,098, t₀ = 22.0 °C, t_n = 24.5 °C). a s = 0.13 m, b s = 0.60 m, c s = 0.78 m

To understand the air movement mechanism of VWAV and investigate the influence of parameters u₀ and s on the ventilation performance, a series of visualization tests were conducted on VWAV in a three-dimensional real-size test chamber (Yin 2012). Figure 2.9 shows the test chamber used, with dimensions of 5.4 m × 7.0 m × 3.16 m. A plenum chamber with a dimension of 2.5 m × 0.5 m × 0.5 m was used to realize the uniform air supply of the slot inlet, whose area is 2.0 m × 0.05 m and the installation height is 2.5 m away from the floor. A flexible connection hose between the air supply duct and plenum chamber is adopted to realize the adjustment of the vertical distance between the slot inlet and vertical wall, which can also eliminate the impact of vibration from air supply plenum chamber on air supply airflow.

Fig. 2.9

Three-dimensional real-size test chamber

The visualization results and airflow patterns of VWAV under different u₀ and s are shown in Figs. 2.10 and 2.11, respectively. The slot inlet locations s/b are equal to 2, 5, 8, and 10, respectively, and the air supply velocity is varied from 1.0 to 2.0 m/s. The airflow pattern at different slot inlet locations is analyzed below.

Fig. 2.10

Visualization of airflow patterns of VWAV. a s/b = 2, u₀ = 1.0 m/s, b s/b = 2, u₀ = 1.5 m/s, c s/b = 2, u₀ = 2.0 m/s, d s/b = 5, u₀ = 1.0 m/s, e s/b = 5, u₀ = 1.5 m/s, f s/b = 5, u₀ = 2.0 m/s, g s/b = 8, u₀ = 1.0 m/s, h s/b = 8, u₀ = 1.5 m/s, i s/b = 8, u₀ = 2.0 m/s, j s/b = 10, u₀ = 1.0 m/s, k s/b = 10, u₀ = 1.5 m/s, l s/b = 10, u₀ = 2.0 m/s

Fig. 2.11

Airflow patterns of VWAV. a s/b = 2, Re = 3320, b s/b = 2, Re = 4980, c s/b = 2, Re = 6640, d s/b = 5, Re = 3320, e s/b = 5, Re = 4980, f s/b = 5, Re = 6640, g s/b = 8, Re = 3320, h s/b = 8, Re = 4980, i s/b = 8, Re = 6640, j s/b = 10, Re = 3320, k s/b = 10, Re = 4980, l s/b = 10, Re = 6640

• (a) When s/b is equal to 2, the air supply velocity changes from 1.0 to 2.0 m/s, and the supply air can form an effective attachment with the wall surface. It can be clearly seen from the figure that when the tracer gas approaches the ground along the vertical wall, the direction turns to the horizontal diffusion flow process after impinging (see Figs. 2.10 and 2.11a–c). In other words, the 0.07 m gap between the slot inlet and the vertical wall does not influence the attachment effect, almost forming a complete attachment ventilation air movement.

• (b) When s/b is equal to 5, the air supply inlet is further away from the attached wall surface. Under the cases of different air supply velocities, after the jet is discharged from the slot inlet, the pressure difference between the two sides of the supply jet drives it to deflect toward the vertical wall gradually. Influenced by the adverse pressure gradient, a vortex is generated at the upper left corner of the room (corresponding to Fig. 2.4). However, once the attachment is formed, the airflow movement is similar to cases of s/b = 2 (Figs. 2.10 and 2.11d–f).

• (c) When s/b is equal to 8, the air supply inlet further deviates from the attached wall surface, and the included angle between the inlet jet and the wall increases significantly. Compared to the case of s/b = 5, in addition to the increase in the included angle of the air supply jet, the distance of the upper deflection zone is also increased, approaching 1/3 of the room height (see Figs. 2.10 and 2.11g–i). It is noted that there is a small amount of air diffusing along the direction of the room height, which will cause a decrease in the effectiveness of attachment ventilation in the occupied zone.

• (d) When the air inlet is far away from the adjacent wall, that is, s/b is equal to 10, it is difficult to form an attachment airflow, see Figs. 2.10 and 2.11j–l, there is an apparent gap between the air supply jet and the vertical wall.

In summary, the slot inlet location s has a crucial influence on the air distribution of VWAV. With the slot location gradually away from the wall, the attachment effect gradually weakened, and the mixing of air supply jet with indoor air intensified. When s is small, the air supply jet deflects toward the wall, almost forming a complete attachment jet, and the flow field distribution in the floor region can be obtained similarly to that of displacement ventilation. With the s/b increase to 10, the air supply jet almost leaves the vertical wall and turns to mixing ventilation, and the VWAV mode fails (Qiu 2008). The test results show that within the design velocity range of the ventilation and air conditioning system, the maximum attachment distance should be maintained within 0.25–0.40 m (i.e., s/b is approximately equal to 5 ~ 8).

When s is determined, increasing air supply velocity is conducive to strengthening the attachment effect and reducing the mixing of the air supply jet with the ambient air. With a further increase of air supply velocity, the jet thickness of the vertical wall and horizontal air reservoir becomes thinner. Thus, it is essential to select the appropriate air supply velocity to ensure the effectiveness of attachment ventilation.

• 3. 2D-PIV flow field test

  Another technique for airflow visualization is particle image velocimetry (PIV), which has become a well-established technique for velocity field measurement in fluid mechanics, power engineering, and other fields (Cao et al. 2014; Wang 2010). It is a non-invasive (indirect) measurement technology for flow field profiles with high spatial resolution, clear flow field observation, and the ability to obtain quantitative flow field velocity from its images (Hosni and Jones 2002).

The 2D-PIV technique can be used to measure the displacement of the tracer particles in a short time interval through the flow trace of the tracer particles, to obtain the transient velocity distribution of the flow field. The methods for testing the flow field are mainly divided into two categories. One is to expose the particle on the flow field cross-section of the sheet light source twice or more times to establish the PIV flow field picture, and use Young's stripe method or related algorithm to read the PIV picture point by point to obtain the flow field on the cross-section of the sheet light source. The other is to use a high-speed CCD camera to directly input the flow field image on the cross-section of the sheet light source to the computer for image processing, and use related software to obtain the flow field velocity.

The essence of the 2D-PIV velocity field test is to obtain the ratio of displacement to time. Let the position of the tracer particle at time t₁ be (x₁, y₁), the position at time t₂ be (x₂, y₂), and the time interval is Δt. The velocity calculation is shown in Fig. 2.12.

Fig. 2.12

2D-PIV velocity field measurement

The velocity component of the tracer particle is shown in Eq. (2.3a, 2.3b).

$$v_x = \mathop {\lim }\limits_{t_2 \to t_1 } \frac{x_2 - x_1 }{{t_2 - t_1 }} = \frac{{{\text{d}}x}}{{{\text{d}}t}}$$

(2.3a)

$$v_y = \mathop {\lim }\limits_{t_2 \to t_1 } \frac{y_2 - y_1 }{{t_2 - t_1 }} = \frac{{{\text{d}}y}}{{{\text{d}}t}}$$

(2.3b)

2D-PIV technology can measure 3500 to 14400 instantaneous velocity vector points in a cross-section with an error range of only 0.1–1% (Fan 2002). For ventilated flow fields, it is generally recommended to use smoke or oil mist particles (Zhao 2004) as tracer particles. In particular, the time difference between the two laser pulses plays a vital role in the PIV tests, and setting a correct value is the key to the success of the PIV tests. Our research shows that the time difference between the two laser pulses (△t, μs) and the maximum velocity (u_max, m/s) in the photographed area vary inversely, and there is the linear logarithm function, that is, ln(△t) = 5.52 − ln(u_max), i.e., △t·u_max = 250. The flow field in the near-wall region can be photographed using a cylindrical lens to obtain a more realistic and clear velocity vector. For the experimental study of the flow field in the near-wall region, the above method can be referred to Qin et al. (2009).

The experimental results of using 2D-PIV to study VWAV (ambient air temperature 24 °C) with different influencing factors (u₀, s) are presented (Qiu 2008). A high-velocity CCD camera is used to capture the flow field images to obtain the velocity field. The dimension of the PIV test chamber is 600 mm × 300 mm × 340 mm, the slot inlet size is 300 mm × 10 mm, and the distance of the slot inlet from the attached wall is adjustable, as shown in Fig. 2.13 (the shooting area is x from 10 to 250 mm, and y from 12 to 330 mm). During the test, a special fixed low turbulence fan is used to supply air to the chamber. To ensure the uniformity and stability of the air supply jet, a plenum chamber and a rectification section need to be set in front of the slot inlet. The tracer particle smoke generator and tracer particle are set in the front section of the fan, to ensure the uniformity of tracer particle concentration and improve the accuracy of the PIV test.

Fig. 2.13

Diagram of 2D-PIV test chamber with attachment ventilation

The results of 2D-PIV tests are given below to analyze the relationship between the air supply velocity u₀, s, and the flow field of the VWAV (Qiu and Li 2010).

• 1. The effect of air supply velocity (u₀)

  Observe changes of the centerline velocity of the attached ventilation by changing the air supply velocity (u₀). See Table 2.1 for the test conditions.

  Table 2.1 Test conditions for different u₀

The flow field distribution of the attachment ventilation is shown in Fig. 2.14. For different u₀, the jet’s centerline velocity decreases with the flow path’s increase. When u₀ gradually increases from 0.3 to 1.5 m/s, the centerline velocity gradually increases, and the effective attachment distance along the vertical wall is extended accordingly.

Fig. 2.14

Velocity field vector of attachment ventilation with different u₀ (s/b = 5, k = 0, t₀ = 24 °C). a u₀ = 0.3 m/s, b u₀ = 1.0 m/s, c u₀ = 1.5 m/s

• 2. The influence of the distance between the air supply inlet and the vertical wall

  Experimental results have shown that the distance between the air supply inlet and the adjacent vertical wall (s), the installation height (h), has a crucial influence on the centerline velocity of the air supply jet (Li 1993).

Table 2.2 shows the test conditions for the variation of s. The 2D-PIV velocity field test results are shown in Fig. 2.15. As s increases from 50 to 100 mm (i.e., s/b = 5–10), the vortex region formed between the jet inlet and the attached vertical wall (see the upper left corner in Fig. 2.15) gradually expanded downwards, hence the location of the attachment point accordingly moved downward. It can be clearly seen that when s/b exceeds a certain threshold value (i.e., s/b = 5–8), the jet attachment phenomenon disappears and finally turns to the ceiling air supply mode (mixing ventilation).

Table 2.2 Test conditions for different s

Fig. 2.15

Velocity field vector of attachment ventilation with different s (u₀ = 0.5 m/s, k = 0, t₀ = 24 °C). a s/b = 5, b s/b = 8, c s/b = 10

In practical ventilation engineering applications, in some cases, there may be a gap between the slot inlet and the attached wall. What will happen to the attached jet? The airflow visualization presents air movement patterns. After leaving the slot inlet for a certain distance, the discharge jet gradually tends to attach to the adjacent wall surface (or column surface) due to the Coanda effect, forming the deflected attachment, see Fig. 2.11. The effect of air inlet location (the distance from the slot inlet to the attached wall) on room air movement is significant. This section analyzes the features of the deflected attached jets (S/b > 0). Here, S is the distance in the direction normal to the vertical wall from a slot’s inner edge or surface adjacent to the wall.

In the jet mainstream section, both the deflected and the complete attachment jet have the same cross-sectional velocity distribution and similar velocity profiles. However, in the initial region, the significant difference between deflected and fully attached jets exists for 0 < S/b ≤ S_max, where S_max is the maximum attachment distance (see initial region in Fig. 2.16b). The location of the attachment point on the attached wall surface will be altered for different distances S. Particularly, as the distance S gradually increases, the airflow in the upper corner shows a larger vortex and further extended entrainment range. When S/b > S_max, the Coanda effect is no longer valid, and the air jet fails to stick to the wall surface, which finally turns out to become the traditional mixing ventilation. According to the experimental results, for ordinary office rooms, the S_max/b is in the range of 4.5–6.5. For attachment ventilation design, the critical value of S_max/b = 4.5 is recommended. In addition, a larger air supply velocity enables the jet to better withstand the adverse pressure gradient, resulting in the attachment point moving upstream (see top-left region, Fig. 2.14).

Fig. 2.16

Schematic diagram of wall-attached jets. a Complete attached jet (S/b = 0), b deflected attached jet (S/b > 0)

2.3 Column Attachment Ventilation Airflow Pattern

• 1. Analysis of CAV

  The rectangular column attachment ventilation (RCAV) mode belongs to a kind of attachment ventilation (Liu 2016; Liu et al. 2017; Yin et al. 2015; Li et al. 2012a, b; Yin et al. 2016a, b, c), where the air movement along the column is similar to VWAV, but the flow field is not exactly the same because of the column’s arris effect. Figure 2.17 gives a schematic diagram of the air distribution of RCAV. After the air supply jet is discharged from the ambulatory-shaped slot inlet installed on the upper part of the rectangular column, it creates a column attachment ventilation similar to VWAV. When the air supply jet moves downwards along the column approaching the ground, then the jet separates from the rectangular column surface and turns to a horizontal airflow (see Fig. 2.17). Different from VWAV, due to the arris effect of the column, while entering the occupied zone, the air supply jets from two adjacent surfaces of the rectangular column can result in the airflow intersection and superposition, further consuming the fluid momentum. As a result, the airflow velocity in the horizontal air reservoir region formed by CAV decays faster than that of WAV (Yin et al. 2016c).

  Fig. 2.17

  

  Air movements of RCAV. a Principle of RCAV, b air distribution in horizontal air reservoir region

CAV is particularly suitable for places with existing columns, such as subway stations, airport terminals, exhibition centers, shopping malls, supermarkets, etc. Making full use of various columns in those buildings can realize the CAV mode (Yin et al. 2016a, b, c).

Four regions can be distinguished for a indoor air movement of RCAV (see Fig. 2.17).

Region I, vertical attachment region

The air supply jet is discharged from a slot inlet with “I”, “L”, or “ ” shapes, and then attached to the column surface and delivered to the control zone. The centerline velocity of jet flow is kept equal to u₀, nearly 10 times s/b. The inlet velocity distribution is one of the major factors in determining the airflow pattern.

The supply air jet vortex motion, affected by the velocity difference between the internal and external fluids, induces a secondary velocity field in the ambient fluid, which draws the non-turbulent ambient fluid into the jet flow. As the vertical jet moves downstream along the column, it is observed from the airflow visualization that local and temporary ejections of low-momentum fluid from the wall outward, inrush movements towards the wall accompanied by sweep movements almost parallel to the column, and alternating with the generation locally of unstable instantaneous velocity-distributions, resulting in turbulence fluctuations, see Fig. 2.20. However, statistically speaking, both in time and in space, there is still a similarity of the time-averaged velocity distributions. Consequently, the centerline and the cross-section velocity can be expressed.

Region II, jet impinging region

When the momentum of the vertical jet flow is not sufficient to overcome the inverse pressure gradient, the jet flow will separate from the column surface and turn into the horizontal flow after impinging on the floor. There exists a sharp change in the turbulence intensity of the jet flow at a horizontal distance of 0.5–1.0 m away from the column surface, and more information can be found in Sect. 2.2.

Region III, horizontal air reservoir region

After impinging the floor, the jet is converted to horizontal flow; it spreads along the floor and carries out more heat exchange in the occupied zone. The air volume and jet thickness increase continuously with the airflow moving forward. This region is the main target zone for the indoor airflow control of RCAV.

Region IV, air reservoir intersection region

As shown in Fig. 2.17, the arris effect of the rectangular column is significant in this intersection region. The momentum of the main flow could be partially depleted by the superimposed mixing of two adjacent flows at the corner or intersection region. Compared to the VWAV, the airflow velocity in this region of RCAV decays more rapidly.

The difference between circular column attachment ventilation (CCAV) and RCAV mainly lies in the curvature effect. For the air supply from the top annular slot, there is a similarity between circular column attachment ventilation (CCAV) and RCAV in the vertical attachment region.

However, in horizontal air reservoir region, the airflow pattern of the jet is slightly different from that of RCAV, showing a 360° radially expanding airflow along the circumference over the floor (Li et al. 2010; Li et al. 2012a, b; Sun 2017). Figure 2.18 illustrates a diagram of the air distribution of CCAV mode.

Fig. 2.18

Principle of CCAV. a Configuration of the test chamber, b schematic representation of column attached-jet

The flow patterns of CCAV can also be divided into the vertical attachment region I, jet impinging region II and horizontal air reservoir region III, as shown in Fig. 2.18 (Sun 2017). Among them, region I, and II are consistent with the RCAV. It is noted that in horizontal air reservoir region III, due to the cylindrical geometric symmetric characteristics of the column, the flow rate and thickness of the jet are increased with the jet radial throw; meanwhile, the centerline velocity is decreased more rapidly.

• 2. Airflow pattern of RCAV

  The full-size visualization test of RCAV is shown in Fig. 2.19. The test room dimensions are 6.6 m × 6.6 m × 3.15 m with a rectangular column of 1.0 m × 1.0 m × 2.5 m (length × width × height). The slot inlet of 0.05 m in width is attached to the column surface, and the installment height of the slot is 2.5 m from the floor. The air supply temperature is identical to the room temperature.

  Fig. 2.19

  

  Flow field visualization test of RCAV

  Fig. 2.20

  

  Visualization of airflow patterns of RCAV with different u₀. a u₀ = 1.0 m/s, b u₀ = 1.5 m/s, c u₀ = 2 m/s

The smoke tracing visualization can be used for a better understanding of the flow process of RCAV qualitatively. The visualization results are shown in Fig. 2.20, and the corresponding airflow patterns are presented in Fig. 2.21. The visualization results demonstrate that an air reservoir flow field similar to displacement ventilation is created in the room RCAV.

Fig. 2.21

Airflow patterns of RCAV with different Re numbers. a Re = 3320, b Re = 4980, c Re = 6640

It can be seen from the vertical attachment region in Fig. 2.20 that when increasing the air velocity from 1.0 to 2.0 m/s, the thickness in the jet extension δ_m becomes thinner, indicating less mixing of the jet with the surrounding air. The three visualized airflow patterns at the floor level show that, after the jet impinges the floor, the airflow diverges at the column’s plinth, almost creating a symmetrically radial flow pattern in the horizontal air-reservoir region.

• 3. Airflow pattern of CCAV

  Similar with the principle of CCAV, the visualization test rigs are shown in Fig. 2.18. A comprehensive series of tests are conducted in a 6.6 m × 6.6 m × 3.15 m chamber. The circular column with a diameter of 1.0 m is set in the center of the room. The air is supplied through a column’s annular slot located 2.5 m above the floor, which is 0.05 m in width. In the vertical attachment region, the CCAV airflow patterns are similar to that of RCAV, as shown in Figs. 2.22 and 2.23. As u₀ increases from 1 to 2 m/s, the thickness of the air reservoir region decreases progressively, whereas the horizontal jet throw further extends accordingly.

  Fig. 2.22

  

  Visualization of airflow patterns of CCAV with different u₀. a u₀ = 1.0 m/s, b u₀ = 1.5 m/s, c u₀ = 2.0 m/s

  Fig. 2.23

  

  Airflow patterns of CCAV with different Re numbers. a Re = 3320, b Re = 4980, c Re = 6640

2.4 Comparison of the Attachment, Mixing and Displacement Ventilation

As described in Sect. 1.1, the essence of air distribution is to control indoor air temperature, velocity, humidity, and pollutant concentration, etc. Scientific design of air distribution can create an indoor environment with low energy consumption and high indoor air quality (Awbi 2003, 2008; Bauman 2003; Li 2000; Ma 1997; Yin and Li 2013). A comparison of the characteristics of different ventilation modes is shown in Table 2.3. It can be seen that attachment ventilation, a new type of air distribution, combines the advantages of mixing ventilation and displacement ventilation. It not only has the advantage of mixing ventilation mode without occupying space (Etheridge and Sandberg 1996), but also has the advantages of displacement ventilation mode with good indoor air quality and high temperature efficiency.

Table 2.3 Comparison of air distributions of VWAV, mixing ventilation, and displacement ventilation