Numerical study on high induction air diffusers for improved indoor environmental quality in vehicles

. Abstract. The amount of time spent in traffic by vehicle occupants has increased dramatically over the last two decades. This is because the average commute length and time spent stuck in traffic without a way out have increased at rates far exceeding population growth. At the same time, the quality of life has increased, in many areas of this world, leading vehicle users to prefer increased thermal comfort conditions inside the cabins when they are in traffic, a better thermal comfort in the vehicle being nowadays an important parameter when choosing a new vehicle. A solution to improve the thermal comfort of vehicle users is to uniformize the airflow inside vehicle by enhancing the mixing of the freshly introduced air with the ambient air. Based on literature previous research, air mixing can be improved by passive means using innovative air diffusers which have the ability of entraining more air than regular air diffusers. A comparative numerical study between different air diffusers will be carried out in the present paper with the help of Ansys Fluent software. The results revealed that a particular air diffuser, based on the lobed shaped cross orifice was found to entertain with 35% more air than a regular air diffuser.


Introduction
Even though it has been studied extensively in the last decades, it is still challenging to describe thermal comfort properly, and it is nearly impossible to come up with a single, all-encompassing definition. This is not because the thermal comfort it is difficult to be understood but because it has very powerful components of human subjectivity.
In fact, to be more precise, it's about the inseparable existence of the two human states which are the physical and the mental state that make it impossible to pinpoint the precise definition of the thermal comfort.
Thermal comfort of a human being is governed by objective and subjective factors. The objective factors are environmental climatic parameters, time of exposure, activity level, clothing, age, gender, health. Subjective factors can be either individual factors like personal sensation from the heat/cold skin receptor, thermal preference, state of emotions, thermal history, etc. or social factors like climate background, companionship, cultural aspects, etc. [1].
Given all this stated above, the most frequently used definition for the thermal comfort is still very general but it's the best way to describe it: the state of a person who would convey a sense of wellbeing in terms of the thermal conditions in an inhabited place.
Most of the studies concerning thermal comfort were oriented mostly to the indoor environment described by buildings. Also, the majority of the scientific research focuses on the physiological aspect, which holds that any person can achieve a state of thermal comfort as long as the excess heat produced by their bodies is dissipated into the surrounding air and their body's thermoregulation system plays a minimal role in the process.
Given all these aspects is almost impossible to have an indoor environment to satisfy all the person inside [2]. Therefore, an environment is considered thermally comfortable when 80-90% of the occupants express no thermal complaints [3][4][5].
In contrast to the buildings indoor environment, however, vehicles have extremely strong thermal transient conditions because of the high air velocities from the HVAC (Heating Ventilations and Air Conditioning) system and a high thermal gradient between the fresh air and the air that exists in the vehicle cabin [6,7]. This creates a significant challenge when studying the thermal comfort for the occupants of the vehicle cabin.
Other differences between the thermal environment in vehicles compared to what can be found in buildings are that the volume of the vehicle cabin is much smaller, when taking into account the resulted space per persons traveling in an automobile when comparing with the people inside the room, the report of glazing area of vehicles per unglazed area is much larger, the vehicle occupants are situated in the proximity of surfaces directly heated form the sun which are reflecting and emitting higher radiant heat fluxes toward the vehicle users. Also, usually the conditioned air is oriented toward the vehicle occupants with higher velocities and turbulence than in a building air conditioning case.
This together with the fact that the amount of time that people are spending in traffic has increased dramatically over the last several decades, and this, along with the recent rise in living standards, has increased people's interest in looking for vehicles more comfortable to ride in [8].
Because of this, experts and researchers in the HVAC area are continually searching for new ways to improve the comfort conditions inside vehicles at the same time without increasing energy consumption of the automobiles, especially in the new context of electrically powered vehicles, where each energy saving is directly translated in the increase of the vehicles range [9][10][11][12][13][14].
There are several techniques to increase the thermal comfort inside confined spaces like automobiles, whether it's the hot or the cold season. For instance, in the summer, it's proved that using the HVAC system is the most crucial approach to cope with the thermal stress, but in certain vehicles, additional systems, such as ventilation seats or even chilled seats, enhance the vehicle user perception of thermal comfort [15][16][17].
In the case of vehicles ambient, taking into account that the ventilated volume it's much smaller than for the case of a regular room and the density of persons per square meter it's mush higher in a vehicle, and at the same time if we think about the fresh airflow rate provided by the HVAC system, it's very simple to draw the conclusion that the air changes per hour are much higher for vehicles. This leads to fact that the airflow rate in a vehicle is a consequence of the thermal load and not directly related with the indoor air quality in the vehicle.
Knowing that the air changes per hour ratio from the vehicle HVAC system is at least 7 times higher than for a regular room, we can see that a better mixing of the fresh air with the ambient air can be a solution in terms of improving the thermal comfort of the vehicle users, considering that the temperature of the fresh air issued by the HVAC system is summer can drop below 10℃ [18], which is enhancing the sensation of draught for the vehicle passengers.
The strategy pursued in this study is to enhance the passive mixing of fresh air with the vehicle ambient air, in a way that it's temperature and velocity to have lower values, these two parameters being decisive parameters for thermal comfort and draught rate sensation.
For this, a comparative numerical study between different air diffusers was performed with the purpose of finding a better passive mixing device than a regular one used in the automotive industry.
The present study it's a small part of a larger study in which the main goal is to develop higher induction air diffusers for automotive industry.

Analysed geometries
The numerical studies performed in the present article were carried out in a framework of a research project between Technical University of Civil Engineering Bucharest and Renault Technologie Roumanie (Renault Group Romania) and were oriented toward Renault Duster vehicle. The dashboard of the studied vehicle is presented in Fig. 1.

Fig. 1 Standard Renault Duster Dashboard
A larger number of innovative geometries for the high induction air diffuser were studied in the research project but in the present article only four are compared, one of them being the reference one, presented in the Renault Duster dashboard.
An innovative fundamental geometry that can entrain more air through passive control was developed by a work emerged at University of La Rochelle, France [19,20], and continued later on at Technical University of Civil Engineering of Bucharest [21,22], and after a decade at the University of Rennes [18]. This approach on the geometry is consisting of various application of the lobed geometry, an example being presented in Fig.  2.

Fig. 2 Example of a lobed geometry
Given the fact that the Renault Duster have round air diffusers (Fig. 3a), the lobed geometry was adapted to fit this round shape (Fig. 3b,c,d).
The analysed geometries were created in SolidWorks software and then imported for the numerical simulations in Ansys 2021 R2.
In Fig. 3a is presented the reference air diffuser located in the Renault Duster (Case A) and in Fig.  3b,c,d, are presented 3 different innovative air diffusers. Fig. 3b (Case B )is presenting a three concentric circles with lobes with guide blades at 0˚, with lobes on the exterior of the air diffuser, Fig. 3c (Case C) is presenting three concentric circles with lobes with guide blades at 0˚, with some turbulence generators on the exterior of the air diffuser, and Case D (Fig. 3d) is presenting three concentric circles with lobes with guide blades at different angles having lobes on the exterior of the air diffuser. Given the fact that the geometry it has a symmetry along the longitudinal direction of the flow, only half of the domain was considered for the numerical simulation. After the generation of the geometries in SolidWorks software, there were imported in Ansys Workbench using Ansys SpaceClaim software.
The entire interest domain can be seen in the following figure. a b The domain part from the top of the air diffuser to outlet has 800mm and a diameter of 1000mm. The duct has a length of 400mm and a diameter of 65mm and the air diffuser it's placed at the end of the duct delivering air into the larger area.

Numerical grid
The numerical grid was generated in two steps. First, a tetrahedral mesh was generated in Ansys Meshing and second, the tetrahedral mesh was transformed into a polyhedral mesh directly using the Ansys Fluent options. From our previous experience [23], we know that using this two-step mesh strategy we can obtain very good quality numerical grids in a limited amount of time.
This is because the polyhedral numerical grid it's bringing a low numerical diffusion in the same way as a hexahedral mesh, and it has an automatic generation in the same way as tetrahedral numerical grids. The major advantage of the polyhedral grids compared with the tetrahedral ones is that are less sensitive to stretching so it leads to better mesh quality, an effect that is leading to an improved numerical stability of the studied case.
A grid independency test was performed on one of the air diffusers -Case B (Fig. 3b) and the resulted numerical grids had the following sizes: 13, 27, 50, 77, 109 million tetrahedral elements. After the numerical results comparison, the 77 million elements mesh settings were used for the rest of the numerical simulations.
For all the numerical grids it was performed a refining of the meshes downstream of the air diffuser using a truncated cone shape and setting up the element size of 0.7mm.
All the air diffuser surfaces were configured for the element size at 0.2mm. The 77 million tetrahedral elements mesh is presented in the Fig. 5.  For all the walls, a boundary layer consisting of 8 layers with a growth rate of 1.1 was configured. The numerical simulation performed with this numerical grid revealed a maximum y+ of 1.09.
All the other meshes were generated using the same configuration and controls like the one presented previously.
Having in mind that the level of details for the studied air diffusers is different, the mesh size was different as well, so for example, the reference Renault Duster air diffuser -Case A (Fig. 3a) mesh consisted of 7 million polyhedral cells (obtained from a 23.5 million tetrahedral cells).
The mesh for the air diffuser presented in Fig. 3c (Case C) has 23 million polyhedral cells (obtained from a 117 million tetrahedral cells) and the mesh for the air diffuser presented in Fig. 3d (Case D) has 23 million polyhedral cells (obtained from a 107 million tetrahedral cells).

Boundary conditions and setup of the numerical case
For the inlet, the chosen boundary condition was the mass flow inlet with a value of 0.010609 kg/s, which corresponds to an airflow rate of 31.18 m 3 /h for the studied air diffuser.
This corresponds to the second speed step (maximum is 4) of the HVAC fan. Renault Duster has 5 air diffusers in the dashboard, and the position of the chose air diffuser is central left. The determined airflow rate through this diffuser is 19% from the total air flow rate of the HVAC system through all the air diffusers in the dashboard.
The plane that split the entire geometry in two halves was imposed as symmetry.
The outlet was imposed as pressure outlet and all the other surfaces were configured as walls.
SST k-ω viscous model, provided by Ansys Fluent was employed for the numerical simulations for these cases since it has the capacity to calculate both near the wall flows and the flows in the mixing zone.
The SST k-ω turbulence model is a two-equation eddy-viscosity model from the RANS family (Reynolds Averaged Navier Stokes) well suited for simulating flow in the viscous sub-layer for predicting flow behaviour in regions away from the wall.
From the previous experience we know that SST kω turbulence model it is a good choice when dealing with these kind of flows [24]. The SST k-ω model stands out in accurately capturing flow separation phenomena, a common occurrence in simulations of low-speed airflow. It offers increased reliability by precisely modelling the transition from laminar to turbulent flows, resulting in improved predictions of separation points. Also, this turbulence model exhibits versatility and adaptability, making it well-suited for simulating a wide spectrum of flow regimes, including laminar, transitional, and turbulent flows. This flexibility makes it an optimal selection for accurately capturing the diverse flow phenomena encountered in simulations of low-speed airflow [24].

Velocity field
The numerical simulation results were postprocessed using both Ansys Fluent and Tecplot software.
The velocity contours in the longitudinal plane for the studied air diffusers can be seen in Fig. 7. a b c d Studying the longitudinal velocity fields, we can see that qualitatively the results are similar, with the observation that the lowest length of the potential core is for the original air diffuser (Case A). Also, we can see that the expansion of the jet for the case A is starting later that for the case B-D.
From the transversal velocity field, we can see that the shape of the jet at 100 mm is quite different among the air diffusers. At 600 mm, we can see that cases B and D have a higher spread that cases A and C, which is preferred because the purpose is to find a higher induction air diffuser.

Entrainment rate
Entrainment rate was obtained using Tecplot software and calculating the air flow rate at 600 mm from the dashboard and is presented in Fig.10.
It can be seen that the air diffuser from the Case D is have a higher induction rate at 600 mm from the dashboard that the other air diffusers.
Considering the Case A as a baseline, it can be seen that the entrainment rate for the Case B is 24% higher, for the case C is 28% higher and for the case D is 35% higher.

Conclusions
The thermal comfort of the vehicle driver is very important mainly because an uncomfortable medium will contribute in a higher manner to build the stress for the driver in a much quicker way, and this can be very dangerous for all the vehicle occupants because higher stress is related with distracted attention and longer reaction times for the driver that can lead to risky situations while driving.
Nevertheless, the thermal comfort inside vehicle is very important for the occupants because a comfortable experience while traveling is something that the majority of people will prefer.
An increase in thermal comfort inside the vehicle can be achieved by better mixing the cool air circulated by the HVAC system in the summer with the air already inside the vehicle thus increasing the air flow entrainment rate. This will lead to a lower temperature difference of the airflow issued from the HVAC system for the vehicle occupants without affecting the needed thermal load.
Numerical results for four air diffuser geometries were compared with each other, one of them being the air diffuser fitted to the Renault Duster. The most capable air diffuser in terms of ambient air entrainment was found to be case D, which had 35% higher air induction than the reference air diffuser.
The present study it's a small part of a larger study in which the main goal is to develop higher induction air diffusers for automotive industry. The research research will continue with the 3D printing of the most capable air diffusers from the point of view of their entrainment and mounting them in a Renault Duster dashboard. After this stage, the airflow will be experimentally measured using modern non-intrusive techniques for determining the 2D and 3D velocity fields, as well as the direct evaluation of the thermal comfort for various human subjects with the help of questionnaires.