CFD INVESTIGATION OF CEILING SHAPE ON AIRFLOW DISTRIBUTION FOR A GENERIC 2-D ROOM MODEL WITH AND WITHOUT PASSIVE CONTROL

International Journal of Mechanical Engineering and Technology (IJMET), ISSN 0976 –
6340(Print), ISSN 0976 – 6359(Online) Volume 5, Issue 1, January (2014), © IAEME
10
CFD INVESTIGATION OF CEILING SHAPE ON AIRFLOW
DISTRIBUTION FOR A GENERIC 2-D ROOM MODEL WITH AND
WITHOUT PASSIVE CONTROL
N.S.Venkatesh Kumar1
, Prof. K. Hema Chandra Reddy2
1
(Principal, Govt. Polytechnic, Pillaripattu, Nagari, (Chittoor Dist.), A.P., India and Research
Scholar, Department of Mechanical Engg, JNTUA, Anantapuram, A.P., India)
2
(Registrar, JNTUA, Anantapuram, A.P., India)
ABSTRACT
Natural ventilation, which provides occupants with good indoor air quality and a high level of
thermal comfort with reduced energy costs, has been drawing utmost importance in sustainable
strategy in building designs. Air flow distribution in a 2-D room with cross ventilation under
different ceiling shape is investigated in this present paper using computational simulations carried
with ANSYS-CFX, commercial CFD software. Investigation of air flow distribution is important to
understand convective flow distribution and to control the flow to achieve its effective distribution
for human comfort condition. The flow inside a building may be characterized by separation,
reattachment, recirculation zone and circulation bubble. Within one room model, airflow patterns are
evaluated with and without the presence of an obstacle. The obstacle creates vortex break down
thereby creating the necessary high and low air speed zones necessary for comfort conditions when
comparison is made for a similar situation without obstacle. The ceiling shape designs investigated in
this present paper are Gable single, Gable-flat, Gutter single and Gutter-flat roof structures with and
without passive control of air. The roof shape effect on air flow distribution is captured and
represented by velocity vectors, streamlines and velocity contours and it is observed that location and
the size of the recirculated air variation is sensitive to changes in the building's roof shape and the
placement of the obstacle.
Key words: Ceiling Shape, Air Flow Distribution, Passive Control and CFD.
INTERNATIONAL JOURNAL OF MECHANICAL ENGINEERING
AND TECHNOLOGY (IJMET)
ISSN 0976 – 6340 (Print)
ISSN 0976 – 6359 (Online)
Volume 5, Issue 1, January (2014), pp. 10-25
© IAEME: www.iaeme.com/ijmet.asp
Journal Impact Factor (2013): 5.7731 (Calculated by GISI)
www.jifactor.com
IJMET
© I A E M E

11
1. INTRODUCTION
The room geometry plays an important role to achieve natural air flow distribution for human
comfort conditions. Good survey about building design criteria, features and practices are given by
Facr [1]. The flow inside a building may be characterized by separation, reattachment, recirculation
zone and circulation bubble. These features can be investigated using simple geometry, two-
dimensional flow that makes a very popular numerical research subject. Several factors influence
airflow pattern inside a building. These include the configuration, location and structure of air inlets
and outlets, the condition of the incoming air and size and shape of building. A recirculation system
can be used to distribute air inside the building. Different arrangements of air inlets and recirculation
air systems would result in different airflow patterns Dr. Ahmed A. Salman et. al. [2] . Bjeg, et. al.
[3] showed that the aspect ratios of the room determined whether the airflow can be two- or three-
dimensional. Iannone [3] showed that the building form and the details of the air outlet and inlet
systems can remarkably affect the ventilation performance. It is possible to start from a simplified
geometry (rectangular room) and modify its form by changing roof slope. Aynsley [4] concluded that
despite the powerful tools in the form of zonal network flow analysis, CFD software and LES
methods used on large naturally ventilated buildings, there is a need for development of simpler
natural ventilation computation method for its use in the preliminary stage of building design of
building shape or orientation. In this case, unresolved issues in computation of natural ventilation for
thermal comfort were discussed.
This paper discusses the airflow pattern, velocity distribution and one of vortex control
systems that might be used inside a room for passive control of air. There are various systems for
changing velocity vectors and reducing vortex strength. As an air stream approaches a solid fence, it
is either forced up and over the obstacle or around it. In buildings with very high ceilings, it is very
difficult to maintain the air velocity near floor level. So, baffles can be installed across the building
width to reconcentrate the airflow along the floor level and to provide personnel comfort.
2. CFD SIMULATIONS
Computational settings and parameters for the reference case are outlined and accordingly the
results for this case are presented. Later on, these settings and parameters will be systematically
modified for the parametric study.
2.1. Computational domain and grid
The dimensions of the generic isolate 2-D building and computational domain were chosen
from [2]. The geometry of the room consists of: two vertical walls, horizontal floor and flat, sloping
or pitched roofs. The buildings had dimensions W x H = 2.4 x 1m2
and depth of the building is 0.25
m to carry simulations in CFX. The computational grid is made of structured mesh with Hexa grid.
The geometry and mesh is generated using ICEM CFD, meshing software.

12
(a) single Gable room model (b) single Gutter room model
(c) single Gutter-Flat room model
(d) single Gable-Flat room model
(e) single Gable room model with verticle obstacle
Figure 1: Structured mesh for different ceiling shape geometries

13
2.2. Boundary conditions
A uniform velocity of 1.725m/s is given at inlet vent as inlet boundary condition. Outlet
boundary condition is applied at bottom vent and it is assumed that average static pressure is zero to
get normal outlet velocity at outlet. For the floor, vertical wall and roof, the no-slip wall functions
are chosen with zero roughness height. The computational domain with boundary condition and
velocity profile at inlet is as shown below.
Figure 2: Applied Boundary Conditions
2.3. Solver settings
The 3D steady RANS equations were solved in combination with the k-ϵ model. Advection
scheme adapted for solving momentum governing equations is High resolution and for turbulence
equations first order scheme. The convergence criteria are taken as 10-6
.
3. RESULTS AND DISCUSSION
Computational investigation of air flow distribution inside a single and double-zone building
of different ceiling shape are presented. Four ceiling shapes are computationally modeled. The
ceiling geometry specifications are summarized in Table 1. The parameter considered for the
computational investigation is roof slope or roof arc length and the inflow and outflow conditions
remains same for all the cases. Room ventilation should provide wind in desired direction and no
recirculation or dead air zone in occupied regions. To achieve such layouts in practice, obstacles of
different height degrees are imposed. For single gable room model, airflow patterns are evaluated
with and without the presence of an obstacle. The obstacle was placed either on the floor or hanged
from the roof at the plane of symmetry in the middle of the room model. The details of vortex
control geometrical arrangement are shown in Table 2.

14
Table 1: Specifications of the case studies
Case
No.
Roof type Room
width [m]
Maximum
height [m]
Roof slope or 1/4 arc angle
[degrees]
1 Single gable roof
2.4
1.105 5
1.32154 15
1.69282 30
2 Single gutter roof 2.4 Same as above
3 Gable - Flat roof 4.8 Same as above
4 Gutter - Flat roof 4.8 Same as above
Table 2: Arrangements for the placement and height of the vertical obstacle
Arrangement No. Obstacle location Obstacle height [m]
1 No obstacle 0.00
2 One obstacle mounted on the floor in the
plane of symmetry
0.43
3 One obstacle mounted on the floor in the
plane of symmetry
0.86
4 Two obstacles mounted on the floor and
hanged from the roof in the plane of
symmetry
0.43 for each one
5 One obstacle hanged from the roof in the
plane of symmetry
0.86
3. 1. Effect of Different Ceiling slope on Single Gable and Gutter Room Models
Computational results presented in this section are for single gable room and single gutter
room with the change of roof slope or arc angle. The air flow distribution and velocity profiles are
presented in the form of vectors, contours and charts. The plots on the right column show contours.
Flow velocity vectors are given in the plot on the left column. In these figures, streamlines that form
closed loops indicate zones of recirculation air or vortices. The fact that these recirculation zones
have areas with high and low flow velocities is also evident in the velocity vector plots. One of these
dominated local vortices occurs in the corner below the inlet edge when the inflow of air is at an
angle of 90 degrees to the windward vertical wall irrespective of the roof shape. As the roof slope
increases, there is a decrease in the recirculation zone that appeared inside the room as shown in
Figures 3-c and 4-c. Results showed that for a gutter with proper height, the air supply flows between
the two circulation zones with opposite directions and spreads along the floor before it is drawn to
the exhaust side and removed. Figure 5a-c shows the normalized horizontal velocity profiles adjacent
to the roof, on the floor and on the plane of symmetry in the middle of the gable (on the left column)
and gutter (on the right column) room models. The velocity profiles adjacent to the roof (Figure 5-a)
show the sensitivity to the roof shape. Increasing the roof slope angle to 30 degrees shows the same
velocity profile for both shapes. Comparison of floor velocity profiles between gable and gutter roof
shape at different pitch angle is given in Figure 5-b. At roof slope equal 5 and 15 degrees, the

15
velocity profiles are almost identical. As the roof angle increased, the reattachment length of the
gable model is smaller than that of the gutter model. Figure 5-c shows that, in the plane of symmetry,
no considerable difference between the different shapes as the pitch angle or roof arc length
increases are observed except at the angle 30 degrees.
(a) Roof slope: 5 degrees
(b) Roof slope: 10 degrees
(c) Roof slope: 15 degrees
Figure 3: influence of roof slope angle on air flow distribution in single Gable room model

16
(a) Arc angle: 20 degrees
(b) Arc angle: 60 degrees
(c) Arc angle: 120 degrees
Figure 4: influence of arc roof angle on air flow distribution in single Gutter room model

17
(a) comparison of roof velocity profiles
(b) comparison of floor velocity profiles
(c) comparison of velocity profiles at the middle vertical plane of the room, x=1.2h
Figure 5: horizontal velocity profile, U/Uin
3. 2. Effect of roof slope or arc angle on Single Gable-flat and Gutter-flat Room Models
The results represented here in form of velocity vector and contour in Figures 6a-c and 7a-c
shows the airflow patterns for three cases of gable-flat and gutter-flat room models. For all models,
one can observe the development of different vortex circulation zones. The flow is attached to the
roof at small roof angles or arc lengths and detaching from the wall nearer to the exit forming large

18
vortex region just below the inlet vent and there is a low velocity circulation region at the floor as it
is observed in Figures 6-a and 7-a. As the roof angle increases, the flow detachment from the top
wall occurred much before exit, it is attached to the floor and there is a recirculation or dead zone
region just below the roof. The three recirculation zones as observed in room are shown in Fig. 6-c
and 7-c. In this type of ceiling shape, most of the action occurred in the gable or the gutter regions,
while the airflow pattern is nearly uniform through the flat region. Further investigation showed that
the velocity vector length increased at the connecting joint between gable or gutter and flat roof
surface. Moreover, subdivision of the flow into near-wall jet and a relatively low-speed circulation
flow is mostly distinct. For both gable and gutter flow models, the within-jet velocity keeps high
values up to the exit from the under gutter roof region and then becomes almost uniform till the exit.
The side-by-side comparison of the velocity profiles at different planes for gable-flat and
gutter-flat room models of three different ceiling shapes are shown in Figure 8a-c. In these figures,
the velocity profiles adjacent to the ceiling, on the floor and at a distance equal 1.2 h from the inlet
vent for these three different arrangements are presented. According to Figure 8-a, the normalized
velocity profiles decrease at constant rate to the exit and there is a slightly increase in profile at the
middle of gable and gutter. As the roof angle or arc length increases, there is a sharp decrease in
velocity ratio until the roof vertex angle of the gable zone and then increase in its value up to
connection of gable and flat roof, and then decrease until towards the exit from the room. Similar
behavior is observed for gutter-flat room model. Figure 8-b shows that at roof slope equal 5 degrees
and 10 degrees, the velocity profiles are almost identical. For both gable-flat and gutter-flat models,
the velocity profiles show no significant effect in the flat ceiling zone. Figure 8-c shows that the
typical shape of velocity profiles along the plane of symmetry is maintained and reflects the size of
the recirculation zones.
(a) Roof slope: 5 degrees

19
(b) Roof slope: 10 degrees
(c) Roof slope: 15 degrees
Figure 6: influence of roof slope on air flow distribution in single Gable-Flat room model

20
(a) Arc angle: 20 degrees
(b) Arc angle: 60 degrees

21
(c) Arc angle: 120 degrees
Figure 7: influence of arc roof angle on air flow distribution in single Gutter-Flat room model
(a) comparison of roof velocity profiles
(b) comparison of floor velocity profiles

22
(c) Comparison of velocity profiles at the middle vertical plane of the room, x=1.2h
Figure 8: horizontal velocity profile, U/Uin
3. 3. Effect of Vertical Obstacle (Passive Control)
Moving air has momentum and it will keep on moving in a particular direction until it is
turned by an obstacle as considered in a passive control system. Attention therefore must be given to
the obstacle location and height, to create the needed high and low air speed zones. Five different
arrangements for the placement and height of the vertical obstacle are given in Table 2. Figures 9a-e
presents the velocity vectors inside the single gable room model with roof slope equal 15o
(left side)
and the velocity contours (right side) adjacent to the roof and on the floor of that model. Figure 10a-c
shows velocity profiles, at the middle plane of the room, for five different arrangements. Due to the
deflection of air through the room, the airflow patterns for the last four arrangements were quite
different from the one obtained without obstacle (arrangement one). Figure 9-b and 9-c show that as
the obstacle is getting higher and higher, the airflow pattern and the original big vortex gets split into
two vortices on the right and left of the obstacle. The velocity vectors and velocity profiles show that
the zone at the left side of the vertical obstacle has different rotational strength than that of the zone
at the right side. The flow patterns for the fifth arrangement are similar to the arrangement of the
other four. However, the centers of the rotary zones are different. When the obstacle hanged from the
roof, as shown in Figure 9-d and 9-e, the big circulation zone near the roof is broken down into small
vortices. Also, the flow patterns form a tunnel inside the room model. However, the center of the
rotary zones was different. Figure 10 shows the horizontal velocity profiles ratio of different obstacle
control arrangements. It can be seen that when air stream strikes the obstacle, the direction and
magnitude of the original flow velocity are altered and cause changes in the airflow patterns. In this
case, the airflow patterns are forced to diverge and pass around the obstacle edge. Behind the
obstacle, the airflow is unable to come together immediately because of the inertia of the air and a
wake is left where they are separated from the obstacle.
In summary, significant differences in air distribution were found with changes in the
obstacle location and height. Within vortex size and location, remarkable differences were found
between obstacles versus without obstacle airflow. Moreover, installing obstacles inside the building
can be used to create the proper circulation of air through the building and to redirect and concentrate
the air flow along the required zone to provide human comfort.

23
(a) Arrangement 1
(b) Arrangement 2
(c) Arrangement 3
(d) Arrangement 4
(e) Arrangement 5
Figure 9: influence of passive control on air flow distribution in single Gutter room model

24
(a) roof (b) bottom (c) mid plane
Figure 10: comparison of velocity profiles in single Gutter room model with passive control
4. CONCLUDING REMARKS
The objective of present paper is to computationally investigate the influence of the ceiling
shape design on the flow pattern and velocity distribution. The computational results presented here
gave the detailed insight about the airflow inside 2-D building model configurations. The following
conclusions are drawn based on the results presented:
• The ceiling shape alters the flow inside the room.
• Air dead zone circulation regions are replaced with low speed recirculation zone with
increase in roof angle.
• The passive control in the form of vertical obstacle resulted in better control of air flow when
compared with the room without passive control.
• The increase in roof angle or arc length created a vertex near the roof acting as an insulator
for the heat flow from outside. Vertical obstacles hung from the roof also provided same
results.
The most noticeable aspect is the development of a second area of recirculation closer to roof
and a third area of recirculation on the floor in addition to the under vent vortex.
The predicted velocity profiles showed that
• At small roof-slope angle, the velocity profiles adjacent to the roof are similar.
• The velocity profile after the gable or gutter in combined ceiling shape with flat roof showed
the same air flow pattern.
• The double-gable and double-gutter ceiling room models had significant effect on velocity
profile.
The factors that affect the airflow pattern through the building are the degree of roof slope
and the height and location of obstacle. The effect of roof shape may be used to ensure that air
movement occurs where it is required, for instance near the floor in the occupied zone of the room.
The correct roof shape will ensure that air movement inside the room is enhanced.

25
5. REFERENCES
[1] Facr, A., 1990, “Cooling building by natural ventilation,” MIL-HDBK-1011/2, AMSC N/A.
[2] Dr. Ahmed A. Salman, “Numerical prediction of airflow patterns in a 2-d room model under
different ceiling shape with and without passive control”, Proceedings of Seventh
International Congress On Fluid Dynamics and Propulsion December 18-20, 2001, Cairo,
Egypt.
[3] Bjeg, B., Svidt, K., Morsing, S., and Zhang, G., "Optimization of the design of two test
rooms by means of CFD,” Dep. of Agricultural Engineering Danish Institute of Agricultural
Sciences, Research Center, Bygholm, Dk-8700 Horsens, Denmark.
[4] Iannone, F. “Natural ventilation and sustainability: Designing with computational fluid
dynamics,” Via Orabona, 4-70125 Bari, Italy, E-mail: f.iannone@libero.it.
[5] Tarun Singh Tanwar, Dharmendra Hariyani and Manish Dadhich, “Flow Simulation (CFD)
& Static Structural Analysis (FEA) of a Radial Turbine”, International Journal of Mechanical
Engineering & Technology (IJMET), Volume 3, Issue 3, 2012, pp. 252 - 269, ISSN Print:
0976 – 6340, ISSN Online: 0976 – 6359.
[6] P.S. Jeyalaxmi and Dr.G.Kalivarathan, “CFD Analysis of Flow Characteristics in a Gas
Turbine- A Viable Approach to Predict the Turbulence”, International Journal of Mechanical
0976 – 6340, ISSN Online: 0976 – 6359.
[7] M.Z.I.Sajid, Dr. K. Hema Chandra Reddy and Dr. E.L. Nagesh, “Design of Vertical Axis
Wind Turbine for Harnessing Optimum Power”, International Journal of Mechanical
0976 – 6340, ISSN Online: 0976 – 6359.
[8] Nitin Kardekar, Dr. V K Bhojwani and Dr. Sane N K, “Numerical Analysis of Air Flow
Velocity Streamlines of Air Curtains”, International Journal of Mechanical Engineering &
Technology (IJMET), Volume 4, Issue 5, 2013, pp. 150 - 155, ISSN Print: 0976 – 6340,
ISSN Online: 0976 – 6359.
[9] N.S. Venkatesh Kumar and Prof. K. Hema Chandra Reddy, “CFD Analysis of Wind Driven
Natural Cross Ventilation for a Generic Isolated Building”, International Journal of
Mechanical Engineering & Technology (IJMET), Volume 4, Issue 5, 2013, pp. 200 - 207,
ISSN Print: 0976 – 6340, ISSN Online: 0976 – 6359.

CFD INVESTIGATION OF CEILING SHAPE ON AIRFLOW DISTRIBUTION FOR A GENERIC 2-D ROOM MODEL WITH AND WITHOUT PASSIVE CONTROL

More Related Content

What's hot (20)

Similar to CFD INVESTIGATION OF CEILING SHAPE ON AIRFLOW DISTRIBUTION FOR A GENERIC 2-D ROOM MODEL WITH AND WITHOUT PASSIVE CONTROL (20)

More from IAEME Publication (20)

Recently uploaded (20)

CFD INVESTIGATION OF CEILING SHAPE ON AIRFLOW DISTRIBUTION FOR A GENERIC 2-D ROOM MODEL WITH AND WITHOUT PASSIVE CONTROL