J. agric. Engng Res.

(1997) 68, 29 – 37

Perforated Ventilation Ducts: Part 2,

Validation of an Air Distribution Model
K. El Moueddeb;1 S. Barrington;1 N. Barthakur2
Departments of 1Agricultural Engineering and 2Natural Resource Sciences, Macdonald Campus of McGill University,
21 111 Lakeshore Road, Ste Anne de Bellevue (Que´ bec) Canada, H9X 3V9

(Receiy ed 8 May 1996; accepted in rey ised form 29 April 1997)

A mathematical model based on energy and mo-

mentum conservation equations was experimentally Notation
validated with four wooden and perforated ventilation Aon outlet normal to the air jet, m2
ducts with aperture ratios of 0?5, 1?0, 1?5, and 2?0. The Ao one outlet area, m2
model facilitated validation because of the elimination A duct cross-section, m2
of the frictional losses achieved by equating expres- Cd discharge coefficient at the outlet,
sions from the conservation equations of energy and dimensionless
momentum. The model accurately predicted the air Cc contraction coefficient at the outlet,
flow distribution parameters along the full length of dimensionless
the perforated ventilation ducts operated under tur- Cy velocity coefficient at the outlet,
bulent flow conditions. The regain coefficient and the dimensionless
energy correction factor were equal to unity, and the Cr regain coefficient in the duct at the
discharge coefficient was 0?65, along the perforated outlet, dimensionless
duct. The outlet air jet discharge angle varied along FL friction losses, J kg21
the entire duct length, and was directly related to the Kei energy correction factor, dimensionless
average velocity and total head inside the duct for Kmi momentum correction factor,
turbulent flow. ÷ 1997 Silsoe Research Institute
dimensionless
Pi air static pressure, Pa
qa measured outlet air flow, m3 s21
1. Introduction Va actual average air outlet velocity, m s21
Vo potential air outlet velocity, m s21
Vi duct average air velocity, m s21
The air distribution pattern in a perforated ventila-
a discharge angle, degree
tion duct has generally been investigated by applying
r air density, kg m23
Bernoulli’s equation, the first law of thermodynamics,
… specific air volume, m3kg21
and the principle of momentum conservation. Since
r aperture ratio: Ao / A
this approach requires the same air velocity for all the
i 5 o, 1, 2
streamlines, two correction factors should be con-
Suffices
sidered. The air velocity profile over the duct cross-
o outlet
section is not constant and a velocity head correction
1 upstream from the outlet
factor is required to compute the kinetic energy and
2 downstream from the outlet
the momentum of the air moving inside the duct.
Since there is an uncertainty in establishing a precise
streamline leaving the perforated outlet, a correction by El Moueddeb et al .,3 using the fundamental equa-
factor is needed to account for the correct down- tions of thermodynamics and fluid mechanics for an
stream energy available. Bajura and Jones,1 and incompressible and inviscid fluid. The model was
Soucek and Zelnick2 have used a regain coefficient to based on Eqn (1) which related the expected pressure
account for the less than perfect axial discharge of downstream (P2) from a single or a pair of outlets to
lateral momentum at the outlet. the duct average air velocity (V ) and to the outlet air
A simplified, one-dimensional model was developed jet discharge angle (a ) and velocity (Vo), and a regain
29
0021-8634 / 97 / 090029 1 09 $25.00 / 0 / ag970177 ÷ 1997 Silsoe Research Institute
30 K . E L M O U E D D E B ; S. B A R R I N G T O N ; N. B A R T H A K U R

coefficient (Cr) of Eqn (2), defined from the energy frame of wooden members measuring 39 mm 3 39 mm
equation, where Ke denotes an energy correction and covered with 6 mm thick presswood panelling
factor. (Fig. 1 ). The perforated duct offered an inside cross-
sectional area of 0?173 m2 [(0?597 m 3 0?292 m) 2
P2 5 r [V1 ((1 / 2)(V1 1 V2)-Vo cos(a )) 1 (V 2o 2 V 22) / 2)]
4(39 mm 3 39 mm)]. Its side panels had rectangular
(1) pairs of outlets, one on each side panel, 12 in number,
and spaced at every 0?61 m. By replacing the side
Cr 5 Ke1 2 [2 / ((V1 1 V2) V1)][(V 2 Ke2 V ) / 2 2 P2 / r ]
2
o
2
2
panels of the perforated duct, the experiment was
(2) repeated, successively, for four aperture ratios
The outlet velocity, Vo, defined by the total duct air (oAo / A) of 0?5, 1, 1?5 and 2?0 (Table 1) while
energy head, does not require an energy correction keeping the perforated length constant.
factor as it represents the potential outlet energy. The experimental duct was equipped with a 0?450 m
The hypotheses upon which the model was built diameter ACME axial fan with a 0?25 kW motor
were: (a) the energy and momentum correction fac- operating at 1600 rev / min. To reduce air swirling at
tors are almost unity for conditions of turbulent flow the outlets, an air straightener, a 1?8 m long tapered
inside the perforated duct, and (b) the friction losses section and a 4?9 m non-perforated section were fitted
expressed by the momentum equation are equivalent between the fan and the 7?3 m perforated duct. The
to those expressed by the thermodynamic energy non-perforated section of 4?9 m had a length equiva-
equation where the gain in internal energy and the lent to more than ten times the hydraulic diameter of
heat transfer to the outside were mainly due to skin the perforated duct.
friction. The second hypothesis was used to combine
the kinetic energy and the momentum equations to
derive Eqn (1), which related the ventilation para-
meters upstream and downstream from the outlet 2.2 . The monitoring equipment
without a friction term. Based on the conservation of
mechanical energy over the average cross-sectional
area of the duct, the model can also demonstrate that The air static pressure was read using a vertical
the regain coefficient is almost unity, and so are the Dwyer micrometer, with an accuracy of Ú0?062 Pa.
energy and momentum correction factors. Through the top panel of the duct and to receive the
The object of this paper is to validate the above wall piezometric taps, one small hole was perforated
hypotheses of the model represented primarily by Eqn half way between each two pairs of outlets and
(1). The model validation can be accomplished by upstream from the perforated section, for a total of 13
measuring the evolution of P1, P2, the outlet discharge holes. A plastic tube 3 mm in diameter was used to
angle (a ) and velocity along the length of the duct. link the piezometric taps to the micrometer and to
This will simultaneously verify whether the regain read static air pressures against that of the atmos-
coefficient and the energy correction factors assumed phere. At each of the 13 locations, static air pressure
unit values. The outlet air jet discharge coefficient can readings were repeated three times and averaged.
then be determined using Eqn (3) of the model, According to El Moueddeb et al .,4 a three-port pitot
tube is required to measure the outlet air jet discharge
qa 5 Cd Aon Vo 5 Cd Ao Vo sin (a ) (3) angle and velocity (Fig. 2 ). Air leaves the outlets at an
angle a (discharge angle) with respect to the wooden
where Vo is the potential outlet air jet velocity and a
side of the duct and must be accurately measured in
will be measured at each outlet of an experimental
order to calculate the true axial lateral air jet momen-
duct. The friction losses along the length of the duct
tum. The three-port pitot instrument was built of
can be verified using the principles of conservation of
three pitot tubes, 3 mm inside diameter, soldered
kinetic energy and momentum.
together. The central tube was used to measure the
dynamic pressure of the outlet air jet by connecting it
to the right port of a micrometer while the left port
2. Methodology was left open to the atmosphere. Thus, the average
outlet air jet velocity was indirectly measured by
2.1. The experimental duct converting the measured dynamic pressure to velocity
using Bernoulli’s equation. The two exterior tubes
The study was limited to a wooden perforated were used to measure the air jet angle by connecting
ventilation duct fed by an axial fan and built of a to the opposite port of a second micrometer and by
 PERFORATED VENTILATION DUCTS: PART 2 31

Static Duct closed

pressure tap side

5.97 m

Velocity traverse

2.95 m

Perforated
7.3 m outlet

4.9 m

1.8 m

Fan

Fig. 1. The experimental duct

aligning the instrument with the jet angle when zero and showed a pitot correction factor of 0?992. During
pressure was recorded. With respect to the duct side the experiment, the accuracy of the air jet angle
panel, the instrument angle of rotation was indicated measurements was found to be Ú2?58 for conditions of
using a horizontal needle moving over a fixed protrac- low air swirling.
tor. The instrument was moved using a grid pattern According to El Moueddeb et al .,4 the outlet air
over the outlet surface to obtain an average value for flow could not be obtained with the grid method
the air jet discharge angle and velocity. applied at the outlet surface because the contraction
The three-port pitot instrument was calibrated in- effect was too small to be measured accurately. A
side the low speed wind tunnel of the Mechanical small contraction of 25 mm on the perimeter of an
Engineering Research Laboratory, McGill University, outlet measuring 25 mm by 100 mm cannot be de-
tected by a set of three, 3 mm pitot tubes which
Table 1 produced an error of 24%. Thus, the outlet air flow
was determined by measuring the air flow inside the
Description of the experimental duct
duct, over its cross-section and by calculating the
Outlet Aperture difference in air flow between each outlet. For this
Duct size , mm ratio , (o Ao / A) purpose, the grid method was applied across the inside
1 145 3 25 0?5
cross-section of the duct, upstream and downstream
2 145 3 50 1?0 from each pair of outlets. Air velocity was measured
3 145 3 75 1?5 using an ALNOR compuflow thermo-anemometer
4 145 3 100 2?0 with an accuracy of Ú3% at the highest indicated
Note: The perforated duct measured 7?3 m in length and reading or 0?05 ms21 over a range from 0?1 to 15 ms21.
in all cases, 12 pairs of outlets were used at a spacing of For each duct cross-section measured, ten consecutive
610 mm. readings were taken over 16 equal rectangular areas
32 K . E L M O U E D D E B ; S. B A R R I N G T O N ; N. B A R T H A K U R

to the
micrometers

Frame

Protractor

Needle

90°

Fig. 2. The three-ports pitot tube used to measure outlet air jet discharge angle and y elocity

and averaged. The air temperature was measured levels of uncertainty were measured and computed
using the ALNOR thermo-anemometer (thermo- according to ANSI / ASME5 and Kline6.
couple sensor) with an accuracy of Ú 18C.

2.4. Verification of the model

2.3. Data accuracy
The model was verified by monitoring the para-
meters in the experimental duct and by comparing the
The maximum uncertainties were 0?8% in static measured static air pressure and outlet air flow data
pressure, 3% in mean axial velocity, Ú18C in air with those predicted from the equations. The charac-
temperature, Ú2?58 in air discharge angle, 1?2% in teristic duct parameters, such as the regain coefficient,
duct cross-section, 4% in the air outlet area, and 0?1% the outlet air jet discharge coefficient, and angle were
in the atmospheric pressure at 95% confidence level. calculated. The variation in static air pressure (P ),
As the air density and the average outlet air jet upstream and downstream from the outlet was also
velocity were calculated from measured parameter calculated using Eqn (1) with measured duct average
[r 5 P / RT and Vo 5 4(2(P / r 1 V 2 / 2)], their maxi- air velocity (V ), the outlet air jet discharge angle (a ),
mum uncertainties were 4 and 5%, respectively. The and potential velocity (Vo). Vo accounted for energy
maximum corresponding uncertainty was 10?3% for friction losses occurring at the outlet level as it was
the discharge coefficient, 10% in the regain coefficient calculated from the measured duct total head using
for Re $ 105, and 10% or Ú2 Pa in the calculated Bernoulli’s equation.
static pressure. The calculated cosine of the discharge The calculated static air pressure values were com-
air angle had a maximum uncertainty of 6%. The pared with those measured using the experimental
 PERFORATED VENTILATION DUCTS: PART 2 33

duct. After validation of Eqn (1), it was used along 80

with the equation derived from the principle of energy 75
conservation to calculate the friction losses along the 70 r = 0·5
length of the duct. The outlet air jet discharge angle 65
observed over the length of the duct was compared 60
with that derived from the total energy head and

Gauge static pressure, Pa

55 r = 1·0
velocity of the air inside the duct. Thus, the effect of 50
friction losses on the discharge angle could be deter-
45
mined. The regain coefficient was calculated using
40
Eqn (2) by comparing the outlet potential velocity
35
with that averaged upstream and downstream from r = 1·5
30
the outlet. The outlet air jet contraction coefficient
was observed by considering the air inside the duct 25
that must contract to exit at an angle. Therefore, it 20
maintained an axial velocity component that reduced 15
r = 2·0
the jet area. The outlet cross-section where the 10
contraction is maximum is called the vena contracta. 5
At this point, the streamlines are parallel throughout 0
0 1 2 3 4 5 6 7
the jet and the pressure is atmospheric (Streeter and
Benjamin7). The ratio of the actual velocity (Va) to the Distance along the duct, X, m
ideal potential velocity given by the Bernoulli’s equa- Fig. 3. Static air pressure along the length of the perforated
tion, is called velocity coefficient (Cy ) which rep- duct , where X is measured from the closed end of the duct. ( r
resents the friction loss at the orifice. The actual is aperture ratio) . j measured pressure; p calculated pressure
discharge (qa), from the orifice is the product of the
actual velocity at the vena contracta (Va), and the area
of the jet. The ratio of the jet area at the vena and the calculated static air pressures confirmed the
contracta to the area of the orifice normal to the jet is assumptions that the air flowing in the duct was
the contraction coefficient (Cc), and the discharge incompressible with negligible viscous effects. Also,
coefficient Cd 5 Cy Cc. The flow at the outlets was the friction loss expressions obtained independently
calculated from the air jet velocity normal to the area from momentum and energy equations are equivalent.
of the orifice using Eqn (3). The energy and momentum correction factors are
almost unity for the turbulent flow inside the perfor-
ated duct.
3. Results and discussion The calculation of the friction losses over the length
3.1. The ey olution of pressure along the duct of the perforated duct from Eqn (4) followed the
validation of Eqn (1) (Moueddeb et al .3).
The predicted evolution of pressure along the
length of the duct compared favourably with the
measurements (Fig. 3 ) for all four aperture ratios of P1 2 P2 5 r [V 22 / 2 2 V 22 / 2] 1 r FL
0?5, 1?0, 1?5 and 2?0. When compared with that
measured, the static pressure was predicted with a 1 r [V 2o / 2 2 V 22 / 2 2 P2 / r ](dmo / dm1)] (4)
maximum uncertainty of 10% (Fig. 3 ). This is equal to
or less than the combined experimental uncertainty
associated with the measurement of the duct average The friction losses were calculated by inserting into
air velocity, outlet air jet discharge angle and air Eqn (4) the static air pressure value for P1 obtained
density. When the aperture ratio was 2?0, a deviation from Eqn (1), obtaining a value for P2 from Eqn (4)
occurred for the first three outlets, located close to the with FL 5 0 and comparing this P2 from Eqn (4) with
fan end. This could be owing to air swirling which P2 obtained from Eqn (1). Thus, FL 5 [P2 ( from Eqn
created some difficulty in accurately measuring the 1) 2 P2 ( from Eqn 4)]. The cumulative effect of static
outlet discharge angle. The number of outlets affected friction losses (Fig. 4 ) was the increased discrepancy
by air swirling increased from zero to three for the between the two static air pressures, starting from the
aperture ratios of 0?5 to 2?0, respectively. fan side where X 5 7?4 m and going towards the
Thus, the good agreement between the measured closed end where X 5 0. The combination of Eqn (1)
34 K . E L M O U E D D E B ; S. B A R R I N G T O N ; N. B A R T H A K U R

80 Also, Vo can be calculated from the total air energy

75 head inside the duct and Eqn (5) can be used to
r = 0·5
70 determine the discharge angle.
65
60
Gauge static pressure, Pa

55 r = 1·0 3.3 . The regain coefficient

50
45
The regain coefficient was calculated from Eqn (2)
40 using the velocities measured inside the duct. The
35 coefficient was also calculated as defined from the
r = 1·5
30 momentum equation of Haerter12
25
20 Cr 5 [2Km1 2 [Vo cos(a ) / ( 1–2 (V1 1 V2))]] (6)
15 r = 2·0
For the four aperture ratios, the calculated regain
10
coefficients, using these equations were both found to
5 be equal to or to be very close to unity (Table 3). The
0 deviation observed near the closed end (X 5 0) of the
0 1 2 3 4 5 6 7
duct resulted from the high experimental error as-
Distance along the duct, X, m sociated with the measurement of air velocities using
Fig. 4. Friction losses measured from the discrepancy be- the Alnor anemometer where, for example, a velocity
tween measured and calculated static air pressure y alues of 0?3 m / s has an error of 0?05 m / s (17%), and from
along the length of the perforated duct , where X is measured the high uncertainty resulting from Eqn (2) for low
from the closed end of the duct. ( r is aperture ratio). j
measured pressure; p calculated pressure for frictionless flow
Re.
The value of one for the regain coefficient also
implied a value of one for the energy correction factor
(Moueddeb et al .3). Based on an energy correction
and (4) therefore defined static air pressure losses
factor of one, the validity of Eqn (1) again held true
between outlets (FL).
and it accurately predicted static air pressure for
turbulent flow observed with all four aperture ratios of
0?5 to 2?0, where Re varied, respectively, from 95000
3.2. Outlet air jet discharge angle to 8000 and 205000 to 23000 (Fig. 5 ). This observation
is in agreement with Streeter,13 where the energy
The outlet air jet discharge angle varied over the correction factor is said to be approximately one for
length of the duct (Table 2). Koestel and Tuve,8 high turbulent flow.
Carpenter9 Bailey10 and Newman11 also found the The unit value for the regain coefficient is a more
outlet air jet discharge angle to vary along both slotted reliable evaluation compared with the values obtained
and perforated ducts. The evolution of the discharge by previous researchers because Eqn (2) was indepen-
angle, observed in the laboratory with the three-port- dent of friction losses. Asheley et al .14 and Jackson15
pitot tube was compared with that calculated from found the regain coefficient to vary as a function of
Eqn (1). The outlet air jet discharge angle was also [V1 / V2] for a single outlet in a smooth pipe. Bailey10
calculated based on the principle of conservation of also found the regain coefficient to vary with distance
mechanical energy, using the duct average air velocity for a perforated duct over its entire length. But in all
and the effect of friction losses as in Eqn (5). these cases, the regain coefficient was determined by
cos(a ) 5 (V1 1 V2) / (2Vo) (5) calculating the frictional losses, according to
Colebrook,16 and from the measured static air pres-
Since the outlet air jet discharge angle was directly sure evolution along the length of a perforated pipe.
related to the ratio of [(V1 1 V2) / 2] and Vo, the effect
of friction losses were studied by comparing the
calculated outlet air jet discharge angles from Eqns (1)
3.4 . Discharge coefficient
and (5) with those measured (Table 2). The frictional
losses on the discharge angle can therefore be as-
sumed negligible for turbulent flow because both the For the entire length of the experimental duct with
calculated and measured values for a were similar. aperture ratios of 0?5, 1 and 1?5, Cd remained at a
 PERFORATED VENTILATION DUCTS: PART 2 35

Table 2
Outlet air jet discharge angle along the length of the perforated duct, where X is measured from the
closed end of the duct.

Perforated outlet number from the duct closed end

1 2 3 4 5 6 7 8 9 10 11 12

Measurement distance from the duct closed end , X (m )

0?44 0?94 1?55 2?15 2?77 3?37 3?98 4?6 5?22 5?82 6?43 7?05

Aperture
ratio Outlet discharge angle (a ) , deg

Meas. a 89 86 85 84 83 83 80 76 76 76 70 69
r 5 0?5 Eqn 1 a 91 87 85 84 83 81 79 77 76 74 72 69
Eqn 5 a 90 86 85 84 82 81 79 77 75 74 72 70

Meas. a 86 82 79 76 72 71 68 65 60 58 54 52
r 5 1?0 Eqn 1 a 89 84 81 77 74 70 66 63 60 56 54 50
Eqn 5 a 88 84 81 77 74 70 67 64 60 57 54 50

Meas. a 84 79 73 70 66 62 57 53 45 42 39 40
r 5 1?5 Eqn 1 a 89 84 78 72 69 63 58 54 49 44 40 35
Eqn 5 a 88 82 78 73 68 63 58 53 49 44 40 35

Meas. a 83 73 70 67 65 59 53 45 38 37 35 35
r 5 2?0 Eqn 1 a 85 77 72 66 61 55 49 43 40 33 31 24
Eqn 5 a 87 81 74 68 62 56 50 45 39 35 30 25

Note: Meas. a is the measured discharge angle.

Table 3
Regain coefficient along the length of the perforated duct, where X is measured from the closed end of the duct

Perforated outlet number from the duct closed end

2 3 4 5 6 7 8 9 10 11 12

Measurement distance from the closed duct end , X (m )

0?64 1?25 1?85 2?46 3?07 3?68 4?30 4?91 5?52 6?13 6?75

Aperture
ratio Regain coefficient (Cr ) dimensionless

r 5 0?5 Mt.Eqn 1?05 0?92 0?98 1?04 1?02 0?99 0?97 1?02 1?02 1?00 0?96
En.Eqn 1?90 0?68 0?98 1?04 1?02 0?99 0?97 1?07 0?97 1?00 0?96

r 5 1?0 Mt.Eqn 0?99 1?05 0?95 1?03 1?01 0?96 0?99 1?00 0?98 1?00 0?99
En.Eqn 0?97 1?05 0?94 1?02 1?01 0?96 0?99 1?00 0?98 1?00 0?99

r 5 1?5 Mt.Eqn 1?25 1?07 0?96 1?05 1?02 0?99 1?00 1?01 0?99 1?01 1?00
En?Eqn 1?17 1?02 0?97 1?04 1?00 0?99 0?99 1?01 0?99 1?00 0?99

r 5 2?0 Mt.Eqn 0?68 0?87 0?93 0?98 0?98 0?97 0?98 1?001 0?98 1?01 0?99
En.Eqn 1?03 0?96 1?01 1?00 0?99 1?00 0?99 1?00 0?99 1?00 1?00

Note: Mt.Eqn is the momentum equation. En.Eqn is the energy equation.

36 K . E L M O U E D D E B ; S. B A R R I N G T O N ; N. B A R T H A K U R

210
McQuiston and Parker.17 For an aperture ratio of 2?0,
the low values of Cd for the first three outlets of the
195
duct from the fan end was attributed to errors in
180 measuring discharge angle because of air swirling
r = 2·0 (Table 2). Beyond the first three perforated outlets,
165
150
for the fourth to the fifth outlets from the fan, the Cd
r = 1·5 values are slightly lower than 0?65, owing to the low
Reynolds number, 103 Re

135
ratio of static to total pressure (Table 4), and to the
120 fact that the fan capacity was exceeded. According to
105 r = 1·0
ASHRAE,18 an axial fan operated at low static air
90
75
r = 0·5
Table 4
60
Discharge coefficients for the four aperture ratios as a
45 function of the static to total pressure ratio varying along
the length of the perforated duct
30
15
Aperture ratios
0
0 1 2 3 4 5 6 7 8 r 5 0.5 r 5 1.0 r 5 1.5 r 5 2.0
Distance along the duct, X, m
Fig. 5. The Reynolds number for the four experimental Static pressure / Discharge coefficient (Cd),
perforated ducts , where X is measured from the closed end of total pressure dimensionless
the duct and r is aperture ratio
0?15 — — — 0?39
0?22 — — — 0?46
constant value of 0?65 (Fig. 6 ), with a maximum 0?29 — — — 0?53
0?31 — — 0?59 —
uncertainty of 10?3% as shown in the section on data 0?35 — — — 0?59
accuracy. This value is equal to the Cd of a simple 0?37 — — 0?66 —
orifice with an outlet fluid jet normal to its area 0?45 — — — 0?59
0?46 — — 0?68 —
0?53 — — 0?69 —
1 0?54 — — — 0?6
0?61 — — 0?65 —
0·9 0?63 — — — 0?61
0?69 — — 0?65 —
0·8 0?72 — 0?66 — —
0?73 — — — 0?63
0?76 0?65
Discharge coefficient, dimensionless

0·7 — — —
0?77 — 0?67 — —
0·6 0?81 — 0?66 — —
0?82 — — — 0?66
0·5
0?83 — — 0?66 —
0?84 — 0?68 — —
0?88 0?68 0?66 — —
0·4
0?89 — — 0?66 —
0?90 — 0?67 — 0?68
0·3 0?91 — 0?67 — —
0?92 — 0?66 — —
0·2 0?93 0?67 — 0?67 —
0?94 0?68 — — —
0·1 0?95 — 0?68 — 0?68
0?96 0?67 — — —
0 0?97 0?67 0?68 — —
0 1 2 3 4 5 6 7 0?98 0?67 0?69 — —
Distance along the duct, X, m 0?99 0?68 0?71 0?66 0?67
Fig. 6. The discharge coefficient of the outlets along the 0?996 0?68 — — —
length of the perforated duct , where X is measured from the 0?998 — 0?71 — —
closed end of the duct and r is aperture ratio. d r 5 0?5; 1 0?999 0?68 — — —
r 5 1; p r 5 1?5; j r 5 2
 PERFORATED VENTILATION DUCTS: PART 2 37

3
pressures has a poor efficiency and a minimum static El Moueddeb K; Barrington S F; Barthakur N Perforated
air pressure is required for its efficient operation. ventilation ducts. Part 1: A model for flow distribution.
Journal of Agricultural Engineering Research 1997, 68,
Thus, Cd of a perforated ventilation duct can be 21 – 27
expected to be constant over the length of the system, 4
El Moueddeb K; Barrington S F; Newman B G
if the fan is operated under a normal static air Evaluation of methods to measure the performance of
pressure range. ventilation ducts. Canadian Agriculture Engineering
1996, 38(3), 207 – 213
5
ANSI / ASME Measurement Uncertainty. ANSI / ASME
4. Conclusions PTC 19.1-1985 Part 1 1986. The American Society of
Mechanical Engineers, 345 East 47th Street New York,
N.Y. 10017
The mathematical model for air flow distribution 6
Kline S J The purposes of uncertainty analysis. Journal of
pattern in a perforated duct was validated by measur- Fluids Engineering 1985, 107, 153 – 160
ing the flow performance in four experimental ducts. 7
Streeter V L; E Benjamin W Fluid mechanics 1981. first
The basic assumptions of the model were upheld. The SI metric edition. MacGraw-Hill Ryerson Limited,
skin friction loss calculated from the momentum 342 – 344
8
Koestel A; Tuve G L The discharge of air from a long
equation was the same as that provided by the energy
slot. Transaction of American Society of Heating and
equation. Thus, the energy equation can be used to Ventilating Engineers 1948, 54, 87 – 100
evaluate the regain coefficient without considering the 9
Carpenter G A The design of permeable ducts and their
friction losses. The regain coefficient and the energy application to the ventilation of livestock building.
correction factor were almost unity for turbulent flow. Journal of Agricultural Engineering Research 1972, 17,
219 – 230
The discharge coefficient was constant and approxi- 10
Bailey B J Fluid flow in perforated pipes. Journal of
mately 0?65, along the full length of the perforated Mechanical Engineering Science 1975, 17, 338 – 347
duct. The frictional losses occurred along the length of 11
Newman B G A hodographic solution for flow leaving a
the duct but had no significant effect on the discharge manifold through a slit. Canadian Aeronautics and
angle. The discharge angle (a ) can be calculated from 12
Space Journal 1989, 35, 205 – 210
Haerter A A Flow distribution and pressure change along
the duct average air velocity (V ) and the potential
slotted or branched ducts. Transactions of the
outlet air jet velocity (Vo). ASHRAE 1963, 69, 124 – 137
13
Streeter V L The kinetic energy and momentum correc-
tion factors for pipes and open channels of great width.
Acknowledgement Civil Engineering 1942, 12(4), 212 – 213
14
Ashley C M; Gilman S F; Church R A; Syracuse N Y
The authors acknowledge the financial contribution Branch fitting performance at high velocity. Transac-
of the Canadian Natural Science and Engineering tions of the ASHRAE 1956, 56, 279 – 294
15
Research Council, the Tunisian Government and Le Jackson K R Branched losses in high velocity duct
systems. Journal Institution Heating and Ventilating
Ministe` re de l’Agriculture, des Peˆ cheries et de Engineers 1969, 4, 208 – 214
l’Alimentation du Que´ bec. 16
Colebrook C F Turbulent flow in pipes with particular
reference to the transition region between smooth and
rough pipe laws. Journal of the Institution of Civil
References Engineers 1938, 4, 133 – 156
17
McQuiston F C; Parker J D Heating, ventilating and air
1
Bajura R A; Jones E H Flow distribution manifold. conditioning analysis and design 1988. Third edition.
Transactions of the ASME 1976, 98, 654 – 665 John Wiley and Sons, 344 – 347
2 18
Soucek E; Zelnick E W Lock manifold experiments. ASHRAE 1993. Equipment Handbook, 1993 Chapter 18.
Transactions of ASCE 1945, 1357 – 1400 Atlanta, Georgia