Understanding Air Release through Air Valves

M. Carlos, Ph.D.1; F. J. Arregui, Ph.D.2; E. Cabrera, M.ASCE3; and C. V. Palau, Ph.D.4

Abstract: Water transients with entrapped air can originate large pressure peaks that can severely damage distribution networks. Entrapped
air can have a damping or amplifying effect on these undesirable pressure peaks. Unfortunately, the complexity of the phenomenon too often
makes it difficult to obtain a fully reliable prediction about when air pockets will mitigate or accentuate water transients. Furthermore, the
value of some of the parameters involved in the conventional numerical models cannot be calculated or measured and need to be determined
through a calibration process. With the aim of overcoming most of the aforementioned uncertainties, this paper summarizes a complete set of
tests conducted at WL | Delft Hydraulics. These tests were simulated by means of a tailored numerical model that includes a set of parameters
whose values were determined by means of a calibration process. The experimental setup, a large-scale facility, consisted of a single steep
pipeline with an air valve installed at its top end. Air release through different air valves was tested under different conditions. DOI: 10.1061/
(ASCE)HY.1943-7900.0000324. © 2011 American Society of Civil Engineers.
CE Database subject headings: Pipelines; Air flow; Water flow; Transient flow; Pipe flow; Sensitivity analysis; Parameters.
Author keywords: Air valves; Entrapped air; Transient analysis.

Introduction Theoretical and experimental analyses of hydraulic transients

with entrapped air have been widely described in the technical
Trapped air, if not properly removed when filling operations take literature. Some do not consider air release through orifices or
place, can cause considerable damage to pipeline systems and other through valves (Abreu et al. 1992), while others take into account
installed elements. This is a well known problem that may give rise both scenarios (Martin 1976; Zhou 2002; DeMartino et al. 2008).
to pipe ruptures with important economic consequences (Chaiko In this line, Lingireddy et al. (2004) proposed a simplified equation,
and Brinckman 2002). Additionally, air pockets may have other which did not consider the compressibility of the air pocket, to
negative effects like reducing pipeline flow capacity, diminishing estimate the pressure surges in pipeline systems resulting from
pump efficiency, hindering filter operation, generating pipe vibra- air release. Despite all the work carried out to date, the influence
tions, damaging mechanical meters, and increasing errors in flow of the dynamic behavior of air valves on water transients is not fully
measurements, among others (Lauchlan et al. 2005). Some of these understood, and works dealing with this topic are still scarce. The
negative effects have been widely reported in technical literature. availability of adequate facilities to conduct the experiments on a
For instance, air pressure was responsible for lifting manhole cov- full-scale level, the economic cost of the tests, the complexity of the
ers and blowing up sewerage pipes in Hamilton, Ontario, Canada physical phenomenon to be modeled, and the best-known and sim-
(Hamman and McCorquodale 1982)—events that were widely plest strategy to avoid problems in the field, i.e., cautiousness,
analyzed two decades later (Zhou 2002). explain why attention is rarely paid to this topic.
In contrast, if adequately managed, entrapped air can have a This work aims to fill this lack of data by providing experimen-
tal results on a set of tests carried out in a full-scale facility at WL |
cushioning effect, damping overpressures when filling operations
Delft Hydraulics. Some initial results from these tests have already
are being conducted. This positive behavior is only achieved when
been published (Arregui et al. 2003; Kruinsbrick et al. 2004). The
pipelines are filled slowly (Izquierdo et al. 1999) and the air valves
setup was built to simulate the expulsion of air through commercial
used to control the air flow, both in and out of the system, are care-
air valves located at the top end of a pipe. The experimental results
fully sized and located (Jönson 1985; Thorley 2004; Yongliang and describe the time history of the main variables recorded during the
Vairavamoorthy 2006). tests: water column velocity, system pressure at different points of
1
the facility, and air valve float displacement.
Associate Professor, Dept. Mechanical Engineering and Construction,
Because two fluids are involved, a numerical model that
Jaume I Univ., Av. de Vicent Sos Baynat s/n. 12071 Castelló de la Plana,
Spain. E-mail: mcarlos@[Link]
describes the phenomenon must take into account both the water
2
Associate Professor, ITA, Universidad Politécnica de Valencia, column equations and the air pocket behavior. Additionally, the
Camino de Vera s/n. 46022 Valencia, Spain. E-mail: farregui@[Link] numerical solution to these transients should also consider a mov-
3 ing boundary, the water–air interface, to handle the variation in the
Professor, ITA, Universidad Politécnica de Valencia, Camino de vera
s/n. 46022 Valencia, Spain. E-mail: ecabrera@[Link] length of the liquid column (Martin 1976). In this particular study,
4
Associate Professor, Dept. of Rural Engineering. Hydraulic Div. Cen- the method of characteristics with a mobile boundary condition
tro Valenciano de Estudios del Riego, Universidad Politécnica de Valencia, (Abreu et al. 1992) was used. However, other alternatives like
Camino de Vera s/n, 46022 Valencia, Spain. E-mail: virpaes@[Link] the one proposed by Malekpour and Karney (2008) could be used
Note. This manuscript was submitted on April 26, 2010; approved on
August 31, 2010; published online on September 7, 2010. Discussion per-
to simulate the movement of the water column.
iod open until September 1, 2011; separate discussions must be submitted To accurately reproduce the physical phenomena, the model
for individual papers. This paper is part of the Journal of Hydraulic En- tailored for this particular case includes several parameters. Some
gineering, Vol. 137, No. 4, April 1, 2011. ©ASCE, ISSN 0733-9429/2011/ of these parameters characterize the fluid-pipe behavior, such as the
4-461–469/$25.00. air compression coefficient, k; the pressure wave celerity, a; and

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011 / 461

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
the pipe friction factor, f . The remaining parameters are directly fied before starting the tests. All instruments, including pressure
linked to the behavior of the air valve, i.e., the air valve discharge transducers and the electromagnetic flowmeter (EMF), were cali-
coefficient, C d ; the volume of trapped air when the air valve begins brated in the laboratory. The dynamic pressure transducers had a
to close, V air ; and the air valve closing time, tc . These parameters, measuring frequency response as high as 50 kHz. However, the
which are related to the behavior of the air valve, cannot be mea- greatest uncertainty was related to the real measuring frequency
sured in the laboratory in a straightforward way. Some of them response of the EMF available in the laboratory. It is well known
have to be estimated indirectly, clearly increasing the uncertainty that there is not an easy procedure for the dynamic calibration of
associated to the estimation of their real value. this type of device. However, this specially designed meter was suc-
Although in practical applications the main target is the calcu- cessfully used in previous studies that were conducted to character-
lation of the maximum pressure peak (Epstein 2008), this work ize the dynamic behavior of check valves (Kruinsbrick 1996). The
aims to contribute to the understanding of the physical phenomena. measurement results confirmed the quick reaction time of the
Subsequently, there are two main objectives of this paper: first, to instrument, with a frequency response slightly higher than 40 Hz.
identify the set of values of the six parameters used in the math- The objective of the experiments was to simulate the filling of a
ematical model that best describe the air expulsion events generated pipe that is partially full of water with an air pocket at one end.
in the laboratory; and second, to identify by means of a sensitivity Three different water column lengths with initial water levels of
analysis which of these values has the greatest influence on the final 1,238, 1,702, and 1,988 mm with respect to the axis of the hori-
results of the tests. zontal pipe were used. At this initial stage, the air valve was open
It is important to underline that the mathematical model pre- and the air inside the pipe and the tank was under atmospheric pres-
sented does not consider the dynamic characterization of the air sure conditions. When the pressure in the tank with water increased
valves. However, Kruinsbrink et al. (2004) made an initial attempt from atmospheric pressure to a controlled pressure Pini , the water
on this subject. The approach used was similar to the one success- column moved to the top end of the pipe. In its movement, the water
fully followed for the dynamic characterization of check valves. column pushed the air pocket located at its end. Because of the fast
Unfortunately, air valve characterization is a much more complex response of the air admission valve (Butterfly Valve 2) installed
matter, and it should include the interaction of two fluids having
between the high-capacity air vessel and the water tank, pressuri-
different phases. Furthermore, most of the time, these devices
zation was almost instantaneous. In order to attain several water
are installed in parallel (although not in the case of the tests
column accelerations, various differential pressure values between
described in this work), increasing the complexity of the interaction
the two tanks, from 0.15 to 0.30 bar, were used in the experiments.
between the two fluids and the system components—air valves can
The large size of the air vessel, 70 m3 , kept the pressure in the
be connected to pipes with many different setups. A considerable
upstream tank almost constant throughout the experiments. When
number of the parameters that can be involved in the dynamic char-
the moving water column reached the top end of the pipe, the air
acterization are exclusively linked to gaseous phase. Although in
this particular case the dynamic characterization has not been con- valve float moved up, and its closing produced a fast change in
sidered, the theoretical and practical results match very well. water velocity with an important rise in pressure. At this final stage,
the pipe was almost full of water with small air pockets.
The movement of water and air generated the transient shown by
Experimental Setup the time history of the pressure (registered by the transducer located
next to the air valve), the water column velocity, and the displace-
The laboratory tests that reproduced the filling of an empty pipe ment of the air valve float, measured by an attached displacement
were conducted at the WL | Delft Hydraulics facilities using transducer (Fig. 2). In this figure, Pm is the maximum pressure
the setup shown in Fig. 1. A DN100 single body air/vacuum release reached in the system, whereas tm is the instant when it occurs;
valve with a stainless steel spherical float (457 g), was installed at tz is the time when the water column velocity is zero. In practice,
the top end of a galvanized steel pipe with an internal diameter of tm and tz are almost equal. With regard to the pressure wave, tp is
489 mm and a thickness of 9.5 mm. the moment when the pressure begins to rise after applying the
The accuracy and frequency response of the instrumentation initial pressure, Pini , from the high-capacity air vessel. Lastly,
used to measure physical variables were carefully selected and veri- tp0 is the time when the compression of the air pocket ends.

Fig. 1. Delft Hydraulics facility scheme

462 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
 Fig. 2. Pressure and velocity evolution in an air expulsion test

Two additional times linked to the movement of the air valve float, entrapped air abruptly begins to expand. From this moment, the
t o and tc , are presented in Fig. 2. These times indicate the instants hypothesis of a water–air interface which is flat and normal to
when the float starts its movement and when the air valve is com- the pipe axis is not met anymore. Water and air are mixed unpre-
pletely closed. The volume of entrapped air, when the air valve dictably and several of the preceding hypotheses cannot be used to
starts closing, can be measured indirectly, with an acceptable accurately reproduce the system behavior.
degree of uncertainty, by integrating the EMF measurements.
Governing Equations
The behavior of a one-dimensional water column can be described
Numerical Model
by the set of equations proposed by Chaudhry (1987) for a simpli-
The system of equations that describes the phenomena models fied elastic model.
the behavior of both fluids inside the pipe. It provides the mass
g dH ∂V g
and momentum conservation equations of the water column and þ þ Vsenα ¼ 0 ð1Þ
also includes the air pocket compression–expansion equation. a2 dt ∂x a
Boundary and initial conditions are also required to solve the sys-
tem of equations.
dV V · jVj ∂H
Basic Hypothesis þf · þg· ¼0 ð2Þ
dt 2D ∂x
The numerical model assumes that
• The water column pushes the air pocket in its upward move- Here, H = the piezometric head; V = water column velocity;
ment. The water–air interface occupies the whole section. It a = fluid celerity; g = gravity acceleration; f = friction factor;
is flat and normal to the axis of the pipe. and D = diameter of the pipe.
• There is a perfect gas evolution. The air pocket evolution (Martin 1976) is characterized by:
• The wall friction factor and wave celerity are constant, both in
time and space. dH
H
d∀a H

¼ k · · k· · Qa ð3Þ
• Fluid–structure interaction effects are negligible. dt ∀a dt ∀a
• The initial pressure applied to the water column (Pini ) is
constant. where H
= absolute piezometric head of the air pocket; Qa =
• Predicting the closing time of the air valve by means of a sim- volumetric air flow discharge through the air valve orifice; ∀a =
plified model like the one used in the present work is unfeasible. air volume; and k = the polytropic coefficient. Changes in the
For this reason, the air valve closing time is an input parameter volume of the air pocket are calculated according to the water
of the model and not a consequence of the forces exerted on the column velocity at each instant.
float by the two-phase fluid. For subsonic conditions (H
=H
0 < 1:89), the volumetric flow
• The air valve starts closing when the entrapped air becomes discharge through the air valve is calculated as
smaller than a predefined volume. For the test facility, this cri- rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
tical volume was typically smaller than 0:08 m2 . Its value was ρ
Qa ¼ Cd · A0 · Y · 2 · g · ðH
 H
0 Þ ð4Þ
estimated during the tests using the EMF data. The final volume ρa
of entrapped air depends on this critical volume and the air valve
closing time. where Ao = orifice area of the air valve; C d = discharge coefficient;
• There is a linear evolution of the discharged coefficient with Y = expansion factor of the air flow through the orifice; ρ = water
time, once the air valve starts closing. density; ρa = air density at the pipe conditions; and H
0 = down-
• The system of equations is solved using the method of charac- stream air pressure, atmospheric pressure in this case. The dis-
teristics, with a mobile boundary condition (Abreu et al. 1992). charge coefficient of the air valve can either be determined
The mathematical model presented provides satisfactory results experimentally or estimated using tables and equations from the
for positive fluid flow velocities. However, once the water column technical literature (AWWA 2001) or from standards such as
acquires a negative velocity and returns to the water tank, the ISO 5167-2:2003.

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011 / 463

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
 According to Martin (1976), the expansion factor can be
expressed as
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  ffi
u
u 
2=n 1  H
0
ðn1Þ=n
u n H0
Y¼t
H
· · H

ð5Þ
n  1 H
1  H 0

The parameter n adopts a value of 1.4 for dry air, and represents
the isentropic coefficient for compressible flow through an orifice.
This coefficient is different from the polytropic coefficient k used in
Eq. (3), which describes the behavior of the air pocket expansion/
compression process. For sonic conditions (H
=H
0 > 1:89 for dry
air), the orifice is choked and the volumetric flow discharge
remains constant, regardless of the internal pressure head of the
air inside the pipe. Fig. 3. Experimental versus simulation results after calibration of the
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
    nþ1 ffi model
ρ 2 n1
Qa ¼ C d · A0 · g · · H
· n · ð6Þ
ρa nþ1
Sensitivity Analysis

A sensitivity analysis was conducted to evaluate the contribution of

Initial Conditions
each parameter in the transient results. The aim of the analysis is to
The instant before the transient starts, the water velocity is zero, the show how pressure and flow velocity during the transient are
air valve is open, and the pressure of the air entrapped in the pipe affected by a controlled variation of a single parameter around
between the air valve and the water column is under atmospheric the value that provided the best fit in the numerical model. For
conditions. the sake of clarity, only one test is used and is presented in Fig. 2.
Nevertheless, the same conclusions can be reached from the
Boundary Conditions remaining tests.
The piezometric head in the water tank is assumed to be constant. The results are presented from least to greatest influence. This
Although the volume of the upstream vessel is considerable, this means that the first are those linked to the characteristics of the pipe
condition was not strictly met, and small variations were measured fluids, followed by the parameters that characterize the dynamic
in the piezometric head at the inlet of the pipe, caused by oscilla- behavior of the air valve.
tions in the water column. The pressure and the initial location of
Fluid-Pipe Parameters
the water–air interface along the pipe is also assumed to be con-
stant. The air pressure is defined by Eq. (3), whereas the position Polytropic Coefficient
of the interface is recalculated at every iteration, taking into account The polytropic coefficient, k, determines the type of air evolution
the water velocity at the previous instant. during the compression and expansion process. The polytropic
coefficient used in the optimal fit (Fig. 3) was k ¼ 1:2. The
explored values correspond to the isothermal (k ¼ 1) and adiabatic
Model Calibration (k ¼ 1:4) evolutions (Fig. 4). In the isothermal case, the transient is
slightly delayed, whereas the opposite applies for the adiabatic
Empirical determination of some of the parameters used in the case. Taking into account the time scale (tenths of a second),
model, like the average friction factor or the polytropic coefficient, the time shift is negligible. The maximum pressure differences
is infeasible in the laboratory; and therefore the uncertainty about are also insignificant, although for the isothermal case the maxi-
which values to use becomes too large. The model needs to be cali- mum pressure calculated is somewhat below the measured value,
brated to obtain the set of numerical values of the parameters that whereas the duration of the air compression–expansion process
best reproduces the laboratory tests. This was an arduous and some- (ΔT ce ) is slightly longer. Because the transient under study is very
times complicated process that was conducted for all the tests— short, and the pressures are relatively low, the value of this coef-
more than 30 for each air valve. The process was carried out by ficient was not found to be relevant. For longer transients (i.e., the
assigning five values to each parameter of the model. Combining air chamber case), this coefficient plays a more important role
the different values of the parameters meant that each test was (Graze 1972).
simulated more than 15,000 times. An error function was defined
to establish the values that best fitted the experimental data. The Wave Celerity
mean squared error (MSE) between simulated and measured values Pressure wave propagation depends on the characteristics of the
for flow velocity and air pressure at the high point of the facility fluid and the pipe. As the amount of air dissolved in the water
within the time interval defined by t p and t p0 (see Fig. 2) was cal- increases and the modulus of elasticity of the pipe material de-
culated for each simulation. The set of parameters providing the creases, the celerity becomes smaller. However, the numerical
minimum MSE was selected as a reference for the test. Additional model presented in this work considers that the water–air interface
details of this analysis can be found in Carlos (2007). is well defined and fluids are not mixed during the transient. For
Fig. 3 presents the set of parameters that provides the best this reason, the model assumes that wave celerity is constant in time
adjustment for one of the tests analyzed. The differential pressure and space.
between the air vessel and the tank with water for the test analyzed As expected and because of the high compressibility of the air
is 0.23 bar. The initial water level in the tank with respect to the axis pocket, wave celerity only becomes significant at the final stage
of the horizontal pipe is 1,238 mm. Pressure at the air valve and when pressure rises to a significant level. Nonetheless, even at
velocity are plotted from t i to t p0 . this stage, the differences corresponding to the extreme values

464 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
 Fig. 4. Effect of the polytropic coefficient on the transient simulation

Fig. 5. Effect of wave celerity on the transient simulation

of wave celerity (Fig. 5) are almost negligible. This is consistent Air Valve Behavior Parameters
with other writers’ formulations that use a rigid model to solve
Discharge Coefficient
the first stage of a water transient with entrapped air (Zhou et al. This is a key parameter for evaluating the air release flow [Eqs. (4)
2002). and (6)]. For the air valve used in the experiments, there is little
uncertainty about its value because it was experimentally deter-
Friction Factor
mined under steady flow conditions, obtaining a mean value of
Owing to the complexity of the dissipative effects in water transi-
0.71 (Arregui et al. 2003). The good agreement between the exper-
ents, the friction factor, f , and its variation in time and space in imental and the theoretical results (Fig. 7) confirms the accuracy of
hydraulic transients have received a great deal of attention over the measurement. As seen in Fig. 7, the results obtained with
the last four decades (Zielke 1968; Brunone et al. 1991; Abreu extreme C d values (0.61 and 0.81) confirm a well known fact,
and Betamio de Almeida 2009). In the case studied here, a fast i.e., that any action slowing down the water column, like a small
transient involving two accelerating fluids, the complexity is even C d value, reduces the impact velocity and hence the pressure peak
greater. Unfortunately, this parameter can have a major impact amplitude. Again, the opposite applies for a bigger C d value. Fig. 7
when its effects are accumulated in time. Because the main shows these differences, which are particularly significant from the
objective of this paper is to analyze the influence of an air valve initial instant, t i , to t p , which is the time when the pressure starts to
on the water transient, attention is only paid to identifying the increase. In conclusion, the control of the water column velocity is
effect of an equivalent friction factor. The best fit for the test in crucial to avoid high overpressures, and this objective can be
Fig. 3 was obtained for f ¼ 0:03. This value is larger than achieved by selecting valves with adequate C d values.
the one derived from the Colebrook-White equation for the
Entrapped Air Volume
average velocity during the transient, which confirms that
The amount of entrapped air that remains in the system depends on
dissipative effects in water transients are higher than those several parameters. Some of these parameters are related to the
corresponding to steady flow. Nonetheless, it is important to under- design of the air valve (i.e., the float size, weight, and shape),
line the great influence of this parameter. With other values whereas others are related to the layout of the pipeline and magni-
(f ¼ 0:02 and f ¼ 0:05), the results are quite different (Fig. 6). tude of the water transient. The first parameters affect the closing
For the lower value, the model predicts that the water column behavior of the valve once the water column has arrived. Conse-
will reach the air valve earlier than in the experiments, whereas quently, for the same installation site and transient conditions, dif-
the peak pressure is higher. Indeed, the opposite applies when ferent valves may produce dissimilar results because the volumes of
f ¼ 0:05 is considered. air that remain inside the pipe can vary. Whereas some air valves

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011 / 465

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
 Fig. 6. Effect of the friction factor on the transient simulation

Fig. 7. Effect of the discharge coefficient on the transient simulation

close shortly after the first drops of water reach the valve, others
expel significant volumes of water before the closure takes
place (Fig. 8).
The transient magnitude can also affect the amount of air in the
system before the air valve starts closing. An estimation of this air
volume can be obtained for the laboratory tests using the measures
from the EMF. Fig. 9 summarizes the results for three types of tests
for the same air valve but with different initial water levels and
several initial pressures applied to the water column (Pini ). For
the facility used, there is an inverse correlation between the velocity
at which the water column impacts the air valve and the volume of
air entrapped in the system when this happens. This may be an
indication that the mixing of water and air also depends on the
magnitude of the transient.
The sensitivity analysis shows that the amount of air that
remains inside the system plays a critical role in the maximum
pressure generated during the transient. Any modification in this
parameter changes the results dramatically. With a smaller value of Fig. 8. Water discharge through an air valve
the sensitivity analysis, 0:028 m3 , the air pocket compression–
expansion is significantly shorter than the one measured in the test, during the closing event, additional air volumes are expelled from
although the maximum pressure and the water column deceleration the pipe. As a result, longer closing times will lead to smaller
are much higher (Fig. 10). As expected, the opposite applies if a volumes of entrapped air.
larger amount of trapped air (0:048 m3 ) is considered. Two additional air valve closing times, 0.02 and 0.06 s, are
Air Valve Closing Time simulated. The faster closing time leads to a larger amount of
The numerical model used here assumes that the air valve starts trapped air within the system after the air valve has completely
closing when the volume of air inside the system falls below a closed. This is reflected in a longer duration of the compres-
predefined value that can be estimated from the test measurements. sion–expansion process. Furthermore, the greater cushioning effect
From that point on, the discharge coefficient of the air valve of the air pocket reduces the maximum pressure (Fig. 11). The
shows a linear reduction with time until it is fully closed. Therefore, results obtained for a longer closing time (t c ¼ 0:06 s) show just

466 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
 are correctly identified for the main parameters involved. Nonethe-
less, in practice, the numerical values of several of the parameters
used in the model cannot be measured or even estimated with the
required degree of confidence.
Consequently, calibration of the model using experimental data
was required. This calibration process provided, for each test, the
combination of values that produced the best fit between the model
and the laboratory data. Afterward, a sensitivity analysis was con-
ducted for one of the tests in which every parameter was changed
one at a time. This methodology allowed for the assessment of the
individual relevance of each parameter.
To simplify the analysis, the parameters of the model were clas-
sified into two groups. The first, k, a, and f , was connected to both
fluids (water and air) and to the pipe. The second group was asso-
ciated with the air valve behavior, C d , V air , and t c . Two of the
Fig. 9. Measured entrapped air volume versus impacting water column parameters in the first group, k and a, were found to exert little
velocity influence (see Figs. 4 and 5); whereas the last parameter, f , was
a key factor in the process because the water column velocity
the opposite behavior with regard to the compression–expansion depended on its value just before impacting the air valve (Fig. 6).
process and the maximum pressure value. Furthermore, a greater The arrival velocity of the water column was also affected by C d
water column deceleration rate is observed, attributable to the (Fig. 7). A smaller value of this parameter is equivalent to a larger
smaller amount of air remaining inside the pipe. friction factor or a lower driving action represented by Pini. Both
effects reduce the water column velocity and, subsequently, the
pressure peak magnitude. This is why air discharge operations
Discussion of the Results
require the selection of air valves with the lowest possible air dis-
The work presented shows how a simplified model can accurately charge capacity. Oversized air valves will very likely lead to large
describe a water transient with entrapped air if appropriate values water column velocities yielding dangerous pressure peaks.

Fig. 10. Effect of the entrapped volume of air on the transient simulation

Fig. 11. Effect of the closing time of the air valve on the transient simulation

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011 / 467

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
 The amount of air remaining in the system was also found to significantly reduce the magnitude of the water transients. Thus, a
exert a critical effect on the overpressure magnitude. As a general correct design of the pipeline layout should help by guiding the air
rule, the greater the volume of entrapped air, the larger the cush- pockets to specific locations where they can be released afterward.
ioning effect. However, the volume of residual air in the system is
difficult to control and depends not only on the dynamic behavior
Conclusions
of the air valve but also on the pipeline layout and the magnitude of
the water transient. Hydraulic transient analysis with entrapped air in pressurized pipes
As a result, the closing time of the air valve only plays a sig- on real field conditions is a very complex exercise. First, the initial
nificant role in short systems like the one used for the tests, in conditions of the system—like the initial volume of entrapped air in
which a closing time of 20 or 40 ms appreciably changes the rel- the pipes and the number and length of isolated water columns—
ative amount of air that remains entrapped in the pipe after the air are not usually known. Second, the physical behavior of the system,
valve has closed completely. For longer pipeline systems, a faster or involving two fluids in different phases and hydraulic devices with
slower closing time within certain limits will not change the moving elements, is extremely difficult to be described by a resol-
air volume in relative terms. In those cases, the amount of air uble numerical model. Finally, some of the parameters needed to
entrapped in the system will instead depend on the pipeline layout solve the model cannot be measured in the field; even if it were
or the water–air mixing process during the water column move- possible, the accuracy of the measurement would not reach the
ment, and not on the closing dynamics of the air valve. Therefore, required level.
in most real systems, the closing time of the air valves will not be a The main achievement of the work presented has been to accu-
critical factor for the pressure peak magnitude. rately reproduce by means of a relatively simple numerical model the
All the preceding comments are summarized in Fig. 12, in measurements taken in a large-scale laboratory setup. Unfortunately,
which the relative change of the parameters is plotted against it must be underlined that in practice such configuration (i.e., the
the relative change of the pressure peak magnitude. The larger pipe having a significant slope, air valve installed at the dead end
the slope of the curve, the greater the influence of the parameter of the duct) is not very common. There are also numerous air
on the maximum pressure generated in the transient. For the system valve designs that can be installed in many different ways that will
studied and the variation range of the parameters considered, all the modify the system behavior. Consequently, the results obtained for
curves are monotonic. However, as discussed in Arregui et al. this particular case, which are correct from quantitative and quali-
(2010), this may not always be the case. tative perspectives, can only be qualitatively extrapolated for other
These results confirm that in order to minimize the negative layouts. Nevertheless, this may not be a minor finding, mainly
effects that entrapped air may have in the system when restarting because some of the conclusions are counterintuitive.
it, it is essential to proceed with extreme care. If this is the case,
water velocity in the pipes can be kept to low values and entrapped
air volumes will be large enough to provide an adequate cushioning Acknowledgments
effect. Moreover, selecting air valves with lower discharge
coefficients and smaller orifice areas will help to mitigate the maxi- The writers would like to acknowledge the valuable contributions
mum pressure peak that may appear in the system. By following of the reviewers of this paper because their comments and sugges-
this sizing criterion, entrapped air volumes will always be larger tions have helped to significantly improve its contents.
and the velocity of the approaching water column smaller, due This research has been possible thanks to the European Commu-
to a higher counterpressure. nity Programme “Access to Major Research Infrastructure” under
Assuming that air valves work as expected, the most dangerous the Fifth Framework Growth Programme (Contract No. GIRT-CT-
situations will be those in which small uncontrolled volumes of air 2002-05069).
remain inside the pipe because they cannot be removed properly
through an air valve. This will very likely lead to pipe bursts. In References
contrast, the presence of large volumes of air at specific locations
cannot be considered undesirable as long as they can be released Abreu, J., and Betamio de Almeida, A. (2009). “Time scale behaviour of
slowly in a controlled manner. Large volumes of entrapped air will the wall shear stress in unsteady laminar pipe flows.” J. Hydraul. Eng.,
135, 415–424.
Abreu, J., Cabrera, E., García-Serra, J., and Izquierdo, J. (1992). “Boundary
between elastic and inelastic models in hydraulic transient analysis
with entrapped air pockets.” Proc., Int. Meeting on Hydraulic
Transients with Water Column Separation, E. Cabrera and M. Fanelli
eds., Universidad Politécnica de Valencia, 159–179.
American Water Works Association (AWWA). (2001). “Air release, air/
vacuum and combination air valves.” Manual of water supply practices,
M51. AWWA, Denver.
Arregui, F. J., García-Serra, J., and Cabrera, E. (2010). “Discussion of
‘Transient Flow Caused by Air Expulsion through an Orifice’ by
G. De Martino, N. Fontana, and M. Giugni.” J. Hydraul. Eng., 136(4),
267–269.
Arregui, F. J., García-Serra, J., Cabrera, E., Fuertes, V. S., Palau, C. V., and
Gascón, L. (2003). “Air valve dynamic behaviour.” Electromechanical
devices and systems applied to urban water management, E. Cabrera
and E. Cabrera Jr., eds., Swets & Zeitlinger, Lisse, Vol. 2, 623–632.
Brunone, B., Golia, U. M., and Greco, M. (1991). “Modelling of fast tran-
Fig. 12. Sensitivity analysis of the parameters involved in the water sients by numerical methods.” Proc., Int. Conf. on Hydraulic Transients
with Water Column Separation, E. Cabrera and M. Fanelli, eds., IAHR,
transient
Valencia, Spain, 273–280.

468 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]
Carlos, M. (2007). “Estudio de la influencia de las ventosas en transitorios pipeline systems.” Ph.D. thesis, City Univ. London, Delft University
hidráulicos con aire atrapado.” Ph.D. thesis, Universidad Politécnica de Press, The Netherlands.
Valencia, Spain. Kruinsbrick, A. C. H., Arregui, F., Carlos, M., and Bergant, A. (2004).
Chaiko, M A., and Brinckman, K. W. (2002). “Models for analysis of “Dynamic performance characterization of air valves.” 9th Int. Conf.
water hammer in piping with entrapped air.” J. Fluids Eng., 124, on Pressure Surges, BHR Group, Chester, UK, 33–48.
194–204. Lauchlan, C. S., Escarameia, M., May, R. W. P., Burrows, R., and Gahan,
Chaudhry, M. H. (1987). Applied hydraulic transients, 2nd Ed., Van C. (2005). “Air in pipelines: A literature review.” Rep. SR 649, HR
Nostrand Reinhold, New York. Wallingford.
De Martino, G., Fontana, N., and Giugni, M. (2008). “Transient flow Lingireddy, S., Wood, D. J., and Zloczower, N. (2004). “Pressure surges
caused by air expulsion through an orifice.” J. Hydraul. Eng., 134, in pipeline systems resulting from air releases.” J. Am. Water Works
1395–1399. Assoc., 96(7), 88–94.
Epstein, M. (2008). “A simple approach to the prediction of waterhammer Malekpour, A., and Karney, B. (2008). “Rapid filling analysis of pipelines
transients in a pipe line with entrapped air.” Nucl. Eng. Des., 238,
using an elastic model. Surge analysis-system design, simulation, mon-
2182–2188.
itoring and control.” 10th Int. Conf. on Pressure Surges, Stephen Hunt,
Graze, H. R. (1972). “The importance of temperature in air chamber oper-
ed., BHR Group, Edinburgh, UK, 539–552.
ations.” 1st Int. Conf. on Pressure Surges, BHRA, Canterbury, UK, K2,
Martin, C. S. (1976). “Entrapped air in pipelines. BHRA.” Proc., 2nd Int.
19–K2, 36.
Conf. on Pressure Surges, Bedford, England.
Hamman, M. A., and McCorquodale, J. A. (1982). “Transient conditions in
the transition from gravity to surcharged sewer flow,” Can. J. Civ. Eng., Thorley, A. R. D. (2004). Fluid transients in pipeline systems, 2nd Ed.,
9(2), 189–196. Professional Engineering Publishing Limited, UK.
ISO. (2003). “Measurement of fluid flow by means of pressure differential Yongliang, Z., and Vairavamoorthy, K. (2006). “Transient flow in rapidly
devices inserted in circular cross-section conduits running full. Part 2: filling air-entrapped pipelines with moving boundaries.” Tsinghua Sci.
Orifice plates.” ISO 5167-2:2003, Geneva. Technol., 11(3), 313–323.
Izquierdo, J., Fuertes, V., Cabrera, E., Iglesias, P., and García Serra, Zhou, F. (2002). Effects of trapped air on flow transients in rapidly filling
J. (1999). “Pipelines start up with entrapped air.” J. Hydraul. Res., sewers, Univ. of Alberta, Edmonton, Canada.
37(5), 579–590. Zhou, F., Hicks, F. E., and Steffler, P. M. (2002). “Transient flow in a rap-
Jönson, L. (1985). “Maximum transient pressures in a conduit with check idly filling horizontal pipe containing trapped air.” J. Hydraul. Eng.,
valve and air entrainment.” Int. Conf. on Hydraulics of Pumping 128(6), 625–634.
Stations, BHRA, Manchester, UK, 55–76. Zielke, W. (1968). “Frequency-dependent friction in transient pipe flow.”
Kruinsbrick, A. C. H. (1996). “The dynamic behaviour of check valves in J. Basic Eng., 90(1), 109–115.

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / APRIL 2011 / 469

Downloaded 19 May 2011 to [Link]. Redistribution subject to ASCE license or copyright. Visit[Link]