Simulation of Accumulator Influence on Hydraulic Shock in Long Pipe Bureček A, Hružík L, Vašina M

ISSN 1339-2972 (On-line)

Simulation of Accumulator Influence on Hydraulic Shock in Long

Pipe
Adam Bureček1 – Lumír Hružík2 – Martin Vašina3
1
VSB – Technical University of Ostrava, Email: adam.burecek@vsb.cz
2
VSB – Technical University of Ostrava, Email: lumir.hruzik@vsb.cz
3
VSB – Technical University of Ostrava, Email: martin.vasina@vsb.cz

Keywords Abstract
Accumulator This paper is focused on experimental measurement and mathematical simulation of liquid dynamics during hydraulic shock. The
Long Pipe mathematical model is created using Matlab SimHydraulics software. The mathematical simulation of pressure response during
Hydraulic Shock hydraulic shock is experimentally verified. The hydraulic shock is caused by a sudden closing of a seat valve, which is located at the
Simulation end of a long pipe. Pressure in front of the seat valve, oil flow and temperature were measured on experimental equipment.
Furthermore the pressure response is mathematically simulated by inserting a membrane accumulator in order to compensate
pressure pulsations.
History Received 20 August 2014 | Revised 2 September 2014 | Accepted 15 September 2014

Category Original Scientific Paper

Article
Bureček A, Hružík L, Vašina M (2014) Simulation of accumulator influence on hydraulic shock in long pipe. Journal of Manufacturing and Industrial Engineering, 14(1-2):1-4,
Citation
http://dx.doi.org/10.12776/mie.v14i1-2.461

INTRODUCTION
Hydraulic shock can be defined as non-stationary flow of
compressible liquid, when pressure changes correspond to
sudden changes of flow velocities. The hydraulic shock is caused
by a change of initial liquid flow velocity, e.g. due to valve
closing [1-4]. After the valve closing, there is positive or negative
liquid acceleration in a given point of a pipe. Periodical pressure
pulsations belong to an accompanying phenomenon of the
hydraulic shock too. They are propagated to pipe walls [1, 5].
Due to these pulsations, the pipe wall can be deformed or the
pipe can be eventually destroyed. It is possible to eliminate this
danger, e.g. by a slow valve closing or by a hydraulic
accumulator, which is able to absorb excessive pressure energy.
A mathematical simulation of the hydraulic shock is performed
using Matlab SimHydraulics software [6] that makes use of a
physical access to mathematical simulations, i.e. the simulation
model consists of single blocks. The blocks present hydraulic
elements that are described by mathematical equations and are
available in software libraries. The aim of the paper is the Figure 1 Schematic diagram of experimental circuit
experimental verification of the mathematical simulation of the
hydraulic shock and the simulation of accumulator influence on
pressure response of a hydraulic system.

EXPERIMENTS
Description of experimental hydraulic circuit – The
schematic diagram of the experimental equipment is shown in
Fig. 1. The hydraulic system consists of the hydraulic pump HP
with possibility of flow control, the check valve CV, the hose H,
the relief valve RV, the tank T and the long pipe P, which is fitted
with the seat valve SV at its end. The flow sensor FS and the
temperature sensor TS are connected behind the seat valve SV.
The pressure sensor PS is connected in front of the seat valve SV
at the end of the pipe P [7].
Figure 2 View of experimental circuit

http://dx.doi.org/10.12776/mie.v14i1-2.461 1
Bureček A, Hružík L, Vašina M Simulation of Accumulator Influence on Hydraulic Shock in Long Pipe

Parameters of the pipe P are the following: outside diameter mathematical model consists of single elements on the basis of
DP = 0.016 m, inside diameter dP = 0.012 m, length lP = 60 m, physical access. The constant flow hydraulic pump HP is defined
Young´s modulus of elasticity EP = 2.1⋅1011 Pa, Poisson number as a source of pressure energy. Oil flows through the check valve
νP = 0.3. Mineral oil of VG46 type was used as the working CV, the hose H, the long pipe P, the seat valve SV and the flow
liquid. Parameters of the hose H are the following: outside sensor FS back to the tank T. The relief valve RV is connected in
diameter DH = 0.02 m, inside diameter dH = 0.012 m, length lH = order to protect the circuit against overload. The mathematical
2.5 m. model also consists of the blocks for pressure measurement (i.e.
Description of experimental equipment: the oil flow Q = 40 the block PS), flow measurement (i.e. the block FS), control of
3 -1
dm ·min inside the pipe P was adjusted by the hydraulic pump the seat valve (i.e. the block Control of SV) and the block Solver
HP for the experimental measurement. The pipe was suddenly Configuration. The parameters of hydraulic oil (i.e. the density
-3 2 -1
closed at its end by means of the seat valve SV. For this reason ρO = 843 kg·m , the kinematic viscosity νO = 28 mm ·s and the
the pressure p was rapidly increased. A time dependence of the bulk modulus KO = 1.45·109 Pa) are defined in the block Custom
pressure was recorded by means of the pressure sensor PS. Hydraulic Fluid and were experimentally determined for the oil
Subsequently, the working liquid began to flow through the temperature tO = 55°C. The long pipe P is simulated as a
relief valve RV that was set to the pressure pRV ≅ 50 bar. The segmented pipe including the wall elasticity [6].
pressure in front of the seat valve was measured by the
pressure sensor PR15 Hydrotechnik with the measuring range of
(0 ÷ 400) bar and the ± 0.5% accuracy. The flow was measured
by the gear flow sensor GFM-70 with the measuring range of
and the ± 0.5% accuracy. The oil
3 -1
(0.7 ÷ 70) dm ·min
temperature was measured by the temperature sensor Pt 100
Hydrotechnik with the measuring range of (-50 ÷ 200) ºC and
the ± 1% accuracy. Time dependencies of the signals from the
sensors were recorded by the measuring equipment
Hydrotechnik M5050 [7]. The measured data were subsequently
processed using Hydrowin and MS Excel softwares.

120

100

80
Pressure [bar]

60

40

20
Figure 4 Mathematical model of experimental equipment
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time [s] Comparison of experimental measurement and
mathematical simulation: the comparison of the mathematically
Figure 3 Measured time dependence of pressure in front of seat valve simulated and the experimentally measured time dependencies
for the flow Q=40 dm3.min-1 and the oil temperature tO=55 °C of the pressure in front of the seat valve at the pipe end are
shown in Fig. 5. It is evident that the simulated and the
The time dependence of the pressure in front the seat valve measured pressure dependencies are very similar in terms of
3 -1
during its closing for the flow Q = 40 dm ·min and the oil the pressure amplitude and the frequency of oscillation.
temperature tO = 55 °C is shown in Fig. 3. It is evident (see Fig. 3)
that the significant hydraulic shock was created. In this case the 120
pressure peak reached nearly 110 bar, i.e. more than double
pressure, which was adjusted by the relief valve. Furthermore it 100
is visible that the pressure oscillates with the frequency f ≅ 4.7
Hz. The pressure is subsequently stabilized at the pressure value 80
Pressure [bar]

pRV, which is adjusted by the relief valve. After the pressure

60
stabilization at the end of the process, the liquid flows only
through the relief valve. The value of the stabilized pressure p =
40
10 bar at the beginning of the process is given by the pressure
Experiment
loss of the long pipe. 20
Description of mathematical model: The mathematical Simulation

model of the experimental equipment created by Matlab 0

SimHydraulics software is shown in Fig. 4 [5, 6]. A simulation of 0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time [s]
the rapid closing of the seat valve at the pipe end was Figure 5 Time dependence of pressure at pipe end during sudden closing
performed on the basis of the mathematical model. The of the seat valve
pressure response was subsequently evaluated in front of the
seat valve. Analogous to the experimental equipment, the

2 http://dx.doi.org/10.12776/mie.v14i1-2.461
Simulation of Accumulator Influence on Hydraulic Shock in Long Pipe Bureček A, Hružík L, Vašina M

Simulation of aacumulator influence on pressure response 120

during hydraulic shock: the aim of this chapter is to describe the without ACCU
100 ACCU - 0/3
influence of an accumulator position lengthwise of the long
ACCU - 1/3
pipe, in this case at the beginning of the pipe, in third and in ACCU - 2/3
80

Pressure [bar]
two-thirds of the pipe length and at the pipe end. Furthermore, ACCU - 3/3
the influence of filling pressure of the accumulator on the 60
pressure response will be described in this chapter.
Simulation of influence of accumulator location lengthwise 40
of pipe: the simulation model (see Fig. 6) was changed in this
manner: It was connected the accumulator ACCU with the 20
3
volume VA = 0.4 dm and the filling pressure pf = 45 bar. The pipe
0
was divided into equal thirds. The mathematical simulations 0 0.2 0.4 0.6 0.8 1 1.2 1.4
were subsequently performed for the following accumulator Time [s]
positions lengthwise of the pipe:
• 60 m in front of the seat valve (i.e. ACCU – 0/3), Figure 7 Simulated time dependence of pressure at the pipe end during
• 40 m in front of the seat valve - in third of the pipe (i.e. sudden closing of the seat valve for accumulator located lengthwise of
ACCU – 1/3), pipe
• 20 m in front of the seat valve - in two-thirds of the pipe
(i.e. ACCU – 2/3), Simulation of influence of accumulator filling pressure:
• 0 m in front of the seat valve – at the pipe end (i.e. ACCU There is investigated the influence of the filling pressure of the
– 3/3). accumulator on the pressure response in this simulation. The
simulations were performed for the accumulator located at the
It is evident (see Fig. 7) that the accumulator ACCU 0/3 pipe end (i.e. ACCU 3/3). The mathematical model corresponds
3
at the beginning of the pipe has practically no influence on the to Fig. 6. The accumulator volume is identical (i.e. VA = 0.4 dm )
pressure response. The pressure amplitude is in general in this case. Only the accumulator filling pressure pf is changed.
decreasing with decreasing the distance between the
120
accumulator and the seat valve (or the pipe end). In this case without ACCU
the frequency of oscillation is increased. It is caused by oil pf = 5 bar
ACCU 3/3 pp
100
pulsations in a short section of the pipe. This is not valid for the pf = 15 bar
ACCU 3/3 pp
accumulator ACCU 3/3 at the pipe end. This accumulator damps pf = 25 bar
ACCU 3/3 pp
80
Pressure [bar]

pf = 35 bar
ACCU 3/3 pp
pressure pulsations at the point of their creation.
pf = 45 bar
ACCU 3/3 pp
60

40

20

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time [s]

Figure 8 Simulated time dependence of pressure at the pipe end during

sudden closing of the seat valve for different filling pressures of
accumulator

It is evident (see Fig. 8) that the filling pressure of the

accumulator has a big influence on the pressure peak during the
hydraulic shock. It can be concluded that the minimum peak
pressure corresponds to the filling pressure pf = 35 bar of the
accumulator.

CONCLUSION AND FUTURE DIRECTION OF RESEARCH

There is described experimental equipment with long pipe
and seat valve at the pipe end in order to realize the hydraulic
shock in this paper. The time dependence of the pressure during
closing of the seat valve at the pipe end was experimentally
Figure 6 Mathematical model of experimental equipment with measured for the oil temperature tO = 55°C and the flow Q = 40
3 -1
accumulator dm ·min .
The mathematical model of the experimental equipment
was created using Matlab SimHydraulics software. This model
was verified by comparison of the measured and the simulated
time dependencies of the pressure during the sudden closing of
the seat valve at the end of the long pipe. It can be concluded
on the basis of this comparison that the mathematical

http://dx.doi.org/10.12776/mie.v14i1-2.461 3
Bureček A, Hružík L, Vašina M Simulation of Accumulator Influence on Hydraulic Shock in Long Pipe

simulation corresponds to the experiment. When after the rapid

closing of the seat valve at the pipe end, the pressure is
suddenly increased. The pressure dependence is subsequently
periodical with the peak pressure p = 110 bar and the frequency
of oscillation f ≅ 4.7 Hz. The frequency of liquid oscillation in the
long pipe depends mainly on its length, also on pipe material,
boundary flow conditions and oil bulk modulus.
The accumulator influence on the pressure response
was simulated too. It was found that the pressure amplitude is
decreased and the frequency of oscillation is increased at
location of the accumulator at the pipe end, i.e. near the seat
valve. This is not valid for the accumulator ACCU 3/3 at the pipe
end. This accumulator damps pressure pulsations at the point of
their creation. It results from the simulation of the filling
pressure of the accumulator that the minimum peak pressure
corresponds to the filling pressure pf = 35 bar of the
accumulator.

ACKNOWLEDGEMENTS: This paper has been elaborated in

the framework of the project Opportunity for young
researchers, reg. no. CZ.1.07/2.3.00/30.0016, supported by
Operational Programme Education for Competitiveness and co-
financed by the European Social Fund and the state budget of
the Czech Republic.
The work presented in this paper was also supported by a
grant SGS "Zkoumání dynamiky hydraulického vedení"
SP2014/208.

REFERENCES
[1] Gjetvaj G, Tadić M (2014) The Effect of Water Hammer on
Pressure Increases in Pipelines Protected by an Air Vessel.
Technical Gazette, 21:479-484.
[2] Tian V, Su G H, Wang S, Qiu S, Xiao Z (2008) umerical
simulation and optimization on valve-induced water
hammer characteristics for parallel pump feedwater
system. Annals of Nuclear Energy, 35:2280-2287.
[3] Himr D (2013) Numerical Simulation of Water Hammer in
Low Pressurized Pipe: Comparison of SimHydraulics and
Lax-Wendroff Method with Experiment. EPJ Web of
Conferences, 45, 01037.
[4] Jablonská J (2014) Compressibility of the fluid. EPJ Web of
Conferences, 67, 02048
[5] Hružík L, Bureček A, Vašina M (2014) Non-stationary Flow
of Hydraulic Oil in Long Pipe, EPJ Web of Conferences, 67,
02042.
[6] The MathWorks, Matlab Simulink User´s Guide,
SimHydraulics User´s Guide (2007).
[7] Hydrotechnik Gmbh, catalogue paper.

4 http://dx.doi.org/10.12776/mie.v14i1-2.461