Energy Conversion and Management: He Wang, Guofang Gong, Hongbin Zhou, Wei Wang
Energy Conversion and Management: He Wang, Guofang Gong, Hongbin Zhou, Wei Wang
Energy Conversion and Management: He Wang, Guofang Gong, Hongbin Zhou, Wei Wang
a r t i c l e i n f o a b s t r a c t
Article history: In this paper, a servo motor operated rotary directional control valve is proposed, and a systematic
Received 1 July 2015 analysis of steady flow torques in this valve is provided by theoretical calculation, CFD simulation and
Accepted 10 November 2015 experimental test. In the analysis, spool rotation angle corresponding to the maximum orifice opening
Available online 13 January 2016
is tagged as 0°. Over a complete change cycle of the orifice, the range of spool rotation angle is symmetric
about 0°. The results show that the direction of steady flow torques in this valve is always the direction of
Keywords: orifice closing. The steady flow torques serve as resistances to the spool rotation when the orifice opening
Steady flow torques
increases, while impetuses to the spool rotation when the orifice opening decreases. At a certain pressure
Rotary directional control valve
Spool rotation angle
drop or flow rate, steady flow torques are approximately equal and opposite when at spool rotation
Pressure drop angles which are symmetric about 0°. When the spool rotates from 0°, at a certain pressure drop, their
Flow rate values increase first then decrease with the spool rotation and reach their maximum values at an angle
corresponding to about 1/2 of the maximum orifice opening, and at a certain flow rate, their values
increase with the spool rotation. The steady flow torques in this valve are the sums of those in the
meter-in and meter-out valve chambers. At a certain spool rotation angle, steady flow torques in
the meter-in and meter-out valve chambers are approximately proportional to the pressure drop and
the second power of the flow rate through the orifice. Theoretical calculation and CFD simulation can
be validated by the experimental test. The proposed calculation formulas can predict the steady flow
torques in the rotary directional control valve correctly.
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction [6]. Other similar progresses include a rotary servo valve with
double-edged flow-metering blade [7] and a rotary valve with spe-
Directional control valves which control the movement direc- cial gear teeth structure [8]. For specific applications, researchers
tion of actuators or connect/disconnect part of a hydraulic circuit presented rotary valves which can realize continuous angular
belong to widely used hydraulic elements [1]. These valves are usu- displacement, such as rotary valve for continuous wave mud pulse
ally used to control force, velocity, or position of the actuators pre- generator [9] and coaxial rotary valve used in pulse detonation
cisely [2–4]. Compared to their slide counterpart, rotary directional combustor [10]. However, these valves are only applicable to
control valves, especially rotary servo/proportional valves, have particular occasion and lack of universality.
received considerable attention recently for their advantages of Inspired by the literatures related to rotary valve structures and
simple structure, exceptional reliability and zero acceleration drift. actuation devices, a servo motor operated rotary directional con-
Some rotary valve structures and actuation devices have been trol valve is proposed. In the proposed valve, the angular displace-
designed. A two dimensional flow control valve controlled by a flow ment of the rotary spool is continuous as the servo motor rotate.
rate pilot is proposed to improve the speed of response, but the flow Compared to the existing directional control valve, the main
range is limited to a low level [5]. For high flow rate applications, an improvement of this valve is that it can change the movement
electrohydraulic rotary valve with bidirectional rotary proportional direction of actuators continuously and rapidly at high flow rate.
actuator using permanent-magnet rotor is designed, but it is not The change of flow direction causes the change of fluid flow
suitable for use when continuous angular displacement is desired momentum when the hydraulic fluid flows through an orifice in
the directional control valves. Then the fluid momentum effects
acting on the spool occur. These momentum effects which provide
⇑ Corresponding author. Tel.: +86 571 8795 2500.
the major interferences to the spool movement have a great effect
E-mail addresses: [email protected] (H. Wang), [email protected] (G. Gong),
[email protected] (H. Zhou), [email protected] (W. Wang). on the performance of the directional control valves. Usually,
http://dx.doi.org/10.1016/j.enconman.2015.11.054
0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
2 H. Wang et al. / Energy Conversion and Management 112 (2016) 1–10
Nomenclature
higher flow rate causes higher fluid momentum effects. They are with spool rotation angle, pressure drop and flow rate through the
generally quite substantial when a large amount of flow is passing orifice.
through the valve. Evaluation of the fluid momentum effects is also
important for proper fit of spool actuation device of the designed 2. Object of study
valve. Thus, it is very necessary to study the fluid momentum
effects in the proposed valve. As shown in Fig. 1, the servo motor operated rotary directional
Whereas, as far as known to us, study of the fluid momentum control valve composed of a servo motor, a rotary spool, a static
effects in rotary valves has not been done yet. The existing research sleeve, a valve body, two valve covers and other accessories is a
related to the fluid momentum effects are almost limited to the slide two-position four-way directional control valve. Port P is pressure
valve field. In slide valves, these momentum effects acting on the port, Port A and B are actuator ports and Port TA and TB are alterna-
spool are familiarly known as ‘‘steady flow forces”. Numerous tive return ports. The rotary spool controlled by the servo motor
attempts have been made to analyze the steady flow forces. For has two identical spool lands. On both sides of the two spool lands,
the valves with flow channels of simple geometry, steady flow forces four rectangular grooves with uniform size are evenly distributed
can be calculate theoretically based on the conservation of fluid along the circumferential direction. The central angle of one groove
momentum for the control volumes [11–13]. In the theoretical cal- is 22.5°, then geometrically, the central angle between two adjoin-
culation, jet angle is a critical parameter cannot be obtained by ing grooves on one side of the spool land is 90°. The rectangular
mathematical method. But with velocity vectors, it can be estimated windows in the sleeve are corresponding to the grooves on the
easily [14,15]. For the valves with flow channels of complex geome- spool land. The overlapped parts of the windows and grooves
try, however, the calculation formula of steady flow forces cannot be which allow hydraulic fluid to flow through serve as the orifices.
obtained. With the development of technology, this task can be facil- With the spool rotation, the orifice area can change within the
itated by computational fluid dynamics (CFD) which is becoming a range of zero to its maximum level.
well-established practice in valve analysis and design. Two- To be convenient for explanation, the orifices are labeled as I, II,
dimensional (2D) approaches are applicable when the fluid flow III and IV, each of which is formed by four grooves on one side of
inside the control volumes can be described with axisymmetric fluid the spool land and their corresponding windows. Orifice I and II
models [16,17]. For the unsymmetrical fluid models, fluid domains are formed on one spool land and orifice III and IV are formed on
are usually described with three-dimensional (3D) models [18– the other. Orifice I and IV are designed to serve as meter-out ori-
20]. Because of structural differences, there is quite a large difference fices and orifice II and III are designed to serve as meter-in orifices.
between the fluid momentum effects in rotary valves and that in Orifice I and III with equal orifice area in all cases form one group
slide valves, e.g., the manifestation of fluid momentum effects shifts and so it is with orifice II and IV. On one spool land, the designed
from force to torque in rotary valves. Therefore, the analysis results stagger angle between the adjacent grooves of the orifices belong-
for slide valves are not applicable to rotary valves. The fluid momen- ing to different groups is 45°. This design make sure that the ori-
tum effects in rotary valves are still unknown. There are no system- fices belonging to one group open simultaneously as the orifices
atic research results can be borrowed directly to analysis the fluid belonging to the other group close. When orifice I and III open,
momentum effects in the proposed valve until now. the hydraulic connection is P ? A and B ? TB and when orifice II
Motivated by the above observations, the present study mainly and IV open, it changes to P ? B and A ? TA. The two hydraulic
concentrates on the fluid momentum effects in rotary directional connections is alternate as the spool rotate.
control valves which can be named as ‘‘steady flow torques”. In this Because the proposed rotary directional control valve has a sym-
paper, a systematic analysis of steady flow torques in the proposed metric structure, the steady flow torques acting on the spool will be
rotary directional control valve is provided by theoretical calcula- exactly equal under the two hydraulic connections theoretically.
tion, CFD simulation and experimental test. The aim is to proposed Therefore, the paper presents a case study in which the hydraulic
calculation formulas of steady flow torques in the rotary directional connection is P ? A and B ? TB. Obviously, the chamber formed
control valves and explore the change law of steady flow torques by the spool and sleeve which connect port P and A is meter-in valve
H. Wang et al. / Energy Conversion and Management 112 (2016) 1–10 3
Fig. 2. Steady flow torques in the meter-in valve chamber: (a) 22.5° < h < 0°, (b) h = 0°, and (c) 0° < h < 22.5°.
chamber, and which connect port B and TB is meter-out valve cham- flow momentum in the circumference direction. The steady flow
ber. The steady flow torques acting on the rotary spool in the meter- torques in the rotary directional control valve are the resultant of
in or meter-out valve chambers are caused by the change of fluid that in the meter-in and meter-out valve chambers.
4 H. Wang et al. / Energy Conversion and Management 112 (2016) 1–10
3. Theoretical calculation direction which is also the direction of the meter-in orifice closing
and serve as impetuses to the spool rotation as 90° < b1 < 180° in
In the theoretical calculation, there are some assumptions: the cases of 0° < h < 22.5°.
incompressible liquid, omitted leakages and constant liquid prop- For the meter-out valve chamber, according to the structure of
erties. Consider first the meter-in valve chamber. According to the proposed rotary directional control valve presented in Figs. 1
the structure of the proposed rotary directional control valve pre- and 3, the fluid flows in through the meter-out orifice usually at
sented in Figs. 1 and 2, the fluid flows in through flow channel P an angle to the spool radial direction, except when the windows
always perpendicularly to the spool axis and out through the and grooves completely coincide and out through flow channel
meter-in orifice usually at an angle to the spool radial direction, TB always perpendicularly to the spool axis. That means the cir-
except when the windows and grooves completely coincide. It cumferential component of fluid flow momentum is usually non-
means that the circumferential component of the fluid flow zero when the fluid flows through the meter-out orifice and zero
momentum is zero when the fluid flows through flow channel P when the fluid flows through flow channel TB. As shown in Fig. 3,
and usually non-zero when the fluid flows through the meter-in in the same way, the steady flow torques in the meter-out valve
orifice. As shown in Fig. 2, based on the theorem of momentum, chamber T2 can be calculated as
the circumferential component of flow forces in the meter-in valve
chamber F1 can be derived as T 2 ¼ qQ 2 v 2 R cos b2 ð3Þ
F 1 ¼ qQ 1 v 1 cos b1 ð1Þ where Q2, v2 and b2 denote the flow rate, fluid velocity and jet angle
at the meter-out orifice, respectively.
where q is the oil density, Q1, v1 and b1 denote the flow rate, fluid As shown in Fig. 3, the geometrical character of the meter-out
velocity and jet angle at the meter-in orifice, respectively. Thus, orifice is same to that of the meter-in orifice and the steady flow
the steady flow torques in the meter-in valve chamber T1 can be cal- torques in the meter-in and meter-out valve chambers are in the
culated as same direction in all cases. Thus, the steady flow torques in the
T 1 ¼ qQ 1 v 1 R cos b1 ð2Þ rotary directional control valve are the sums of those in the
meter-in and meter-out valve chambers and in the direction of
where R is the spool land radius. the orifice closing. According to Eqs. (2) and (3), the steady flow
It can be seen from Fig. 2 that there are three relative position torques in this valve Ts can be expressed as
relationships between the grooves and corresponding windows
which form the meter-in orifice as the spool rotate counterclock- T s ¼ qRðQ 2 v 2 cos b2 Q 1 v 1 cos b1 Þ ð4Þ
wise. In Fig. 2(b), the spool rotation angle h is tagged as 0° when Employing Bernoulli equation, the fluid velocity at the meter-in
the grooves and the corresponding windows completely coincide. and meter-out orifices can be expressed as
At this time, the orifice opening reach its maximum size. Because sffiffiffiffiffiffiffiffiffiffiffiffiffi
the central angle of one groove is 22.5°, the meter-in orifice goes 2
through a complete change cycle when the spool rotation angle h v1 ¼ Cv Dp1 ð5Þ
q
increases from 22.5° to 22.5°. Counterclockwise direction is
defined as the positive direction in this paper. Therefore, relative sffiffiffiffiffiffiffiffiffiffiffiffiffi
2
to the corresponding windows, the grooves are in clockwise direc- v2 ¼ Cv Dp2 ð6Þ
tion at 22.5° < h < 0° shown in Fig. 2(a) and in counterclockwise q
direction at 0° < h < 22.5° shown in Fig. 2(c). As the spool rotate,
where Cv is the velocity coefficient, Dp1 and Dp2 denote the pres-
the meter-in orifice opens gradually within the range of
sure drop through the meter-in and meter-out orifices, respectively.
22.5° < h < 0°, reaches a maximum opening at h = 0° and closes
The flow rate at the meter-in and meter-out orifices can be
gradually within the range of 0° < h < 22.5°.
expressed as
The flow directions at the meter-in orifice are shown in Fig. 2. In
the cases of 22.5° < h < 0°, as the jet angle is in the range of sffiffiffiffiffiffiffiffiffiffiffiffiffi
0° < b1 < 90°, the steady flow torques in the meter-in valve cham- 2
Q 1 ¼ C d1 A Dp1 ð7Þ
ber are in the clockwise direction which is the direction of the q
meter-in orifice closing and serve as resistance to the spool rota- sffiffiffiffiffiffiffiffiffiffiffiffiffi
tion because their values are negative according to Eq. (2). Simi- 2
larly, the steady flow torques in the meter-in valve chamber are Q 2 ¼ C d2 A Dp2 ð8Þ
q
zero as b1 = 90° in the case of h = 0° and in the counterclockwise
Fig. 3. Steady flow torques in the meter-out valve chamber: (a) 22.5° < h < 0°, (b) h = 0°, and (c) 0° < h < 22.5°.
H. Wang et al. / Energy Conversion and Management 112 (2016) 1–10 5
Window defined from Eqs. (7) and (8), in which the pressure drop and flow
Flow area
rate can be obtained by CFD simulation. The jet angle b1 and b2 can
be estimated from velocity vectors at the meter-in and meter-out
orifices obtained by CFD method [14,15].
x
4. CFD simulation
y The first step for CFD simulation is to build grid models of the
Groove
computational domains. On the basis of 3D model of the proposed
Fig. 4. Orifice area formed by one groove and its corresponding window. rotary directional control valve, computational domains of the
meter-in and meter-out valve chambers can be obtained. The
where A is the orifice area of the meter-in and meter-out orifices ANSYS ICEM CFD software, dedicated to creating meshes for the
with the same geometrical character, Cd1 and Cd2 denote the CFD, is used to build the grid models. In a grid model, while the
discharge coefficient of the meter-in and meter-out orifices, tetrahedral grids exist in the majority area, a finer pentahedral
respectively. prism grid scheme is adopted near the boundary, which meets
In Fig. 4, dashed lines represent the groove and solid lines its requirement of mesh in the boundary layer [21].
represent the window, then geometrically, the orifice area formed Fig. 5 shows grid models for the meter-in and meter-out valve
by one groove and its corresponding window is shown by the chambers at h = 15° regarded as a typical case. Computational
shaded portion. Thus, the orifice area of the meter-in and meter- domains for both valve chambers have symmetric structures. In
out orifices can be expressed as the adopted coordinate system, X-axis coincides with the spool
axis, thus it is the axis of the steady flow torques in CFD simulation.
A ¼ 4xy ð9Þ There are four side faces labeled as wall 1 and another four side
faces labeled as wall 2 in each grid model. The torques on wall 1
where x is the axial length of the groove and y is the width of the over-
and 2, the sums of which are the steady flow torques, are calcu-
lapped area of one groove and its corresponding window. In Figs. 2
lated. Each of the grid models has four inlets and outlets.
and 3, the central angle of one groove a = 22.5°, then the central angle
To ensure that the results are independent from the mesh reso-
of the overlapped area c can be expressed as c = 22.5° |h| in the
lution, a mesh independence (grid sensitivity) analysis is performed
range of 22.5° < h < 22.5°. Thus, y can be calculated as
in the construction and analysis of the grid models. In the mesh
22:5 jhj independence study, the case of h = 15° and Dp = 6 MPa is used to
y ¼ 2R sin ð10Þ
2 carry out a test on the performance of the mesh with quite reason-
From Eqs. (9) and (10), the orifice area can be expressed as ably computational effort. The globe element size decreases from
1.2 mm to 0.6 mm at a reduction of 0.2 mm. The information of grid
22:5 jhj
A ¼ 8xR sin ð11Þ model and the steady flow torques under different globe element
2 sizes are shown in Table 2. Obviously, the change of steady flow tor-
By applying Eqs. (4)–(10), during a complete change cycle of the ques in the results is very small as the globe element size vary from
meter-in and meter-out orifices, the steady flow torques in this 0.8 mm to 0.6 mm. Therefore, it can be concluded that the grid
valve can be written as model with globe element size of 0.8 mm is adequate, and it
22:5 jhj
requires less computational effort than the 0.6 mm.
T s ¼ 16R2 C v xðC d2 Dp2 cos b2 C d1 Dp1 cos b1 Þ sin ð22:5 < h < 22:5 Þ
2
ð12Þ 4.2. Turbulence model
Inlet Outlet
Outlet Inlet
Wall 1
Wall 1
Wall 2 Wall 2
(a) (b)
Fig. 5. Grid models of the computational domains at h = 15°: (a) meter-in valve chamber, and (b) meter-out valve chamber.
Table 2
The results obtained from the mesh independence studies at h = 15° and Dp = 6 MPa.
Turbulent viscosity lt is calculated as incompressible liquid, omitted leakages and roughness of channel
2 walls, constant liquid properties and thermal equilibrium condition.
k
lt ¼ C l q ð16Þ
e
where u is the velocity, l is the dynamic viscosity, Sij is the mean 5. Experimental test
component of deformation rate of a fluid element in the turbulent
flow. The standard k–e model constants are C1e = 1.44, C2e = 1.92, The experimental system of the servo motor operated rotary
Cl = 0.09, rk = 1.00 and re = 1.30. directional control valve is illustrated in Fig. 6. The tested valve
is fixed on the test bench. The test bench can measure the flow rate
4.3. Boundary conditions and solving strategies through the valve by its own flow meters with a measurement
error of 0.3% and display the data on panel. Servo controller sends
Boundary conditions must be defined to carry out the calcula- out command signals to the servo motor to make the spool rotate
tion with ANSYS/Fluent software. The boundary condition type of to a specific angle required for test and remain still when the fluid
outlet is set to pressure-outlet with constant pressure 0.1 MPa. flow through the valve. Angle indicator can display the spool rota-
The boundary condition type of inlet is set to pressure-inlet when tion angle in real time. With the servo controller and angle indica-
analyzing the steady flow torques at different pressure drops or tor, the error of spool rotation angle can be limited to less than 0.1°.
velocity-inlet when investigating the steady flow torques at differ- Torque sensor with a measurement error of 0.1% of the full scale
ent flow rates. All boundary walls are assumed to possess a no slip (5 N m) is mounted coaxially between the servo motor and valve
boundary condition. Standard wall function is used in the near wall body. In order to measure the inlet and outlet pressures more accu-
treatment. rately, the pressure sensors are directly installed on the valve body.
The CFD code is used with the Semi-Implicit Method for Pressure- The measurement error of the pressure sensor is 0.3% of the full
Linked Equations (SIMPLE) velocity–pressure coupling algorithm scale (25 MPa). Data acquisition instrument is employed to collect
with the second order discretization scheme. In addition, the density the data of torques and pressures and send them to the computer.
of hydraulic oil is 880 kg/m3, and dynamic viscosity is about Three different hydraulic circuits can be realized by changing the
0.036 kg/(m s). The absolute criterion for convergence is set to oil supply way in the test: only P ? A, only B ? TB, or both. By
0.000 01 for the scaled residuals of all the parameters. There are also means of regulating the displacement of the variable pump, the
some assumptions in this simulation: steady-state condition, specific flow rates and pressures required for test can be obtained.
H. Wang et al. / Energy Conversion and Management 112 (2016) 1–10 7
Fig. 6. Experimental system of the servo motor operated rotary directional control valve: 1 – cooler, 2 – relief valve, 3 – oil tank, 4 – electric motor, 5 – filter, 6 – variable
pump, 7 – check valve, 8 – flow meter, 9 – servo controller, 10 – servo motor, 11 – torque sensor, 12 – valve body, 13 – pressure sensor, 14 – angle indicator, 15 – data
acquisition instrument, 16 – computer.
Window
Sleeve
Spool
Groove
Fig. 7. Spool and sleeve of the rotary directional control valve.
Fig. 9. Flow rate through the meter-in and meter-out orifices at pressure drop of
1 MPa.
Fig. 8. Orifice area of the meter-in and meter-out orifices calculated from Eq. (11).
Fig. 10. Discharge coefficient of the meter-in and meter-out orifices calculated
from Eqs. (7) and (8).
Fig. 11. Velocity vectors of the meter-in orifice: (a) h = 15°, (b) h = 7.5°, (c) h = 0°, (d) h = 7.5° and (e) h = 15°.
Fig. 12. Velocity vectors of the meter-out orifice: (a) h = 15°, (b) h = 7.5°, (c) h = 0°, (d) h = 7.5° and (e) h = 15°.
Fig. 14. Steady flow torques versus spool rotation angle at a certain pressure drop
Fig. 13. Jet angle at the meter-in and meter-out orifices. (1 MPa).
6. Results and discussion rates increase with increase of the orifice area and reach their
maximum values at h = 0°. The discharge coefficient of the meter-
During the analysis, the spool rotation angle increases from in and meter-out orifices can be calculated with Eqs. (7) and (8),
20° to 20° at increment of 2.5°, the pressure drop through the and the results are indicated in Fig. 10. The discharge coefficients
meter-in and meter-out orifices increase from 1 MPa to 7 MPa at are in the range of 0.61–0.71, and symmetric about the zero spool
increment of 1 MPa, and the flow rate through the meter-in and rotation angle. Due to the same geometrical character of the two
meter-out orifices increase from 20 L/min to 140 L/min at orifices, the discharge coefficient values of the meter-in and
increment of 20 L/min. Obviously, the range of spool rotation angle meter-out orifices are very close.
is symmetric about 0°. The discharge coefficient and jet angle Some typical cases of velocity vectors of the meter-in and
which are very essential in steady flow torques calculation must meter-out orifices are shown in Figs. 11 and 12, respectively. It
be defined first. can be seen that the flow direction vary from the location of the
The orifice area of the meter-in and meter-out orifices are velocity vector in the orifice at a certain spool rotation angle.
always equal to each other. As is shown in Fig. 8, the orifice area ver- When estimating the jet angle, the direction of the velocity vec-
sus the spool rotation angle either tends to be linear, or has a trian- tor in the middle of the orifice is defined as the flow direction.
gular waveform during the range of spool rotation angle and the Then, the jet angle can be obtained by geometry method. The jet
value reaches the maximum of about 72.6 mm2 at h = 0°. The flow angle at the meter-in and meter-out orifices are shown in Fig. 13.
rate through the meter-in and meter-out orifices at pressure drop The jet angle at the meter-in orifice is in the range of 78–102°,
of 1 MPa obtained by CFD simulation are shown in Fig. 9. The flow and at the meter-out orifice, the jet angle is in the range of 70–
H. Wang et al. / Energy Conversion and Management 112 (2016) 1–10 9
Table 3
The error of theoretical calculation and experimental test comparing with the CFD simulation results in Figs. 14–17.
the estimation of jet angle and assumptions in the theoretical cal- References
culation. The experimental values are a little larger than the theo-
retical values, but have the same variation trend (maximum error [1] Manring ND. Hydraulic control systems. New York: John Wiley & Sons; 2005.
[2] Karpenko M, Sepehri N. Electrohydraulic force control design of a hardware-in-
is about 19% and average error is about 12%). The error may be the-loop load emulator using a nonlinear QFT technique. Control Eng Pract
caused by the interference of friction, pressure fluctuation and 2012;20:598–609.
measurement error. The theoretical calculation and CFD simulation [3] Yang CF, Huang QT, Han JW. Computed force and velocity control for spatial
multi-DOF electro-hydraulic parallel manipulator. Mechatronics
can be validated by the experimental test. The proposed calcula- 2012;22:715–22.
tion formulas can predict the steady flow torques in the rotary [4] Chiang MH. A novel pitch control system for a wind turbine driven by a
directional control valve correctly. variable-speed pump-controlled hydraulic servo system. Mechatronics
2011;21:753–61.
[5] Ruan J, Burton RT, Ukrainetz P. An investigation into the characteristic of a two
7. Conclusions dimensional ‘‘2D” flow control valve. J Dyn Syst – Trans ASME
2002;124:214–20.
[6] Cui J, Ding Fan, Li QP. Novel bidirectional rotary proportional actuator for
In this paper, a servo motor operated rotary directional control electrohydraulic rotary valves. IEEE Trans Magn 2007;43:3254–8.
valve is proposed, and a systematic analysis of steady flow torques [7] Leonard MB. Rotary servo valve. United States patent US 5954093; 1999 Sep 1.
in this valve is provided by theoretical calculation, CFD simulation [8] Hochrein BG. Rotary valve. United States patent US 6499507; 2002 Dec 31.
[9] Yan ZD, Wei CM, Geng YF, Shao J, Hu XF, Li Y. Design of a rotary valve orifice for
and experimental test. The conclusions are summarized as follows. a continuous wave mud pulse generator. Precis Eng 2015;41:111–8.
[10] Matsuoka K, Esumi M, Ikeguchi KB, Kasahara J, Matsuo A, Funaki I. Optical and
(1) At a certain pressure drop or flow rate, steady flow torques thrust measurement of a pulse detonation combustor with a coaxial rotary
valve. Combust Flame 2012;159:1321–38.
are approximately equal and opposite when at spool rota-
[11] Manring ND. Modeling spool valve flow forces. 2004 ASME International
tion angles which are symmetric about 0° (the angle corre- Mechanical Engineering Congress and Exposition. November 13–20, California,
sponding to the maximum orifice opening). When the USA; 2004.
[12] Valdes JR, Miana MJ, Nunes JL, et al. Reduced order model for estimation of
spool rotate from 0°, at a certain pressure drop, their values
fluid flow and flow forces in hydraulic proportional valves. Energy Convers
increase first then decrease with the spool rotation and Manage 2008;49:1517–29.
reach their maximum values at an angle corresponding to [13] Yuan QH. Modeling and experimental study of flow forces for unstable valve
about 1/2 of the maximum orifice opening, and at a certain design. 2003 ASME International Mechanical Engineering Congress, November
15–21. Washington, USA; 2003.
flow rate, their values increase with the spool rotation. [14] Yuan QH, Li PY. Using steady flow force for unstable valve design: modeling
(2) The direction of steady flow torques is always the direction and experiments. J Dyn Syst – Trans ASME 2002;127:451–62.
of the orifice closing. The steady flow torques serve as resis- [15] Amirante R, Vescovo GD, Lippolis A. Flow forces analysis of an open center
hydraulic directional control valve sliding spool. Energy Convers Manage
tances to the spool rotation when the orifice opening 2006;47:114–31.
increases, while impetuses to the spool rotation when the [16] Amirante R, Vescovo GD, Lippolis A. Evaluation of the flow forces on an open
orifice opening decreases. center directional control valve by means of a computational fluid dynamic
analysis. Energy Convers Manage 2006;47:1748–60.
(3) The steady flow torques in the rotary directional control valve [17] Amirante R, Moscatelli P, Catalano L. Evaluation of the flow forces on a direct
are the sums of steady flow torques in the meter-in and meter- (single stage) proportional valve by means of a computational fluid dynamic
out valve chambers. At a certain spool rotation angle, steady analysis. Energy Convers Manage 2007;48:942–53.
[18] Vescovo GD, Lippolis A. Three-dimensional analysis of flow forces on
flow torques in the meter-in and meter-out valve chambers
directional control valves. Int J Fluid Power 2003;4:15–24.
are approximately proportional to the pressure drop and the [19] Lee GS, Sung JH. Flow force analysis of a variable force solenoid valve for
second power of the flow rate through the orifice. automatic transmissions. J Fluid Eng – Trans ASME 2002;124:214–20.
[20] Lisowski E, Czyzycki W, Rajda J. Three dimensional CFD analysis and
(4) Under the same outside conditions, the values of steady flow
experimental test of flow force acting on the spool of solenoid operated
torques in the meter-out valve chamber are always larger than directional control valve. Energy Convers Manage 2013;70:220–9.
those in the meter-in valve chamber. The reason is that the [21] ANSYS/Fluent: Users guide. 13th ed.; 2011.
fluid flow further from vertical at the meter-out orifice than [22] Pan X, Wang G, Lu Z. Flow field simulation and a flow model of servo-valve
spool valve orifice. Energy Convers Manage 2011;52:3249–56.
at the meter-in orifice in the vertical surface of the spool axis. [23] Qian JY, Wei L, Jin ZJ, Wang KJ, Zhang H, Lu AL. CFD analysis on the dynamic
(5) Theoretical calculation and CFD simulation can be validated flow characteristics of the pilot-control globe valve. Energy Convers Manage
by the experimental test. The proposed calculation formulas 2014;87:220–6.
[24] Song XG, Cui L, Cao MS, Cao WP, Park Y, Dempster WM. A CFD analysis of the
can predict the steady flow torques in the servo motor oper- dynamics of a direct-operated safety relief valve mounted on a pressure vessel.
ated rotary directional control valve correctly. Energy Convers Manage 2014;81:407–19.
[25] Aung NZ, Yang QJ, Chen M, Li SJ. CFD analysis of flow forces and energy loss
characteristics in a flapper–nozzle pilot valve with different null clearances.
Energy Convers Manage 2014;83:284–95.
Acknowledgements [26] Lisowski E, Czyzycki W, Rajda J. Multifunctional four-port directional control
valve constructed from logic valves. Energy Convers Manage 2014;87:905–13.
[27] Versteeg HK, Malalasekera W. An introduction to computational fluid
The authors would like to give their acknowledgement to the
dynamics. New Jersey: Prentice Hall; 2007.
National Natural Science Foundation of China (No. 51275449), the
National Basic Research Program of China (No. 2013CB035404)
and the Funds for Creative Research Groups of China (Grant No.
51221004) for the financial support on the paper.