Available online at www.sciencedirect.

com
Available online at www.sciencedirect.com
ScienceDirect
ScienceDirect
Energy Procedia 173 (2017) 000–000
Available
Available onlineatatwww.sciencedirect.com
online www.sciencedirect.com
Energy Procedia 173 (2017) 000–000 www.elsevier.com/locate/procedia
ScienceDirect
ScienceDirect
www.elsevier.com/locate/procedia

Energy Procedia
Energy 12600
Procedia (201709) 1147–1154
(2017) 000–000
www.elsevier.com/locate/procedia
72nd Conference of the Italian Thermal Machines Engineering Association, ATI2017, 6-8
72nd Conference of the ItalianSeptember
Thermal Machines Engineering
2017, Lecce, Italy Association, ATI2017, 6-8
September 2017, Lecce, Italy
A computational model of axial piston swashplate pumps
A computational modelSymposium
The 15th International of axialonpiston swashplate
District Heating pumps
and Cooling
F. Fornarelliaa, A. Lippolisaa, P. Orestaaa, A. Posabb *
AssessingF.the feasibility
Fornarelli a of using
, A. Lippolis the heat
, P. Oresta demand-outdoor
, A. Posa *
Dipartimento di Meccanica, Matematica e Management – Politecnico di Bari
temperature function for a long-term district heat demand forecast
b

b
Marine Technology
a
Research
Dipartimento Institute –Matematica
di Meccanica, Italian National Research –Council
e Management (CNR-INSEAN),
Politecnico di Bari Roma
Marine Technology Research Institute – Italian National Research Council (CNR-INSEAN), Roma

I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc

Abstract
Abstract
a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b
Variable cdisplacement hydraulicVeolia Recherche & Innovation,
machines offer a 291 Avenue
very Dreyfousalternative
promising Daniel, 78520 andLimay, France
energy saving solution for many
Département
Variable indisplacement Systèmes Énergétiques
hydraulic et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France for many
applications mobile machines, mobilemachines
robots and offer
othera applications.
very promising In thealternative
present paperandanenergy
axial saving
piston solution
swashplate pump will
applications
be theoretically in mobile
analyzedmachines, mobileusing
and explained robotstheand other applications.
software AMESim inIn the present
order paper
to estimate theanpiston
axial friction
piston swashplate pump will
force and volumetric
be theoretically
efficiency analyzedhardworking
loss without and explained using the software
experimental tests. TheAMESim
present in order
paper is to estimate
aimed the piston
at analyzing thefriction
forces force
actingand
onvolumetric
the swash
efficiency
plate
Abstract loss without
in stationary hardworking experimental
and non-stationary conditions, in tests. The
order to present
optimize paper is aimed
the main at analyzing
design parameters theofforces acting actuators
the control on the swash
of a
plate in stationary
variable displacement andpump.
non-stationary conditions,
The behavior in order to
of the machine optimize and
is analyzed the main design
presented parameters
at different of thevelocities
angular control actuators of a
and pressure
variable
District displacement
regimes. heating networks pump.are The behavior addressed
commonly of the machinein theis literature
analyzed as andone
presented at different
of the most angular
effective velocities
solutions and pressure
for decreasing the
greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat
regimes.
©sales.
2017 Due to the changed
The Authors. Published climate conditions
by Elsevier Ltd. and building renovation policies, heat demand in the future could decrease,
©prolonging
2017 TheThe Authors. Published
the investment returnby Elsevier Ltd.
© 2017
Peer-review Authors.
under Published
responsibility ofperiod.
by Elsevier Ltd. committee of the 72nd Conference of the Italian Thermal Machines Engineering
the scientific
Peer-review under responsibility of the scientific committee of the 72nd nd Conference of the Italian Thermal Machines Engineering
The main scope
Peer-review
Association. underofresponsibility
this paper is toof assess the feasibility
the scientific committee of using
of thethe72heat demand –of
Conference outdoor temperature
the Italian Thermalfunction
Machines forEngineering
heat demand
Association
forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665
Association.
buildingsAxial
Keywords: that piston
vary in bothSwash
pump, construction periodPerformance
plate behaviour, and typology. Three weather scenarios (low, medium, high) and three district
optimization.
renovation
Keywords: scenarios
Axial wereSwash
piston pump, developed (shallow,Performance
plate behaviour, intermediate, deep). To estimate the error, obtained heat demand values were
optimization.
compared with results from a dynamic heat demand model, previously developed and validated by the authors.
1.The results showed that when only weather change is considered, the margin of error could be acceptable for some applications
INTRODUCTION
1.(the
INTRODUCTION
error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation
scenarios,
Since 50s severalthe errormanufacturers
value increasedhave up todeveloped
59.5% (depending
swash plateon thecontrol
weatheraxial
and piston
renovation
pumps scenarios combination
and several considered).
authors tried to
The
Since value of slope coefficient increased on average within the range of 3.8%
propose accurate models of performance for hydraulic piston machines, which are especially appreciated tried
50s several manufacturers have developed swash plate control axial up to
piston 8%
pumpsper decade,
and that
several corresponds
authors fortothe
the
to
decrease
propose
high power in the
accurate number
models
density of heating
ratio. of hours
performance
In 1981 of 22-139h
Zarotti during the
for hydraulic
and Nervegna [1]heating
piston seasonnon-linear
machines,
introduced (depending
which are on theforcombination
especially
terms and offlow
appreciated
force weather
for and
the
losses.
renovation
high
Several power
worksscenarios
density considered).
due toratio.
Hooke In (see,
1981Onfortheinstance,
other
Zarotti andhand,
[2])function
Nervegna intercept
[1]
are focused on increased
introduced for
of 7.8-12.7%
non-linear
the study terms per
the slippers decade
forconnecting
force and(depending
flow
the on to
losses.
piston the
coupled scenarios). The values suggested could be used to modify the function parameters
Several works due to Hooke (see, for instance, [2]) are focused on the study of the slippers connecting the piston to for the scenarios considered, and
improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
* Corresponding author. Antonio Lippolis Tel.: +39 0994733317; fax: +0-000-000-0000
Cooling.
E-mail address:author.
* Corresponding [email protected]
Antonio Lippolis Tel.: +39 0994733317; fax: +0-000-000-0000
E-mail address: [email protected]
Keywords: Heat demand; Forecast; Climate change
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review underThe
1876-6102 © 2017 responsibility of theby
Authors. Published scientific
Elsevier committee
Ltd. of the 72nd Conference of the Italian Thermal Machines Engineering
Association. under responsibility of the scientific committee of the 72nd Conference of the Italian Thermal Machines Engineering
Peer-review
Association.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 72nd Conference of the Italian Thermal Machines Engineering Association
10.1016/j.egypro.2017.08.314
1148 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154
2 F. Fornarelli et al./ Energy Procedia 173 (2017) 000–000

the swash plate. Ivantysynova investigated extensively axial pistons swashplate pumps [3] and especially the
problem of the control of the swash plate [4]. Manring [5] analyzed carefully the force components moving the
slipper away from the swash plate. The work by Heon and Hyoung [6] discussed in detail the force losses in such
class of pumps. Roccatello et al. [7] adopted the AMESim software to simulate an axial piston pump. More recent
works by Borghi et al. [8] e [9] are focused on the analysis of the motion of the slippers over the swash plate and on
the critical condition causing the slipper to move away from the swashplate, respectively.
However, unfortunately so far there are no exhaustive models able to accurately describe the behavior of swash plate
axial piston pumps throughout the whole operating range, i.e., at different rotational speeds, different delivery
pressures and especially different displacement volumes. Furthermore, in almost all cases available literature
focuses very accurately on specific details. For instance, Authors dealing carefully with mechanical details of the
problem, utilize simplified approaches to handle volumetric losses and viceversa. In the present paper a new model,
developed using AMESim software, is proposed. Our target in the present paper is to report a comprehensive
computational tool, involving some simplifications, but accurate enough to help manufacturers in designing new
axial piston pumps. Section 2 describes the geometry of the pump analyzed here and section 3 presents its
theoretical analysis. In section 4 the model developed using AMESim software is shortly presented. Sections 5 and
6, respectively, discuss the volumetric and torque losses. Finally, in section 7 charts of pump performance are
presented.

Nomenclature Qtr leakages flow rate

A surface of viscous friction R piston radius
a piston axial acceleration r radial coordinate from cylinder axis
D rotating body theoretical diameter t time variable
d piston diameter Vp piston axial velocity
Fac frictional forces associated to centrifugal forces Vtr lubricating oil velocity
Fp pressure force on the piston y vertical piston position
Fµ viscous friction forces z piston axial position
fa static friction coefficient Greek
g cylinder/piston clearance α swash plate inclination
L piston/cylinder interference in axial direction θ piston angular position
Mp piston mass θ0 angular gap between pistons
N number of pistons µ dynamic viscosity
n pump rotational speed ρ oil density
p oil pressure inside cylinders ω pump angular speed
Qm delivery flow rate

2. PUMP GEOMETRY

The simulation model reported here is general. Nonetheless, in order to provide details about the geometry of the
pump analyzed in the present paper, Fig. 1 shows pictures of some pump elements: the rotating body, the
distribution plate, a piston and a slipper connecting the piston to the swash plate. The rotating body is composed of 9
cylinders accommodating the relative pistons. However in the present work simulations with different numbers of
cylinders will be reported. The axes of the cylinders stand on a circumference of diameter D = 70 [mm], while the
axial length of each cylinder is equal to 60 [mm]. All pistons, as well as the groove within each cylinder, have a
diameter d = 17 [mm] and the clearance between them is equal to 200 [µm]; the axial length of all pistons is equal to
42 [mm]. As shown in Fig.1, each piston presents a spherical head with a diameter of 12 [mm], connected to the
cylindrical body with a smooth fillet. The overall piston length, including its spherical head, is equal to 56 [mm].
The spherical head is coupled with the brass slipper in Fig. 1, which moves over the swash plate, keeping almost
aligned with it. In contrast, the relative motion between the piston head and the slipper is over a sphere and is
characterized by an amplitude depending on the effect of the friction forces. The main rotation of the slipper relative
to the piston head is due to the rotation of the swash plate, affecting the actual pump displacement. In the present
case it is a rotational motion on a plane around an axis parallel to that of rotation of the swash plate. The spherical
 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154 1149
Fornarelli-Lippolis-Oresta-Posa/ Energy Procedia 173 (2017) 000–000 3

head is hollow and has a hole, aligned with the piston axis, which allows the oil within the cylinder to lubricate both
the region between head and slipper and that between slipper and swash plate. The lubricating oil from the axial hole
in the piston head goes into the leakage hole of the slipper, visible in the slipper on the side of the swash plate. Such
oil generates a pressure level almost constant beneath the central circular groove with a height of 0.4 [mm] and a
diameter of 11.5 [mm]. An outer circular crown with a diameter of 21 [mm] and in contact with the swash plate
creates a very thin meatus causing significant losses and a conical evolution of pressure up to the pump body.

Fig. 1 – Pictures of the pump components

In Fig. 1 also the distribution plate is represented, connecting all cylinders with the suction and delivery channels.
The distribution plate presents two port connection. The left one is composed of a single cavity, while the right one
is divided in 5 holes, in order to increase the stiffness of the plate. The left port, connected with the suction channel,
is characterized by a circumferential length of about 74 [mm], equivalent to 124°. Taking into account that the
circumferential extent of each cylinder hole, coupled with the distribution plate, is equal to 17 [mm], that is 28°,
each cylinder needs a rotation of 14°, after reaching the top dead centre, to be connected with the suction channel.
The same applies for the bottom dead centre: each cylinder loses connection with the suction channel 14° before
reaching such configuration. The 5 holes connected to the delivery channel have a circumferential length of 12 [mm]
and a distance between each others of 4 [mm]. In such a case each cylinder needs a rotation of 13° in order to
connect to the delivery channel from the bottom dead centre and the same rotation amplitude is required to each
cylinder between losing connection with the delivery channel and reaching the top dead centre.

3. THEORETICAL ANALYSIS

The offset of the temporal scale was chosen in such a way the first piston at t = 0 is horizontal, therefore its
angular position, relative to the horizontal direction, is defined as below:
θ = 2 π n t [ rad ] = 360 n t [ ° ] (1)
where n is in [Hz] and t in [s]
All other pistons have angular positions with increments of:
θ 0 = 2 π / N [ rad ] = 360 / N [ ° ] (2)
The angular position allows us to compute the vertical position of each piston axis as:
y = D / 2 sin ( θ ) = D / 2 sin ( 2 π n t + θ 0 ) (3)
The axial position of each piston, depending on the inclination of the swash plate, ranging between 0° e 18°, is
evaluated with reference to the rotation axis of the piston:
z = D / 2 sin (θ ) tan( α ) = D / 2 tan( α ) sin ( 2 π n t + θ 0 ) (4)
First and second time derivatives provide velocity and acceleration for each piston:
V p = π n D tan( α ) cos ( 2 π n t + θ 0 ) (5)
a = −2π 2
n 2 D tan( α ) sin ( 2 π n t + θ 0 ) (6)
Finally, flow-rate is given by the piston cross-section and velocity, while the force of each piston on the swash
plate is given by the piston cross-section and the oil pressure:
d 2
Qm = π Vp
4 (7)
1150 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154
4 F. Fornarelli et al./ Energy Procedia 173 (2017) 000–000

d 2
F p = pπ (8)
4

4. PUMP SIMULATION MODEL

Each piston presents the same behavior, therefore the AMESim model was developed only for one piston in a
supercomponent. Thus, the entire pump consists of the assembly of a number of supercomponents equal to the
number of pistons. In Fig.2 the scheme of a single supercomponent in the hypothesis of ideal conditions is
presented, with oil compressibility, leakages and friction forces neglected. The icon 1 stands for the number of
pistons N. The icons 2 and 3 give the output of the system, which are torque and flow-rate, respectively. The icon 4
stands for the angular velocity of the pump in [RPM]. The icon 5 represents the number of the analysed cylinder, in
the range 0÷N-1. In fig.3 a scheme of a three pistons pump is reported, where each black icon represents one of the
cylinders. The parameter in the top left area represents the angular velocity that is, indeed, constant for each
cylinder. The parameter on the left bottom area represents the number of pistons, that is equal to 3 in this case. The
only parameter linked to each supercomponent is the number of the corresponding cylinder. In particular, in this
case it goes from 0 to 2. The first piston (0) has 0° phase shift, therefore cylinders 1 and 2 have 120° and 240° phase
shift, respectively.

Fig. 2 – Ideal simulation model of a piston in AMESim.

Fig. 3 – Model of a 3-pistons pump.

5. OIL COMPRESSIBILITY AND LEAKAGE LOSSES

The actual behavior of the pump differs from the ideal one because of the friction losses, compressibility of the
oil and leakages. In this section the two latter effects are discussed, being the main ones influencing the volumetric
efficiency of the pump.
At the beginning of the delivery stroke, the volume of the cylinder decreases, causing the increase of the oil
pressure. This phenomenon is positive because the outlet pressure is very high, compared to the suction pressure and
the pressure jump cannot be too abrupt. It is convenient also because of the design requirement to avoid a sharp
passage from the suction to the outlet pressure and vice versa. The reduction of the flow rate due to the
compressibility is just one of the causes. Another effect reducing the flow rate is the leakage due to the internal gaps.
We can distinguish three main leakages: piston/cylinder, slipper/swash plate and distribution plate leakages. The
leakage due to piston/cylinder clearance is caused by two main sources. One of them is the piston movement
pushing the oil into the gap between piston and cylinder, which can be modelled by means of a linear velocity
gradient, due to the small gap, equal to zero at the cylinder surface and equal to the piston velocity at the piston
surface. The velocity is positive during the delivery phase, corresponding to the stroke from the bottom dead centre
to the top dead centre, and negative during the suction phase. The second source of leakage is due to the pressure
gradient through the gap between the internal and external regions of the cylinder. This effect is negligible during
the suction phase, but it is important during the delivery phase, when the pressure gradient is maximum. It is
possible to evaluate the flow rate associated to the linear velocity profile assuming a mean velocity equal to the
piston mean velocity:
 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154 1151
Fornarelli-Lippolis-Oresta-Posa/ Energy Procedia 173 (2017) 000–000 5

1 (9)
Q tr = π d g V p
2
where g represents the gap between cylinder and piston, d is the cylinders diameter, Vp is the piston velocity, having
a sinusoidal behavior from the bottom dead centre to the top dead centre and vice versa.
We have to take into account the parabolic profile associated to the pressure jump between internal and external
regions of the cylinder. Due to the force balance along the axial direction of a hollow cylinder with thickness dr,
considering the viscous forces on the lateral surfaces and the pressure forces on the top and bottom surfaces, we can
state that:
∂ 2 V tr ∆p (10)
=
∂r 2 µ L
where µ is the dynamic viscosity, ∆p is the pressure gap between the internal and external regions and L is the length
of the lateral surface of contact between piston and cylinder. Integrating two times equation (10), we can find the
flow rate equation:
1 ∆p g 2 (11)
Qtr = − πdg
12 µL
In the AMESim model the oil compressibility and the leakages are included as reported in fig. 4. The red part on
the left is equal to that reported in fig.2 and is utilized to determine the position and the velocity of the piston. These
data are the input of the right part of the model. The cylinder volume is simulated with an icon similar to a cylinder
and the central sphere with a label Ch. The first element with a moving cylinder has a variable volume, whereas the
sphere has a constant volume. The constant volume is about equal to the internal volume of the cylinder when the
piston is at the top dead centre position. The variable volume models the cylinder stroke.
The central sphere is linked with other two spheres in order to take into account the inlet zone (the top sphere)
and the outlet zone (the bottom sphere). The inlet zone was kept at 12 [bar] in all tests and was linked to the cylinder
volume by means of a check valve allowing flow rate at suction, but not at delivery. Also a throttling valve is
considered in order to model leakages. The pressure in the outlet zone is variable and here was chosen equal to 200
[bar]. The outlet region is connected to the cylinder volume by means of a check valve allowing the outlet, but
avoiding the inlet during the suction phase. Downstream of the check valve there is a throttling valve to simulate the
pressure losses through the distribution plate.

Fig. 4 – Simulation model in AMESim of a delivery piston. Fig. 5 – Cylinder volume performance, 0.5 [s]-5 Cycles.

In figure 5 the internal volume of the cylinder is shown, whose evolution is sinusoidal. In contrast, the inlet flow
rate is not sinusoidal, as reported in figure 6: the detail of this effect is reported for just one cycle from the top dead
centre to the bottom dead centre. The main source of this behavior is the cylinder pressure, keeping too high after
the top dead centre to allow the suction phase to start. Thus for a period slightly smaller than 5 ms, corresponding to
18°, there is no suction. For 14° the cylinder is not in communication with the inlet, but this is not a problem
because suction would not be possible. In figure 7 the outlet flow rate is provided (values are negative because
outward from the cylinder). There, leakage effects are more important. Thus, for clarity, only leakages are reported
in more detail in figure 8. Leakages are important during the outlet phase and reach 3% of the outlet flow rate.
Leakages are almost constant, whereas the flow rate is dependent on the piston velocity. Pressure within the cylinder
is rather low during the suction phase and instead high during the delivery phase, as reported in figure 9. We point
1152 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154
6 F. Fornarelli et al./ Energy Procedia 173 (2017) 000–000

that the variation between the two values is not instantaneous, because of the oil compressibility. Such phenomenon
is more obvious in figure 10, showing that oil compressibility causes a delay of 6 [ms], corresponding to 22° of
pump rotation.

Fig. 6 – Suction flow performance between top and bottom dead points. Fig. 7 – Delivery flow performance between bottom and top
dead points.

Fig. 8 – Flow loss performance, 0.5 [s]-5 Cycles Fig. 9 – Cylinder pressure performance, 0.5 [s]-5 Cycles

6. MECHANICAL PERFORMANCE

The mechanical performance depends on several components, as below:

1) The oil pressure forces inside the cylinder;
2) The inertia forces;
3) The viscous friction forces;
4) The friction forces due to centrifugal force;
5) The viscous friction forces between the slipper and the swath plate.
The oil pressure forces were reported in equation (8).

Fig. 10 – Pressure trend between suction and delivery phases. Fig. 11 – Simulation model in AMESim of a piston with friction
losses and volumetric losses.
 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154 1153
Fornarelli-Lippolis-Oresta-Posa/ Energy Procedia 173 (2017) 000–000 7

The inertia forces are calculated as the mass times the axial acceleration of the piston, where the overall mass is
composed of the piston mass and the internal piston oil mass. The total mass was estimated to be approximately 70
[g].
The viscous friction force is the resistance of the oil flow on the piston and is proportional to the velocity
derivative along the direction normal to the fluid motion. The fluid shear on the piston can be associated to two
separate phenomena: the parabolic trend due to the pressure gradient and the linear trend due to the piston velocity.
These phenomena were already presented in the last paragraph, equations (9)-(11). Using the overlapping effects
principle, the viscous friction forces can been calculated as sum of two components.
The viscous friction force due to the pressure gradient is reported below:
F = µ A
∂ V tr
= µ π d L
∆ p
g = π d g
∆ p (12)
µ 1
∂n 2 µ L 2
In the final equation dynamic viscosity µ and length L do not appear.
The viscous friction force due to the piston speed is given by the equation below:
∂ V tr V p (13)
Fµ 2 = µ A = µ π d L
∂n g
Centrifugal force tends to push the piston outward. The centrifugal force is balanced by the cylinder reaction: all
these forces act along the radial direction. The frictional forces in the axial direction, associated to the radial forces,
are expressed by:
F ac = f a M p ω 2 R = f a M p (2 π N ) R (14)
2

Fig. 11 shows the full AMESim supercomponent, including both volumetric leakages and mechanical losses. All
forces applied to the piston converge to the green element shown below. It is worth stressing the presence of
pressure transducers, measuring both the pressure inside the cylinder and the pressure in the pump body.

Fig. 12 – Piston axial force trend. Fig. 13 – Secondary components of the piston axial force.

Fig. 12 shows the trends of all force components and their sum: the blue line, below in the caption, shows the
total force. The red line is representative of the pressure force and the green line is representative of the viscous
friction force due to the pressure gradient, while the other force components are obviously less significant. The other
force components are shown enlarged in Fig. 13. The red line represents the viscous friction force due to the
pressure gradient, already visible in Fig. 12. This force is reported during the suction phase, while the delivery phase
is outside the figure range. The green line represents the viscous friction force due to the piston motion, whose
evolution is strictly a sine-wave, as that of the piston speed. In figure this force is always less than 1 [N], but it
should be noted that this loss is more important at high speeds, so at the maximum pump speed, 3000 [rpm], it
would be 5 times higher. The blue line shows the inertia forces. It is worth noting that the inertia forces are
proportional to the square of velocity and they do not contribute to the pump power. Indeed the energy accumulated
during acceleration is returned during deceleration. The friction forces due to the centrifugal force are represented
by the pink line. They are constant and approximately equal to 1 [N]. Also such forces are proportional to the square
of the speed and would reach 25 [N] at the maximum pump speeds.
1154 A. Posa et al. / Energy Procedia 126 (201709) 1147–1154
8 F. Fornarelli et al./ Energy Procedia 173 (2017) 000–000

7. PUMP PERFORMANCE

The AMESim supercomponents allow computing pump performance as a function of pistons number, rotational
speed and operating pressure. Fig. 14 shows the pump volumetric performance as function of rotational speed for
three different values of operating pressure, 30, 150 and 300 [bar], respectively. The same pump volumetric
performance is shown for three different rotational speeds, 300, 1500 and 3000 [RPM], respectively, in Fig. 15.

Fig. 14 – Volumetric efficiency diagram for 3 different pressure tests. Fig. 15 – Volumetric efficiency diagram for 3 different speed tests.

8. SUMMARY

The present work analyzes the main phenomena affecting the volumetric and mechanical performance of a swash
plate axial piston pump. Therefore, the methodology reported here can be considered a practical tool in order to
reduce the considerable cost of testing before manufacturing. Nonetheless a final stage of comparison between
simulations and experiments is still recommended before finalizing the new design.

REFERENCES
[1] Zarotti, G.L. and Nervegna, N. – “Pump efficiencies approximation and modelling” - Int Fluid Power Symp, 6th;
Cambridge, Engl; 8-11 April 1981, Pages 145-164
[2] Koç, E. and Hooke, C.J. - “Investigation Into the Effects of Orifice Size, Offset and Overclamp Ratio on the
Lubrication of Slipper Bearings”, Tribology International, Elsevier Science, Vol. 29, No. 4, pp. 299 – 305, 1996.
[3] Ivantysyn, J. and Ivantysynova, M. – “Hydrostatic Pumps and Motors – Principles, Designs, Performance,
Modelling, Analysis, Controll and Testing” – Academia Books International, New Delhi, 2000.
[4] Grabbel, J. and Ivantysynova, M. – “An investigation of swash plate control concepts for displacement
controlled actuators” – International Journal of fluid Power, vol. 6, N. 2, pagg. 19-36, 2005.
[5] Manring, N. D., “Predicting the Required Slipper Hold-Down Force an Axial-Piston Swash Plate Type
Hydrostatic Pump”, Proc. of 2001 ASME Int. Mechanical Engineering Congr. and Exp., Novemb. 11-16, 2001, NY.
[6] Heon, S. J. and Hyoung, E. K. – “On the Istantanius and Average Piston Friction of Swash Plate Type Hydraulic
Axial Piston Machines” - KSME International Journal, Vol. 18 No. 10, pp. 1700-1711, 2004
[7] Roccatello, A., Mancó, S., Nervegna, N. - "Modelling a variable displacement axial piston pump in a multibody
simulation environment" – J. of Dynamic Systems, Measurement and Control, Trans. of the ASME, 129 (4), 2007.
[8] Borghi, M., Specchia, E., Zardin, B. - “Numerical Analysis of the Dynamic Behaviour of Axial Piston Pumps
and Motors Slipper Bearings”, 2009-01-1820, Accepted for Publication in the SAE 2009 International Powertrains,
Fuels and Lubricants Meeting, June 15-17, 2009, Florence.
[9] Borghi, M., Specchia, E., Zardin, B., Corradini, E. - "The critical speed of slipper bearings in axial piston swash
plate type pumps and motors" - (2010) Proc. of the ASME Dynamic Systems and Control Conf. 2009, DSCC2009.