Barrel tipping in axial piston pumps and motors

Peter Achten, Sjoerd Eggenkamp

INNAS BV, Nikkelstr. 15, 4823AE Breda, the Netherlands

E-mail: pachten@innas.com, seggenkamp@innas.com

Abstract
In axial piston pumps and motors, the interface between the cylinder block and the port plate is
one of the most complicated design details. This interface is a combination of a thrust bearing
and a face sealing. The gap height, i.e. the thickness of the oil film, is only allowed to be a few
micrometers. A smaller gap height results in high friction losses and possibly in metal-to-metal
contact, causing wear and particle generation. On the other hand, a large gap height causes high
leakage. In most cases, the barrel or cylinder block will be tipping instead of lifting, thereby
causing a variable gap height and combining both high friction losses (where the gap height is
small) and high volumetric losses (where the gap height is large). This paper investigates the
factors that determine the barrel force and torque balance. Two axial piston pump principles are
investigated and compared: a swash plate, slipper type machine and a floating cup pump/motor.

Keywords: Axial piston pump, barrel tipping

1. Introduction the barrel from the port plate, thereby allowing the barrel to
move a few micrometers in the z-direction, and rotate or tip
Positive displacement machines (by definition) have sliding around the x- and y-axis. In the end, these small micro mo-
interfaces: machine components sliding on each other, with tions determine the volumetric and mechanical losses of the
a thin oil film separating the two components. Some exam- interface between the barrel and the port plate, but also
ples are, pistons moving up and down in cylinders, piston wear, durability and stability of the pump or motor opera-
slippers running on swash plates and gears running in gear tion.
pumps. These interfaces often perform a combination of a
bearing and a sealing function. The gap height (i.e. the
height of these oil films) needs to be small, but preferably
not too small, mostly in the order of a few micrometers. This
poses a challenging task for the designer, especially since
thermal expansion, pressure deformation and production
tolerances can have a substantial effect on the size and
geometry of these gaps.

An often-used design principle and solution to this problem

is to allow the sliding components to find their own relative
position. This solution is used in axial piston machines, for
instance, in the sliding interface between the rotating barrel
and the stationary port plate, in which the cylinder block is
free to find its own position on the port plate (Figure 1). The
barrel is connected to the shaft by means of a spline, on
which, to some extend, it is allowed to move in the z-direc-
tion, and rotate around the x- and y-axis in point A. A spring
(not shown in Figure 1) is mounted which pushes the barrel !
to the surface of the port plate after assembly. Furthermore,
the sealing lands of the barrel ports are dimensioned as such Fig. 1: Barrel and port plate of an axial piston, slipper type
that there is a remaining hydrostatic force pushing the barrel machine
even harder to the port plate when the pressure level in-
creases. In bent axis machines (Figure 2), the tipping around the x-
and y-axis is often prevented by means of a central pin. This
When being positioned on the port plate, it seems that the is possible, since bent axis machines don’t have a through
only remaining degree of freedom is the rotation around the drive; the drive shaft stops at the drive flange. Moreover, the
large tilt angle of the barrel creates a machine having a rela-
z-axis. However, on a micro scale, the oil film still separates

Non-reviewed paper. 381 SICFP2017 7-9 June 2017

Linköping, Sweden
tively long piston stroke and a small barrel diameter. The chines, and even smaller than of slipper type machines. The
small diameter of the barrel and the barrel ports is important small tilt angle results in a multi-piston design, having two
since the pin reduces the demand for perpendicularity to the rings of pistons, arranged in a back-to-back configuration.
surface of the port plate.
Like in slipper type machines, the shaft is going through the
entire length of the machine. Another similarity with the
central pin to counteract barrel tipping slipper type machine is the relatively large diameter of the
barrels. The barrels are allowed to rotate around the pivot
points, indicated with A in Figure 4.

port plates
barrels

Fig. 2: Cross section of a bent axis pump or motor

Slipper type pumps and motors (Figure 2) have a shaft going

through the entire length of the machine. This has the advan-
tage that a second machine can be mounted on the back of
the first, but it eliminates the opportunity to apply a central
barrel pin. Furthermore, due to the limited swash angle
(which is about half of the tilt angle of bent axis machines) ! pivot points A
the pistons have a shorter stroke and the barrel diameter is
larger, which makes it harder to accomplish the perpendicu- Fig. 4: Cross section of a floating cup pump or motor
larity between the pin and the sealing area of the barrel
ports. As a consequence, the solution of a central barrel pin The floating cup design belongs to the family of axial piston
is not used in slipper type machines, and the barrel is free to principles: the barrels are free to find its position on the port
tip around the x- and y-axis shown in Fig. 1. plates, thereby being positioned after assembly by a barrel
spring. And, just like in other axial piston designs, the force
barrel and torque balancing is a difficult design challenge.
barrel spring
port plate This paper discusses the factors that influence the tipping of
the barrel in axial piston machines, in particular the amount
of spring force which is needed to avoid barrel tipping at
high revolutionary speeds and low operating pressures. The
paper will only discuss slipper type and floating cup pumps
and motors.

2. Tipping torque research

There are many different loads acting on the barrel, which
result in a torque load around the x- and y-axis of the barrel:

• Hydrostatic pressure forces inside the cylinder;

! • Hydrostatic pressure forces acting on the barrel ports
and sealing lands;
Fig. 3: Cross section of a slipper type pump or motor • Centrifugal forces;
• Friction forces between the pistons and the cylinders;
Floating cup pumps and motors (Figure 4) have a different • Friction between the barrel and the port plate;
construction, in which the pistons are press-fitted into the • Lateral piston forces and torques;
rotor. Instead, the barrel cylinders have been isolated from • Force of the axial barrel spring;
the barrel. Each piston is combined with a cup-like cylinder • Impulse forces of the oil flow.
which is floating on the barrel plate. Like in the bent axis
design, the barrel has a tilted position relative to the main The hydrostatic forces from the sealing lands are dependent
axis. But the tilt angle is much smaller than in bent axis ma- on the pressure profile in the gap between the barrel and the

Non-reviewed paper. 382 SICFP2017 7-9 June 2017

Linköping, Sweden
port plate. The pressure profile is strongly influenced by Yamaguchi [5, 6] was one of the first to measure the gap
elasto-hydrodynamic and thermal deformations, as well as height and tipping of a barrel-like structure running on a
by the variation of the viscosity of the oil, while passing the valve plate. Other experimental research in this area has
gap, due to temperature and pressure variations [1-4]. The been performed by Ivantysynova and her team [3, 4, 14-16,
gap profile is also influenced by the precise barrel position 20], Bergada, et al [21-23], Wegner, et al [24, 25], Shin, et al
on the valve plate, and hence by the tipping itself. [17], Han, et al [26] and Zhang, et al [27]. Most of these
studies measured a gap height between 0 and 20 µm. The
Many of the underlying physical phenomena are, up to now, experiments also showed a tilted barrel position, creating a
still not completely understood. The first modeling attempts wedge shaped fluid film with a varying gap height.
started in the early 80’s [5-7]. These first models were based
on rigid-body models [8, 9]. By the turn of the century, the Extensive measurements were performed by Bräckelmann
increased capacity and performance of computers allowed [18] on a relatively small 40 cc swash plate pump. The gap
the addition of elasto-hydrodynamic (EHD) effects [10-12]. height was measured to be 2 to 14 µm, depending on the
At first, only the deformation of the barrel and the port plate rotational speed and the oil pressure. Furthermore, Bräckel-
was calculated. In 2006 [13], it became clear that the defor- mann measured a tipping of the barrel of up to 0.07°. The
mation of the entire housing strongly affected the bearing tipping angle proved to be about linearly dependent of the
interface between the barrel and the port plate. Most recent rotational speed. Furthermore, a larger swash angle and a
simulations also include the thermal expansion and deforma- higher pressure level also increased the barrel tipping.
tion of the components [3, 14, 15], and the non-flatness and
micro-geometry of the bearing surfaces [16, 17]. The calcu- The influence of the rotational speed on the barrel tipping
lation of the dynamic micro-motion of the cylinder block was also investigated by Noah Manring [28-30]. According
has become a complex, multi-domain simulation effort, to Manring, ‘Tipping the cylinder block within an axial-pis-
combining EHD with CFD-analysis, thermal analysis and ton swash-plate type hydrostatic machine is a phenomenon
multi-body mechanical analysis. Yet, despite this enormous that results in a momentary and sometimes permanent fail-
progress in simulation techniques, the theoretical analysis ure of the machine’ In his paper from 2000 [29] he presented
does not yet include all relevant aspects, like production a design criterion for the required force of the central spring
tolerances, surface roughness and wear. of the cylinder block:

⋅ tan (α ) ⋅ (1+ tan 2 (α ))

Aside from the analytical complexity, it is almost impossible
N ⋅ M ⋅ r 2 ⋅ ω max
2
to get a direct, experimental verification of the analytical ! Fsp > (1)
results. Often, separate test benches are built, in which com- 2R
ponents, like the cylinder block, the pistons and the port
plate, are isolated from the rest of the pump. Figure 5 shows in which Fsp is the spring force, N is the number of pistons,
an example of such a test bench, as was developed by M is the mass of piston and slipper, ωmax is the maximum
Bräckelmann [18, 19]. Consequently, important effects, like rotational speed, r is the piston pitch radius, R is the outer-
the deformation of the housing, are neglected. Other factors, most radial point of contact on the cylinder block, and α is
like the bending of the main shaft, are often enlarged, since the swash plate angle.
a special shaft is often constructed for these benches.

Fig. 5: Test bench for measuring the interface between the barrel and port plate of a swash plate pump [18]

Non-reviewed paper. 383 SICFP2017 7-9 June 2017

Linköping, Sweden
In the paper from 2014 [30], Manring, et al, adapted this on these components. Only two pistons are shown, whereas
equation: in reality most swash plate machines have nine pistons. Fig-
ure 6 is also a simplification since all vectors are only shown
2 ⋅ Fsp ⋅ R in a two-dimensional representation, whereas, some of these
!ω 2 < (2) forces also have a component in the third, missing dimen-
( )
N M p + M s r 2 ⋅ tan (α ) sion.

Mp and Ms are the mass of the piston, respectively of the A model has been developed in which all these forces and
slipper. This equation can be rewritten as: components are combined. The model does not include the
slipper friction Fsf. According to Bräckelmann [18], this
force is negligible compared to the hydrostatic lateral piston
! Fsp > N ⋅ M ⋅ r ⋅ ω max ⋅ tan (α )
2 2
(3) load.
2R
In swash plate machines, the pistons are severely loaded due
which is essentially different from his earlier criterion to the kinematic principle. The high lateral loads in the con-
(equation (1)). Manring does not offer an explanation for tact between the pistons and the cylinders cause substantial
this difference. Manring also simplifies the reaction loads friction loads on the barrel [31].
from the pistons on the cylinder block. As a result of the
angled position of the slipper on the swash plate, the slipper The dimensions and design parameters of the A4FM28/A4-
creates a lateral load on the piston. This load results in a FO28 have been determined by means of disassembling an
torque load on the cylinder block. Manring seems to ignore existing machine and measurement of the dimensions,
this torque load. He also neglects the friction between the weights and other relevant parameters (Table 1). The force
pistons and the cylinders, which also create a tipping torque of the barrel spring has been measured. The data about the
on the barrel. Finally, he does not include the friction be- friction between the barrel and the valve plate are taken
tween the barrel and the valve plate and the friction between from the thesis of Bräckelmann [18]. The commutation in
the slippers and the swash plate. Due to the variation of the the top and bottom dead centers is assumed to be instanta-
thickness of the oil film in between the barrel and the valve neous.
plate, the friction force will also vary around the circumfer-
ence of the barrel sealing lands, which will cause another Table 1: A4FM28/A4FO28 parameters
tipping torque. Also, the fluid film underneath the slippers
varies in height.
geometric displacement 28 cc/rev
This paper presents a new analysis of the torque loads acting
on the cylinder block. The aim of the study is to determine Number of pistons 9 -
the axial force of the cylinder block that is required to pre-
vent the barrel from tipping. Instead of detailed deformation Swash angle 19.2 °
analysis and CFD-calculations of the bearing gaps, simple
friction models will be used with constant friction coeffi- Piston diameter 13.5 mm
cients. Following Manring’s analysis, the analysis will be
based on a simple rigid-body analysis, and will therefore not Piston and slipper mass 56.2 g
include any elasto-hydrodynamic deformations, thermal
expansions, or detailed CFD-analysis. However, unlike Piston length 45 mm
Manring’s analysis, this study also includes the torque loads
of the pistons on the barrel, as well as the friction between Piston pitch radius 31.2 mm
the pistons and the cylinders.
Port plate outer edge radius 28.45 mm
Two machines will be considered and compared: friction coefficient between the barrel
0.1 -
and the valve plate
• A swash plate type constant displacement pump/motor friction coefficient between the piston
from Bosch Rexroth (A4FO28 or A4FM28); 0.1 -
and the cylinder
• A floating cup constant displacement pump/motor from
INNAS. Barrel spring 41 N/mm

Both machines have a geometric displacement of 28 cc per compression of the barrel spring 9 mm
revolution.
force barrel spring 369 N
3. Swash plate pump or motor
Figure 6 shows the pistons, the barrel and the barrel spring
as isolated components, including some of the loads acting

Non-reviewed paper. 384 SICFP2017 7-9 June 2017

Linköping, Sweden
!

Fig. 6: Loads acting on the pistons, the cylinder block and the barrel spring
of a swash plate type pump or motor, showing only 2 pistons,
assuming only a pressure load on the upper piston

Fp hydrostatic force acting on the piston

Fsl Hydrostatic load of the slipper
Fc Centrifugal force of the piston and slipper
Ff Friction force between the piston and the cylinder
Fsf Friction between the slipper and the swash plate
Fsb Friction between the barrel and the valve plate
Fr,1, Fr,2, Fr,3, Fr,4 Reaction force, resulting from the contact between the piston
and the cylinder, acting on the piston
Fp,1, Fp,2, Fp,3, Fp,4 Reaction force, resulting from the contact between the piston
and the cylinder, acting on the cylinder
Fop,1 Hydrostatic force acting on the bottom of the hydraulic cylinder
Fop,2 Hydrostatic force acting on the barrel ports and sealing lands
Fsp Force of the central barrel spring
Fr,5 Reaction force between the barrel and the port plate to
counteract the tipping torque
Fr,6 Sum of all radial piston loads acting on the spline of the barrel

The parameters from Table 1 can be substituted in Equation In order to include the pressure load, a new simulation
2 to calculate the maximum rotational speed before, accord- model has been made, which includes all the loads shown in
ing to Manring, et al [30], the cylinder block starts to tip: Figure 6. The coordinate system for these calculations is
defined in figure 7. The torque load acting on the barrel will
! nmax = 3342 rpm (4) make the reaction force move out of the centre. If this force
moves outside the circle with radius R, the barrel will tip.
Equation 2, however, does not contain the pump pressure as
a parameter influencing the tipping torque. But, it is certain
that a higher pump pressure creates a higher torque load of
the pistons on the cylinders of the cylinder block, and, in
addition, a higher friction torque.

Non-reviewed paper. 385 SICFP2017 7-9 June 2017

Linköping, Sweden
 100 bar
200 bar
300 bar
400 bar

m
m
5
.4

y [mm]
28

1000 rpm

4000 rpm

! x [mm]

a) Calculated for a hydrostatic axial force balance of 100%

#

Fig. 7: Top view and cross section of the valve plate of the
A4FO28-pump. The red dotted line shows the outmost ra- 100 bar
dius to support the cylinder block against tipping. 200 bar
300 bar
In a first calculation, the hydrostatic axial force balance is 400 bar
assumed to be 100%: the size of the seal lands is chosen as
such that the counteracting hydrostatic force is exactly equal
to the hydrostatic force, which is pushing the barrel towards m
m
the valve plate. In this theoretical situation, only the barrel 5
.4
28
spring is pushing the cylinder block to the valve plate. The
calculated position of the reaction force is displayed in Fig-
y [mm]

ure 8, for two different operating speeds and four different

pump pressures. 1000 rpm

The simulation results (Figure 8a) show the strong influence

of the pump pressure on the tipping torque. Actually, the
barrel will tip at almost all operating conditions. Only for
low operating pressures, below 100 bar, the pump can be
operated up till about 3000 rpm, without tipping of the bar- 4000 rpm
rel.

In a second simulation, the hydrostatic, axial balance has

been changed to 93%. In other words, 7% is no longer com- ! x [mm]
pensated by the hydrostatic pressure field in between the
barrel and the port plate. This corresponds with the actual
dimensions of the components of the A4FO28. b) Calculated for a hydrostatic axial force balance of 93%

This analysis (Figure 8b) shows that a hydrostatic force is

needed to prevent the barrel from tipping. At a pressure lev-
el of 400 bar, this hydrostatic barrel force has an average Fig. 8: Position of the axial reaction force, necessary to
value of 1850 N. The strong spring force is needed to pre- counteract the tipping torque, calculated for te slipper type
vent the barrel from tipping at low pump pressures in com- machine for two different hydrostatic balances. The dotted
bination with a high rotational speed. red circle is the outer edge of the port plate to support the
barrel. If the reaction force is outside this circle, the barrel
will tip.

Non-reviewed paper. 386 SICFP2017 7-9 June 2017

Linköping, Sweden
4. Floating cup pump or motor mains to push the barrel towards the port plate. Compared
to the slipper type pump, the barrel spring force relatively
A floating cup machine is essentially different from a bent low, being about 27% of the spring force in the slipper type
axis or a slipper type pump or motor. Figure 10 shows a pump. This force is also counteracted by the new hydrostat-
cross section of the main rotating parts. Figure 11 shows the ic bearing.
main loads acting on the barrel. Table 2 gives the main pa-
rameters of the 28 cc floating cup pump or motor. The cups make a small relative movement on the barrel
plate [34]. Consequently, the position of the hydrostatic cup
Table 2: 28 cc FC-pump or motor parameters force also moves relative to the compensating hydrostatic
force of the barrel port and the corresponding seal lands.
28 cc/rev This results in an additional torque load on the barrel, which
geometric displacement
is pressure dependent.
Number of pistons 24 -
Figure 9 shows the calculated position of the resulting bar-
Swash angle 8 ° rel force, which is needed to counteract the tipping torque of
the barrel. The force trajectories are calculated for two dif-
Piston diameter 12.5 mm ferent rotational speeds and 4 different pressure levels. The
calculation is performed assuming a constant pressure in the
Piston and slipper mass 5.6 g pockets of the seal lands [33]. In reality, the pressure in the
pockets varies depending on the local gap height. As a re-
Piston pitch radius 34 mm sult, the pockets create an additional torque load on the bar-
rel, which counteracts the barrel tipping.
Port plate outer edge radius 34.28 mm
friction coefficient between the barrel
0.1 - 100 bar
and the valve plate
200 bar
friction coefficient between the piston
0.1 - 300 bar
and the cylinder
400 bar
force barrel spring 100 N 4000 rpm 1000 rpm

m
m
8
.2
34
In floating cup machines, the pistons are press-fitted into the
rotor, and can therefore not create a centrifugal load on the
y [mm]

barrel. Instead, the cups are free to move and rotate around
the piston crown. The position of the centre of gravity of the
cup mostly differs from the midpoint of the piston crown,
and consequently the cups will create a centrifugal torque
load on the barrel. This torque load is counteracted by the
forces Fr,1 and Fr,2. The number of cups per barrel (12 in
this case) is larger than the number of pistons in most axial
piston machines. Also the piston pitch radius is somewhat
larger. Both factors increase the centrifugal load on the bar-
rel. On the other hand, the weight of the cups is much
smaller, about 10% of the piston of the axial piston pump.
Also the stroke length is much smaller. In total, the cen- ! x [mm]
trifugal load on the barrel in the floating cup machine is
93% smaller than in the slipper type machine.
Fig. 9: Position of the axial reaction force, necessary to
counteract the tipping torque, calculated for the 28 cc float-
Another difference concerns the piston friction. Due to the ing cup pump or motor. The dotted red circle is the outer
hydrostatic balanced design of the cup, there is almost no edge of the barrel sealing lands to support the barrel. If the
friction between the pistons and the cups [32]. The only
reaction force is outside this circle, the barrel will tip.
friction is caused by the centrifugal force of the cup and its
oil contents. This (small) friction force has been included in
the calculation of the barrel tipping torque.
The calculation shows that the barrel in the floating cup
The hydrostatic balance of the barrel is also essentially dif- machine does not tip at 4000 rpm, even at high operating
ferent. A new hydrostatic bearing is applied in the floating pressures. All curves stay within the outer diameter of the
cup machine [33]. The new hydrostatic bearing creates a largest seal land.
variable, gap height dependent pressure load in the sealing
lands of each barrel port. In principle it creates a perfect
axial hydrostatic balance. Only the barrel spring force re-

Non-reviewed paper. 387 SICFP2017 7-9 June 2017

Linköping, Sweden
 !

Fig. 10: Cross section of the rotating parts of a floating cup pump

Fig. 11: Loads acting on the cups, the cylinder block and the barrel spring
of a floating cup pump or motor, showing only 2 pistons,
assuming only a pressure load on the upper piston

Fcup axial hydrostatic force created by the cup

Fc Centrifugal force of a cup
Fr,1, Fr,2 reaction force to counteract the centrifugal torque of the cup
Fsp Force of the barrel spring
Fbp hydrostatic force create by the barrel port and seal lands
Fbr Reaction force of the barrel to counteract the tipping torque

Non-reviewed paper. 388 SICFP2017 7-9 June 2017

Linköping, Sweden
5. Comparison of the tipping torque The calculation does not include any hydrodynamic effects,
nor any elasto-hydrodynamic-deformation which could in-
The simulations show that the tipping torque is strongly fluence the gap geometry and the bearing capacity. The cal-
influenced by the design principle. For both design princi- culated force is therefore not necessarily a contact force but
ples, the total tipping torque load on the barrel is given in can be regarded as the force that needs to be overcome by
Table 3, as is calculated for 16 different operating condi- for instance hydrodynamic effects.
tions. On average, the tipping torque of the floating cup de-
sign is only 3.3% of the barrel tipping torque of a slipper 8
type design. The strong reduction is the result of a combina- 30 bar
tion of improvements. Most important are, the reduction of 50 bar
100 bar
the centrifugal forces by about 93% and the elimination of 150 bar
the lateral loads between the pistons and the cylinders. 6

barrel friction torque [Nm]

200 bar

Table 3: Calculated average tipping torque in Nm

for a 28 cc slipper type (ST) and floating cup machine (FC) 4

1000 rpm 2000 rpm 3000 rpm 4000 rpm

p [bar] 2
ST FC ST FC ST FC ST FC

100 bar 14.3 0.37 15.2 0.43 17.6 0.70 22.1 1.21
0
0 500 1000 1500 2000 2500 3000
200 bar 28.7 0.75 29.3 0.84 31.0 1.06 34.3 1.48
! n [rpm]

300 bar 43.0 1.13 43.6 1.22 45.0 1.41 47.8 1.78
Fig. 13: Friction torque between the barrel and the port
400 bar 57.4 1.50 57.9 1.59 59.2 1.77 61.6 2.11 plate, measured for a 56 cc slipper type machine

The hydrodynamic lubrication is demonstrated by Bräckel-

6. Contact force between the barrel and the mann [18] by building a separate test bench to measure the
friction torque at various operating conditions (Figure 5).
valve plate Figure 13 shows the measured friction torque for a 56 cc
The slipper type pump needs a strong axial barrel force to slipper type machine. The measurements are consistent with
keep the barrel from tipping. Even a relatively strong barrel the Stribeck-effect, showing mixed lubrication conditions
spring is not sufficient to avoid barrel tipping (see Figure and high friction at low operating speeds. For operating
9a). Only when adding a substantial hydrostatic force, the speeds of 500 rpm and higher, the friction losses are more or
unit can be operated to rotational speeds of about 3000 rpm. less linear with the operating speed. Nevertheless, the opera-
ting pressure creates an additional friction torque, indicating
coulomb friction at these conditions.
For this last situation, the calculated contact force between
the barrel and the valve plate is displayed in the diagram of In the floating cup machine, the axial load between the bar-
Figure 12. The diagram shows the calculated contact force rel and the port plate is almost negligible compared to the
for four different pressure levels. Due to the simplified slipper type. As a consequence, the floating cup principle
model of the instantaneous commutation, the calculated does not have as much friction losses between the barrel and
force is not influenced by the rotational speed. the valve plate, especially at startup and low operating
speeds.

7. Conclusions
In axial piston machines, the operating conditions can be
constrained because of tipping of the cylinder block or bar-
rel. Manring [29, 30] has already analysed the relationship
between operating speed and barrel tipping. However, in
these publications, the torque balance on the cylinder block
is strongly simplified, and does not include pressure depen-
dent loads.

This paper has shown, that the operating pressure has a

stronger influence on the barrel tipping than the operating
!
speed. In order to avoid tipping in slipper type machines, the
barrel and port plate geometry needs to be dimensioned as
Fig. 12: Calculated contact force between the barrel and the such that the barrel is not completely balanced in the axial
valve plate direction. Of the total hydrostatic force exerted by the axial
hydrostatic force, only 93% can be compensated by the bar-

Non-reviewed paper. 389 SICFP2017 7-9 June 2017

Linköping, Sweden
rel ports and seal lands. The remaining 7% is needed to suf- 7. Yamaguchi, A., et al., Bearing/Seal Characteristics of
ficiently press the barrel against the port plate, thereby coun- the Film between a Valve Plate and a Cylinder Block of
teracting the tipping torque created by the pistons. Neverthe- Axial Piston Pumps (3rd Report- Effects of Fluid Types
less, despite the strong axial force, the maximum rotational and Theoretical Discussions). Hydraulics & Pneumat-
speed is constraint by the high centrifugal forces and the ics, 1987. 18(7): p. 543-550.
strong tipping torque created by the pistons.
8. Wieczorek, U., Simulation of the gap flow in sealing
In this respect, the floating cup principle performs much and bearing gaps of axial piston machines. Proc. of 1st
better. The centrifugal load is decimated due to the low cup FPNI-PhD Symposium Hamburg, 2000.
mass and the short stroke. Furthermore there are hardly any
other loads created by the piston-cup-interface. The total 9. Wieczorek, U. and M. Ivantysynova, Computer aided
tipping torque is reduced by about 97% compared to the optimization of bearing and sealing gaps in hydrostatic
slipper type machine. The force of the central barrel spring machines - the simulation tool CASPAR. International
can be reduced by more than 70%. Nonetheless, floating cup Journal of Fluid Power, 2002. 3(1).
pumps and motors can be operated at higher operating
speeds and pressures without any barrel tipping. 10. Ivantysynova, M., A new approach to the design of
sealing and bearing gaps of displacement machines.
This paper also shows that, in slipper type machines, barrel Proc. 4th Int Symposium on Fluid Power, 1999.
tipping is related to the efficiency of the pump or motor. In
order to prevent barrel tipping, the barrel needs to be pushed 11. Huang, C. and M. Ivantysynova, A new approach to
stronger in the direction of the valve plate, which increases predict the load carrying ability of the gap between
the friction and reduces the efficiency. In the floating cup valve plate and cylinder block. Proc. Bath Workshop on
principle, this can be avoided completely. The reduced axial Power Transmission & Motion Control, PTMC 2003,
force load and tipping torque results in a strong reduction of Bath, UK, 2003.
the friction losses between the cylinder block and the port
plate. This results in an increased efficiency and higher 12. Ivantysynova, M., Prediction of pump and motor per-
startup torque. formance by computer simulation. 1st Int. Conf. on
Computational Methods in Fluid Power Technology,
References Melbourne, Australia, 2003, 2003.
1. Pelosi, M. and M. Ivantysynova, The influence of pres- 13. Achten, P.A.J. and M.P.A. Schellekens, Deformation
sure and thermal deformation on the piston/cylinder effects on the load carrying capacity of the barrel bear-
interface film thickness. In: Proceedings of the 52nd ing in axial piston pumps and motors. Proc. IME-
National Conference on Fluid Power 2011, NCFP, CE2006, 2006 ASME International Mechanical Engi-
2011. I11-9.3. neering Congress and Expo November 5-10, 2006,
Chicago, Illinois, USA, 2006. IMECE2006-13223.
2. Pelosi, M. and M. Ivantysynova, Heat Transfer and
Thermal Elastic Deformation Analysis on the 14. Schenk, A., M. Zecchi, and M. Ivantysynova, Accurate
Piston:Cylinder Interface of Axial Piston Machines. Prediction of Axial Piston Machine’s Performance
Journal of Tribology, 2012. 134(041101): p. 1-15. Through a Thermo-Elasto-Hydrodynamic Simulation
Model. Proc. of the ASME/BATH 2013 Symposium on
3. Zecchi, M., A novel fluid structure interaction and Fluid Power & Motion Control, FPMC2013, October
thermal model to predict the cylinder block/valve plate 6-9, 2013, Sarasota, Florida, USA, 2013.
interface performance in swash plate type axial piston
machines. Dissertation Purdue University, 2013. 15. Chacon, R. and M. Ivantysynova, An investigation of
the impact of micro surface on the cylinder block/valve
4. Zecchi, M. and M. Ivantysynova, Spherical valve plate plate interface performance. Proc. of the 8th FPNI
design in axial piston machines - A novel thermo-elas- Ph.D Symposium on Fluid Power June 11-13, 2014,
to-hydrodynamic model to predict the lubricating inter- Lappeenranta, Finland, 2014(FPNI2014-7837-0).
face performance. The 8th International Conference on
Fluid Power Transmission and Control (ICFP 2013), 16. Ivantysynova, M. and J. Baker, Power Loss in the Lu-
Apr. 9-11, 2013, Hangzhou, China, 2013: p. 325 - 329. bricating Gap between Cylinder Block and Valve Plate
of Swash Plate Type Axial Piston Machines. In-
5. Yamaguchi, A. and M. Tsuchimoto, Bearing Seal ternational Journal of Fluid Power, 2009. 10(2): p.
Characteristics of the Oil Film between a Valve Plate 29-43.
and a Cylinderblock of Axial Piston Pumps. Hydraulics
& Pneumatics, 1982. 13(1): p. 55-60. 17. Shin, J.-H. and K.-W. Kim, Effect of surface non-flat-
ness on the lubrication characteristics in the valve part
6. Yamaguchi, A., et al., Characteristics of Fluid Films of a swash-plate type axial piston pump. Meccanica,
Between a Valve Plate and a Cylinder Block of Axial 2014. 49(5): p. 1275-1295.
Piston Pumps and Motors. Hydraulics & Pneumatics,
1984. 15(4): p. 64-72.

Non-reviewed paper. 390 SICFP2017 7-9 June 2017

Linköping, Sweden
18. Bräckelmann, U., Reibung, Steifigkeit und Dämpfung 30. Manring, N.D., et al., Scaling the Speed Limitations for
in Schrägscheiben-Axialkolbenpumpen und -motoren. Axial-Piston Swash-Plate Type Hydrostatic Machines.
Thesis Ruhr-Universität Bochum, 2006. Journal of Dynamic Systems, Measurement, and Con-
trol, 2014. 136(3): p. 031004.
19. Bräckelmann, U. and F. Jarchow, Reibung, Steifigkeit
und Dämpfung in Schrägscheiben-Axialkolbenpumpen 31. Jeong, H.-S. and H.-E. Kim, On the instantaneous and
und –motoren. Wissensportal Baumaschine.de, 2006. average piston friction of swash plate type hydraulic
axial piston machines. KSME International Journal,
20. Pelosi, M., M. Zecchi, and M. Ivantysynova, A fully 2004. 18(10): p. 1700-1711.
coupled thermo-elastic model for the rotating kit of
axial piston machines. Proc. Bath ASME Symposium 32. Achten, P., T.v.d. Brink, and M. Schellekens, Design of
on Fluid Power and Motion Control FPMC, 2010: p. a variable displacement floating cup pump. Proc.
217 - 234. SICFP’05, June 1-3, 2005, Linköping, Sweden, 2005.

21. Bergada, J.M., J. Watton, and S. Kumar, Pressure, 33. Achten, P.A.J., T.L.v.d. Brink, and G.E.M. Vael, A ro-
Flow, Force, and Torque Between the Barrel and Port bust hydrostatic thrust bearing for hydrostatic ma-
Plate in an Axial Piston Pump. Journal of Dynamic chines. Proc. 7.IFK, March 22-24, 2010, Aachen, Ger-
Systems, Measurement, and Control, 2008. 130(1): p. many, 2010: p. 100-112.
011011.
34. Achten, P.A.J., T.L.v.d. Brink, and J.W. Potma, Move-
22. Bergada, J.M., et al., Experimental investigation in ment of the Cups on the Barrel Plate of a Floating Cup,
axial piston pumps barrel dynamics. Proc. Flucome Axial Piston Machine. Int. Journal of Fluid Power,
2009, 10th Int. Conf. on Fluid Control, Measurements, 2004. 5(2): p. 25-33.
and Visualization, August 17-21, 2009, Moscow, Rus-
sia, 2009.

23. Bergada, J.M., et al., The effect of oil pressure and tem-
perature on barrel film thickness and barrel dynamics
of an axial piston pump. Meccanica, 2011. 47(3): p.
639-654.

24. Wegner, S., et al., Experimental Investigation of the

Cylinder Block Movement in an Axial Piston Machine.
Proc. ASME/BATH 2015 Symposium on Fluid Power
and Motion Control, FPMC2015, Oct. 12-14, 2015,
Chicago, Ill., USA, 2015(FPMC2015-9529).

25. Wegner, S., F. Löschner, and S. Gels, Validation of the

physical effect implementation in a simulation model
for the cylinder block/valve plate contact supported by
experimental investigations. Proc. 10th IFK In-
ternational Fluid Power Conference, Dresden , Ger-
many , March 8-10, 2016. 1(269-282).

26. Han, L., S. Wang, and C. Zhang, A partial lubrication

model between valve plate and cylinder block in axial
piston pumps. Proceedings of the Institution of Me-
chanical Engineers, Part C: Journal of Mechanical En-
gineering Science, 2015. 229(17): p. 3201-3217.

27. Zhang, C., et al., A new dynamic seven-stage model for

thickness prediction of the film between valve plate and
cylinder block in axial piston pumps. Advances in Me-
chanical Engineering, 2016. 8(9).

28. Manring, N., Torque on the cylinder block of an axial-

piston swash-plate type hydrostatic pump. Retrospec-
tive Theses and Dissertations. Paper 11164, 1996.

29. Manring, N.D., Tipping the Cylinder Block of an axial-

piston swash-plate type hydrostatic machine. Transac-
tions of the ASME, 2000. 122: p. 216-221.

Non-reviewed paper. 391 SICFP2017 7-9 June 2017

Linköping, Sweden