Vous êtes sur la page 1sur 10

The Role of Each Groove on the Behavior of Twin Axial Groove Journal Bearings

F.P. Brito
(1)
, A. S. Miranda
(1)
, J. C. P. Claro
(1)
, M. Fillon
(2)

francisco@dem.uminho.pt asm@dem.uminho.pt, jcpclaro@dem.uminho.pt, fillon@lms.univ-poitiers.fr

(1)
Mechanical Engineering Department, University of Minho, Guimares, Portugal
(2)
Solid Mechanics Laboratory, University of Poitiers, Futuroscope Chasseneuil, France



Abstract
An experimental study of the role of each groove on the thermal behavior of a 50mm plain journal
bearing with two axial grooves located at 90 to the load line was carried out. The temperature profiles
at the oil-bush interface, oil outlet temperature, oil flow rate through each groove and shaft locus, were
measured for variable applied load and oil supply pressure.
It was found that the cooling effect of the downstream groove is small for low eccentricities, becoming
more relevant as eccentricity increases. The opposite phenomenon occurs at the upstream groove. The
cooling effect is related to the flow rate in each groove. Under high load / low supply pressure negative
flow rate at the upstream groove was detected. This phenomenon was found to affect dramatically the
bearing performance. Increasing supply pressure yielded a decrease in shaft eccentricity along with a
temperature decrease, especially for high loads.



1. Introduction

Hydrodynamic journal bearings are widely used to support rotating shafts, especially under high
applied load and shaft speed. Bearing behavior is greatly influenced by the way the oil is supplied to the
bearing gap. It is frequent to perform the oil supply through a pair of axial grooves located at 90 to the
load line. This geometry allows shaft rotation in both directions, being also generally accepted that a pair
of grooves provides a more efficient cooling of the bearing than one single groove.
Although experimental work on journal bearings has been carried out since several decades from now,
not always the oil supply conditions were given sufficient attention. Nevertheless, there were several
works on this matter [1-4].
Often theoretical bearing analyses do not take into account the real supply conditions or use
algorithms that do not always assure flow continuity or energy consevation, making it difficult to
correctly assess the influence of supply conditions on bearing behavior. Authors like Floberg [5] and
Elrod [6] proposed mass conserving algorithms that have been sucessfully used by some authors who
took into account the real influence of supply conditions [7-10].
The theoretical and experimental work done on this matter shows that oil supply conditions often play
a vital role on bearing performance and their influence should not be neglected in bearing analysis and
design.
The aim of the present work was to adress the need of a better understanding of the role of oil grooves
on bearing behavior. The twin groove plain journal bearing has been the bearing geometry chosen for the
present work, as its performance could be particularly influenced by supply conditions. Furthermore, the
existing theoretical models for this bearing geometry not always acurately predict the temperature
evolution in the lubricant film, particularly at the ruptured film region. Oil flow rate is also under-
estimated in most analyses.
The poor correlation between theoretical results and experimental data often observed for this bearing
geometry could be partly due to the location of one of the grooves at the divergent portion of the film and
the simplified way that most theoretical models treat this region. Furthermore, hydrodynamic pressure
generation often extends to the vicinity of the grooves. This would affect strongly the local flow patterns
and the thermal behavior.
This points out the need for a better understanding of the bearing behavior, in particular how the oil
supply conditions affect flow patterns.
Recent work by the authors on a different test rig has already started to adress this matter [11, 12].
Focus has been put on the influence of operating conditions (speed and load) and supply conditions
(supply pressure and supply temperature) on bearing performance. The lack of information concerning
To cite this work:
F.P. Brito, A. S. Miranda, J. C. P. Claro, M. Fillon
The Role of Each Groove on the Behavior of Twin Axial Groove Journal Bearings, IV
Iberian Congress of Tribology (IBERTRIB 2007), Bilbao (Spain), June 21-22th, 2007
the individual flow rate at each groove proved to be somewhat limitative of the discussion of the results.
For instance, certain tests pointed out to a possible occurrence of negative flow rate at one of the grooves,
but this could not be confirmed experimentally. The authors did not find in the literature any experimental
study adressing this particular problem, except in the discussion of the publication of Fillon et al [13] on
tilting-pad journal bearings, where they demonstrated that the temperature at the inlet zone of a high
loaded pad could be greater than the temperature of the exit of the previous low loaded pad, due to the
large recirculating flow. Therefore, the present work describes a thorough investigation on the flow rate
distribution on each groove and its influence on the bearing performance.

2. Test rig

Figure 1(a) shows a general view of the experimental test rig existing in the Tribology Laboratory of
the University of Minho. This apparatus has already been used on several experimental works having
been modified for the present work. The experimental layout has already been presented in published
works [7, 14] and therefore only a brief description will be made here.
The rig allows the regulation of rotational speed, applied load, oil supply pressure and supply
temperature.
The measured performance parameters were the temperature at the oil-bush interface, the oil outlet
temperature, the oil flow rate at each groove, the total oil flow rate, and the shaft locus.
The shaft was driven by a 0.95kW variable speed motor via a transmission belt. The speed was
regulated through an inverter drive and kept within a range of 10 rpm of the nominal speed. The shaft,
shown in Fig. 1(b), is made of X22-CrNi17 stainless steel, was rigidly mounted on by two precision
preloaded conical rolling bearings that assured a good stiffness to the system. The bush, also shown in
Fig. 1(b), was made of RG5G-CuSn5ZnPb bronze. The bush diameter, the shaft diameter and cylindricity
were measured using a Mitutoyo BHN706 coordinate measuring machine with a resolution of 10
-7
m.
Roughness was measured with a Mahr Perthen S5P machine. The oil used was an ISO VG 32 (Galp
Hidrolep 32).


(a) (b)

Figure 1 General view of (a) the test rig and (b) the shaft and the bush.

The loading system relies on a cantilever system on which dead weights are applied. The cantilever
acts on the bush body through a closed loop steel wire supported by needle bearings. The loading system
was calibrated using a high precision load cell with an error of less than 0.5kN .
The supply pressure (P
f
) was regulated by a restrictor valve and monitored by pressure transducers
located at the interior of each groove. P
f
was kept within an interval of variation of 0.004MPa.
The supply temperature (T
f
) was regulated via a thermostatic bath with outer circulation passing
through a plate heat exchanger in order to heat the oil that was being supplied to the bearing. The
temperature of the bath was regulated so that T
f
was kept within a range of 0.4C from the set point. T
f

was monitored by three thermocouples, one located in the main feeding pipe, just upstream of the point
where the flow is separated in two branches to feed each groove and two other thermocouples, one on
each groove, just before the groove entrance.
The measurements were always made under a steady-state regime. In order to achieve this, start-up
times were set for thermal stabilization. Between tests, parameters such as temperature and flow rate were
monitored until stabilization occurred.
The oil flow was measured by three gear flow meters (repeatability 0.03%), suitable to low flow rate
measurements, linked to the data acquisition system.One flowmeter was attached to the main feed line,
while.the other two were located in each branch in order to measure partial flow rates. To ensure accurate
flow rates, measurements were performed during 35 seconds. The difference between the total flow rate
and the sum of the partial flow rates was below 1.5% in most of the cases.
The temperature field was monitored by type K thermocouples attached to a data acquisition system.
The accuracy of the measured values was within 1C. The temperature at the oil-bush interface was
measured at the locations depicted in Fig. 2. The thermocouples were placed inside fully drilled holes,
flush with the inner bush surface. Another set of thermocouples was positioned so as to measure the oil
outlet temperature and the environment temperature.



Figure 2 Angular location of the thermocouples at the inner surface of the bush.

The relative shaft position was obtained with the help of two pairs of Eddy current proximity probes
located at 45 to the load line, on both sides of the bearing. The system was calibrated, and sensitivities
around 7mV/m were obtained. The accuracy of the measurements was found to be affected by bush
elastic deformation. Therefore, it was not possible to analyze the influence of applied load on shaft locus.
As an alternative, the locus of the shaft center was relative to a reference test where the absolute position
of the shaft has been obtained from a theoretical model presented by Costa et al. [9] adapted for twin
groove bearings. For each applied load a specific reference test has been adopted.
The geometric parameters, operating and supply conditions, as well as lubricant properties, are
presented in Table 1.

Table 1 Main bearing characteristics, lubricant properties and operating conditions.

Parameters Units Values
Geometrical bearing characteristics
- Inner bush diameter (nominal) d mm 50
- Outer Bush diameter D mm 100
- Bush width / diameter ratio b/d 0.5
- Groove lenght / diameter ratio a/b 0.8
- Groove circunferential extension / diameter ratio w/d 0.2
- Bearing diametral clearance (at 20C) C
d
m 54
Oil properties ISO VG 32
- Dynamic viscosity at 30C
30
Pa.s 0.0467
- Dynamic viscosity at 75C
75
Pa.s 0.0083
- Specific Heat C
p
J/kgC 1943
- Specific mass kg/m
3
875
- Thermal conductivity k W/mC 0.13
Operating conditions
- Ambient temperature T
amb
C 33 - 40
- Rotational speed N rpm 1000 - 4000
- Applied load W kN 0.4 - 5
Supply conditions
- Oil supply pressure P
f
kPa 50 - 350
- Oil supply temperature T
f
C 30



10 20 20 20 20 20 20 10 10 20 20 20 20 20 20 10 20 20 20 20

U
p
s
t
r
e
a
m

g
r
.

W
U
p
s
t
r
e
a
m

g
r
.

Loaded Arc Unloaded Arc
D
o
w
n
s
t
r
e
a
m

g
r
.

3. Results and discussion

Various results obtained for different working conditions and oil supply parameters will now be
presented and discussed. In order to evaluate the behavior of each oil supply groove, a parametric study
was performed on two major parameters that are known to clearly affect the oil supply regime: applied
load and oil supply pressure.
The results presented here correspond to a shaft speed of 3000rpm, except for Fig. 4, where speeds
were in the range of 1000 4000rpm. Six different load values in the range 0.4kN to 5kN were tested for
a fixed supply pressure of 0.1MPa. The influence of variation on oil supply pressure (from 0.05 to
0.3MPa) was analyzed for both low (0.4kN) and high (5kN) loads. Oil supply temperature was imposed
to 30C.


3.1. Influence of Applied Load

Oil Flow Rate
Figure 3 shows the variation of oil flow rate with increasing load, for constant shaft speed (3000rpm)
and oil supply pressure (0.1MPa). The total oil flow rate curve shows a typical evolution with load as
presented in other works [15, 16], with an initial increase, a stabilization and a smooth decline.

-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Applied Load [kN]
O
i
l

F
l
o
w
r
a
t
e

[
l
/
m
i
n
]
Total Flow Rate
Upstream Groove Flow Rate
Downstream Groove Flow Rate
N = 3000rpm
Pf = 0.1MPa


Figure 3 Influence of applied load on oil flow rate.

It is interesting to note that the two grooves display a totally different behavior with loading
conditions. As load increased, the flow rate at the downstream groove began with a low value
(0.033 l/min) then increased consistently up to 0.182 l/min. This behavior can be explained by the fact
that, as load increases, greater hydrodynamic pressure values tend to occur on the loaded arc of the
bearing. This causes an increase in oil leakage with its subsequent renewal at the groove located
immediately downstream of the pressure buildup zone. Furthermore, as load increases the changes in
shaft locus cause the gap at the downstream groove to increase, while the pressure buildup zone tends to
develop on a smaller angular extension, away from the downstream groove region, further facilitating oil
entrance from this groove.
On the contrary, at low loads the flow rate at the upstream groove was higher than the flow rate at the
opposite groove, but when load started increasing above 1kN, a linear fall was observed. Furthermore, for
loads higher than 4kN the flow rate from the upstream groove was negative. This means that, rather than
flowing into the bearing clearance, oil flowed out of the bearing through the upstream groove. The
decrease of this flow rate with increasing load is associated with the corresponding change of location and
magnitude of the hydrodynamic pressure field. As the load increases, there is an increase in pressure
generation and a decrease in attitude angle, causing the pressure buildup zone to move upstream, closer to
the groove region. Above some critical load, the pressure generated in the upstream groove region
exceeds the value of the supply pressure. This causes some of the oil to flow out of the bearing through
the oil groove.
This behavior is undesirable in various senses. Concerning lubrication needs, the region near the location
of minimum film thickness (H
min
) is the critical one. When flow rate at the upstream groove is negative,
oil is being supplied only at the non-active part of the film, where it will increase power loss due to
viscous dissipation. At the same time, oil is being retrieved from the film instead of being supplied to it,
just upstream of the location where it is mostly needed. Furthermore, this happens under severe operating
conditions (high applied loads / eccentricities) rising the failure risk of the system.
The impact of the occurrence of negative oil flow rate can be further assessed by analyzing the
temperature field, which is deeply affected by the way the oil is delivered to the system.

Temperature field
A global view of the temperature profiles at the midplane of the film-bush interface for a wide set of
operating conditions is presented in Fig. 4. The present discussion is focused on the 3000rpm tests,
detailed in Fig. 5.

35
40
45
50
55
60
65
70
75
0 60 120 180 240 300 360
Angle from center of upstream groove []
T

[

C
]
1000
2000
3000
4000
W = 0.4kN
Pf = 0.1MPa
N
35
40
45
50
55
60
65
70
75
0 60 120 180 240 300 360
Angle from center of upstream groove []
T

[

C
]
1000
2000
3000
4000
W = 2kN
Pf = 0.1MPa
N
35
40
45
50
55
60
65
70
75
0 60 120 180 240 300 360
Angle from center of upstream groove []
T

[

C
]
2000
3000
4000
W = 5kN
Pf = 0.1MPa
N

(a) (b) (c)

Figure 4 Temperature profiles at the midplane of the bush-film interface for (a) W=0.4kN, (b) W=2kN
and (c) W=5kN.

When the bearing is lightly loaded (0.4kN) it can be seen in Fig. 5(a) that the maximum temperature
(T
max
) occurs near the downstream grooves edge, at an angular distance of about 20. This is so because
lightly loaded bearings tend to display high attitude angles that cause the minimum film thickness be
located near the downstream groove region. There is a slight temperature decrease of 1C before the
groove zone, followed by a temperature fall across this groove which is also very low (1C). The fact that
the cooling effect of the downstream groove for low loads is small should be due to the low flow rate at
this groove (see Fig. 3). This causes the maximum temperature at the unloaded arc to be only 1.5C lower
than that of the loaded arc. In previous work performed on a different test rig [11, 12], the authors
obtained several results concerning lightly loaded bearings where T
max
occurred at the unloaded arc of the
bearing. Slightly after the downstream groove the temperature profile displayed a decreasing trend. This
temperature decrease might have several causes. On one hand, the heat losses by conduction to the bush
body may overcome at some point the heat generation within the film, which is small in this region. Fresh
oil backflow coming from the upstream groove can also be responsible for some temperature decrease in
its vicinity. Finally, a coolling effect of about 3.5C across the upstream groove was observed, being 3.5
times stronger than that of the downstream groove. This is in good agreement with the proportion of flow
rate values between the two grooves, which is also 3.5 times aproximately (Fig. 3). In order to facilitate
comparisons, the temperature change across grooves is detailed on Fig. 5c.

40
45
50
55
60
65
70
0 60 120 180 240 300 360
Angle from center of upstream groove []
T

[

C
]
0.4
1.0
2.0
3.0
4.0
5.0
N = 3000rpm
Pf = 0.1MPa
W

40
45
50
55
60
65
70
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Load [kN]
T

[

C
]
Oil outlet
temperature
Tmax (loaded arc)
Tmax (unloaded arc)
N = 3000rpm
Pf = 0.1MPa

-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Load [kN]
T

[

C
]
T across upstream groove
T across downstream groove
N = 3000rpm
Pf = 0.1MPa

(a) (b) (c)

Figure 5 - (a) Influence of applied load on the temperature profile at the midplane of the bush-film
interface and (b) on the maximum temperature at each arc and on oil outlet temperature; (c) Temperature
fall across each groove.

As load increased, it was observed an increase in T
max
at the loaded arc (Fig. 5a) caused by an increase
in heat dissipation. The location of T
max
further upstream could be caused by the decrease in attitude
angle: Due to this, the region where heat dissipation is more intense occurs further upstream. The
temperature fall observed before the downstream groove became more pronounced as load increased. The
greater temperature fall across the downstream groove as load increased (from 1C for 0.6kN to 7.8C for
5kN) was mainly due to flow rate increase at this groove. Linked to this, a general lowering of the
unloaded arc temperature level was observed as load began to increase. However, there is a gradual
change from decreasing temperature profiles at low loads to increasing temperature ones at high loads.
Again, the flow rate pattern can explain this behavior: as flow rate at the downstream groove increases
with increasing load (Fig. 3), more fresh oil is present inside the bearing, thus promoting viscous
dissipation and temperature rise. At the same time, as already seen in Fig. 3, flow rate at the upstream
groove decreased with increasing load. As a consequence, the observed temperature decrease in the
vicinity of the upstream groove tends to diminish and eventually disappear. Above 2kN the cooling effect
of the upstream groove was so low that there was no noticeable temperature fall across this groove region.
Surprisingly, the occurrence of negative flow rate at the upstream groove for the highest load tested (5kN)
had consequences at the opposite groove. As shown in Fig. 6, the hot oil (57.8C) that flowed out of the
bearing through the upstream groove mixed with the new oil inside the supply system piping. This
induced a temperature rise of more than 5C to the oil supplied to the opposite groove. This supply
temperature rise is noticeable in Fig. 5a, where it can be seen that an increase of load from 4kN to 5kN
yielded an increase of 2C in temperature after the groove instead of a decrease like in all other cases.
This effective rise in supply temperature caused by negative flow rate further increases bearing seizure
risk.
Concerning oil outlet temperature, it can be observed in Fig. 5b that for low loads it was lower than
T
max
occurring at any of the arcs decreasing slightly with increasing load. However, as load increased
above 1kN the outlet temperature started to increase, becoming higher than the maximum temperature in
the unloaded arc.



Figure 6 Oil supply temperature increase due to negative flow rate at the upstream groove.


3.2. Influence of Supply Pressure

Oil Flow rate and Shaft Locus
Figure 7 shows the evolution of flow rate with the increase of oil supply pressure P
f
for the lowest
(0.4kN) and highest (5kN) load tested.
Under low load (Fig. 7a), the total flow rate increased 2.1 times for a sixfold increase of P
f
, from
0.05MPa to 0.30MPa. It can be seen that this increase was mainly due to the 127% increase in flow rate in
the upstream groove. Although the flow rate in the downstream groove has increased 70% over the entire
P
f
range, its weight on the total flow rate is reduced from 25% to 20%.
Under high load (Fig. 7b) the flow rate was supplied to the bearing mainly through the downstream
groove. This flow rate increased only 6% as P
f
increased sevenfold (from 0.05 to 0.35MPa). On the
contrary, the flow rate at the upstream groove was strongly dependent on the value of supply pressure.
Low values of P
f
induced a negative flow rate in this groove, with all the implications associated with
this, as already explained. For supply pressures above 0.15MPa, the negative flow rate ceased to occur
and flow rate increased linearly with P
f
, exceeding in 25% of the total flow rate for the highest value of P
f

tested (0.35 MPa).

-0.100
-0.050
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
Supply Pressure [MPa]
O
i
l

F
l
o
w
r
a
t
e

[
l
/
m
i
n
]
Total Flow Rate
Upstream Groove Flow Rate
Downstream Groove Flow Rate
W = 0.4kN
N = 3000rpm

-0.100
-0.050
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Supply Pressure [MPa]
O
i
l

F
l
o
w
r
a
t
e

[
l
/
m
i
n
]
Total Flow Rate
Upstream Groove Flow Rate
Downstream Groove Flow Rate
W = 5kN
N = 3000rpm

(a) (b)

Figure 7 Influence oil of supply pressure on oil flow rate for (a) 0.4kN and (b) 5kN.

Figure 8 shows the influence of supply pressure on the shaft locus, for both the lightly loaded and the
heavily loaded bearing cases. The tests with a supply pressure of 0.15MPa were chosen to be the
reference, with preimposed eccentricity ratio () and attitude angle () (see explanation in chapter 2). For
a load of 0.4kN, the reference position was set to be = 0.26 and = 58.5, while for a load of 5kN
50.2 47.2 49.9 52.8 56.4
0.0 53.4 54.0 56.3 58.2 60.4
0.0 51.9 49.7 49.7 50.9 53.9
47.4 46.7 48.0 49.2 51.0
50.2 47.2 49.9 52.8 56.4
0.0 -3.4 -0.9 0.5 1.2 2.3
0.0 -1.0 -3.2 -5.2 -6.7 -7.6
25
30
35
40
45
50
55
60
65
70
0.0 1.0 2.0 3.0 4.0 5.0
Applied Load [kN]
T

[

C
]
Tfeed at upstream groove
Tfeed at downstream groove
Tfeed before entering system
N = 3000rpm
Pf = 0.1MPa
reference position was set to be = 0.77 and = 33.0. It can be observed that higher values of P
f
yielded
lower eccentricities. The impact of P
f
on shaft eccentricity is especially visible for low values of P
f
(below
0.15MPa). Under high load (5kN) this corresponds to the tests where negative flow rate occurred. For
instance, when P
f
decreased from 0.15 to 0.05MPa eccentricity increased from 0.77 to 0.95, a value that is
already unsafe for bearing operation. This is three times the eccentricity rise observed for a similar
decrease in P
f
from 0.25 to 0.15MPa.

-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.0 0.2 0.4 0.6 0.8 1.0
E
c
c
e
n
t
r
i
c
i
t
y
W = 0.4kN
W = 5kN
N = 3000rpm
Pf = 0.3MPa
Pf = 0.05MPa
Pf = 0.3MPa
Pf = 0.05MPa


Figure 8 Influence of supply pressure on the shaft locus for two different applied loads (0.4 and 5kN).

Temperature field
The temperature profiles at the midplane of the bush-film interface for the lowest (0.4kN) and the
highest load tests (5kN) and supply pressures of 0.05MPa and 0.30MPa are presented in Fig. 9.
Information concerning oil outlet temperature and maximum temperature at each arc of the bearing is also
detailed in Fig. 9b. Although P
f
undoubtedly affected the temperature level of the bearing, the
temperature profiles corresponding to the different values of P
f
showed a very similar trend, being nearly
parallel to each other. The decrease of T
max
with increasing P
f
was more pronounced at high load (8.6C
over the entire P
f
range as seen in Fig. 9c). The decrease of outlet temperature with increasing P
f
was
higher than the corresponding decrease of T
max
, especially at low load. For instance, when P
f
increased
from 0.05 to 0.30MPa, the oil outlet temperature decreased almost 10C, while T
max
dropped only 4C, as
seen in Fig. 9b.

40
45
50
55
60
65
70
0 60 120 180 240 300 360
Angle from center of upstream groove []
T

[

C
]
W=0.4kN Pf=0.05MPa
W=0.4kN Pf=0.30MPa
W=5kN Pf=0.05MPa
W=5kN Pf=0.30MPa
N = 3000rpm

40
45
50
55
60
65
70
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
Supply Pressure [MPa]
T

[

C
]
Oil outlet
temperature
Tmax (loaded arc)
Tmax (unloaded arc)
W = 0.4kN
N = 3000rpm

40
45
50
55
60
65
70
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Supply Pressure [MPa]
T

[

C
]
Oil outlet temperature
Tmax (loaded arc)
Tmax (unloaded arc)
W = 5kN
N = 3000rpm

(a) (b) (c)

Figure 9 (a) Influence of supply pressure on the temperature profile at the midplane of the bush-film
interface for two different applied loads (0.4 and 5kN); (b) Influence of supply pressure on the maximum
temperature at each arc and on oil outlet temperature for 0.4kN and (c) for 5kN.
A diagram representing the temperature at the oil supply system and at the entry of each groove is
presented in Fig. 10. The cases for which a negative flow rate occurred can be easily detected because
both grooves displayed temperatures that exceeded the supply temperature (30C), as explained in the
previous section. For instance, in the test with the lowest P
f
(0.05MPa), hot oil that was flowing out of the
upstream groove with a temperature of 63.3C could be detected. After mixing with the fresh supply oil
(30C) at the supply system pipes, it caused a 9C increase in the supply temperature.



Figure 10 - Oil supply temperature increase due to negative flow rate at the upstream groove.


5. Conclusions

In a hydrodynamic journal bearing with two axial grooves located at 90 degrees to the load line the oil
flow rate entering each groove has been measured individually, and its effect on the performance of the
bearing has been analyzed for different sets of operating and supply conditions.

In the range of applied load studied, the total flow rate in the bearing showed only slight variations.
On the contrary, the partial flow rates through the upstream and the downstream grooves varied
dramatically.
For low applied loads, the flow rate in the upstream groove was higher than in the downstream
groove. The latter groove had, therefore, a poor contribution to the bearing cooling. Under certain
conditions, the occurrence of maximum bearing temperature can even move from the loaded to the
unloaded arc of the bearing.
For high applied loads the flow entering the upstream groove is significantly lower than that on the
downstream groove. For the highest load tested the measured upstream flow rate was negative, meaning
that oil flowed out of the bearing through this groove. Under such conditions, no cooling effect was
obtained from the upstream groove. On the contrary, recirculating oil suffered an increase in temperature
of around 2 C.
It was observed that under the high loading conditions mentioned above the oil supply pressure could
have an important role in preventing the risk of bearing failure due to the occurrence of high eccentricity.
For an applied load of 5 kN the eccentricity decreased from 0.95 to 0.77 as the supply pressure increased
from 0.05 to 0.15 MPa.


Acknowledgements
The present work was funded by the project POCTI/39202/EME/2001 and the PhD grant
SFRH/BD/22278/2005) funded by FCT - Fundao para a Cincia e a Tecnologia (Portugal) and the
European Union fund FEDER. The work has been carried out at the Tribology Laboratory of Minho
University. The authors greatly acknowledge the support of these institutions.


References

[1] Mitsui, J., Hori, Y., and Tanaka, M., 1983, Thermohydrodynamic Analysis of Cooling Effect of
Supply Oil in Circular Journal Bearings, ASME Journal of Tribology, 105, pp. 414421
0.0 46.6 47.5 50.0 51.9 53.9 56.6
0.0 45.1 44.5 45.7 48.0 49.9 52.4
42.5 42.6 44.4 45.9 49.2 51.8
44.3 44.2 46.8 49.5 51.3 53.6
0.0 -0.5 0.6 1.5 1.6 1.8 1.9
0.0 -0.8 -1.9 -3.2 -3.9 -4.2 -4.4
25
30
35
40
45
50
55
60
65
70
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Supply Pressure [MPa]
T

[

C
]
Tfeed at upstream groove
Tfeed at downstream groove
Tfeed before entering system
W = 5kN
N = 3000rpm
[2] Gethin D. T., and El-Deihi, M. K. I., 1990, Thermal Behavior of a Twin Axial Groove Bearing
Under Varying Loading Direction, Proc. ImechE, 204, pp. 7790.

[3] Costa L., Fillon, M., Miranda, A. S., and Claro, J. C. P., 2000, An experimental Investigation of
the Effect of Groove Location and Supply Pressure on the THD Performance of a Steadily
Loaded Journal Bearing, ASME Journal of Tribology, 122, pp 227231.

[4] Ma, M. T., and Taylor, C. M., 1995, Effects of Oil Feed Temperature on the Performance of an
Elliptical Bore Bearing, Lubricants and Lubrication Proc. of the 21st Leeds-Lyon
Symposium on Tribology, Elsevier, pp 143151.

[5] Floberg, L., 1974, Cavitation Boundary Conditions with Regard to the Number of Streamers
and Tensile Strength of the Liquid, Cavitation and Related Phenomena in Lubrication,
ImechE, England, pp. 3136.

[6] Elrod, H. G., 1981, A cavitation algorithm, ASME Journal of Lubrication Technology, 103,,
pp. 350354.

[7] Claro J. C. P., Miranda A. S., 1993, Analysis of hydrodynamic journal bearings considering
lubricant supply conditions, Proc. Inst. Mech. Engs., Part C, 207, pp. 93101
[8] Vijayaraghavan, D., 1996, An efficient numerical procedure for thermohydrodynamic analysis
of cavitating bearings. ASME Journal of Tribology, 118, pp. 555563.

[9] Costa L., Miranda A. S., Fillon M., Claro J. C. P., 2003, An analysis of the influence of oil
supply conditions on the thermohydrodynamic performance of a single groove journal
bearing, Proc. Inst. Mech. Engs., Part J: Journal of Engineering Tribology, 217, pp. 133144.

[10] Pierre I., Fillon, M., 2000, Influence of geometric parameters and operating conditions on the
thermohydrodynamic behaviour of plain journal bearings, Proc. Inst. Mech. Engs., Pt J,
Journal of Engineering Tribology, 214, pp. 445457.

[11] Brito, F. P., Miranda, A. S., Bouyer, J., and Fillon, M., 2006 Experimental investigation on the
thermal behavior and performance characteristics of a twin axial groove journal bearing as a
function of applied load and rotational speed, 5th International Conference On Mechanics &
Materials in Design - M2D'2006, Symposium on Wear and Lubrication In Design, FEUP,
Porto, 24-26 July 2006. CD-ROM (ISBN 972-8826-10-9).

[12] Brito, F. P., Miranda, A. S., Bouyer, J., and Fillon, M., 2007 Experimental Investigation of the
Influence of Supply Temperature and Supply Pressure on the Performance of a Two Axial
Groove Hydrodynamic Journal Bearing, ASME Journal of Tribology, 129, pp. 98105.

[13] Fillon M., Bligoud J. C., Frene J., 1992, Experimental study of tilting-pad journal bearings -
Comparison with theoretical thermoelastohydrodynamic results, ASME Journal of Tribology,
114, pp. 579588.

[14] Costa, L., Miranda, A. S., Claro, J. C. P. ., and Fillon, M., 2003, Temperature, Flow, and
Eccentricity Measurements in a Journal Bearing with a Single Axial Groove at 90 to the Load
Line, Lubrication Science, 15-2, pp. 147161.

[15] Fitzgerald M. K., and Neal P. B., 1992, Temperature distributions and heat transfer in journal
bearings, ASME Journal of Tribology, 114, pp. 122130.

[16] Ma M. T., Taylor C. M., 1992, A theoretical and experimental study of thermal effects in a
plain circular steadily loaded journal bearing, IMechE Seminar: Plain Bearings - Energy
Efficiency and Design, Mech. Engs. Publications Ltd., London