Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 81 (2015) 794 – 804

69th Conference of the Italian Thermal Machines Engineering Association, ATI2014

A Critical Analysis on the Lubrication Circuit of a non-road Diesel

Engine by Adopting a 3D and 1D Approaches
Emma Frosinaa,*, Adolfo Senatorea, , Dario Buonoa,
a
Department of Industrial Engineering, University of Naples Federico II, Via Claudio 21, Naples 80125, Italy

Luigi Arnoneb,
b
Lombardini s.r.l. Kohler Group, Via Cav. Del Lav. Adelmo Lombardini 2, Reggio Emilia 42124 , Italy

Abstract

The aim if this paper is the analysis of a lubrication circuit adopting two simulation models to study the influence
of three different lubrication oils.
First, the oils pump performance were analyzed with a tri-dimensional simulation model, realized with the
commercial code PumpLinx®, which allowed to obtain the performance curves of the oil pump, varying the oils. Then
the lubrication circuit was studied with a 1D model, realized with the commercial code LMS AMESim®, and the
behaviour of the lubrication circuit was analyzed in terms of pressure drops and components oil consumption.
The adopted oils were: 0W30; 5W40 and10W60. In previous activities, both models showed good correlation with
experimental data.
Analyzing the models results, it was noted that the pressure behavior in the lubrication circuit is highly influenced
by the oil, especially for low engine speed and high temperatures.
The adopted methodology, that couples mono-dimensional and tri-dimensional models, allowed to analyze the oil
influence in a lubrication circuit, avoiding excessive computational cost, but with an acceptable accuracy

©
© 2015
2013TheTheAuthors. Published
Authors. by by
Published Elsevier Ltd.Ltd.
Elsevier This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Selection and peer-review under responsibility of ATI NAZIONALE.
Peer-review under responsibility of the Scientific Committee of ATI 2014
Keywords: 3D CFD and 1D Oil Pump Modeling and Lubrication Circuit

* Corresponding author. Tel. +39-081-768-32-85

E-mail address: emma.frosina@unina.it

1876-6102 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the Scientific Committee of ATI 2014
doi:10.1016/j.egypro.2015.12.086
 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804 795

1. Introduction

The engine lubrication circuit is a very complex system that presents several issues and have to be studied with
different techniques and approaches. A well designed lubrication circuit must ensure a correct oil supply to all circuit
components in all engine running conditions, in particular varying the engine rpm and oil temperature. One of the
circuit components that most influences the circuit behavior is the oil pump, which presents different issues respect to
circuit components.
There are different techniques to evaluate and analyze the functioning of an automotive engine lubrication circuit.
Experimental tests allow to evaluate the pressure in some engine points but cannot evaluate the oil consumption of
the different circuit users. A 3D model simulation can well reproduce the functioning of the oil pump, the oil
consumption and the pressure drop in the circuit galleries and ducts while a 1D model, can well analyze the whole
lubrication circuit, but needs a lot of data, that non often are available.
Furthermore, several procedures were implemented to analyze the performance of a lubrication circuit: Chun and
Soltani et all [1, 2] analyzed the circuit as a network, studying the pressure drop in the several circuits ducts and the
components oil consumption. Mancò et all [3] performed a 1D simulation model of an oil pump model, to estimate
the pump volumetric efficiency and oil leakage. Some authors [4], on the other hand, realized dedicated 3D simulation
models to study the oil pump performance and there are examples of 3D model of some circuit components [5, 6], that
estimates the oil consumption in different condition.
In recent years, many experiences were made by this research group about automotive engine lubrication circuits,
building up several simulation models [7, 8, 9], with different techniques and by experimental campaign [10, 11].
In this paper the Authors show an analysis of the influence of three types of lubrication oil on a Diesel engine
circuit. The aim of this activity was to highlight the influence of the oil characteristics on the lubrication circuit, varying
the engine running conditions.
This analysis was done using two different simulation models: a mono-dimensional model of the entire lubrication
circuit and a tridimensional model of the oil pump (both validated with experimental data in previous activities [9,
11]), and the results were coupled in order to obtain data that allow to analyze the whole circuit in several engine
conditions. The models were used to analyze the pressure behavior at high oil temperature and low engine speeds, as
these are the most critical conditions for the lubrication circuit.

2. Pump Model

The gerotor oil pump consists of two gears with the inner rotor rotating off-centre, creating 13 vanes evolving
during one complete revolution from a minimum to a maximum volume during the suction phase and vice versa during
the discharge phase. The rotors have 12 and 13 teethes respectively, the outer rotor rotates with an angular velocity
that is equal to 12/13 of the inner rotor velocity (the transmission ratio between the rotors is 12/13). The 3D simulation
model was built up with the commercial CFD program PumpLinx®; this code solves numerically the fundamental
conservation equations of mass, momentum and energy and includes accurate physical models for turbulence and
cavitation.
From the pump 3D CAD geometry, the fluid volume has been extracted, starting from the surfaces wet by the
lubricant itself. The obtained geometry has then been meshed with the PumpLinx® grid generator [7, 8].
Figure 1 shows the mesh in a section plane of the considered pump. In the boundary layer the grid density increases
on the surface, without excessively increasing the total cell count.
796 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804

Fig 1. Binary Tree Mesh – Pump Cross Section

The code allows the simultaneous treatment of moving (rotors) and stationary (intake and delivery ports) fluid
volumes. Different techniques are available for the treatment of a moving mesh. For positive displacement pumps, it
is necessary to use a moving/sliding methodology whereby the stationary and moving volumes are meshed separately.
Each moving volume connects to the others via a common interface, which, due to deformations and motion, must
be updated at each time-step. PumpLinx® Mismatched Grid Interface is a very efficient implicit algorithm that
identifies the overlap areas and matches them without interpolation. After that, this area is treated as the common face
connecting cells on both sides of the interface.
The model of the considered pump consists of a 302.000 cells mesh (450.000 nodes). Figure 2 shows pictures that
illustrate the grid differences between stationary and moving parts.

Fig 2. Pump Mesh

®
The PumpLinx fluid model accounts for liquid compressibility. This is critical to accurately model pressure wave
propagation in liquid. The liquid compressibility is found to be very important for a high-pressure system and the
systems in which water hammer effects are relevant.
Once created the model, a large number of simulations have been run to obtain a pump characteristic map, in terms
of delivery flow rate as a function of pressure, varying shaft angular velocity and oil temperature. Figures 3 and 4
show the pressure distribution in the pumping volume and the ports and the flow streamlines colored by the flow mean
velocity magnitude [m/s].
 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804 797

In figures 3 and 4 it is possible to observe the suction port, through which the oil is sucked from the oil pan, and
the 13 fluid vanes where the oil is pressurized and sent into the lubrication circuit. If the pressure is too high and the
velocity is low (see figure 4), a pressure relief valve recirculates a portion of the delivered oil flow rate.
A preliminary pump model validation activity has been performed, using data from an experimental campaign held
at the Department of Industrial Engineering of University of Naples “Federico II” [7].

Fig. 3. Pressure Distribution

Fig. 4. Typical streamline colored by velocity magnitude

A 5W40 grade oil that complies with API SL, JASO MA-2, and ACEA A3 specifications, was used for the tests.
The experimental tests allowed to determine the full pump characteristics map, in terms of pressure/flow rate curves
as a function of rotational speed and at an oil temperatures of 110°C.
The model validation has been carried out comparing results from the PumpLinx® simulations with experimental
data at different rotational speeds. Graphs in figure 5 show the comparison between numerical model results (dashed
lines) and experimental ones (continuous line).
798 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804

1.2

1
3100rpm

2800rpm

0.8 2400rpm

Normalized flow rate

2000rpm
0.6

1500rpm

0.4
1000rpm

700rpm
0.2
500rpm

0.2 0.4 0.6 0.8 1 1.2

Normalized pressure

Fig 5. Model/Experiment Results Comparison at 110°C

For all the operating conditions the maximum difference between the numerical model flow-rate estimate and the
experimental data is 9%.
Starting from the validated model a lot at simulations were made changing the oil type and temperature, in order to
obtain data to implement the pump characteristic map for the analyzed oils.
These analyses were made at the oil temperature of 80°C, 110°C, 140°C and 150°C, using the following oils:

x 5W40
x 0W30
x 10W60

The following graphs (figures 6, 7 and 8) show comparison between the model results varying the pump speed for
three different oil temperatures. All graphs were normalized to the maximum value, for each shaft rpm.

1.01
0W30 1.01
5W40 1.01
10W60

0.99 0.99 0.99

NormalizedFlowͲRate[Q/Qmax]
NormalizedFlowͲRate[Q/Qmax]
NormalizedFlowͲRate[Q/Qmax]

0.97 0.97 0.97

0.95 0.95 0.95

0.93 0.93
0.93

80°C 80°C 80°C

0.91 110°C 0.91 110°C 0.91 110°C

140°C 140°C 140°C

0.89 0.89
0.89
2 2.2 2.4 2.6 2.8 3 3.2 2 2.2 2.4 2.6 2.8 3 3.2
2 2.2 2.4 2.6 2.8 3 3.2
Pressure[bar] Pressure[bar]
Pressure[bar]

Fig 6. Model Results at 1000rpm

 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804 799

0W30 5W40
1.01 1.01

1.01
10W60
0.99 0.99
0.99

NormalizedFlowͲRate[Q/Qmax]
NormalizedFlowͲRate[Q/Qmax]

NormalizedFlowͲRate[Q/Qmax]
0.97 0.97
0.97

0.95 0.95
0.95

0.93 0.93
80°C 80°C 0.93
80°C
110°C 110°C 110°C
0.91 140°C 0.91 140°C 0.91 140°C

0.89 0.89
0.89
2.8 3 3.2 3.4 3.6 3.8 4 2.8 3 3.2 3.4 3.6 3.8 4
2.8 3 3.2 3.4 3.6 3.8 4
Pressure[bar] Pressure[bar]
Pressure[bar]

Fig 7. Model Results at 1800rpm

1.01
0W30 1.01
5W40 10W60
1.01

0.99 0.99
0.99
NormalizedFlowͲRate[Q/Qmax]
NormalizedFlowͲRate[Q/Qmax]

NormalizedFlowͲRate[Q/Qmax]
0.97 0.97 0.97

80°C 80°C 80°C

0.95 110°C 0.95 110°C 0.95
110°C
140°C 140°C 140°C

0.93 0.93 0.93

0.91 0.91 0.91

0.89 0.89 0.89

3.5 3.7 3.9 4.1 4.3 4.5 4.7 3.5 3.7 3.9 4.1 4.3 4.5 4.7 3.5 3.7 3.9 4.1 4.3 4.5 4.7

Pressure[bar] Pressure[bar] Pressure[bar]

Fig 8. Model Results at 2800rpm

Model results confirm that for each case, with an increase of oil temperature, the flow-rate value decreases. This is
more evident at low pump speeds in fact at 2800rpm the differences between the lines, varying the oil temperatures,
are low this phenomena probably depends by the internal pump leakages.
Figure 9 show the viscosity trend as a function of the oil temperature for each analyzed oil type. The oil 5W40
viscosity line is between oil 10W60 and oil 0W30. As shown in the previous graphs (figures 6, 7 and 8), the model
takes into account the viscosity variation with the oil type and temperature. In fact, the flow-rate values of the oil
5W40 lines (yellow line in figures 6, 7 and 8) are always between the 0W30 lines and the 10W60 oil lines.
For the three shaft speeds analyzed, the highest oil flow rates were obtained with the 10W60 oil, but it is necessary
to highlight that differences between the three oils are slight while the oil temperature influences much more the pump
performance in terms of delivered oil flow rate.
30

25 0W30

5W40
20
10W60
Viscosity[cP]

15

10

0
60 80 100 120 140 160
OilTemperature[°C]

Fig 9. Oil Viscosity in Function of Oil Temperature

800 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804

3. Lubrication Circuit Model

The complete oil path of the lubrication circuit will be described through the engine crankcase and head-lines
towards the different lubrication users.

5 10
8
7

9
4
11 6
3

Fig 10. Oil Path

The lubricant fluid goes from the oil pan to the suction pipe (point 1 in figure 10) and it is then processed by the
pump (3). A by-pass is positioned on the pump delivery port, with a pressure relief valve (4).
Once passed the valve, the oil lubricant goes through the filter- exchanger (5) and arrives in the main gallery. Five
branches depart from the gallery, to feed oil into the five crank-shaft bearings (6), the four piston cooling jets (7) and
the middle distribution gear (11).
At the exit of the crank-shaft bearings, through internal ducts, the oil lubricates the 4 cam-shaft bearings (8). This
engine, in fact, uses a rod and rocker arm distribution system. The last crank-shaft bearing, lubricates a plain shaft
bearing (9). From the second cam-shaft bearing, starts the engine head feeding line. This duct also feeds the
turbocharger (10).
The oil lubrication circuit model was built with the mono-dimensional AMESim® code and, as previously said, the
3D PumpLinx® model results (flow-rate vs pressure characteristic) were implemented in the AMESim® pump sub-
model.
The AMESim® model reproduces all the ducts and circuit components: filter, cooler, piston cooling jets, bearings
and tappets. About the bearings and tappets, they were modelled assuming an average tolerance and taking into account
the thermal expansion, depending on the oil temperature.
In figure 11 the lubrication circuit model and the entire model are shown; the complete model presents more than
300 state variables. It requires two minutes of calculation time with a desktop PC (Xeon E5 1650 with 16 Gb RAM),
in steady state conditions at a fixed engine rpm
 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804 801

Fig. 11. Lubrication circuit model

The validation phase of the entire model was made starting from experimental data, performed by the engine
manufacturer on an engine test bench. The experimental tests were carried out at different rotational speeds (from
1000 to 2800 rpm) at full load, measuring oil pressure in the main gallery, near the oil filter.
In figure 12 the comparison between the model results and experimental data is shown: the oil pressure behavior
versus rotational speed, in the main gallery, was monitored. The data are presented normalizing the oil pressure with
respect to the experimental maximum value.
1.1

Model
1.05 Experimental

0.95
Normalized pressure

0.9

0.85

0.8

0.75

0.7

0.65

0.6

600 900 1200 1500 1800 2100 2400 2700 3000

Engine speed [rpm]

Figure 12. Model validation

The error is always less than 2%, for all the rotational speeds, and there are no significant differences varying form
idle conditions to maximum speed.

4. Model Application: Oil influence

In this paragraphs the results of the oil type influence, at different temperatures, will be shown. In particular it
will be shown the results at 1000 and 1800 engine rpm because these velocities are the most critical at high
temperatures.
The simulations were made at the oil temperature of 80°C, 110°C, 140°C and 150°C using the following oil type:
- 5W40,
- 0W30
- 10W60.
802 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804

In the following figures the pressure behavior, near the filter, varying the oil temperature for each oil type will be
presented.

1.1
1000RPM 1.1
1800RPM
NormalizedPressure[p/pmax]

NormalizedPressure[p/pmax]
1 1
0W30
0.9 0.9
5W40
0.8 10W60 0.8

0.7 0.7
0W30
0.6 0.6
5W40
0.5 0.5
10W60
0.4 0.4

0.3 0.3
60 80 100 120 140 160 60 80 100 120 140 160
Oiltemperature[°C] OilTemperature[°C]
Figure 13. Oil Influence

At 1800 rpm, at oil temperature lower than 100°C, the pressure is influenced by the bypass valve, which makes
the oil pressure constant at its maximum value, while at 1000 rpm the oil pressure level is lower than the by-pass valve
opening value. So at 1000 rpm, for all the analyzed cases, the pressure highly decreases varying the temperature; the
model shows that at 140 and 150 °C, the pressure is about 50% lower than at 80°C. And this trend can be noted for
all three analyzed oils, but for the 10W60 case, the absolute pressure value is the higher.
In the previous engine conditions, the crank shaft bearing consumption was evaluated by the 1D model simulation,
in order to better analyze the oil influence on the lubrication circuit behavior. In figure 14, three histograms report the
oil consumption, varying the oil type for 1000 and 1800 engine speeds: the oil consumption is higher for the 0W30
case while is lower for the 10W60 case.

1.2
80°C 110°C 150°C
1.2 1.2

1000rpm
1 1000rpm
1000rpm 1 1
NormalizedFlowͲRate[Q/Qmax]

NormalizedFlowͲRate[Q/Qmax]
NormalizedFlowͲRate[Q/Qmax]

1800rpm 1800rpm
1800rpm
0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2

0 0 0
0W30 5W40 10W60 0W30 5W40 10W60 0W30 5W40 10W60
OilType OilType OilType

Figure 14. Crank-Shaft Bearings – Oil Consumption

Analyzing the results in figure 14 is possible to observe that the crank-shaft bearing has the highest oil
consumption with the 0W30 oil and the lower oil consumption with the 10W60 oil for all speed running conditions
and the histograms allow to evaluate the differences among the analyzed cases.
We can conclude that the simulation models, both validated by experimental data, allow to numerically evaluate
the lubrication circuit behavior, in terms of oil pressure and component consumption, varying the adopted oil type.
 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804 803

5. Conclusions

This paper presents an analysis on the lubrication circuit of a Diesel engine for non-road applications. The pump
was modelled with the commercial tridimensional code Pumplinx®, while the entire lubrication circuit with the
commercial mono-dimensional code AMESim® .The Pumplinx® model results were used as input to the AMESim®
pump sub-model.
The models were used to critically analyze the circuit behavior varying the adopted oil, at different temperatures.
In particular, the circuit was analyzed at low engine speed, which, for high oil temperature, is the most critical
conditions. The analyzed engine speeds were 1000 and 1800 rpm, while the oil types were 0W30, 5W40 and 10W60.
The models results highlight that, at 1000 engine rpm, the higher level pressure can be reached with the 10W60 oil,
while the lower crank shaft bearings consumption is obtained with the 0W30 oil.
In conclusion, the adopted methodology, that couples the results of a mono-dimensional and tridimensional model,
can be very useful to analyze and study a lubrication circuit and can help designers on the oil type choice, in function
on the engine running conditions.

Acknowledgements

A sincere thanks to OMIQ s.r.l. and Bsim s.r.l. for technical support.

References

[1] S.M. Chun, “Network analysis of an engine lubrication system” 2003 Elsevier Science Ltd., Tribology International 36 (2003) 609–617,
doi:10.1016/S0301-679X(02)00266-9
[2] R. Soltani, j. Ghorbanian, “A New Approach to Flow Network Analysis of an
Engine Lubrication System”, 2010, The Journal of Engine Research/Vol. 18 / Spring 2010, www.sid.ir
[3] S. Mancò, N. Nervegna, M. Rundo, G. Armenio, “Modelling and simulation of Variable Displacement Vane Pumps for IC Engine Lubrication”,
2004 SAE Tecnhical Paper, 2004-01-1601, DOI 10.4271/2004-01-1601, 2004 SAE World Congress and Exhibition.
[4] Y. Jiang, M. Furmanczyk, S. Lowry, D. Zhang, C. Perng, “A Three-Dimensional Design Tool for Crescent Oil Pumps” 2008 SAE TECHNICAL
PAPER, doi:10.4271/2008-01-0003, 2008 SAE World Congress and Exhibition.
[5] T. Sreenivasulu, S. Kumar, M. Garg, “Numerical Analysis of Lubrication System Using Acusolve”, HTC 2012, TVS, Simulation Driven
Innovation, 2012 India HyperWorks Technology Conference.
[6] B. Kim, K. Kim, “Thermo-Elastohydrodynamic Analysis of Connecting Rod Bearing in Internal Combustion Engine, Transactions of the
ASME, Vol. 123, JULY 2001 DOI: 10.1115/1.1353181
[7] Frosina, E., Senatore, A., Buono, D., Monterosso, F. Olivetti, M., Arnone, L., Santato, L., "A Tridimensional CFD Analysis of the Lubrication
Circuit of a Non-Road Application Diesel Engine," SAE Technical Paper 2013-24-0130, doi:10.4271/2013-24-0130, Capri, 2013.
[8] Frosina, E., Senatore, A., Buono, D., Manganelli, M. U., “A Tridimensional CFD Analysis of the Oil Pump of an High Performance Motorbike
Engine”, Energy Procedia Journal, paper nr EGYPRO7688, 2014.
[9] Frosina, E., Senatore, A., Buono, D., Olivetti, M., “A Tridimensional CFD Analysis of the Oil Pump of an High Performance Engine”, Detroit
SAE Congress, paper nr. 2014-01-1712, 2014.
[10] E. Frosina, A. Senatore, D. Buono, A. De Vizio, P. Gaudino, A. Iorio “A Simulated Analysis of the Lubrication Circuit of an In-line Twin
Automotive Engine” Detroit SAE Congress, paper nr. 2014-01-1081, 2014.
[11] Frosina, E., Senatore, A., Buono, Santato, L., “Analysis and Simulation of an Oil Lubrication Pump for the Internal Combustion Engine”, San
Diego ASME Congress, paper nr. 63468, 2013.
[12] D. Zhang, C.Y. Perng, M. Laverty, “Gerotor Oil Pump Performance and Flow/Pressure Ripple Study” SAE Technical Paper 2006-01-0359,
2006.
[13] Y. Jiang, C.Y. Perng, "An Efficient 3D Transient Computational Model for Vane Oil Pump and Gerotor Oil Pump Simulation," SAE-970841.
[14] D. Buono, M. Cardone, A. Senatore, L. Fabbri “Optimization Methodology of High Performance Motorcycle Engine Cooling System” FISITA
2006 World Automotive Congress, Yokohama, Japan, October 22-27, 2006;
[15] D. Buono, M. Cardone, A. Senatore, A. Dominici “Fluid-dynamic Analysis of a High Performance Engine Lubricant Circuit” SAE paper n.
2007-01-1963 JSAE/SAE International Fuels and Lubricants Meeting, Kyoto, Japan, July 23-27, 2007;
[16] M. Cardone, A. Senatore, D. Buono, M. Gustato, W. Scattolin “Simulated Analysis of a Motorbike High Performance Lubrication Circuit”
SAE paper n. 2008-01-1647, SAE Powertrains, Fluid, and Lubricants Congress, Shanghai, China, 23-25, June, 2008;
[17] D. Buono, A.Senatore, M. Cardone, C. Lombardi, “A simulation model of an engine motorbike lubrication circuit”, CONAT 2010
International Congress on Automotive and Transport Engineering, 27-29 October, Brasov, Romania;
804 Emma Frosina et al. / Energy Procedia 81 (2015) 794 – 804

[18] Q. Zhou, “Engine Lubrication System Analysis and Oil Pump design Optimization”, 2008, Proceedings of CIST2008 & ITS-IF0MM,
Bejing, China
[19] R. Burke, C. Brace, R. Stark, I. Pegg, “Investigation into the benefits of reduced oil flows in internal combustion engines, International
Journal of Engine Research May 28, 2014 1468087414533954, doi: 10.1177/1468087414533954
[20] R. Coy, “Practical applications of lubrication models in engines”, Tribology International Vol. 31, No. 10, pp. 563–571, 1998, DOI:
10.1016/S0301-679X(98)00077-2
[21] T. Arata, N. Novi, K. Ariga, A. Yamashita, G. Armenio, “Development of a Two-Stage Variable Displacement Vane Oil Pump”, 2012 SAE
TECHNICAL PAPER, doi: 10.4271/2012-01-0408, 2012 SAE World Congress and Exhibition.

Emma Frosina
Department of Industrial Engineering, University of Naples Federico II - Via Claudio 21, Naples, Italy

emma.frosina@unina.it
www.dii.unina.it