KSME International Journal, Vol. 16 No. 8, pp.

1095~ 1101, 2002 1095

Load Characteristics of Engine Main Bearing" Comparison

Between Theory and Experiment
Cho, Myung-Rae*, Oh, Dae-Yoon, Ryu, Seung-Hyuk
Power Tra#1 R & D Center, Hyundai Motor Co., Kyunggi-do 445-706, Korea
Han, Dong-Chul
School o f Aerospace and Mechanical Engineering, Seoul National University,
Seoul 151- 742, Korea

The load characteristics of engine main bearing are very important in the design of crankshaft
and engine block. The stiffness of crankshaft and block, or the optimal dimension of the bearing
can be determined according to the load level. This paper presents the load characteristics of
engine main bearing. Two components of the main bearing load are measured during engine
tiring and motoring. The vertical and horizontal load components are measured by using the
dynamic load cell mounted in each main bearing cap bolt. The measured main bearing loads are
compared with calculated results by using the statically determinate method. The theoretical
results, provided in this study, agreed well with the experimental results. The presented results
are very useful for achieving optimal design of engine.

Key W o r d s : L o a d , Main Bearing, Crankshaft, Statically Determinate Method, Bearing Cap,

Moving System

Nomenclature in engine has been steadily increasing. Hence, the

A : Phase angle of rotating mass operating conditions of crankshaft have become
F : Force more and more severe.
L : Length The loads on the crankshaft provide very
M : Mass important information to analyze the oil film
R : Rotating radius thickness and behavior of the crankshaft. Also,
0 : Crank angle they can be used as design criterion of crankshaft,
w : Angular velocity bearing and engine block, as well as boundary
conditions in FEM model.
I. Introduction Regarding the crankshaft load, the loads on the
crank pin can be simply obtained by using the
As the main source of vibration and noise in cylinder pressure and the inertia force of the
engine, the crankshaft plays a very important role piston and the connecting-rod. However, it is
in engine performance, and influences on the very difficult to obtain the loads on the crank
reliability and durability of the engine. In recent journal by analysis, because crankshaft is a
years, the tendency of engine design has become statically indeterminate system supported by mul-
small size and low weight, but the power required tiple bearings. Also, the experiment is not easy
due to the constraint of space and the difficulty of
* Corresponding Author. mounting the sensor.
E-mail : formell@hyundai-motor.com The statically determinate method (Cho, et al.,
TEL : -+-82-31-369-4517;FAX : +82-31-369-4503
Power Train R&D Center, Hyundai Motor Co., 2000) is commonly used to calculate the main
Kyunggi-do 445-706, Korea. (Manuscript Received bearing load. In this method, all the bearings are
November 26, 2001;Revised May 16, 2002) treated independently of each other. The other

2opyright (C) 2003 NuriMedia Co., Ltd.

 1096 Cho, Myung-Rae, Oh, Dae-Yoon, Ryu, Seung-Hyuk and Han, Dong-Chul

method is that a crankshaft is treated as a flexible

shaft supported by a spring and a damper, or a
plain journal bearing (Gross, and Husmann,
1966; Eglof, 1970; Tinaut, et al., 2000). The
finite element method is also used for more accu-
rate analysis (Mourelatos, 1995; Prakash, et al.,
1998). Fig. 1 Schematic diagram of 8-balance crankshaft
The loads on the main bearings can be gener-
ally understood to measure the tensile or com-
pressive strain of the bolt in the main bearing
cap. However, it is very difficult to attach the
strain gage to the engine block, and calibrate the
signal. In recent years, the method to measure the
vertical load by using the ring type load cell was 2 2 2 2 2 2 2 2

introduced (Ishihama, et al., 1981). In this method,

it is impossible to measure the horizontal load.
Therefore, Tinaut, et al. (2000) measured the L.,,,
vertical load, and then they indirectly derived the l
R
horizontal load from the vertical component by
using the multilineal regression method. However,
there are some differences between the theory and Fig. 2 Statistically determinate model of in-line 4
the experiment, and it is impossible to know the crankshaft system
horizontal load, if the load cell cannot detect the
negative force.
pin-jointed at each main bearing (see Fig. 2). In
In this study, vertical and horizontal load com-
this test engine, the crankshaft is composed of
ponents are simultaneously measured by using
four bays, and a bay consists of three webs. It is
two different ring type load cells. The experi-
also assumed that the pressure of each cylinder
mental results are compared with the theoretical
has only phase difference and the crankshaft is
results by using the statically determinate method.
symmetric with respect to the main 3 bearing.
The theoretical results agreed well with experi-
Therefore, the load of each bearing can be
mental results.
obtained by analysis of half side of the crankshaft
The usage of statically determinate method is
only.
very helpful for saving the computational time
From the above model, the force equilibrium of
and cost to verify the performance of crankshaft
No. 1 main bearing is defined as follows:
and main bearing in the early stage of design.
F.x,,(Lb~y,r L.,t) F.x,2(Lb~,: Lc.,=)
FmX,I
2. Model of Crankshaft System Lbay.l Lbay, l
+ ~ M~,b,,~R~.b.,.~sn(Aw.b,i.+0)[L~y,l-L,~.b,,.]=0 (1)
ir~=l tbay,t
Figure 1 shows the schematic diagram of the
F.y,l(L~y,l--L.,l) F.y.2(Lb~.:L..2)
crankshaft system used in this study. The crank- F,,y.~
Lb~.i Lb~,t
shaft has eight balance weights, and is supported + _ (2)

by five main bearings.

im=l tbay, l
In this study, the loads on the main bearings
are calculated by using the statically determin- From the 2nd to the n n - I th bearings, the force
ate method. In the case of in-line 4 engine, the equilibrium can be expressed as the following
crankshaft is treated as four independent systems form

2opyright (C) 2003 NuriMedia Co., Ltd.

 Load Characteristics of Engine Main Bearing." Comparison Between Theory and Experiment 1097

Fc,.l(Lbaya-Lcra) Fcry,z(Lba~,z-Lcra)
Fmy,2
Lb~ Lba~a
+ ~ Mw.b,,,R,~.b,,,co~cos(A,~.b,,,,+8)
[L~,z-L,.,,b,,,,,] (3)
im=l ' ' ' Lbaya
""i~m=lMweb, 2
i,Rzoeb,imOkjcos(Aweb,i.4"O). ~Lwebim
: 0

F~,,l(Lb~,2-L.,l) F.x~(L~,2-L.,2)
F.~:
Lb~a Lbay,z
+ ~ Mweb,imRweb,i.tO~sire(A..b,,. + O) [Lbaya- L..b,,m] (4)
im=l tbay,2

nil L " "

2 , tg~o,t m
+ ~ M~.b,,.R~,b,i.oJ~sin (Aw.~,~m+ 0 ) - r ~ = 0
ira=l Lbay, I

Im!

where 'nn' means the third main bearing in the

case of L-4 engine. load cell
The force equilibrium of nnth bearing can be Fig. 3 Schematic diagram of load cell mounting
written as
0111W g l e
F~,I (L~,~.- L~r,I) Fc~a(L~,..- L~,z)
F.=,.. L~,r~ L~,..
nn
(5)
+ ~ M~b ,.~g~ i.c0~s n (A.b ,m+ 0) L~b,i.
F~,I (L ~,,~- L~,I) F~,,z(L ~,.~- L~r,z)
F.~,.. L~,.~ L~,..
(6)
+ ,.=t
~ M~b.,~b,i.c0~cos(A~,i. + 8)L='::
=0,.

In Eqs. (1)-(6), the loads acting on the crank

pin can be calculated from the cylinder pressure
and the translating and rotating inertia forces of
the piston, connecting-rod and the crank pin.
3. Experiment
In this study, the main bearing loads are
measured by using the ring type quartz dynamic
load cell. The compressive and shear load cells
are simultaneously mounted in the bolts of the
main bearing cap as can be seen in Fig. 3. The
load cells are pre-loaded by using the assembling
torque of the cap bolt of the main bearing. The
vertical load is measured at No. I-4 main bear-
ings, and the horizontal load is only measured at
No. 3 main bearing.
In order to know the conversion coefficient
between the actual force and the output voltage of
load cell, a dummy shaft was inserted in the crank
journal hole and loaded with different weights
Controller Char~ DSP
Fig. 4 Schematic diagram of engine test system
and at various angles. By adding or extracting
the output signal of each load cell, the actual
force can be obtamed.
Figure 4 shows the schematic diagram of the
engine test rig and the data acquisition system.
The experiment under the firing and motoring
conditions was performed using the l l0kW DC
dynamometer. A 1.6L-DOHC engine was used
for the test. During the test, the temperature of the
cooling water was kept at 90C by using the water
cooler. The output signal from the load cell was
amplified by using the charge amplifier, and then
saved in DSP system. The cyclic variation of the

2opyright (C) 2003 NuriMedia Co., Ltd.

 1098 Cho, Myung-Rae, Oh, Dae- Yoon, Rvu Seung-Hyuk and Han, Dong Chul

Table 1 Specification of test engine bearing load was averaged during 300 engine
Engine type L4/1.6D cycles. The specification of the test engine is
described in Table 1.
Bore dia. (ram) 76.5
Stroke (mm) 87 4. Results and Discussion
r
Con rod ~width (mm~ 18
bearing [dia. (ram) 50 Figure 5 shows the comparative results between
theory and experiment at each bearing under the
Main width (ram) 16
full load condition at 1500 rpm. The actual
bearing [dia. (mm) 45 cylinder pressure is measured to use in load cal-
Piston Assy. mass (kg) 0.290 culation. The crank angle of 0 degree in the cycle
Crankshaft mass (kg) 13.4 indicates the cylinder 1 firing TDC position
The main 1 bearing is mainly influenced by
Con-rod mass (kg) 0.450
cylinder 1 (0 degree). The measured load is smaller
WOT (wide open throttle) than the calculated results. This is because the left
Load condition Part Load
end side of the crankshaft is supported by the
Motoring
timing belt rather than by bearing cap 1.

1200O i

--exp ~exp
- - -cal - - - - -eal
10000: main1, 1500rpm,WOT ma,n2, 1500rprn, WOT

lOOOoI

8OO0

~. 6ooo

200O 4O0O

0 20OO
,'<

-2000 I I , I ,
180 360 540 720 180 360 540 720
Crank angle(deg) Crank angle(deg)

(a) Main 1 (b) Main 2

12000 12OOO
Fx-cal
1O0OO - -- -Fx4e=p
. . . . Fy-cai 1OO00 main4, 1500q~, WOT
- - - - - Fy-exp
80OO rnam3,1,S(~rpm, WOT
880O

6000
Z
J;
,~
Jt
t~. ~" 8ooo
ttII
4O0O
2000
.; "-:.. ,,. ,,:
20O0 ii
-200( 0
180 360 540 720 180 360 540 720
Crank angle(deg) Crank angle(deg )

(c) Main 3 (d) Main 4

Fig, 5 The experimental and calculated bearing load in each main bearing

Copyright (C) 2003 NuriMedia Co., Ltd.

 Load Characteristics o f Engine Main Bearing." Comparison Between Theory and Experiment 1099

The main 2 bearing is influenced by cylinders
1 and 2 (540 degree), simultaneously. The effect
of cylinder 1 is relatively larger than cylinder 2.
This is because of the deviation of the cylinder
pressure between each cylinder and deformation
Fig. 8, there are some differences in the absolute
magnitude and position of the maximum load.
This is caused by the difference in the combustion
120OO i i

of crankshaft. Fx-cal
10000 -- - - Fx-exp
The main 3 bearing is at the symmetric position .~ .... Fy-cal
"'t -main3,
- - - - F2y0- ~0 0p r p m WO T
of the crankshaft and it is influenced by the
combustion pressure of all cylinders. But the
~ 6ON}

effects of cylinders 2 (540 degree) and 3 (180 g I

"
degree) are relatively greater than cylinders I
and 4 (360 degree)_ The cyclic variation of the
horizontal force is caused by the inertia force of
the rotating mass. The magnitude of the horizon- 0

tal force should increase as the engine speed

180 360 540 720
increases.
Crank engle(deg]
The main 4 bearing is affected by cylinders 3
(a) 2000 rpm
and 4, and has a phase difference with respect to
12000. . , . , . , .
main 2 bearing.
t Fx-cal
As can be seen in Fig. 5, the maximum load of 10000 I" -- -- -Fx-eXp
I- .... Fy-~I l
main 2 is the greatest of all bearings and the L ~' --- Fy-e~p

magnitude of actual force is different the accord- r !1 '

ing to the bearing position. As mentioned above, .....

this is because of the pressure deviation between 4ooo i#,: it # I/',.. ;~ .

the cylinders, deformation of the crankshaft, and I ~ I ~:1 ",/~-~ i i',: r

the effect of timing belt and flywheel, which are
0
not considered in this study.
The effects of engine speed on the main 3
A
bearing loads are shown in Fig. 6. At high
0 180 360 540 720
engine speed, the negative vertical force is not Crank angle(deg)
detected from the load cell, but the measured (b) 4000 rpm
results agree well with calculation. At low engine
speed, the cylinder pressure is the main source
16000 l - -- -- - Fx- e xp
of bearing force. As engine speed increases, the .... Fy-c~

effect of inertia force increases and the explosive

tbrce term decreases. The effect of inertia force
L ----- Fpexp

.....
,.. ,: t
is dominant at high engine speed in main 3 bear-
ing.
Figure 7 shows the effect of engine speed in the ,

main I and 2 bearings. The maximum force sig-

J
nificantly increases as the engine speed increases " i i ~,j~

due to increment of the cylinder pressure. The

-8000 i " TI , I ", *t
negative force is not detected in main 1 bearing. 0 180 360 540 720

The variation of inertia force is relatively small in Crank angle(deg)

main 2 bearing. (c) 6000 rpm
Figure 8 shows the comparative results between Fig. 6 The effect of engine speed on the bearing load
motoring and firing conditions. As can be seen in in the main 3 bearing

3opyright (C) 2003 NuriMedia Co., Ltd.

 1100 Cho, Myung-Rae, Oh, Dae- Yoon, Ryu, Seung-Hyuk and Hart, Dong-Chul

100O0 i 5O i

80OO

6000
-- -- -2O(]oq0m

main 1. VVOT
~5~0mm
4O - II
It
i t
I
I
1
- - -~tno
WOT, 2(X)Orpm

1
~" 4000 I
I

/
20 t
29OO
". ;v" . .. -".
I
~. . o,

-300(]
180 360 540 720 0
0 180 36O 540 720
Crank angle(daD)
Crank angle(deg)
(a) Main 1
Fig. 9 Cylinder pressure variation according to the
16000 1 i i load condition

12000 i t "-] - - h
II
12000

~
!

BMEP 2bar
i i

It
I
10000 .... BIdEP e b a r
4. - - --- BMEP 8bar

h~
2,
4 I1~, k IAr k ,
0 - - lit'; _.11~t;. z
0 180 360 540 720
Crank angle(dag) 20o0 F \t.~. !..11 \~>,-,.z~',
(b) Main 2
ol . ~'~ . . ',
0 180 300 540 720
Fig. 7 The effect of engine speed on the bearing
Crank angle(deg)
force in the main 1 and 2 bearing
(a) Experiment

12000 ! i

l BMEP 2bar
12(}O0 | i
L -- -- - BMEP 4bar
10000 F I . . . . BMEP el~r
~ motonng-~xp ~ - - - - - BMEP 8bar
10000 -- -- " ~ hi'i main2, 200Orpm
.... firin~p 8000 ==u~ i.!
B000
~
~
----- flring~
mare3, 2000rpm.WOT ~, t r,t
(.
i:!
i "!
b ( I.

" : ~ " -,. .'/ " ~ " "-'9

20O0
0

0 Crank angle(deg)
0 180 36O 540 72O
Crank angle(deg)
(b) Calculation

Fig. 8 The effect of engine load on the bearing Fig. 10 The effect of load on the bearing reaction
reaction force in the main 3 bearing force in the main 2 bearing

3opyright (C) 2003 NuriMedia C o . , Ltd.

 Load Characteristics of Engine Main Bearing." Comparison Between Theory and Experiment
process according to the operating condition
and can be confirmed in Fig. 9.
Figure 10 shows the effect of engine load on the
bearing reaction force. The cylinder pressure
increases with increasing BMEP (break mean
effective pressure). As the engine load increases,
the maximum force increases due to increasing
cylinder pressure. However, the inertia term is not
influenced by the load change.
5. Conclusion
To investigate the load characteristics on the
main bearings, the vertical and horizontal load
components were measured in the main bearing
caps by using the ring type load cells. Also, the
main bearing loads were analyzed by using the
statically determinate method, and then compared
with the experimental results. The measured
maximum load was generally greater than the
calculated results except main I bearing, and the
load cell could not detect negative vertical forces
in main 1 and 3 bearings. However, the theo-
retical results, provided in this study, agree well
with the experimental results.
The presented results in this study will be
helpful to understand the characteristics of crank-
shaft load and behavior.
, opyright (C) 2003 NuriMedia Co., Ltd.
1101
References
Cho, M. R., Shin, H.J. and Han, D. C., 2000,
"A Study on the Circumferential Groove Effects
on the Minimum Oil Film Thickness in Engine
Bearings," KSME International Journal, Vol. 14,
No. 7, pp. 737--743.
Eglof von Schnurbein, 1970, "A New Method
of Calculating Plain Bearings of Statically Inde-
terminate Crankshafts," SAE paper 700716.
Gross, W. and Hussmann, A. W., 1966, "Forces
in the Main Bearings of Multicylinder Engines,"
SAE paper 660756.
Ishihama, M., Hayashi, Y. and Kubozuka, T.,
1981, " A n Analysis of the Movement of the
Crankshaft Journal During Engine Firing," SAE
paper 810772.
Mourelatos, Z. P., 1995, " A n Analytical Inves-
tigation of the Crankshaft Flywheel Bending
Vibrations for a V6 Engine," SAE paper 951276.
Prakash, V., Aprameyan, K. and Shrinivasa, U.,
1998, "An F E M Based Approach to Crankshaft
Dynamics and Life Estimation," SAE paper
980565
Tinaut, F. V., Melgar, A., Gimenez, B., Fernan-
dez, L. and Huidobro, H., 2000, "A Method to
Determine the Two Components of the Crank-
shaft Load on a Bearing Cap in Firing Engines,"
SAE paper 2000-01-1340.