CH 23

Download as pdf or txt
Download as pdf or txt
You are on page 1of 18

Subramanyan, P. K.

“Crankshaft Journal Bearings”


The Engineering Handbook.
Ed. Richard C. Dorf
Boca Raton: CRC Press LLC, 2000

© 1998 by CRC PRESS LLC


23
Crankshaft Journal Bearings
23.1 Role of the Journal Bearings in the Internal Combustion Engine
23.2 Construction of Modern Journal Bearings
23.3 The Function of the Different Material Layers in Crankshaft Journal
Bearings
23.4 The Bearing Materials
23.5 Basics of Hydrodynamic Journal Bearing Theory
Load-Carrying Ability
23.6 The Bearing Assembly
Housing • The Bearing Crush • Other Factors Affecting Bearing Assembly
23.7 The Design Aspects of Journal Bearings
23.8 Derivations of the Reynolds and Harrison Equations for Oil Film Pressure

P. K. Subramanyan
Glacier Clevite Heavywall Bearings

In modern internal combustion engines, there are two kinds of bearings in the category of
crankshaft journal bearingsnamely, the main bearings and the connecting rod bearings.
Basically, these are wraparound, semicylindrical shell bearings. Two of them make up a set and,
depending on the position in the assembly, one is called the upper and the other the lower bearing.
They are of equal sizes. The main bearings support the crankshaft of the engine and the forces
transmitted to the crankshaft from the cylinders. The connecting rod bearings (or, simply, rod
bearings) are instrumental in transferring the forces from the cylinders of the internal combustion
engine to the crankshaft. These connecting rod bearings are also called big end bearings or crank
pin bearings. Supporting the crankshaft and transferring the pressure-volume work from the
cylinders to the pure rotational mechanical energy of the crankshaft are accomplished elegantly
with minimal energy loss by shearing a suitable lubricating medium between the bearings and the
journals. The segment of the crankshaft within the bounds of a set of bearings, whether main
bearings or rod bearings, is called the journal. Consequently, these bearings are called journal
bearings.

23.1 Role of the Journal Bearings in the Internal Combustion


Engine
The crankshafts of internal combustion engines of sizes from small automotive to large slow-speed
engines run at widely varying rpm (e.g., 72 to 7700). When the internal combustion engine

© 1998 by CRC PRESS LLC


continues to run after the start-up, the crankshaft, including the crank pins, is suspended in the
lubricating oila fluid of very low friction. In such a condition, it is conceivable that
precision-machined, semicylindrical steel shells can function as good bearings. However, there are
stressful conditions, particularly in the case of automotive, truck, and medium-speed engines, when
the crankshaft remains in contact with the bearings and there is little or no lubricating oil present.
This condition corresponds to the initial and subsequent start-ups. The oil pump is driven directly
by the engine and it takes several revolutions of the crankshaft before a good oil film is developed,
as shown in Fig. 23.1, so that the journals are completely lifted and suspended. During the
revolutions prior to the formation of a sufficiently thick oil film, the journal contacts the bearing
surface. In such situations, the bearings provide sufficient lubrication to avoid scuffing and
seizure. Another stressful situation, but not as critical as the start-up, is the slowing down and
shutting off of the engine when the oil film reduces to a boundary layer.

Figure 23.1 Schematic representation of the hydrodynamic lubricant film around a rotating journal in its
bearing assembly. (Source: Slaymaker, R. R. 1955. Bearing Lubrication Analysis. John Wiley & Sons,
New York. With permission.)

In the case of slow-speed engines, the oil pump, which is electrically driven, is turned on to
prelubricate the bearings. This provides some lubrication. Nonetheless, bearings with liners and
overlays are used to avoid seizure, which can result in costly damage.
Essentially, the function of journal bearings can be stated as follows:Development of the
hydrodynamic lubricating oil films in the journal bearings lifts the journals from the surfaces of
the bearings and suspends the entire crankshaft on the oil films by the journals. [Theoretical
aspects of this will be considered later.] The lifting of the crankshaft or, equivalently, lifting of the
journals is in the range of 30 to 1000 micro-inch in the entire range of IC engines. This process

© 1998 by CRC PRESS LLC


allows the crankshaft to rotate with minimal energy loss. The journal bearings make it possible so
that the internal combustion engine can be started, utilized, and stopped as many times as needed.

23.2 Construction of Modern Journal Bearings


The majority of modern crankshaft journal bearings have three different layers of metallic
materials with distinct characteristics and functions. Conventionally, these are called trimetal
bearings. The remaining bearings belong to the class of bimetal bearings and have two different
metallic material layers. Bimetallic bearings are becoming very popular in the automotive
industry.
All crankshaft journal bearings have a steel backing, normally of low-carbon steels. Steel
backing is the thickest layer in the bearing. The next layer bonded to the steel backing is the
bearing liner. This is the layer that supports the load and determines the life of the bearing. The
third layer bonded to the bearing liner is the overlay. Generally, this is a precision electrodeposited
layer of (1) lead, tin, and copper, (2) lead and tin, or (3) lead and indium. A very thin
electrodeposited layer of nickel (0.000 05 in.) is used as a bonding layer between the liner and the
lead-tin-copper overlay. This nickel layer is considered a part of the overlay, not a separate layer.
Construction of a trimetal bronze bearing is illustrated in Fig. 23.2.

Figure 23.2 Schematic representation of the construction of a trimetal bearing.

© 1998 by CRC PRESS LLC


There are two classes of bearing liners in widespread use nowadays. These are the leaded
bronzes and aluminum-based (frequently precipitation-strengthened) materials, such as
aluminum-tin and aluminum-silicon. Bimetallic bearings have the advantage of being slightly more
precise (about 0.0002 to 0.0003 in.) than the trimetal bearings. The bimetal bearings have a bored
or broached internal diametral (ID) surface. The electrodeposited layer in the trimetal bearings is
applied onto the bored or broached surface. The nickel bonding layer is applied first onto the liner,
followed by the deposition of the lead-tin-copper overlay. The electrodeposited overlay introduces
a certain degree of variation in the wall thickness of the bearings. In a limited application, babbitt
overlays are centrifugally cast on bronze liners for slow-speed diesel engine journal
bearings.
Another class of bearings is the single layer solid metal bearingsnamely, solid bronze and
solid aluminum bearings. These bearings are not generally used as crankshaft journal bearings.
However, solid aluminum is used in some of the medium-speed and slow-speed diesel
engines.
The most popular copper-tin-based leaded bearing liner in current use has 2 to 4% tin, 23 to 27%
lead, and 69 to 75% copper (all by weight). This material is applied directly on mild steel by
casting or sintering. The aluminum materials are roll-bonded to steel. The material as such is
produced by powder rolling as a strip or by casting and rolling.

23.3 The Function of the Different Material Layers in


Crankshaft Journal Bearings
The bulk of modern crankshaft journal bearings is mild steel (1008 to 1026 low-carbon
steels). This is the strongest of the two or three layers in the bearing. It supports the
bearing liner, with or without the overlay. The bearing liner derives a certain degree of
strength from the steel backing. The function of the steel backing is to carry the bearing
liner, which on its own is weaker, much thinner, and less ductile. With the support of
the steel backing, the bearings can be seated with a high degree of conformance and
good interference fit in the housing bore (steel against steel).
The bearing liners in automotive and truck bearings have a thickness in the range of
0.006 to 0.030 in. In the case of the medium-speed and slow-speed engines, the
thickness of the liner ranges from 0.010 to 0.080 in. The liner material contains
sufficient amounts of antifriction elements, such as lead and tin. Lead is the most
valuable antifriction element in the current materials and is present as a separate phase
in the matrix of copper-tin alloy in the leaded bronze materials. Similarly, tin is present
as an insoluble second phase in the matrix of aluminum-based materials. Lead is also
insoluble in the aluminum matrix. The liner materials play the most critical role in the
bearings. Once the liner material is damaged significantly, the bearing is considered
unfit for further use. In a trimetal bearing, when the overlay is lost due to wear or
fatigue, the bronze liner will continue to support the load and provide adequate
lubrication in times of stress. The friction coefficient of liner materials is designed to be
low. Besides, the soft phases of lead (in bronze) and tin (in aluminum) function as sites
for embedment of dirt particles.

© 1998 by CRC PRESS LLC


The overlay, which by definition is the top layer of the bearing surface, is the softest layer in the
bearing. Its functions are to provide lubrication to the journal in the initial start-up situations, adjust
to any misalignment or out-of-roundness of the journal, and capture dirt particles by embedment.
The overlay provides sufficient lubrication during the subsequent start-up and shut-down
conditions also. The journal makes a comfortable running environment in the bearing assembly
during the initial runs by "bedding in." As a result of this, the wear rate of the overlay is higher in
the beginning. As long as the overlay is present, the phenomenon of seizure will not occur. Once
the wear progresses through the overlay, the bearing liner will provide adequate lubrication during
start-up and shut-down conditions. However, if the oil supply is severely compromised or cut off
for more than several seconds to a minute or so, seizure can take place once the overlay is gone,
depending on the nature of the bearing liner and the load.

23.4 The Bearing Materials


All modern crankshaft journal bearing materials are mainly composed of five elementsnamely,
copper, aluminum, lead, tin, and silicon. These elements account for the leaded bronze and
aluminum-tin, aluminum-lead, and aluminum-silicon materials. Indium is used as a constituent of
the overlays. Antimony is used in babbitts. Silver is a bearing material with good tribological
properties, but it is too expensive to use as a bearing liner in journal bearings. However, it is used
in special applications in some locomotive engines. An important characteristic of a good bearing
material is its ability to conduct heat. Silver, copper, and aluminum are, indeed, good conductors of
heat. Silver has no affinity for iron, cobalt, and nickel [Bhushan and Gupta, 1991]. Therefore, it is
expected to run very well against steel shafts. Both copper and aluminum possess a certain degree
of affinity for iron. Therefore, steel journals can bond to these metals in the absence of antifriction
elements, such as lead and tin, or lubricating oil. Aluminum spontaneously forms an oxide layer,
which is very inert, in the presence of air or water vapor. This suppresses the seizure or the
bonding tendency of aluminum. Besides, the silicon particles present in the aluminum-silicon
materials keep the journals polished to reduce friction.
The microstructure of the most widely used cast leaded bronze bearing liner is shown in Fig.
23.3. This has a composition of 2 to 4% tin, 23 to 27% lead, and 69 to 75% copper. Another
material in widespread use, especially in automotive applications, is aluminum with 20% tin. A
typical microstructure of this material is shown in Fig. 23.4. It can be used as the liner for both
bimetal and trimetal bearings. The copper-tin-lead material shown in Fig. 23.3 is mainly used in
trimetal bearings.

© 1998 by CRC PRESS LLC


Figure 23.3 SEM photomicrograph of a typical cross section of the cast leaded bronze diesel locomotive
engine bearing material manufactured by Glacier Clevite Heavywall Bearings. The nominal composition is
3% tin, 25% lead, and 72% copper. The light gray, irregular spots represent lead in a matrix of copper-tin.
This material is bonded to mild steel at the bottom. (Magnification 50£.)]

Figure 23.4 SEM photomicrograph of a typical cross section of aluminum-tin material roll bonded to
mild steel, manufactured by Glacier Vandervell Ltd. The nominal composition is 20% tin, 1% copper, and
79% aluminum. The light gray, irregular spots represent tin in the aluminum-copper matrix. Below the
aluminum-tin layer is a layer of pure aluminum which functions as a bonding layer to the mild steel
underneath. (Magnification 210£ .)

© 1998 by CRC PRESS LLC


23.5 Basics of Hydrodynamic Journal Bearing Theory
Load-Carrying Ability
As mentioned previously, when running in good condition, the journal which was initially lying on
the surface of the bearing is lifted and surrounded by the lubricant. It becomes suspended in the
surrounding film of lubricating oil. If the engine keeps running, the journal will remain in its state
of suspension indefinitely. The inertial load of the crankshaft and the forces transmitted from the
cylinders to the crankshaft are supported by the lubricant films surrounding the main bearing
journals. The oil film surrounding the rod bearing journal supports the gas forces developed in the
cylinder and the inertial load of the piston and connecting rod assembly. Around each journal, a
segment of the oil film develops a positive pressure to support the load, as shown in Fig. 23.5. In
the following brief theoretical consideration, the process that develops this load-carrying positive
pressure will be illustrated.

Figure 23.5 Schematic representation of the profile of the load supporting pressure in the oil
film. (Source: Slaymaker, R. R. 1955. Bearing Lubrication Analysis. John Wiley & Sons, New
York. By permission.)

As a background to the theoretical considerations, the following assumptions are


made. The flow of the lubricating oil around the journal at all speeds is assumed to be
laminar. The length of the bearing L is assumed to be infinite, or the flow of the
lubricant from the edges of the bearing is negligible. The lubricant is assumed to be
incompressible.

© 1998 by CRC PRESS LLC


Consider a very small volume element of the lubricant moving in the direction of rotation of the
journalin this case, the x direction. The forces that act on this elemental volume and stabilize it
are shown in Fig. 23.6. Here, P is the pressure in the oil film at a distance x. It is independent of
the thickness of the oil film or the y dimension. S is the shear stress in the oil film at a distance y
above the bearing surface, which is at y = 0 . The length L of the bearing is in the z direction. The
equilibrium condition of this volume element gives us the following relationship [Slaymaker,
1955; Fuller, 1984]:
· µ ¶ ¸ · µ ¶ ¸
dP dS
P+ dx dy dz + S dx dz ¡ S + dy dx dz ¡ P dy dz = 0
dx dy

(23:1)

Therefore,
µ ¶ µ ¶
dS dP
= (23:2)
dy dx

Equation (23.2) represents a very important, fundamental relationship. It clearly shows how the
load-carrying pressure P is developed. It is the rate of change of the shear stress in the direction of
the oil film thickness that generates the hydrostatic pressure P. As we shall see from Eq. (23.3), the
shear stress is directly proportional to the shearing rate of the oil film (dv=dy) as (dv=dy)
increases, (dS=dy) must increase. Since the thickness of the oil film decreases in the direction of
rotation of the journal, a progressive increase in the shearing rate of the oil film automatically
occurs because the same flow rate of oil must be maintained through diminishing cross sections
(i.e., decreasing y dimension). This progressive increase in the shearing rate is capable of
generating very high positive hydrostatic pressures to support very high loads. A profile of the
pressure generated in the load-supporting segment of the oil film is shown in Fig. 23.5. By
introducing the definition of the coefficient of viscosity, we can relate the shear stress to a more
measurable parameter, such as the velocity, v, of the lubricant, as
µ ¶
dv
S=¹ (23:3)
dy

Figure 23.6 Schematic representation of the forces acting on a tiny volume element in the hydrodynamic
lubricant film around a rotating journal.

© 1998 by CRC PRESS LLC


Substituting for (dS=dy) from Eq. (23.3) in Eq. (23.2), we obtain a second order partial
differential equation in v. This is integrated to give the velocity profile as a function of y. This is
then integrated to give Q, the total quantity of the lubricant flow per unit time. Applying certain
boundary conditions, one can deduce the well-known Reynolds equation for the oil film
pressure:
µ ¶
dP 6¹V
= (h ¡ h1 ) (23:4)
dx h3

where h is the oil film thickness, h1 is the oil film thickness at the line of maximum oil film
pressure, and V is the peripheral velocity of the journal. The variable x in the above equation can
be substituted in terms of the angle of rotation µ and then integrated to obtain the Harrison equation
for the oil film pressure. With reference to the diagram in Fig. 23.7, the thickness of the oil film
can be expressed as

h = c(1 + " cos µ) (23:5)

where c is the radial clearance and " is the eccentricity ratio. The penultimate form of the Harrison
equation can be expressed as
Z 2¼ Z 2¼ · ¸
6¹V r" cos µ ¡ cos µ1
dP = dµ = P ¡ P0 (23:6)
0 0 c2 (1 + " cos µ)3

where P0 is the pressure of the lubricant at µ = 0 in Fig. 23.7, and µ1 is the angle at which the oil
film pressure is a maximum. Brief derivations of the Reynolds equation and the Harrison equation
are given in section 23.8.

Figure 23.7 Illustration of the geometric relationship of a journal rotating in its bearing assembly.
(Source: Slaymaker, R. R. 1955. Bearing Lubrication Analysis. John Wiley and Sons, New York. By permission.)

© 1998 by CRC PRESS LLC


For practical purposes, it is more convenient to carry out the integration of Eq. (23.6)
numerically rather than using Eq. (23.14) in section 23.8. This is done with good accuracy using
special computer programs. The equations presented above assume that the end leakage of the
lubricating oil is equal to zero. In all practical cases, there will be end leakage and, hence, the oil
film will not develop the maximum possible pressure profile. Therefore, its load-carrying
capability will be diminished. The flow of the lubricant in the z direction needs to be taken into
account. However, the Reynolds equation for this case has no general solution [Fuller, 1984].
Hence, a correction factor between zero and one is applied, depending on the length and diameter
of the bearing (L/D ratio) and the eccentricity ratio of the bearing. Indeed, there are tabulated
values available for the side leakage factors for bearings with various L/D ratios and eccentricity
ratios [Fuller, 1984]. Some of these values are given in Table 23.1.

Table 23.1 Side Leakage Correction Factors for Journal Bearings

Eccentricity Ratio
L/D Ratio 0.80 0.90 0.92 0.94 0.96 0.98 0.99
0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
2  0.867 0.88 0.905 0.937 0.97 0.99
1 0.605 0.72 0.745 0.79 0.843 0.91 0.958
0.5 0.33 0.50 0.56 0.635 0.732 0.84 0.908
0.3 0.17 0.30 0.355 0.435 0.551 0.705 0.81
0.1  0.105 0.115 0.155 0.220 0.36 0.53

Booker [1965] has done considerable work in simplifying the journal center orbit calculations
without loss of accuracy by introducing new concepts, such as dimensionless journal center
velocity/force ratio (i.e., mobility) and maximum film pressure/specific load ratio (i.e., maximum
film pressure ratio). This whole approach is called the mobility method. This has been developed
into computer programs which are widely used in the industry to calculate film pressures and
thicknesses. Further, this program calculates energy loss due to the viscous shearing of the
lubricating oil. These calculations are vital for optimizing the bearing design and selecting the
appropriate bearing liner with the required fatigue life. This is determined on the basis of the peak
oil film pressure (POFP). In Booker's mobility method, the bearing assembly, including the
housing, is assumed to be rigid. In reality, the bearings and housings are flexible to a certain
degree, depending on the stiffness of these components. Corrections are now being made to these
deviations by the elastohydrodynamic theory, which involves finite element modeling of the
bearings and the housing. Also, the increase in viscosity as a function of pressure is taken into
account in this calculation. The elastohydrodynamic calculations are presently done only in very
special cases and have not become part of the routine bearing analysis.

© 1998 by CRC PRESS LLC


23.6 The Bearing Assembly

Housing
The housing into which a set of bearings is inserted and held in place is a precision-machined
cylindrical bore with close tolerance. The surface finishes of the housing and the backs of the
bearings must be compatible. Adequate contact between the backs of the bearings and the surface
of the housing bore is a critical requirement to ensure good heat transfer through this interface. The
finish of the housing bore is expected to be in the range of 60 to 90 ¹in: (Ra ) (39.4 ¹in: = 1
micron). The finish on the back of the bearings is generally set at 80 ¹in: maximum. Nowadays,
the finishes on the housing bore and the backs of the bearings are becoming finer. The finish at the
parting line face of bearings of less than 12 in. gage size is expected to be less than 63 ¹in . For
larger bearings, this is set at a maximum of 80 ¹in . The bearing backs may be rolled, turned, or
ground. All the automotive and truck bearings have rolled steel finish at the back. The housing can
be bored, honed, or ground, but care must be taken to avoid circumferential and axial
banding.

The Bearing Crush


The term crush is not used in a literal sense in this context. A quantitative measure of the crush of
a bearing is equal to the excess length of the exterior circumference of the bearing over half the
interior circumference of the bearing housing. Effectively, this is equal to the sum of the two
parting line heights. When the bearing assembly is properly torqued, the parting line height of each
bearing in the set is reduced to zero. In that state, the back of the bearing makes good contact with
the housing and applies a radial pressure in the range of 800 to 1200 psi (5.5 to 8.24 MPa).
Thereby, a good interference fit is generated. If the bearings are taken out of the assembly, they are
expected to spring back to their original state. Therefore, nothing is actually
crushed.
The total crush or the parting line height of a bearing has three componentsnamely, the
housing bore tolerance crush, the checking load crush, and the engineering crush. The housing
bore tolerance crush is calculated as 0:5¼(D2 ¡ D1 ) , where D1 and D2 are the lower and upper
limits of the bore diameter, respectively. Suppose a bearing is inserted in its own inspection block
(the diameter of which corresponds to the upper limit of the diameter of the bearing housing). The
housing bore tolerance crush does not make a contribution to the actual crush, as shown in Fig.
23.8 (high limit bore). If load is applied on its parting lines in increasing order and the values of
these loads are plotted as a function of the cumulative decrease in parting line height, one may
expect it to obey Hooke's law. Initially, however, it does not obey Hooke's law, but it does so
thereafter. The initial nonlinear segment corresponds to the checking load crush. The checking
load corresponds to the load required to conform the bearing properly in its housing. The final
crush or the parting line height of the bearing is determined in consultation with the engine
manufacturer.

© 1998 by CRC PRESS LLC


Figure 23.8 Schematic illustration of the components of crush of a bearing in the thinwall bearing
inspection block, before application of load (i.e., in the free state). The magnitude of the crush components
is exaggerated.

Other Factors Affecting Bearing Assembly


These factors are (1) freespread, (2) bore distortion, (3) cap offset or twist, (4) misalignment of the
crankshaft, (5) out-of-roundness of the journal, and (6) deviation of the bearing clearance. The
outside diameter of the bearing at the parting lines must be slightly greater than the diameter of the
housing bore. This is called the freespread. It helps to snap the bearings into the housing. The
required degree of freespread is determined by the wall thickness and the diameter. In the case of
wall thickness, the freespread is inversely proportional to it. For a wide range of bearings, the
freespread is in the range of 0.025 to 0.075 in. Bearings with negative freespread are not used
because, when bolted, the side of the parting lines could rub against the journal and lead to
possible seizure while running. It is possible to change the freespread from negative to positive by
reforming the bearing. Bore distortion, cap offset or twist, and misalignment of the crankshaft can
lead to the journal making rubbing contacts with the bearing surface. The conformability of the
bearings can take care of these problems to a certain degree by local wearing of the overlay in a
trimetal bearing or by melting the soft phase in a bimetal bearing, which results in the two-phase
structure crushing and conforming. In severe cases, the liner materials in both cases are
damaged.
By developing high oil film pressures on the peaks of the lobes, out-of-roundness in the journal
can accelerate fatigue of the bearing.
If the clearance is not adequate, the bearing will suffer from oil starvation and the temperature
will rise. In extreme cases, this will lead to bearing seizure and engine damage. On the other hand,
if the clearance is excessive, there will be increased noise and increased peak oil film pressure,
which will bring about premature fatigue of the loaded bearing.

© 1998 by CRC PRESS LLC


23.7 The Design Aspects of Journal Bearings
Even though the journal bearings are of simple semicylindrical shape and apparently of
unimpressive features, there are important matters to be taken into account in their design. The
bearing lengths, diameters, and wall thicknesses are generally provided by the engine builder or
decided in consultation with the bearing manufacturer. A journal orbit study must be done to
optimize the clearance space between the journal and the bearing surface. This study also provides
the minimum oil film thickness (MOFT) and the POFP (Fig. 23.9). Values of these parameters
for the optimized clearance are important factors. The MOFT is used in the calculation of the oil
flow, temperature rise, and heat balance. According to Conway-Jones and Tarver [1993], about
52% of the heat generated in connecting rod bearings in automobile engines is carried away by the
oil flow. Approximately 38% of the remaining heat flows into the adjacent main bearings via the
crankshaft. The remaining 10% is lost by convection and radiation. In the case of main bearings,
about 95% of the total heat is carried away by the oil flow, which is estimated to be more than five
times the flow through the connecting rod bearings, which were fed by a single oil hole drilled in
the crank pin. The POFP is the guiding factor in the selection of a bearing liner with adequate
fatigue strength or fatigue life.

Figure 23.9 Journal center orbit diagram of two-stroke cycle medium-speed (900 rpm) diesel engine
main bearings (no. 1 position). The inner circle represents the clearance circle of the bearings. It also
represents the bearing surface. The entire cross section of the journal is reduced to a point coinciding with
the center of the journal. The upper main bearing has an oil hole at the center with a circumferential groove
at the center of the bearing represented by the dark line. Maximum unit load: 1484 psi. MOFT: 151 ¹in. @
70/166. POFP: 11 212 psi @ 55/171. Oil: SAE 30W. Cylinder pressure data given by the manufacturer of
the engine. Clockwise rotation. The journal orbit analysis done at Glacier Clevite Heavywall Bearings.
 ¤ 0−180 crank angle,  + 180−360 crank angle, @ crank angle/bearing angle. Arrow indicates the
location of MOFT.

The bearing must be properly located in the housing bore. This is achieved by having a notch at
one end of the bearing at the parting line. There must be provisions to bring in the lubricant and
remove it. Therefore, appropriate grooves and holes are required. The best groove to distribute the

© 1998 by CRC PRESS LLC


lubricant is a circumferential groove with rounded edges, centrally placed in both bearings. If this
is a square groove, the flow will be diminished by 10%. If these grooves are in the axial direction,
the oil flow is decreased by 60% with respect to the circumferential ones. Having a circumferential
groove in the loaded half of the bearings does increase the POFP. In the case of large slow-speed
diesel engines, the POFPs are generally very low compared to the pressures in automotive, truck,
and medium-speed diesel engines. Therefore, central circumferential grooves are best suited for
slow-speed engines.
In the automotive, truck, and medium-speed engines, the loaded halves of the bearings do not
have circumferential grooves. However, the other halves have the circumferential grooves. Some
of the loaded bearings have partial grooves. Otherwise, some type of oil spreader machined in the
location below the parting line is desirable in the case of larger bearings. If the oil is not spread
smoothly, the problems of cavitation and erosion may show up. The end of the partial groove or
the oil spreader must be blended.
The edges of all the bearings must be rounded or chamfered to minimize the loss of the lubricant.
Edges are also chamfered to eliminate burrs. A sharp edge acts as an oil scraper and thereby
enhances oil flow in the axial direction along the edges, which is harmful. Finally, bearings have a
small relief just below the parting lines along the length on the inside surface. This is meant to
protect the bearings in case of slight misalignment or offset at the parting
lines.

23.8 Derivations of the Reynolds and Harrison Equations for Oil


Film Pressure
The background for deriving these equations is given in section 23.5 of the text. The equilibrium
condition of a tiny volume element of the lubricating oil (Fig. 23.6) is represented by the following
equation [Slaymaker, 1955; Fuller, 1984]:
· µ ¶ ¸ · µ ¶ ¸
dP dS
P+ dx dy dz + S dx dz ¡ S + dy dx dz ¡ P dy dz = 0
dx dy

(23:7)

Therefore,
µ ¶ µ ¶
dS dP
= (23:8)
dy dx

Now, by introducing the definition of the coefficient of viscosity ¹ , we can relate the shear stress
to a more measurable parameter, like the velocity v of the lubricant, as
µ ¶
dv
S=¹ (23:9)
dy

Substituting for (dS=dy) from Eq. (23.9) in Eq. (23.8), a second order partial differential equation
in v is obtained. This is integrated to give an expression for the velocity profile as

© 1998 by CRC PRESS LLC


µ ¶
V 1 dP
v= y¡ (hy ¡ y 2 ) (23:10)
h 2¹ dx
In Eq. (23.10), V is the peripheral velocity of the journal and h is the oil film thickness. The
boundary conditions used to derive Eq. (23.10) are (1) v = V when y = h , and (2) v = 0 when
y = 0 (at the surface of the bearing). Now applying the relationship of continuity, the oil flowing
past any cross section in the z direction of the oil film around the journal must be equal. The
quantity Q of oil flow per second is given by
Z h
Q=L v dy (23:11)
0

where L is the length of the bearing which is in the z direction. Now substituting for v from Eq.
(23.10) in Eq. (23.11) and integrating,
· µ ¶¸
Vh h3 dP
Q=L ¡ (23:12)
2 12¹ dx

The pressure P varies as a function of x in the oil film, which is in the direction of rotation of the
journal. At some point, it is expected to reach a maximum. At that point, (dP=dx) becomes zero.
Let h1 represent the oil film thickness at that point. Therefore,

LV
Q= h1 (23:13)
2

Now we can use Eq. (23.13) to eliminate Q from Eq. (23.12). Hence,
µ ¶
dP 6¹V
= (h ¡ h1 ) (23:14)
dx h3

Equation (23.14) is the Reynolds equation for the oil film pressure as a function of distance in the
direction of rotation of the journal. The variable x in Eq. (23.14) can be substituted in terms of the
angle of rotation µ and then integrated to obtain the Harrison equation for the oil filmpressure.
With reference to the diagram in Fig. 23.7, the oil film thickness h can be expressed as
q
h = e cos µ + (r + c)2 ¡ e2 sin 2 µ ¡ r (23:15)

Here, e is the eccentricity, c is the radial clearance, and e = c" , where " is the eccentricity ratio.
The quantity e2 sin 2 µ is much smaller compared to (r + c)2 . Therefore,

h = c(1 + " cos µ) (23:16)

Now, (dP=dx) is converted into polar coordinates by substituting rdµ for dx . Therefore, Eq.
(23.14) can be expressed as
µ ¶ · ¸
dP 6¹V r" cos µ ¡ cos µ1
= (23:17)
dµ c2 (1 + " cos µ)3

© 1998 by CRC PRESS LLC


where µ1 is the angle at which the oil film pressure is a maximum. Integration of Eq. (23.17) from
µ = 0 to µ = 2¼ can be expressed as
Z 2¼ Z 2¼ · ¸
6¹V r" cos µ ¡ cos µ1
dP = dµ = P ¡ P0 (23:18)
0 0 c2 (1 + " cos µ)3
where P0 is the pressure of the lubricant at the line of centers (µ = 0) in Fig. 23.7. If (P ¡ P0 ) is
assumed to be equal to zero at µ = 0 and µ = 2¼ , the value of cos µ1 , upon integration of Eq.
(23.18), is given by

3"
cos µ1 = ¡ (23:19)
2 + "2

and the Harrison equation for the oil film pressure for a full journal bearing
by

6¹V r" sin µ(2 + " cos µ)


P ¡ P0 = (23:20)
c2 (2 + "2 )(1 + " cos µ)2

Acknowledgment
The author wishes to express his thanks to David Norris, President of Glacier Clevite Heavywall
Bearings, for his support and interest in this article, and to Dr. J. M. Conway-Jones (Glacier Metal
Company, Ltd., London), George Kingsbury (Consultant, Glacier Vandervell, Inc.), Charles
Latreille (Glacier Vandervell, Inc.), and Maureen Hollander (Glacier Vandervell, Inc.) for
reviewing this manuscript and offering helpful suggestions.

Defining Terms
Boundary layer lubrication: This is a marginally lubricating condition. In this case, the surfaces
of two components (e.g., one sliding past the other) are physically separated by an oil film
that has a thickness equal to or less than the sum of the heights of the asperities on the
surfaces. Therefore, contact at the asperities can occur while running in this mode of
lubrication. This is also described as "mixed lubrication." In some cases, the contacting
asperities will be polished out. In other cases, they can generate enough frictional heat to
destroy the two components. Certain additives can be added to the lubricating oil to reduce
asperity friction drastically.
Crush: This is the property of the bearing which is responsible for producing a good interference
fit in the housing bore and preventing it from spinning. A quantitative measure of the crush is
equal to the excess length of the exterior circumference of the bearing over half the interior
circumference of the housing. This is equal to twice the parting line height, if measured in an
equalized half height measurement block.
Hydrodynamic lubrication: In this mode of lubrication, the two surfaces sliding past each other
(e.g., a journal rotating in its bearing assembly) are physically separated by a liquid lubricant
of suitable viscosity. The asperities do not come into contact in this case and the friction is
very low.
Minimum oil film thickness (MOFT): The hydrodynamic oil film around a rotating journal
develops a continuously varying thickness. The thickness of the oil film goes through a

© 1998 by CRC PRESS LLC


minimum. Along this line, the journal most closely approaches the bearing. The maximum
wear in the bearing is expected to occur around this line. Therefore, MOFT is an important
parameter in designing bearings.
Peak oil film pressure (POFP): The profile of pressure in the load-carrying segment of the oil
film increases in the direction of rotation of the journal and goes through a maximum (Fig.
23.5). This maximum pressure is a critical parameter because it determines the fatigue life of
the bearing. This is also called maximum oil film pressure (MOFP).
Positive freespread: This is the excess in the outside diameter of the bearing at the parting line
over the inside diameter of the housing bore. As a result of this, the bearing is clipped in
position in its housing upon insertion. Bearings with negative freespread will be loose and
lead to faulty assembly conditions.
Seizure: This is a critical phenomenon brought about by the breakdown of lubrication. At the core
of this phenomenon is the occurrence of metal-to-metal bonding, or welding, which can
develop into disastrous levels, ultimately breaking the crankshaft. With the initiation of
seizure, there will be increased generation of heat, which will accelerate this phenomenon.
Galling and adhesive wear are terms which mean the same basic phenomenon. The term
scuffing is used to describe the initial stages of seizure.

References
Bhushan, B. and Gupta, B. K. 1991. Handbook of Tribology. McGraw-Hill, New York.
Booker, J. F. 1965. Dynamically loaded journal bearings: Mobility method of solution. J. Basic
Eng. Trans. ASME, series D, 87:537.
Conway-Jones, J. M. and Tarver, N. 1993. Refinement of engine bearing design techniques. SAE
Technical Paper Series, 932901, Worldwide Passenger Car Conference and
Exposition, Dearborn, MI, October 25−27.
Fuller, D. D. 1984. Theory and Practice of Lubrication for Engineers, 2nd ed. John Wiley & Sons,
New York.
Slaymaker, R. R. 1955. Bearing Lubrication Analysis. John Wiley & Sons, New York.

Further Information
Yahraus, W. A. 1987. Rating sleeve bearing material fatigue life in terms of peak oil film pressure.
SAE Technical Paper Series, 871685, International Off-Highway & Powerplant Congress
and Exposition, Milwaukee, WI, September 14−17.
Booker, J. F., 1971. Dynamically loaded journal bearings: Numerical application of the mobility
method. J. of Lubr. Technol. Trans. ASME, 93:168.
Booker, J. F., 1989. Squeeze film and bearing dynamics. Handbook of Lubrication, ed. E. R.
Booser. CRC Press, Boca Raton, FL.
Hutchings, I. M. 1992. Tribology. CRC Press, Boca Raton, FL.
Transactions of the ASME, Journal of Tribology.
STLE Tribology Transactions.
Spring and Fall Technical Conferences of the ASME/ICED.

© 1998 by CRC PRESS LLC

You might also like