Cylinder Liner Bore Distortion Estimation During Assembly of

Diesel Engine with FEA

Pratik B Mirajkar1, Kedar A Kanase2 & Narendra K Chhapkhane3

1&3
R.I.T. Sakharale, Sangali, 415414 [email protected]
2
Senior Manager of R&E (kirloskar oil engines limited)

cylinder bore distortion is likely to occur under dynamic

Abstract - The trend in an IC engine industry today is
toward a shorter product development cycle and faster stress (heat and pressure). The challenge is to
time to market with increased emphasis on up front understand how these changes take place and to
analysis to design, develop and optimize a reliable and establish procedures that will attempt to minimize these
durable product. The stability of components of engines is changes. As the block ages and is exposed to thermal
essentially affected by the existing inner distribution of changes, the casting‟s molecular structure will change
stresses. Today, the calculation of stability of critical slightly, which will affect bore geometry. In addition,
components like e.g. the cylinder head and the cylinder distinct changes in bore shape will take place as the
block is state of art. However, the decisive factor that leads block is assembled. The clamping forces that result from
to accurate simulation results is to know the boundary
the installation of cylinder heads will cause the bores to
factors as precisely as possible for the calculation. Cylinder
bore distortion in IC engines has been identified as a cause “morph” in most cases as the fasteners pull and squeeze
for fuel and oil consumption, thus it influences efficiency metal adjacent to the bores. Sometimes these changes
and emissions. In particular it has a significant effect on are insignificant, while in other situations, the changes
the performance of the tribological system consisting of can be so dramatic as to cause measurable ring dragging
piston, into the crankcase. It also causes an increase in and subsequent loss of power, fuel economy and sealing
hydrocarbon emissions and can deteriorate the emission due to both out-of-roundness and frictional heat. In
control system. Due to these reasons, the evaluation of bore some engines, these problems may be compounded if
deformation and the optimization of the cylinder bore the specific block is severely affected by additional
shape is considered significant subjects in modern engine
distortion that results from clamping forces caused by
research. For present project work, the distortion levels in
the block due to clamping loads are used to measure with bell housing bolts, water pump bolts, motor mount bolts,
FEA analysis using CAE tool ANSYS. Accordingly can be etc.
checked with excremental setups. A. Literature Review
Key words-Bore Distortion, Liner Bore, Engine analysis.
Prof. Paul Gibbs[1] has given the equations and
concepts used for “Piston ring conformability”. Also he
I. INTRODUCTION
has given the harmonic distortions in cylinder. A first
Bore Distortion means deviation from ideal circular order distortion appears as an offset of the whole section
shape (roundness), or bore out of roundness. Initially it from the main axis, a second order appears as an „ova
is understood that manufacturing tolerances can cause ling‟ of the section, a third as a three lobed deformation,
the bore to deviate from ideal condition. a fourth as four lobes, and so on. Theoretically an
infinite number of orders combine to make up the final
The main types of distortions that occur in an engine are
shape, the magnitudes and directions of these orders are
1. Cylinder liner bore distortion. calculated by using the Fourier series and applying it to
measurements of the bore. He has given the information
2. Main bearing bore distortion. about LOTUS tolerances and how these are used for
3. Connecting rod bore distortion. acceptability of bores.
Amongst which cylinder liner bore distortion is Teimuraz Bardzimashvili,James F. Kelly and Elene
major as compared to the others. Even with the best of Romelashvili et.al.[2] have given the details of piston
intentions and preparation, an unavoidable degree of rings. They have stated the causes of bore distortion
such as imperfections in design and assembly, gas

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

66
 International Journal on Mechanical Engineering and Robotics (IJMER)

pressure and thermal variations, the cylinder may distort as four lobes, and so on. Theoretically an infinite
before and during operation. Hence, the circular piston number of orders combine to make up the final shape,
ring must conform to a slightly imperfect cylinder bore. the magnitudes and directions of these orders are
Also they have stated the three equation of GOETZE , calculated by using the Fourier series and applying it to
Dunaevsky and Tomanik. Another approach to the ring measurements of the bore.
conformability problem is Finite Element Analysis
(FEA). In order to test the FE data and evaluate the
effectiveness of each bound, wrote a MATLAB program
called RINGPACK.
Lawrence E.Bird[4] has described common
procedure used to define bore distortion. Orders of
distortion are also classified as zero to Fourth order and
its effect on geometry of bore liner .Assembly load
causes fourth order distortion due to which four lobe
deformation of liner takes place.
K.Loenne[5] has explained the causes of (a):2nd order. (b): 3rd order (c):4th order
deformation. Deviation from the cylinder shape through
the production process. The possibility of influencing
this factor lies solely with engine manufacturer cylinder Causes of Distortions:
distortion through the assembly of the cylinder block
with the cylinder head and peripheral components called The cylinder block, as cast, has only a rough shape with
static initial clamping distortion. These elastic tolerances of around 0.5-1mm, but machining of the
deformations may be due to the design changes to the surfaces aims to give as perfect a cylinder as possible
installed components. The gas pressure acts on the (first with a rough drilling and then with a
cylinder wall in restricted areas. Distortion of the polishing/honing process to reach the final tolerances in
cylinder due to combustion pressure is significant only the region of tens of microns). Even after this there will
in highly rated diesel engines with thin walled wet be a small, but measurable deviation from the ideal
liners. The magnitude of the thermal expansion is far shape, but the design of the piston rings allows them to
larger than the deformation caused by clamping bolts. conform easily to these small distortions.
Bolt Loads:
Cylinder heads, bearing shells and bearing carriers all
require bolting to the cylinder block and the bolts used
will unavoidably deform the block. Heads are the
primary cause of fourth order distortions, bearing
carriers cause second order and the unsymmetrical
supports on the outside bore of blocks cause third order
distortions. As expected, the higher the bolt loads, the
greater the distortion, but placement of the bolt threads,
and the depths to which they sink can affect distortion to
a great extent. With cylinder liners, supports, cooling
jackets, head gaskets, external mountings and other
considerations, the design of blocks to minimise
distortion is a complex subject, and is beyond the scope
B. Harmonic Distortions: of this report –which aims only to find simple method of
This kind of distortion causes a regular, sinusoidal measuring, quantifying and comparing distortions.
deformation in the cylinder, which can be derived Temperature:
mathematically from measurements of the surface of the
cylinder. The ‘harmonic’ refers to the building up of the At present the most accurate measures of the shape of
real shape from a combination of many sinousidal the bore can only be made with the engine disassembled
waves of different orders (see below). A first order as the harsh conditions during combustion give rise to
distortion appears as an offset of the whole section from serious (but not unsurpassable) problems when trying to
the main axis, a second order appears as an ‘ovaling’ of measure the bore. As a result, most distortions are
the section, a third as a three lobed deformation, a fourth measured cold (25°C). Since the temperature gradient
during operation can range from below zero °C outside

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

67
 International Journal on Mechanical Engineering and Robotics (IJMER)

(at start-up) to over 2000°C inside the bore, deformation Engine seize:
of the bore by thermal expansion is common (with the
Due to bore distortion in the engine temperature rise
greatest expansion occurring, naturally, towards the
occur which causes engine seize problem. so these are
combustion face of the block).
the some major problems occurs due to bore distortion.
C. Bolting concept : so to reduce these problems it is necessary to find out
the bore distortion in initial stages.
The connection between block and head can be
done either conventionally or by the through-bolt And it is the basic motivation of this project.
technique. In case of conventional bolting, the cylinder
head and grey iron bearings are bolted directly to the II. FEM APPROACH
block resulting in high stresses in the vicinity of the
thread and the bolt head. This bolting concept reaches It is not always possible to obtain the exact
analytical solution at any location in the body, especially
its limits in case of high loaded direct injection diesel
for those elements having complex shapes or
engines. In order to prevent high tensile stresses in the
geometries. Always matters are the boundary conditions
engine block, the main bearings and the cylinder head
and material properties. In such cases, the analytical
can be connected directly by long bolts which penetrate
the whole block and head, thus setting them under solution that satisfies the governing equation or gives
compressive stress only (through-bolt concept). The extreme values for the governing functional is difficult
to obtain. Hence for most of the practical problems, the
drawback of this solution is a more complicated
engineers resort to numerical methods like the finite
assembly because bearing caps and cylinder head are
element method to obtain approximate but most
not any more independent of each other, i.e. the final
probable solutions.
assembly of bearings and heads has to be carried out at
the same time. This problem can be solved by screwing Finite element procedures are at present very
in the through-bolts so that head and bearing caps can be widely used in engineering analysis. The procedures are
mounted separately while maintaining the load-bearing employed extensively in the analysis of solids and
benefit of the through-bolt. structures and of heat transfer and fluids, and indeed,
finite element methods are useful in virtually every field
D. Need Of This Project:
of engineering analysis.
Air-cooled Diesel engine, for instance, is commonly
used in heavy-duty transport fleets applications due to Description of the method:
their high performance, efficiency, and low fuel In any analysis we always select a mathematical
consumption. But when bore deformation occurs the model of a physical problem, and then we solve that
following problem may arises: model. Although the finite element method is employed
to solve very complex mathematical models, but it is
Blow by:
important to realize that the finite element solution can
Blow by is the loss of combustion gases from the never give more information than that contained in the
combustion chamber, past the piston and into the mathematical model.
crankcase. This is undesirable as the gas takes energy
Physical problems, mathematical models, and the finite
with it, reducing the pressure on the piston face and
element solution:
reducing the power output of the engine.
The physical problem typically involves an actual
Oil consumption:
structure or structural component subjected to certain
If the piston (or piston rings) were in direct contact loads. The idealization of the physical problem to a
with the cylinder wall then the combustion chamber mathematical model requires certain assumptions that
would be sealed off perfectly, but it would require a together lead to differential equations governing the
huge force to move the piston against the friction mathematical model. The finite element analysis solves
generated. To overcome the friction the piston rings are this mathematical model. Since the finite element
designed to ride on an oil film, which, although very solution technique is a numerical procedure, it is
thin, provides a small amount of friction. If the rings are necessary to access the solution accuracy. If the
designed badly, then instead of riding over this film, accuracy criteria are not met, the numerical solution has
they scrape it away - either into the combustion chamber to be repeated with refined solution parameters (such as
or out through the exhaust (giving rise to high finer meshes) until a sufficient accuracy is reached. It is
hydrocarbon emissions and reducing the life of catalytic clear that the finite element solution will solve only the
converters), or down into the sump where it is recycled selected mathematical model and that all assumptions in
through the engine. this model will be reflected in the predicted response.

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

68
 International Journal on Mechanical Engineering and Robotics (IJMER)

Hence, the choice of an appropriate mathematical model 3. The density of the mesh must be controllable, to
is crucial and completely determines the insight into the allow trade-off between accuracy and solution time.
actual physical problem that we obtain by the analysis.
4. The grid density will vary depending on local
Once the mathematical model has been solved
accuracy requirements, but any variations must be
accurately and the results have been interpreted, we may
smooth to reduce or eliminate numerical
well decide to consider next a refined mathematical
diffusion/refraction effects.
model in order to increase our insight into the response
of the physical problem. Furthermore, a change in the 5. There are some requirements on the shape of
physical problem may be necessary, and this in turn will elements. In general, the elements should as
also lead to additional mathematical models and finite equiangular as possible in equilateral triangles &
element solutions. The key step in engineering analysis regular tetrahedron.
is therefore choosing appropriate mathematical models.
6. Highly distorted elements (long, thin triangles,
These models will clearly be selected depending on
squashed tetrahedron) can lead to numerical
what phenomena are to be predicted.
stability problems caused by round-off errors. This
Advantages and limitations of FEA: requirement is modified for boundary layers, where
highly stretched elements are desired and facilitated
Planning the analysis is arguably the most
in the FEM formulation. The min-max-angle
important part of any analysis, as it helps ensure the
property is still required in this case.
success of the simulation. Oddly enough, it is usually
the one analyst‟s leave out. The purpose of an FE Element mesh Parameters:
analysis is to model the behavior of a structure under a
Usually there are certain parameters that determine
system of loads. In order to do so, all influencing factors
must be considered and determined, whether their the quality of the results. The Engineer has to ensure
effects are considerable or negligible on the final result. that these parameters are maintained to the minimum
required levels in the FE model for obtaining good
The degree of accuracy to which any system can be
results. These are called mesh quality parameters which
modeled is very much dependent on the level of
are as discussed below:
planning that has been carried out.FEA is an
approximate way of simulating the system behavior. But 1. Warpage: This is applicable to only Hexahedron
the results can be quite close to actual testing values. and shell elements only. Any three nodes on any
FEA can never replace actual physical testing all the face of the Hexahedron element define a plane. If
times. This is due to the fact, the information required the fourth node on the same face is away from the
for FEA simulations, like material properties emanates plane beyond a certain angle, the results of analysis
from physical testing. obtained by using these kinds of elements shall be
FEA results by themselves can never be taken as erroneous.
complete solution. Usually at least one prototype testing 2. Maximum Angle: If the obtuse angle between two
is necessary before the design guided/validated through edges in a Quadrilateral is more than a prescribed
FEA can be certified. But when effectively used FEA limit, then erroneous results can be the outcome of
can predict the results/behavior quiet close to reality and an analysis.
can reduce the design lead times as well as the number
of prototypes to be tested. Also there are some situations 3. Minimum Angle: If the acute angle between two
like gears in contact, which cannot be simulated exactly edges in a Quadrilateral/Triangle is less than a
using FEA techniques. Under such situations some work prescribed limit, then erroneous results can be the
around such as simulating the worst condition that can outcome of an analysis.
happen can be followed. 4. Aspect Ratio: If the ratio of maximum to minimum
Mesh requirements: length in an element is more than a prescribed limit,
the results of the analysis can be erroneous.
The Finite Element Method (FEM) has certain
requirements on a mesh: 5. Jacobian: The differential of the matrix obtained
during the formulation of a stiffness matrix is called
1. The mesh must be valid, (no holes, self- the Jacobian. If the element is formed with a poor
intersections, or faces joined at two or more edges). shape, this differential can become negative
resulting in a negative stiffness matrix. Hence care
2. The mesh must conform to the boundary of the
domain. should be taken to avoid poor shaped elements to
avoid Jacobian with low values.

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

69
 International Journal on Mechanical Engineering and Robotics (IJMER)

III. WORK DONE -FORCE CALCULATION: IV. COMPANY MODEL

Bolt Pretension Calculation: Crank Case:
The Basic equation considered for bolt pretension
calculation is,
Pi = T/ (K D)
Where, Pi = Bolt preload
T = Bolt installation torque
K = Torque coefficient
D = Bolt nominal size
Torque coefficient K,
K= {[0.5 p/π] + [0.5 μt (D-0.75 p sinα)/sinα] + [0.625
μc D]}/D
D = Bolt nominal size =16 mm Gasket:
p = thread pitch=1.25 mm
α = thread profile angle = 60° (for M, MJ, UN, UNR,
and UNJ thread profiles)
β = thread profile half angle = 60°/2 = 30°
μt = thread coefficient of friction = 0.17
μc = collar coefficient of friction = 0.12
For our case K= 0.2233
So, Pi = 255060/ (0.2233 x 16)
Pi = 255.060/3.57
Pi = 71445.37 N – Bolt Pretension Load/bolt
No of Bolts=6
6 x 71445.37=428672 N
Liner:

Nonlinear analysis:
There are three types of non-linearity.
1. Material non-linearity
2. Geometric non-linearity
3. Contact non-linearity

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

70
 International Journal on Mechanical Engineering and Robotics (IJMER)

Non-linear Structural Analysis: STRUCTURAL STEEL:

When structure response (deformation, stress & strain)

is linearly proportional to magnitude of load then the
analysis of such a structure is known as linear analysis.
When the load to response relationship is not linearly
proportional, then the analysis falls under non linear
analysis. For example when a compact structure made of
stiff metal is subjected to a load with relatively lower in
magnitude as compared to strength of material, the
deformation in structure will be linearly proportional to
the load and the structure is known to have subjected to
linear static deformation. But most of the time either
material behavior will not linear in the operating
condition or geometry of the structure itself keep it from
responding linearly. Due to cost or weight advantage of In above problem there are two non-linearity‟s
non metal (polymers, woods, composites etc.) over takes places, material and contact. In material non
metals, nonmetals are replacing metals for variety of linearity‟s there two types again. 1) Plastic non-linearity
applications, which have nonlinear load to response 2) Intrinsic non-linearity.
characteristics, even under mild loading conditions.:
MESH MODELS:
 Material non linearity ( Strains beyond the elastic
limit (plasticity)) Crankcase:

 Geometric non linearity( Large deflections, such as

with a loaded fishing rod)

 Boundary non linearity Contact between two

bodies.

V. MATERIAL PROPERTIES:
Material Property:
Material property must frequently be approximated
in FEM. Indeed of course there is an accepted degree of
variation in the young‟s modulus of even the most
standard of engineering material.

Material properties in Ansys: Gasket:

GREY CAST IRON:

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

71
 International Journal on Mechanical Engineering and Robotics (IJMER)

Liner: Convergence Study:

VII. RESULTS AND DISCUSSION:

VI. CONTACTS:
Overall Deformation:
Contact needs to be defined wherever necessary. It
allows proper transmission of forces. Contact occurs
when the element surface penetrates one of the target
segment elements on a specified target surface. If more
than one target surface will make contact with the same
boundary of solid elements, you must define several
contact elements that share the same geometry but relate
to separate targets (targets with different real constant
numbers), or you must combine the two target surfaces
into one (both having the same real constant number).

Types of contacts:

1. Rigid-flex:
Bodies of vastly different stiffness
Steel against rubber seals Von mises stress:

2. Flex-flex:
Bodies of comparable stiffness
Metal contacting metal
3. Self contact:
Body folds over itself
Column buckling
Large sliding with friction for all:
We have four Flex-Flex type contact in the case of
cylinder bore distortion analysis. In Flex-Flex type of
contacts, both surfaces are flexible in nature i.e. both
bodies are having comparable stiffness.

ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

72
 International Journal on Mechanical Engineering and Robotics (IJMER)

Experimental validation and conclusion: VIII. REFERENCES

Above all results were plotted in the excel sheet. In [1] K.Loenne, Ron Ziemba, “ The GOEETZE
this the comparison of perfect bore with the actually Cylinder Distortion Measurement Systems and
distorted bore and FEM results were takes place. The the Possibilities of Reducing Cylinder Distortions
green bore shows the perfect bore, red shows the “, SAE Paper, 880142,1999
experimental measured bore and blue shows the Fem
[2] Franz Maassen, Franz Koch and Markus
results. As discussed in earlier chapter distortion were
Schwaderlapp,Timo Ortjohann and Jürgen
measured at 10 planes. The first 5 planes show the much
Dohmen “Analytical and Empirical Methods for
accurate bore shape but as moving towards bottom the
Optimization of Cylinder Liner Bore Distortion”,
shape was not matching as upper portion.
SAE paper –2001-01-056
[3] Eysion A. Liu and Mick WinshipSimon Ho, Kuo-
Ting Hsia, Mitchel Wehrly and William F.
ReshDaimlerChrysler, “Engine Cambore
Distortion Analysis From Design to
Manufacturing”, SAE paper 2004-01-1449
[4] Lawrence E. Bird and Robert M. Gartside “
Measurements of Bore Distortion in Firing
Engine“, SAE Paper, 2002-01-0485.
[5] Fagan, M.F “Finite Element Analysis Theory and
Practice”,University of Hall, Logman Scientific
and Technology, 2001.
The experimental results are provided by company.
Readings are measured at different levels. Maximum
Distortion is 15% to 20 % more than Ansys results in
Liner Bore.



ISSN (Print): 2321-5747, Volume-1, Issue-2, 2013

73