0% found this document useful (0 votes)
21 views13 pagesGrinding Review 1992
Grinding review
Uploaded by
anderdrumerCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
21 views13 pagesGrinding Review 1992
Grinding review
Uploaded by
anderdrumerCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 13
MANUFACTURING PERSPECTIVE
An Overview of Applied Mechanics
in Grinding
Liangehi Zhang
Department of Mechanical Engineering, The University of Sydney, NSW 2006, Australia ZHANG [email protected]
Tetsuya Suto, Hosei Noguchi, and Tohru Waida
Machining Technology Division, Manufacturing Systems Department, Mechanical Engineering Laboratory, 1-2 Namiki,
‘Tsukuba, Ibaraki 305, Japan [email protected]
Grinding has been a major manufacturing process for many decades, but itis still known
‘throughout industry as a skill-controlled, not-highly-automated process. Published technical
Papers do not directly provide useful grinding methods, so that industry experts usually prefer to
use their own empirical or semi-empirical methods to design particular processes. For that
‘matter, the results published in technical papers are sometimes in conflict with practical
observations. However, the rapid development of high technology, especially associated with the
application of advanced materials, creates greater requirements for grinding processes. Thus,
empirical methods become powerless in solving grinding application problems. What is needed is
increased reliability, greater economy, higher performance, and the design of fully automatic
grinding production lines. Attaining these objectives requires more research into grinding
Processes to reveal their mechanisms and to provide practical guidelines for industry.
With these goals in mind, this article critically reviews the main problems of mechanics in
grinding processes, discusses the importance of mechanics analysis in the improvement of
grinding technology, and emphasizes some new and unconventional research approaches for the
‘automation of grinding operations. The authors attempt to explain the profoundness of various
Problems in very simple terms, so that even nonexpert readers can easily understand the
problems and take them as starting points for further research.
—_— Y—
INTRODUCTION Of all the conventional machining processes, grinding is
ssl se mot xen er velone aoe oo
Ging one emote metois of macnng sult hol rene rsa le fb ror
for providing precise geometry. Although it was used meinly may constitute a significant portion of total cost (e.g. iypi-
‘as a finishing process just a decade ago, a bulk removal grind- cally 25 to 50 percent for rolling element bearings). How-
ing process (¢.g., [1, 2]) has been developed that combines ‘ever, rapid developments in various areas of high technology,
the advantages of a large material removal rate — character- ‘such as space, optics, atomic power, and electronics, create
istic of cutting processes — with high precision and surface further requirements for the quality of ground elements. Be-
‘quality—characteristics of grinding. It appears evident that in sides low roughness and high accuracy, the traditional re-
‘the near future increasing proportions of total material removal quirements for grinding, these technologies need further
will be accomplished by grinding, thereby increasing the im- guarantees of optimal residual stress distribution and reason-
portance of grinding in the overall manufacturing scheme. able microstructures of the surface material. This, n turn,
Manactring Review vol 5 0.4, December 192 a (©1992 Ameria Socety of Mechanical Eaineers demands greater productivity, higher quality, and lower costs
from grinding processes. Hence, manufacturers must know
the principle factors affecting the surface quality ofthe
‘workpiece as well as the proper setting of grinding param-
«ters. Unfortunately, grinding processes are so complex and
dificult to study that their governing parameters are difficult
to define. For instance, it has long been known that residual
stress in machined surfaces, caused by thermal cycling, me-
chanical deformation, and microstructural transformation of
the surface layers, plays an important role in surface integrity
and the working life of a workpiece. To study this problem,
however, many significant factors must be involved: grain
sizes, grain type, number of active grains per unit area, feed
rate, depth of cut, length of contact arc, properties of the
work material, geometric structure and grade ofthe wheel,
coolant chemistry, coolant supply conditions, and so on.
Simply put, the dominant variables are difficult to determine.
Practical grinding processes are still being performed empiri-
cally or semi-empirically, and are known through industry as
Black Art [3] This occurs because technical papers pub-
lished on grinding research usually do not directly provide
useful methods for grinding engineers. In fact, their results
are sometimes in conflict with practical observations. How-
eves, empirical and conventional methods become powerless
in solving complicated grinding problems. A new approach
must be found to meet the imperative needs of higher perfor-
‘mance, greater economy, and reliability.
‘With these objectives in mind, itis the goal of this article
10 ctitically review the main problems of mechanics in grind-
ing processes and to emphasize the importance of mechanics
analysis in the improvement of grinding technology. Based
om @ careful discussion, the article proposes some new and
‘unconventional approaches for further research on the auto-
mation of grinding processes. The authors try to explain the
profoundness of various grinding problems in very simple
terms, so that even nonexpert readers can easily understand
the problems presented and use them to develop future re-
search strategies.
UNDERSTANDING GRINDING AND
MECHANICS:
Some Grinding Basics
A grinding wheel essentially consists of a large number of
abrasive particles, called grains, held together by a suitable
agent, called bond, as shown in Fig. 1. The cutting process of
« grinding wheel may be regarded as a multipoint cutting tool
with a cutting action similar to that of a milling cutter except
thatthe cutting points on the grinding whee! are irregularly
shaped and are randomly distributed over the active face of
the wheel
‘Those grains atthe surface of the wheel that actually
perform the cutting operation are called active grains. The
FIG.1. Grains and bond
percentage of active grains relative to the bonding agent sig-
nificantly influences grinding efficiency. Take peripheral
sinding as an example, as shown in Fig. 2a. Each active
rain removes a short chip of gradually increasing thickness
regulated by the relative feed rate of the workpiece. Because
of the irregular shape of the grains, however, there is consid-
crable interference between each active grain and the new
work surface. The interference results in progressive wear,
causing the formation of worn areas, thus increasing friction,
leading to an increase in the force acting on the grain. Even-
tually, this force becomes large enough either to tear the
wom grain from the bonding agent, thus exposing a new un-
‘wom grain orto fracture the wom grain producing new cut
ting edges. From the angle of the work material under grind-
ing, as shown in Fig. 2b, horizontal force leads to severe
plastic shear deformation and forms a chip (e.g, in a cutting
process). Therefore, vertical force is necessary to keep the
depth of cut ofthe grain. Unlike tool cutting, the deformation
of achip is unpredictable. Before it tears or wears away, 2
chip may bend, break, or cram into the gap between neigh-
boring grains. For ductile work materials, if the grinding
wheel is not dressed ina timely manner, such a cram will
result in a sharp change of interface forces and lead to insta-
bility of the grinding process.
In steady grinding, when the interface tangential and
normal grinding forces F, and F, have reached their equilib
rium value, the elastic deformation rate of the wheel—
workpiece system must become zero. Thus for a steady state,
if only the grinding interface is examined, the elastic deforma-
tion of the grinding machine does not need to be considered,
asin the deflection of a wheel spindle.
Correlation of Mechanics with Grinding
Mechanics analysis is predominantly involved in two aspects
of a grinding system, grinding wheel selection and surface
integrity of ground components. Figure 3 summarizes the
Manufacturing Review vol. 5, no. 4, December 1992 FIG.2. a) A schematic diagram of grinding and b) deformation of a chip
relationship between mechanics analysis and the principal
factors in grinding processes: the thick lines indicate strong
correlation and dependence, while the thin lines imply rela-
tively weak connections. The figure shows that a grinding
process is influenced by many factors. Among these, heat,
interface forces, surface cracking, residual stress, and the
Droperties of the work material ae the five most important
factors related to surface integrity and the selection of
Wheels. Obviously, itis unrealistic to take all of the factors
into account a the same time. Therefore, for convenience,
the investigations into wheel selection and the surface integ-
tity of workpieces are performed individually. Interactions
between them are simplified into comprehensive effects un-
er certain acceptable assumptions.
Findings in stresses and strains are the main direct out-
puts from mechanics analysis. Specifically, wheel deform:
tion, wheel—workpiece interaction, (residual) stress predic
tion, and the stability ofa grinding operation are the
objectives of mechanics research. Usually, the complexity of
analyzing a process depends on how many factors are in-
volved. To predict residual stresses in a ground component,
for instance, the total stain in governing equations of me-
cchanies (e.g.,[4, SP) should be a summation of different parts
‘rom various sources: mechanical deformation, thermal de-
formation, and phase transformation,
Numerous technical papers and monographs regarding
srinding have been published in the last 40 years. The litera-
ture search performed for this article showed that the number
Of papers that discuss grinding recorded in the Engineering
Index from January 1970—April 1991 is 15,604. OF these,
161 investigated residual stresses (with explicit keywords),
while only a small number studied grinding from the angle
cof mechanics. This dearth of research on the mechanics of
arinding has resulted because the process is too complicated
for theoretical modeling, and primary empirical data must
‘aang etal: An Overview of Applied Mechanics in Grinding
bbe accumulated from actual processing — which is difficult
‘and expensive to accomplish — to understand grinding
empirically.
RESEARCH PROGRESS, METHODS, AND
COMMENTS:
‘There are many open questions in every branch of the flow
chart shown in Fig. 3. However, only those that are impor-
tant and dependent on mechanics analysis will be discussed
individually.
Interaction Between a Grinding Wheel and
AWorkpiece
Interaction between an Individual Grain and
AWorkpiece
An abrasive cutting process is closely related to the interec-
tion of negative rake tools with a workpiece. These interac-
tions are also closely related to the mechanics of indentation
‘and work hardening of material in the theory of plasticity.
Geometric Models of Grains
Abrasive grains embedded on bonded abrasive wheels have
ismegular profiles. In order to develop a model of the cutting
process for a grain, it is essential to simulate the grain with @
simple geometric shape that will approximately but reason-
ably reflect its real characteristics. In the literature, a cone,
sphere, and a pyramid with or without a pointed tip have
been applied. Usually, one would choose a specifically
shaped grain to investigate specific effects, such as rake
angle effect truncation effect, and so on. Sessoms | =
=
Tnprevonen of Greig Teor
FIG.3. Relationship between mechanics and grinding
A Machining Model of Grains
Experiments have indicated that an abrasive grain performs
three distinct processes on @ workpiece in grinding metals
(€-8. {6, 7): 1) rubbing, in which the grain rubs on the
‘workpiece causing elastic and plastic deformation without
‘material removal; 2) ploughing, in which the grain causes
plastic flow in the material and extruded material is thrown
up and broken off along the sides of the groove resulting in
Jow rates of bulk removal; and 3) cutting, in which 2 fracture
takes place in the plastically stressed zone just ahead of the
rubbing grain, causing formation of a chip, resulting in fairly
rapid stock removal rates. Generally, cutting and ploughing,
‘ora combination of these elements are the dominant type of
materials lows. In some cases, it may not be unrealistic to
concentrate attention purely on cutting or ploughing. For ex-
ample, based on the available results, Malkin discussed cer-
tain behaviors of an individual grain in the machining pro-
cess and its relationship with grinding energy (power), and
‘categorized the behaviors into grinding mechanisms (8). His,
conclusions, however, were rather qualitative.
FIG. 4. Chips crammed into a wheel surface [13]
Methods of Analysis and Mechanisms of Cutting
Ductile and Brittle Materials
‘The mechanism of chip formation has a significant effect oa
the surface integrity of the ground components. Four main
‘methods have been used in the analysis of grain-workpiece
interaction: experimental investigation, energy, slip-line, and
upper bound. A number of studies have used these methods
to examine grain workpiece interaction. Childs, while study-
ing experiments on the sliding of conical indenters over duc~
tile materials, analyzed theoretically the forces involved [9]
He used the principle of minimum energy in plastic works to
predict the sliding forces and claimed that his force predictions
‘were within 20 percent of the observed values. However, his
shape predictions of the slide ridge were not satisfactory.
Kida et al. conducted tests with conical tools simulating
single abrasive grains [10]. They suggested five deformation
regions that is, the region removed as chip ahead of the tool:
the stagnant region beneath and adjacent to the tool; the se-
verely deformed region that followed down along the side of
the stagnant one; and finally, an undeformed region in front
of all the others. The stagnant region acted as a cutting edge,
and the initial positions of such a region determined whether
or not a chip could be formed. Kudo and Tsubouchi tested
lubricated carbon steel ploughing and concluded that the re-
lationships between the forces and the depth of penetration
of the tool were almost parabolic, while the ratio of tangen-
tial to normal force was approximately constant withthe
change in depth of cut 6]. Challen and Oxley used slip-line
field theory to explain the deformation of soft asperity by a
hard asperity [11]. De Vathaire etal. proposed an upper
bound method to study the ploughing action ofa rigid pyra-
mid indenter [12]. The model predicted the geometric shape
of track made by the indenter, the normal and tangential
forces, the strain, and the rate of strain of the ploughed mate-
rials. Simply put, in grinding ductile materials, chips are
‘Maaiufacturing Review vol 5, no. 4, December 1982 FIG. 5. Schematic model of chip formation process in grinding brittle materials [14)
formed by plastic shear deformation so that the processes of
cutting, ploughing, and rubbing are quite clear. Chips usu-
ally like to cram and load, in an unpredictable way, into the
pores or grain gaps on the wheel surface as shown in Fig, 4
[13], Surface roughness is mainly characterized by the mi-
‘ro-structure of ploughed grooves.
‘In grinding brittle materials, however, chips are mainly
produced through the mechanism of micro-brittle fracture
[14]. The chips are fragments produced through micro-
cracking the surface materials by grain indentation, as
shown in Fig. 5 [14]. Plastic deformation becomes very
ight. Unlike in grinding ductile materials, ploughing britie
materials may also lead predominantly tothe creation of
chips. The meterials adjacent to the grain subjected to
ploughing may undergo a process of compression, crack
formation, crack development, unloading, and chip splitting
[23]. The existence of surface micro-cracks is then a major
Problem among the factors determining surface integrity
f13-17).
Anew technology called ductile-regime grinding is be-
ing developed to avoid the damage of micro-cracking in
singing brittle materials [74]. It was found that ductile chip
formation is also possible when a grinding unit becomes
very small, on the order of nanometers. The transition from
brittle chip to ductile formation was thought of as a transi-
tion process of favorable material removal energy. In fact,
the ratio of material removal energies, plastic flow energy to
fracture energy, is approximately proportional to the depth
of cut, d,..As d, decreases to & specific value, plastic flow
will become an energetically more favorable material re~
‘moval mechanism. It is the transition phenomenon between
cracking and plastic flow and should be independent of the
materials under grinding [75]. In principle, a successful
‘hang etal: An Overview of Applied Mechantes in Grinding
application of the ductile-regime grinding technique is to
make fall use ofthe brite-ductile transition. However, fur
ther research is necessary to explore characteristic transition
parameters for different materials
Elastic Modulus of Grinding Wheels
It is well-known that a macroscopically measured, elastic
‘modulus of a grinding wheel, £, isa physically well-defined
Wheel criterion because it reflects most of the effects of the
‘wheel's microscopic components. It also significantly influ-
éences the interface condition between the wheel and the sur-
face under grinding. In addition, the elastic modulus could
meet the needs of both manufacturers and end-users
Researchers have found thatthe elastic modulus of a
‘grinding wheel is an extremely complex function of numer-
us factors: temperature, 8; specific weight, p; hardness
grade, hi, concentration, c; grain grade, G,; total eppareat vol-
ume, V; Volume of grain, V; volume of bond material, V, vol-
ume of pores, V,;type of grain, t,; type of bond material, 1;
and the mean diameter of grains, d. It seems to be rather dif
cult to reveal clearly the dependence of E upon these factors.
‘Typical investigations into this problem in the early stages
included those published in [8, 18-25]. The first study looked
atthe relationship between E and k, while the latter group
Alscussed the variation of E with c,d ty V, Vi Vp and V,
in series of eight papers, which expressed their results with
hhundreds of curves.
Progress has recently been achieved by Zhang et al. in
[26]. They dug out the nature of E withthe aid of dimensional
analysis. They found that hc, V, and G, are not governing
Parameters, so that for a class of grinding wheels with specific types of grain and bond material, the nondimensional elastic
‘modulus is only a function of non-dimensional temperature t
and a governing dimensionless variable:
E =00,X), X= (0X0). %=t (1)
where E= E((dp), v,= VJV,v,= Vy, and D = av,
Consequently, in coordinates E— X, al the experimental
points obtained under the same dimensionless temperature, x,
should lie on a single curve, if no measurement errors have
‘occurred, Their formula is extremely well confirmed by the
‘experimental results from different sources (21-23, 27], as
shown in Fig. 6.
This formula is @ general relationship between
ondimensional Young's modulus, E, and goveming factors.
¥, ¥jp and D. For different measuring temperatures, it will
‘produce different curves. The significance of this revelation
lies in that it reveals the nature of E and tha it can greatly
reduce experimental work and make theoretical analysis con-
venient.
‘The nondimensional elastic modulus, E, implics certain
Paysical meanings. E/p corresponds to the square ofthe lon
gitudinal wave speed of the material, v (under the condition
of isotropy). Obviously, grain diameter, d relates to the
anisotropy of the wheel and, in turn, influences v. On the
other hand, Ep is also a measure of specifi stifiness, and
‘appears in the formula forthe bending of beams under their
‘own weight. As luck would have it, people are using the
‘methods of sonic test and bending test [18] to determine the
modulus of elasticity, although the latter presently does not
involve p.
The bending cest method produces the so-called static
modulus, E,,., and the sonic test method gives a dynamic
modulus, £,.. Comparisons have shown that these two meth-
ods lead to almost the same values of elastic modulus for a
specific wheel [18]. It is therefore not valuable to distinguish
Eq and E,, in studying related problems.
Interface Forces and Contact Length
Investigations into the deformation of wheel-workpiece sys-
tems have been performed in different ways, including ex-
perimental and theoretical as well as microscopic and macro-
scopic approaches. Microscopically, an active grain will be
deflected because ofthe forces exerted during grinding; mac-
roscopically the wheel might be considered as a thick circular
dis pressed against the curved surface of a workpiece. Studies
done before 1980 inthis field were summarized individually
in [28, 29]. They concluded that observations concerning the
influence of various parameters on the magnitude of contact
Geflections were diverse and conflicted with each other, pro-
ducing a paradox. Thus their results are still open to ques-
tions and further research,
FIG. 6. Nondimensional elastic modulus
Some researchers, (eg. [30)), adopted microscopic ap-
roaches for investigating the deflection effect. They wanted
to understand precisely the grinding wheel contact deforma-
tion by studying the local elastic deflection of a single grain
or set of grains. Although there is no doubt that the nature of
the work material should have a huge effect on the behavior
of grains, these researchers did not consider such interaction.
‘Actually, their qualitative conclusions were only a repeat of|
existing solutions in the theory of elasticity — the solutions
of a half space subjected to boundary tractions. Therefore,
they are not useful toa real grinding process. To carry out
investigations of practical interest on a microscopic level, in
the authors” opinion, researchers should try to generate an
extremely detailed description of the grain’s topography and
then determine how much of the grain cuts the work mate-
rial, how much ploughs it, and how much rubs it.
Usually, it seems that because the shapes of grains are
irregular and the depths of grain cut are different from each
other, the distributions of interface forces could be very ir-
regular and discontinuous. However, Zhang etal. argued that
itis quite reasonable to assume that the whee! and the
workpiece are in continuous contact for the sake of macro-
scopic analysis [26]. To a great extent, except for randomly
distributed active and inactive grains, bond material and
loaded chips can also be in contact with the work material, as
shown, for example, in Figs. 2b and 4. This type of grinding
force can not be distinguished from that of the cutting,
ploughing, or rubbing force of active grains, because they
always come together in the chip formation process. Based
‘on this viewpoint, they proposed a macroscopic model to
investigate interface forces that considered the cutting pro-
cesses of grains comprehensively and where contact defor-
‘mation of both the workpiece surface and the wheel were
stressed, Their analysis showed that actual pressure distribu-
tion over the grinding bite deviates greatly from a Hertzian
profile, as shown in Fig. 7. This conclusion was strongly
‘confirmed by experimental measurements [31]. Snoeys and
‘Wang also tried to apply continuum mechanics to grinding
‘force analysis (32]. Unfortunately, they did not take micro-
Manufacturing Review vol. 5,0. 4, December 1992 —
