INFLUENCE OF GRINDING PRODUCTIVITY ON
THE BURN INTENSITY OF THE GROUND SURFACE
D. Ciglar
R. Cebalo
T.Udiljak
Dr.sc. D. Ciglar, University of Zagreb, FSB, I.Lučića 5, 10000 Zagreb
Prof.dr.sc. R.Cebalo, University of Zagreb, FSB, I.Lučića 5, 10000 Zagreb
Doc.dr.sc. T.Udiljak, University of Zagreb, FSB, I.Lučića 5, 10000 Zagreb
Keywords: grinding productivity, grinding defects, grinding burn intensity, grinding surface
lightness interval, burn intensity of the ground surface
ABSTRACT: The productivity increase of the grinding process expressed by reduced
material removal rate can result in incorrect grinding and the consequence is the ground
surface burn and occurrence of other grinding defects on the ground surface. The paper
studies the burn intensity of the ground surface and other grinding defects in different
material samples, depending on the increase of grinding productivity. The ground surface
burn intensity has been determined by means of the surface lightness interval, with the
upper and lower value, and measurements were carried out on the Image Analysis
System LECO2001. The results have shown that the increase in grinding productivity on
all the ground material specimens results in an increase of burn intensity, reduction of the
surface lightness interval value, and an increase in the change of surface layer hardness
and the layer depth of variable workpiece hardness. Therefore, the surface lightness
interval values, i.e. the burn intensity of the ground surface can determine the correctness
of the grinding process and approximate limited productivity.
1. INTRODUCTION
Very small dimensions of separated particles, the characteristic that breaking of
one cutting edge does not affect the stability of the process, and the self-sharpening
process of the grinding wheel, these are all significant advantages that may insure the
future of the grinding processes. The tendency is to achieve the maximum productivity of
the grinding process, thus reducing the machining time, and thus making the grinding
process a more economic and for the industry a more acceptable one, at the same time
meeting the quality requirements of the treated surfaces.
The increase of productivity of the grinding process can result in incorrect grinding, i.e.
in undesired generation of heat on the workpiece surface, which is much greater than the
heat taken away so that the temperature of the workpiece surface layer will rise quickly.
210
D.Ciglar; R.Cebalo; T.Udiljak
High temperature affects the integrity of the ground surface, and the notion was
introduced in the seventies by M. Field and J. Kahles 1 describing the total quality of the
surface which affects the product characteristics, and the most important include: impact
strength, fatigue strength (dynamic durability), corrosion resistance, tribological behaviour,
dimensional stability, residual stresses, surface roughness, etc. Except influencing the
mentioned characteristics of the product, grinding temperature primarily causes thermal
and temperature damaging of the ground surface, and the consequences are defects in
grinding that occur on the surface or in the sub-surface layer of the workpiece. The
grinding defects according to 2,3,4,5,6 may include: burn or oxidation, change in
microstructure, change in hardness, residual stresses and generation of thermal cracks.
In rough grindings there is a less demanding requirement regarding the quality of the
ground surface and little burn of the workpiece surface is allowed, since the resulting
thermally damaged layer can be removed by finish grinding, if necessary. However,
thermal cracks on the ground workpiece surface are not allowed, since according to [2,6]
cracks have the tendency to penetrate into the depth and are very dangerous for the
dynamically loaded parts, substantially reducing the dynamic strength of the workpiece,
with the possibility of fracture as result of material fatigue.
Little intensity of surface burn though, allowed in less demanding grinding processes,
will determine the critical conditions of grinding, i.e. the point where the incorrect grinding
starts, if there is indication that there is significant dependence between the burn intensity
of the ground surface and other grinding defects. For some material specimens, therefore,
the dependence will be studied of the ground surface burn intensity and other grinding
defects on the increase of grinding productivity.
2. GRINDING DEFECTS
One of the first defects in grinding that occurs on the workpiece ground surface is
burn or surface oxidation. According to 7 all metals and alloys (except gold) form an
oxide layer, but in many cases at lower temperatures this reaction is very slow. In alloys
the oxidation products are much more complex, since alloy components can form
chemical compounds and the resulting oxide layer is a heterogeneous mixture of different
particles at the alloy surface, i.e. chemical composition of the workpiece marginal layer is
changed. It is difficult to say which oxide compound will be formed on the surface of the
alloy, but it is certain that the thickness of the oxide layer affects the absorption and
reflection of light, and thus the oxide colour changes with the increase of the oxide layer
thickness. It has been proven that the oxide layer thickness increases with the increase in
duration and temperature of oxidation. According to 8, the change in colour of the
workpiece surface during grinding is caused by surface oxidation due to high grinding
temperatures. This leads to the conclusion that with higher burn intensity the grinding
temperature is higher, oxide layer thicker and the colour of the machined surface darker.
Influence of Grinding Prod. on the Burn Intensity of the Ground Surface
39
In the work, according to 9, the burn intensity of the ground surface is determined
based on the colour of the machined surface of the material specimen, i.e. by quantifying
the lightness interval of the surface after grinding, carried out on the Image Analysis
System LECO2001, Figure 1.
Figure 1 – Image Analysis System LECO 2001.
The LECO 2001 device is installed at the Laboratory for Material Science at the
Faculty of Mechanical Engineering and Naval Architecture, Zagreb, and according to [10],
the basic components of the Image Analysis System are: light microscope, two CCD
cameras (black&white), Leco 386 computer, Leco Image processor, colour image monitor,
colour text monitor (VGA card), printer, 2001 main program. By means of the black&white
camera the device has the possibility to record the material specimen surface, to analyse
its lightness and to use the histogram in order to present the grey scale level of the
recorded surface. The value of lightness interval, according to the grey scale level of
certain points of the recorded and analysed surface can range, by standardised 8 bit BMP
format, between 0 and 255, with the value 0 denoting black, and the value 255 denoting
white of a certain point. The histogram provides readings of the grey scale level values
which encompass the majority of the points from the analysed surface and these values
determine the interval of the surface lightness. The lower value of the surface lightness
interval describes the darker colour in the image, and the upper value is the lighter colour.
In order to determine the burn intensity of a ground surface the lower value of the surface
lightness interval is more relevant since it describes the more burned ground surface of
the material specimen.
Another defect in grinding considered in this paper is the change in the hardness
of the surface layer of the ground surface. As in the case of burn, the change occurs due
to high temperatures of surface heating and due to quick cooling of the surface by
conducting heat towards the interior of the workpiece. Workpiece heating and cooling
results in the change of hardness of the surface layer and in structural transformations in
410
D.Ciglar; R.Cebalo; T.Udiljak
the solid state. These structural transformations are basically the same as with the
thermal processing treatment, only that the heating rate is greater in grinding. Faster
cooling of the workpiece layer is accomplished by self-cooling, since basic conditions are
met that the cold workpiece core is a sink large enough for transmission of heat into the
interior of the workpiece. Due to these changes, in grinding of tempered steel, depending
on the level of grinding temperature, the hardness can be increased compared to the
reference value (hardening) or decreased in relation to the reference value (softening)
The increase in the hardness of the ground workpiece surface layer is a consequence of
either creating an untempered martensite (UTM), i.e. re-tempering or conversion of
residual austenite into the secondary martensite, i.e. secondary hardening. The reduction
of the surface layer hardness of the ground workpiece resulted from producing tempered
martensite (OTM), and it is present at lower grinding temperatures.
The thickness of the layer of variable hardness is approximately proportional to the
volume of heat supplied to that surface, and it is related to the heat affected zone (HAZ). It
is important to know the HAZ thickness because this is, in fact, really the thickness of the
damaged workpiece layer.
3. EXPERIMENTAL RESEARCH
The study of burn intensity of the ground surface and other grinding defects
regarding productivity of the grinding process, i.e. in various values of reduced material
removal rate was carried out on four different materials Č3840, Č6980, Č4150 and Č4770.
These tool steels were used to produce test tube specimens of 15x10x30mm dimensions,
i.e. 10x10x30mm for material specimen Č6980. Test tubes of selected material specimens
were thermally treated (hardened and tempered) according to the manufacturer’s
instructions. The specimens were ground by straight peripheral down cut grinding without
cooling, on a surface grinding machine “DOALL G – 10”. For the pre-selected material
specimens, according to [11], the surface grinding wheel of the company SWATY was
selected, with the grinding wheel designation PA120/1EF14/5V40. The dressing of
grinding wheel cutting surface was done by single-point 2 Ch diamond, with the edge
radius of 0.2mm, in three passings. The depth of diamond penetration was measured at
every passing, it is constant and amounts to 0.03mm, and the diamond cutting feed is
0.1mm/revolution. Using the above conditions the topography of the grinding wheel cutting
surface was obtained described by the number of static cutting edges which amounts to
Ns=3.59mm-2. The speed of the machining surface during the study was 0.0685m/s, and
the peripheral wheel speed was constant and amounted to 40m/s.
The change in the reduced material removal rate value during the grinding process
was achieved by changing the value of the grinding depth. With the grinding depths of
0.01mm, 0.05mm, 0.1mm and 0.18mm, the following values of reduced material removal
rate were obtained: Qbr=0.685 mm3/smm, Qbr=3.425 mm3/smm, Qbr=6.85 mm3/smm and
Qbr=12.33 mm3/smm. By grinding of selected material specimens, of each one separately,
Influence of Grinding Prod. on the Burn Intensity of the Ground Surface
59
under the mentioned conditions and values of the reduced material removal rate, the
surface of the material specimens was obtained as presented in Figure 1.
a)
b)
c)
Figure 1. Surfaces of the ground material specimens.
d)
In Figure 1 (a to d), the material specimens are arranged top downwards in the
following order: Č3840, Č6980, Č4150 and Č4770. It is clear in Figure 1 that grinding with
higher value of reduced material removal rate provides stronger surface burn in all
material specimens, which is seen from the darker surface of the specimens.
Image Analysis System LECO2001 was used to determine the burn intensity of the
ground surface of the material specimens in Figure 1, and the quantified values of the
lightness interval of the ground surfaces of material specimens are presented in Table 1.
Table 1 – Lightness interval values of the ground surfaces of material specimens.
first
measurement
SURFACE LIGHTNESS INTERVAL
second
third
fourth
measurement
measurement
measurement
WORKPIECE
MATERIAL
195 – 215
155 – 180
170 – 195
135 – 160
60 – 95
40 – 60
55 – 90
20 – 40
Č3840
Č6980
185 – 205
180 – 200
120 – 145
150 – 185
55 – 75
65 – 85
20 – 40
35 – 65
Č4150
Č4770
Table 1 shows that the surface lightness interval values of the ground material
specimens, that were ground with low value of reduced material removal rate, are several
times greater than those in which grinding was carried out with greater productivity. It may
be concluded that the surface lightness interval values match well the burn intensity of the
610
D.Ciglar; R.Cebalo; T.Udiljak
ground surface, so that unburned workpieces have higher value of surface lightness
interval and lower value of surface lightness interval describes the more burned ground
surface of the material specimens.
In order to study the connection between burn intensity of the ground surface and
other defects, i.e. in order to state which specimen surface presented in Figure 1 shows
the correctly and which the incorrectly ground material specimens, other indicators will be
studied that show incorrect grinding conditions. The change in surface layer hardness of
the ground surface and the depth of the layer of variable hardness i.e. the HAZ depth
were studied.
In the first grinding, where Qbr=0.685 mm3/smm, i.e. where high values of surface
lightness interval of the ground material specimens were determined, the Vickers
hardness test measurement did not show change in the surface layer hardness on any of
the material specimens, and the grinding is considered correct.
Vickers hardness test measurement on material specimens in Figure 1b shows
negligible change in surface layer hardness. In all material specimens the softening of the
ground surface occurred. In Č3840 approximately 2.9%, in Č6980 approximately 1%, and
in Č4150 and Č4770 approximately 2.4%. The depth of the variable hardness layer, i.e.
the softening depth in all the material specimens is less than 0.03mm. Since change of
hardness and depth of the variable hardness layer are small, this grinding at Q br=3.425
mm3/s mm can be regarded as correct.
On material specimens ground at Qbr=6.85 mm3/smm (Figure 1c), the change in
hardness and depth of variable hardness layer are not negligible. In material specimens
Č3840, Č6980 and Č4770 surface softening was determined, whereas material specimen
Č4150 showed hardening of the surface itself, but the layer beneath it softened. Reason
for the reduction in surface layer hardness of material specimens Č3840, Č6980 and
Č4770 is the tempering of the primary martensite due to the grinding temperature. The
specimen material Č3840 has the maximum depth of the softened layer approximately
0.13 mm and surface softening of 18.3%. The least softening of the surface of 5.2% and
the least depth of the softened layer of approximately 0.05mm was obtained from material
specimen Č6980, whereas in the material specimen Č4770 the softening was 8.2%, and
the HAZ depth approximately 0.12mm. In specimen of material Č4150 the grinding
temperature at the very surface of the specimen caused the conversion of the residual
austenite into the secondary martensite and thus hardening of 3.8%, but a little further
from the grinding surface only the tempering of the primary martensite occurred, i.e. OTM
generated, and this layer softened by approximately 7.7%. The total depth of the variable
hardness layer is approximately 0.11mm. It may be noted that the percentage of the
hardness change compared to the initial hardness and depth of the variable hardness
layer are different in ground material specimens. This is understandable, since regardless
of the fact that all the selected material specimens belong to the group of tool steels,
according to 12,13,14 their chemical composition, microstructure, thermal-physical
properties and behaviour during grinding certainly differ. In grinding of all material
Influence of Grinding Prod. on the Burn Intensity of the Ground Surface
79
specimens at Qbr=6.85 mm3/s mm, the hardness of the surface layer of the ground area
significantly changed and since depths of variable hardness layers are large, this grinding
may be regarded as incorrect.
In the fourth measurement in grinding the material specimens with maximum value
of reduced material removal rate of Qbr=12.33 mm3/smm, (Figure 1d), again the hardness
of the surface layer was reduced, i.e. the ground surface softened in materials Č3840,
Č6980 and Č4770. The specimen material Č3840 has the highest percentage of surface
layer softening of 25.7% and the greatest depth of the softened layer and the HAZ depth
amount to approximately 0.14mm. Specimen material Č4770 has a lower percentage of
softening 9%, and approximately same depth of the softened layer of 0.14mm, whereas
specimen material Č6980 has the least percentage of softening 5.8% and the least depth
of softened layer of approximately 0.07mm. Only in the material Č4150 the surface of the
specimen hardened, since the hardness of the surface layer increased compared to the
initial hardness by approximately 13.6%. The depth of the variable hardness layer is also
quite large and amounts to approximately 0.1mm. Since in this measurement also the
hardness of the surface layer hardness and the depth of the variable hardness layer
changed significantly in all the material specimens, it may be concluded that the surface of
the ground specimens in Figure 1d represents again an incorrect grinding process.
This is confirmed by the study of the occurrence of thermal cracks on the ground
surfaces of material specimens. The magnetic method using black particles has
determined the presence of thermal cracks only in the third and fourth measurement, i.e.
in grinding with the value of reduced material removal rate of Qbr=6.85 mm3/smm and
Qbr=12.33 mm3/smm, and only on the surface of material specimen Č4150. Figure 2
shows the recorded thermal cracks on the ground surface of the material specimen Č4150
in the third and fourth measurement.
Figure 2. Cracks on the material specimen Č 4150 ground at
Qbr=6.85 mm3/smm and Qbr=12.33 mm3/smm.
Figure 2 shows that the thermal cracks on the ground surface of the material
specimen Č4150 are in fact perpendicular to the direction of grinding and have a form of a
net. They tend to penetrate into the depth of the workpiece, and therefore are very
dangerous and cannot be allowed in grinding. Only with material specimen Č4150 are
810
D.Ciglar; R.Cebalo; T.Udiljak
Surface lightness
interval
180
160
140
120
100
Č3840
80
60
40
20
0
3,425
6,85
100
90
80
70
60
50
40
30
20
10
0
Change of hardness, %
thermal cracks understandable, since only this grinding surface hardened, shown by
hardness measurements. Testing of thermal cracks has confirmed the inaccuracy of the
grinding process of material specimen Č4150 in the third and fourth measurement, i.e.
grinding with the value of reduced material removal rate of Q br=6.85 mm3/smm and
Qbr=12.33 mm3/smm.
Based on the results of previous measurements, the next four diagrams show for
each material specimen the dependence of the burn intensity of the ground surface i.e.
surface lightness interval and the change in hardness of the subsurface layer of the
ground specimen on the value of the reduced material removal rate. The burn intensity of
the ground surface is presented by the lower surface lightness interval value, and the
change in hardness is determined with relation to the initial hardness and expressed in
percentages.
12,33
3
Reduced material removal rate, mm /s mm
160
Č6980
Surface lightness
interval
140
120
100
80
60
40
20
0
3,425
6,85
100
90
80
70
60
50
40
30
20
10
0
Change of hardness, %
Diagram 1. Dependence of the surface lightness interval and the change of
hardness on reduced material removal rate , Č3840.
12,33
3
Reduced material removal rate, mm /s mm
Diagram 2. Dependence of the surface lightness interval and the change of
hardness on reduced material removal rate , Č6980.
140
Č4150
Surface lightness
interval
120
100
80
60
40
20
0
3,425
6,85
100
90
80
70
60
50
40
30
20
10
0
99
Change of hardness, %
Influence of Grinding Prod. on the Burn Intensity of the Ground Surface
12,33
3
Reduced material removal rate, mm /s mm
160
Č4770
Surface lightness
interval
140
120
100
80
60
40
20
0
3,425
6,85
100
90
80
70
60
50
40
30
20
10
0
Change of hardness, %
Diagram 3. Dependence of the surface lightness interval and the change of
hardness on reduced material removal rate , Č4150.
12,33
3
Reduced material removal rate, mm /s mm
Diagram 4. Dependence of the surface lightness interval and the change of
hardness on reduced material removal rate , Č4770.
Diagrams 1 to 4 show that in all the material specimens the increase in the value
of reduced material removal rate causes reduction in the lower surface lightness interval
value, and increase in hardness change. Since the lower surface lightness interval value
represents the burn intensity of the ground surface, this means that only in higher burn
intensity of the ground surface of material specimens substantial change in the hardness
of subsurface layer occurs, and also great depth of the variable hardness layer. In that
case, the ground surface of the material specimen Č4150 shows also thermal cracks. It
1010
D.Ciglar; R.Cebalo; T.Udiljak
has thus been confirmed that there is a dependence between the burn intensity of the
ground surface and other grinding defects, so that the lower value of the surface lightness
interval can determine the correctness of the grinding process and the approximate limited
productivity, i.e. approximate value of the reduced material removal rate satisfying the
quality of the ground surface.
4. CONCLUSION
The results of research lead to the conclusion that the grinding process with low
value of reduced material removal rate (low productivity), causes little surface burn of the
material specimen, and high values of surface lightness interval. By substantial increase
in the productivity of the grinding process, all the material specimens show greater burn,
have darker surface and substantially lower lightness interval values. It may be concluded
that the surface lightness interval matches well the burn intensity of the ground surface,
since non-burned workpieces have greater value of the surface lightness interval,
whereas lower value of the surface lightness interval describes a more burned ground
surface of the material specimen.
It may be further concluded that there is a significant dependence between the
intensity of the ground surface burn and other grinding defects, since the results show that
only in the case of greater burn of the ground surface is there substantial change in the
hardness of the surface layer and large depth of the variable hardness layer, and in
material specimen Č4150 it came even to thermal cracks on the ground surface.
Finally, it may be concluded that ground surface lightness interval can be used to
determine the correctness of the grinding process. That is, it may be concluded that there
is a certain critical lightness interval of the ground surface and a critical burn of the ground
surface on the basis of which the approximate limited productivity can be determined, up
to which the grinding process is still correct.
5. LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
M.Field, J.Kahles ; Rewiew of Surface Integrity of Machined Components, Annals
of the CIRP Vol. 20/2/1971., str. 153-163.
R. Cebalo, Suvremena tehnologija brušenja, Zadarska tiskara, Zagreb 1993.
M. C. Shaw, A.Vyas, Heat Affected Zones in Grinding Steel, Annals of the CIRP
Vol. 43/1/1994., str. 279 –282.
International standard ISO 6507-3, first edition 11.01.1989.
H. K. Tönshoff, E. Brinksmeier, Determination of the Mechanical and Thermal
Influences on Machined Surfaces by Microhardness and Residual Stress Analysis,
Annals of the CIRP Vol. 29/2/1980., str. 519 – 530.
A. Gilardoni, A. Orsini, M. Tacconi, Gilardoni Handbook – Nondestructive Testing
NDT, Gilardoni S.p.A., Mandello Lario (Como), Italy, 1981.
Influence of Grinding Prod. on the Burn Intensity of the Ground Surface
119
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
O.Kubaschewski, B.E.Hopkins; Oxidation of Metals and Alloys, Butterworths,
London, 1962.
W.B.Rowe, S.C.E.Black, B.Mills, H.S.Qi, M.N.Morgan, Experimental Investigation
of Heat Transfer in Grinding, Annals of the CIRP Vol. 44/1/1995., str. 329 – 332.
D.Ciglar, Doktorat, Zagreb 1999.
LECO 2001 Image Analysis System Operator’s Manual, Version 2.01, Kirchheim
1992.
Katalog Swaty – Maribor
Katalog Željezare Ravne
Physical Constans of some commercial steels at elevated temperatures, The
british iron and steel research association, Butterworths scientific publications,
London 1953.
Böhler – Edelstahlhanduch 2.0, Böhler – Edelstahl GMBH, 1989.