LIBRARY
Michigan State
University

 

 

 

This is to certify that the

thesis entitled
LOCAL DETERMINATION OF TOOL NEAR

DURING TURNING OPERATIONS

presented by
BULENT DOGRUYOL

has been accepted towards fulfillment
of the requirements for

NE degree in W

91am
Major professor
Date g/i/gb if

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

MSU

LIBRARIES
as.

s.’

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

LOCAL DETERMINATION OF TOOL WEAR

DURING TURNING OPERATIONS

By

Bulent Dogruyol

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Metallurgy, Mechanics and Materials Science

1985

ABSTRACT

LOCAL DETERMINATION OF TOOL WEAR
DURING TURNING OPERATIONS

By

Bulent Dogruyol

A cutting tool that was instrumented with strain gages was used
to machine a low carbon steel. From the measured strains, the cutting
forces at the tool tip were calculated. These forces were input into
a finite element analysis that was employed to determine the stresses
throughout the tool. During machining the tool was periodically exa-
mined in a scanning electron microscope to investigate the wear beha-
vior. It was possible to correlate the stresses and forces on the

tool with its wear.

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to Professor
John F. Martin for his support and guidance throughout this project.
Thanks are also due to Dr. Karen Baker from the Center for Electron

Optics for her help concerning the electron microscopy aspects.

ii

TABLE OF CONTENTS

Page

LIST OF TABLES. ........ ...... ...................... . ....... ........ iv
LIST OF FIGURES.... ....... . ............................. . .......... v
1. INTRODUCTION............. OOOOOOOOOOOOOOOOOOOOOOOOOOO . ........... 1
2. EXPERIMENTAL TECHNIQUES.................. .................. ..... 3
2 1. Experimental Setup and its Elements.. . ................... 3
2.1.1. The Lathe... ................ O ........... ....... ..... 3

2.1.2. The Workpiece.. ................ .. ....... .. .......... 3

2.1.3. The TOOlooooooo ......................... ... ......... 5

2.1.4. The Strain Gages...... ....... ...... ....... .......... 9

2.2. Experimental Procedure ............... ..... ............. .... 11
2.3. Experimental Results ................................. . ..... 12

3. ANALYTICAL PROCEDURES. . . . . ............... . ..... . ........... . . . . . 16
3.1. Calculation of Cutting Forces .............................. 16
3.2. Finite Element Analysis ............................ . ....... 23
3.2.1. Tool Model...... ...... . .............. . .......... .... 23

3.2.2. Stress Distribution in the Tool ...... . ......... ..... 27

4. ELECTRON MICROSCOPIC INVESTIGATIONS .......... . .................. 32
4.1. Wear Mechanisms on High Speed Steel Tools........ ...... .... 32
4.1.1. Plastic Deformation by Shear at High Temperature.... 35

4.1.2. Plastic Deformation under Compressive Stress........ 35

4 . 1 . 3. Diffusion wear 0 . . . . . . . . . . . ........ . . . . . . . . . . . . . . . . . . 38

4 . 1 . 4. Attrition Wear . . . . . ....... . ..... . .......... . . . . . . . . . 38

4 . 1 . 5 . AbraSive wear. . . . . . . . . . ...... . . . .......... . . . . . . . . . . 39

4.1.6. Wear under Sliding Conditions.... ..... . ............. 39

4.2. Results of Electron Microscopic Investigations... .......... 42

5. CORRELATION OF DATA AND SUMMARY OF RESULTS ...................... 52
6. DISCUSSION ........ . ....... . ..................................... 54
LIST OF REFERENCES. . . . . . . . . O . . . . . . . . . . .......... . . O . . . . . . . . . O . . . O . O 56

iii

LIST OF TABLES
Table Page

1. Chemical and Mechanical Properties of the Workpiece Material.... 5

2. Chemical Composition and Hardness of the Tool Material ........ .. 6
3. Strain Readings............ ......................... . ........ ... 14
4. Calculated Cutting Forces at the Tool Tip ............. .......... 19
5. Compressive Stresses at the Cutting Edge.... .......... .... ...... 27

iv

LIST OF FIGURES

Figure Page
1. Experimental Setup and its Elements. ............ . .......... ..... 4
2. Geometry of the Tool Used in Experiments ................. .. ..... 7
3. Geometry Definitions of a Conventional Tool... ........ . ......... 8
4. Strain Gages and Cutting Forces ................................. 10
5. Strain vs. Time for One Pass.. ....................... . .......... 13
6. Strain vs. Cumulative Cutting Time .............................. 15
7. Cutting Forces vs. Cumulative Cutting Time.... .................. 20
8. Deformation of the Cutting Edge under Compressive Stress ........ 22
9. 3-D Isoparametric Element .......................... . ............ 24

10. Finite Element Model of the Tool ............................ .... 25

11. Compressive Stress vs. Cumulative Cutting Time .......... . ....... 29

12. Compressive Stress Distribution on the Rake Face ................ 3O

13. Chip Formation ............ . ........................... . ......... 33

14. Seizure and Intermittent Contact Areas on the Rake Face ......... 36

15. Wear Mechanisms on HSS Tools... ..... .... ........... .... ........ 37

16. Built-up Edge Formation ......................................... 41

17. Tool Tip after 5 min. of Cutting.... ................ .. .......... 43

18. Tool Tip after 10 min. of Cutting... ...... ...................... 43

19. Tool Tip after 20 min. of Cutting ...... . ........................ 44

20. Tool Tip after 20 min. of Cutting (blow-up of Fig. 19) .......... 44

21. Built-up Edge Adhering to the Rake Face of the Tool ..... . ..... .. 45

22. Side Cutting Edge after 24 min. of Cutting ...................... 45

23. Rake Face after 26 min. of Cutting.. ......................... ... 48

24. Rake Face after 26 min. of Cutting (blow—up of Fig.23). ...... ... 48

25. Rake Face after 28 min. of Cutting .............................. 49

26. Rake

27. Rake

28. Side

29. Rake

Face after 28 min. of Cutting (blow—up of Fig. 25) .......... 49

Face after 28 min. of Cutting (blow-up of Fig. 26).. ........ 50
Cutting Edge after 29 min. of Cutting.. ..................... 50
Face after 30 min. of Cutting ...... ..-. ..... . ........ . ...... 51

vi

1. INTRODUCTION

Despite the complicated nature of the many interacting parameters
in a machining environment, it is the behavior of a very small volume
of metal around the cutting edge that determines the performance of
tools. During any cutting operation, the interface between tool and
work material is largely inaccessible to observation. However, indirect
evidence concerning stresses, temperatures and chip formation can be
obtained. This study will attempt to evaluate the tool wear and the
cutting forces in the case of a turning operation by monitoring the
parameters while the process is in progress. The outline that it will
follow is shown in the following diagram and can be divided into three

parts:

 

 

 

   

 

 

 

 

 

 

 

Lathe Strain XrY

Tool/Workpiece Gages Recorder
Scanning Electron Microscope Cutting Finite Element
Wear Analysis Forces Analysis of Tool

 

 

 

 

 

Correlation of Experimental [‘
and Analytical Data J‘

 

 

Experimental Techniques: Machining experiments to evaluate the cutting
forces that will eventually dull the cutting
tool completely. Since the final goal of this
work is to correlate wear behavior of the tool

with cutting forces, the machining parameters
1

Analytical Procedures

Electron Microscopic

Investigations

were selected so as to guarantee the gradual

wearing and final destruction of the tool.

Determination of cutting forces and finite
element analysis to evaluate the stress and

strain state throughout the cutting tool.

Use of a scanning electron microscope for

analysis of wear on the cutting edge of

the tool.

2. EXPERIMENTAL TECHNIQUES

2.1. Experimental Setup and its Elements

The experimental part of this thesis consists of a single set of
turning operations on a 1athe.Strain gages were mounted on three sides
near the nose of the tool. These gages were placed in positions such
that the forces on the tool tip could be calculated directly using
simple beam theory. They were used to continuously measure the strains
experienced by the cutting tool during the cutting passes and these
strain values were recorded with two X—Y recorders. These measured
strains, then, became the input into the beam theory to determine the

cutting forces. The elements of this setup are shown in Figure 1.

2.1.1. The Lathe

The experiments were performed on an 8-speed, 13-inch South Bend
engine lathe with the following characteristics:
Chuck :‘9 in. with 3 jaws
Spindle speed range : 30—685 RPM

Bar capacity : 1 in.

2.1.2. The Workpiece

A low carbon steel, SAE—AISI 1018, was used as the workpiece,
material to ensure low temperatures near the tool tip where the strain

gages were located. SAE-AISI 1018 is a very common material and this

 

JAW
111 1 . GAGE 1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LATHE 2 - GAGE 2
3 - GAGE 3
CHUCK WORKPIECE
1\
Fa . 3
TOOL
(N
a,
X-Y RECORDER

 

 

 

 

 

Figure 1. Experimental Setup and its Elements

was a further criterion in selecting the workpiece material. The
specimens were cut from cold drawn bar stock with a diameter of 1.0 in.
and were 7 in. long. Table 1 summarizes the chemical and mechanical

properties of the workpiece material (1,2):

Table 1. Chemical and Mechanical Properties of the Workpiece Material

 

SAE-AISI C Mn P S Tensile Yield Elongation
Number (Z) (Z) (Z) (Z) Strength Strength (Z 2 in.)
max max (psi) (psi)
1018 .18 .60 .04 .05 68,000 55,000 32.0

Reduction of Hardness
Area (Z) BHN RB

 

66.5 137 77

2.1.3. The Tool

A high speed steel (HSS) of type M2 (2) was chosen as the cutting
tool. This material constitutes a good compromise between strength at
high temperature and the toughness requirements for machining relative—
ly soft steels like SAE—AISI 1018. These tools are now the most common-
ly used of the high speed steels, because they make possible the cut*
ting of steel and other high melting-point materials at much higher
rates of metal removal than can be achieved with carbon steel tools.
The improved performance is made possible by their retention of hard-
ness and compressive strength to higher temperatures. The composition

and hardness of the tool material is given in Table 2 (1,2,3):

Table 2. Chemical Composition and Hardness of the Tool Material

 

SAE—AISI C Mn W Mo Cr V Fe Hardness
Number (Z) (Z) (Z) (Z) (Z) (Z) (Z) HV RC
M2 .85 .30 6.0 5.0 4.0 2.0 Balance 840 65

To enable an uncomplicated 3-D modeling of the tool for the finite
element analysis, a simple geometry was desirable; yet it had to be
sophisticated enough to machine the workpiece and simulate some

actual machine shop conditions. Finally its overall length was kept
less than three inches to facilitate its fitting into the specimen
chamber of the scanning electron microscope that was used to inves-
tigate the wear patterns on the tool tip. Figures 2 and 3 show the
geometry of the tool used in these experiments and of a conventional
tool geometry that is used in machine shops. The list of various
angles comprising the tool geometry is called the "signature" and has
been standardized by the American Standards Association. In accordance
with the above considerations, the cutting tool.employed in this study

was ground to the following geometry (1,2,3,4);

Experimental Recommended
Tool Geometry
Back Rake Angle = 0 (10)
Side Rake Angle = 0 (12)
End Relief Angle = 10 ( 8)
Side Relief Angle = 10 (18)
End Cutting Edge Angle = 10 ( 8)
Side Cutting Edge Angle = 0 (15)
Nose Radius = 1/64 in. (1/32 in.)

The numbers in parantheses indicate the recommended angles for single

 

 

 

 

END CUTTING
EDGE ANGLE

-— --

I
/|

SIDE RELIEF
“as?“

Figure 2. Geometry of the Tool Used in Experiments

 

TOP VIEW
SHANK
FACE sun: comma
%// EDGE ANGLE

 

7’" ‘

END CUTTING NOSE RADIUS

EDGE ANGLE

SIDE RAKE ANGLE BACK RAKE ANGLE
\\
.\_ - - SIDE

 

 

Iii-:5:
HEEL //

   

 

 

 

 

FLANK
: Rfflgryd’
ANGLE '
FRONT VIEW RIGHT VIEW

Figure 3. Geometry Definitions of 3 Conventional Tool

point HSS tools when machining plain low carbon steel.

The deviations of the back rake, side rake and side cutting edge
angles from the recommended values will all result in an increased
cutting force, decreased tool life and worsened surface finish. These
effects, besides the goal of a simplified geometry, were also intended
to accelerate the wearing of the tool and to eventually result in

failure.

2.1.4. The Strain Gages

Three strain gages, as shown in Figure 4, were mounted on three
faces of the tool to monitor the strain state caused by the forces Fx,
Fy and F2. The main purpose of measuring the strains was to use them
in determining the cutting forces Fx, Fy and F2 at the tool tip.

Gage 1 was placed on the neutral axis and far enough away from the tip
so as to not be affected by the high temperature that develops in the
vicinity of the tool tip. Gage 2 was located along the neutral axis of
the flank face and gage 3 was placed off the neutral axis and at the
lower edge of the right face. All three gages had the same perpendicu-
lar distance to the tip (Figure 4). By this configuration the forces
Fx, Fy and F2 at the tip can be determined by simple beam theory.

The specifications of the gages are listed below:

Manufacturer: Micro-Measurements

Gage Type : EA—06-075AA-120
Resistance : 120.0 (+/-) 0.15Z ohms
Gage Factor : 2.005 (+/—) 0.5% at 75 F.

10

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l—Y—i

I.
l 10'
CROSS-SECTION or TOOL AT x

Figure 4. Strain Gages and Cutting Forces

11

2.2 Experimental Procedure

The machining experiments were carried out on round bar work-
pieces that were 5.5 in. long which allowed for a cutting time of
60 sec. for each cutting pass at a feed rate of .012 inches per re-
volution (ipr). At the end of each pass the tool was disengaged, the
workpiece was replaced with a new bar with the same diameter and a
new pass was started. On one hand, this gave the tool enough time to
cool down, thus avoiding the temperature distortion of strain gage
readings; on the other hand, the method is not a realistic simulation
of continuous cutting operations under shop conditions where the tool
remains in contact with the workpiece for longer periods of time. To
be in line with the recommended values of cutting speeds and cutting
depths when turning low carbon steels, the following values were

decided on:

Cutting depth d=0.10 in.

Cutting speed v=118 surface-feet per minute (sfpm)

The magnitude of the cutting speed was obtained by:

v = II*D*N/12 where D=1 in. workpiece diameter
N=450 RPM spindle speed

The experiments were performed dry.

12

2.3. Experimental Results

During each cutting pass, two X-Y recorders simultaneously plotted
the output from the three strain gages. Altogether 30 passes of 60 sec.
duration were carried out on 30 bars resulting in a cumulative cutting

time of 30 min. for the tool.

Figure 5 shows the original data trace of strain vs time for one
pass. The rapid initial increase in strain for gages 1 and 2 was due to
the sudden engagement of the tool with the workpiece. A sudden jump in
strains in gage 1 can also occur if a hot chip hits this gage, as
indicated in Figure 5. Table 3 shows the average strains obtained from
the gages for each pass. Since the strain gradient within a 60 sec pass
was small, the table contains the median values measured at the 30 sec.
period for each pass. From the strain table, it can be observed that
gages 1 and 2 are in tension whereas gage 3 is in compression.This
situation is expected since the effect of the larger forces Fz and Fy
would produce tensile strains on gages 1 and 2 and compressive strains
on gage 3. The strain readings also indicate that during the first half
of tool life (first 15 passes) the deformations in the tool increase
linearly and slowly. This trend changes in the second half of the tool
life (last 15 passes) when the strains rise progressively in an exponen
tial manner meaning a corresponding increase in the forces acting at

the tool tip. A plot of these data is shown in Figure 6.

13

AA
5 HOT CHIP
T /
R
A
I
N

 

[’AS]

 

 

l L l
I I I
20 40 60
T I M E ISEC.)

Figure 5. Strain vs Time for One Pass

14

Table 3. Strain Readings

WOrkpiece : SAE-AISI 1018 Feed : .012 ipr
Tool : HSS-M2 Cutting depth : .10 in.
Speed : 118 sfpm Duration of each pass : 60 sec.
Pass Strain (Microstrains)

e1 e2 e3

1 187 143 -246
2 188 145 -248
3 190 147 -250
4 192 149 -252
5 194 152 -255
6 195 153 -256
7 196 154 -257
8 198 155 -258
9 200 157 -259
10 202 159 -261
11 203 160 -262
12 204 161 -263
13 205 162 -265
14 206 163 -267
15 208 165 -269
16 210 167 -271
17 212 169 -274
18 215 172 —277
19 218 175 -281
20 222 179 -285
21 226 181 -288
22 231 183 -292
23 236 186 -297
24 242 190 -302
25 249 194 -307
26 256 201 -319
27 266 209 -334
28 278 218 -349
29 294 227 -360

30 321 241 —379

15

A
350 2L
'5
R 250 -_ GAGE 1
A 200 ..
I
160 .4
N
100 __
(p51
50 l-

l

‘I—

r I
6 10 15 20 26

CUTTING TIME mm.)

1
T

 

Figure 6. Strain vs Cumulative Cutting Time

 

.—

 

N7

3. ANALYTICAL PROCEDURES

3.1. Calculation of Cutting Forces

By using simple beam theory the forces at the tool tip that cor-
responded to each strain state were calculated. Fz is the tangential
cutting force due to rotational relative motion between the tool tip
and the workpiece. This is normally the largest cutting force compo—
nent and acts in the direction of the cutting velocity. Fy, the feed
force, is generated by the longitudinal feeding motion of the tool
with respect to the workpiece. The magnitude of Fy, in general, ranges
between 30Z and 60% of Fz. Fx, the radial force, is the least signifi-
cant of all cutting force components and is produced by the thrusting
action of the tool tip against the work material. Usually this force,Fx
is neglected for the purposes of analysis of cutting forces in simple

turning (5,6,7).

A summary of the beam theory employed for this analysis is given

as follows:

Strains on Gage 1

 

Fx FX
Compression due to Fx: S1 = - --A_ = E e1 thus el = _ E-A
Bending due to Fy : e1 = O gage on neutral axis

M Fz z 2
Bending due to F2 : 81 --E = (X) I = E e1

I (1/12) z3 y

 

16

Strains on Gage 2

 

 

l7

 

 

 

. Fx _ Fx
Compression due to Fx. 82 A - E e2 thus e2 - E—A
Bending due to Fy : 82 M—C = Fy (x) ng = E e2
I (1/12) y 2
thus e = --§—§- Fy
2
E y 2
Bending due to F2 : e2 = O gage on neutral axis
Strains on Gage 3
C ssi d t F ° S = - -E§- = E th = — EE-
ompre on ue o x. 3 A e3 us e3 E A
Bending due to Fy : S3 = - M_c = - Fy (x) ng — E e3
I (1/12) y 2
thus e = — --§-§- F
3 2
E y 2
Bending due to F2 : S3 = — §_E = - Fz (x) ng E e3
I (1/12) 2 y
thus = - -_§_§- F2
3 2
E z y

Linearly adding strains for each gage results in:

Ex

e1 = - E‘K + + —-—§—- Fz
E z y
_ Fx 6 x
%- EA+ ‘2 W
E y z
Fx 6 x 6 x
e = - --- — -—--—- F — ------ F2
3 E A E 2 z E 22 y

Solving for Fx, Fy and F2:
E A

Fx = - -§- (el + e2 + e3)

Fy =

 

18

F = §_EE_Z ( 2 - _ )
18 e1 e2 83
With x = .250 in.
y = .500 in.
z = .500 in.
A = .228 1n2
E = 30x106 psi. (see Figure 4)

one obtains:

= 6
Ex 2.2800x10 ( e1 + e2 + e3 )
Fy = 0.8333x106 (~e1 + 2e2 - e3 )
Fz = 0.8333x106 ( 2e1 - e2 - e3 )

Based on this analysis, the values of the cutting forces at each
pass are given in Table 4. A plot of these data is presented in Figure
7. On this plot also the most striking feature of cutting force beha-
vior with respect to cutting time is the rapid exponential increase in
cutting forces in the last one third of the accumulated cutting time.
Within the first two thirds of cutting time (first 20 min. ), the rise
in cutting forces is linear with a shallow slope; in fact, the tangen-
tial cutting force increases only by 15% within this time period.
However, the increase within the last 10 min. amounts to a total of
42Z increase over the starting value. This pattern is not a result of
a variation in cutting parameters since they were all kept constant.
The increase in these forces can be attributed to the changes in
geometry and cutting conditions which resulted from wear in the tool/

work contact area as well as an increase in temperature.

A comparison of the magnitudes of feed and tangential cutting
forces shows that the feed force has a relatively high value, i.e.

above 70Z of F2 throughout 30 min. This is indicative of a large

19

Table 4. Calculated Cutting Forces at the Tool Tip

Pass Fz (lb) Fy (1b) Fx (lb)
1 398 287 -192
.2 399 292 -194
3 402 295 -198
4 406 298 -203
5 409 304 -207
6 411 306 -210
7 413 308 -212
8 416 309 -219
9 418 311 -223
10 421 314 -228
11 423 316 -230
12 425 318 -231
13 427 320 -232
14 430 323 -233
15 433 326 -237
16 437 329 -241
17 441 333 -244
18 446 338 -251
19 452 344 -255
20 458 351 -264
21 466 353 -271
22 476 356 -278
23 486 361 -285
24 497 367 -296
25 509 372 -310
26 525 388 -315
27 548 405 -321
28 573 422 -335
29 601 433 -367

30 649 452 -419

20

 

 

600 .‘E
F
0 500 -I. Fz
R 400
C
E 300
S
200 __
(LB)
100 ..
J l i L L A
I I I T I T V
5 10 15 20 25 30
_L. CUTTING TIME mm.)

 

 

Figure 7. Cutting Forces vs Cumulative Cutting Time

21

chip/tool contact area on the rake face(see Figure 14), since the feed
force is a measure of the drag which the chip exerts as it flows away
from the cutting edge across the rake face (5,6). The analysis produ—
ces negative values for the radial force Fx which means that it is
acting away from the tool tip in a radial direction as a tensile force.
This is not realistic. However, this effect might have been caused due
to extensive deformation into the workpiece of the cutting edge under
high tangential and feed forces causing the gages to interpret the
radial force as being tensile (see Figure 8). This and all other afore—
mentioned effects will be discussed in more detail when the results of

the electron microscopic investigations are presented.

//////////////

%

23

3.2. Finite Element Analysis

A finite element model of the tool was used to obtain an estimaé
tion of stresses acting throughout the tool. The input to the analysis,
which used the commercially available software package ANSYS by
Swanson Analysis Systems, Inc. (8), was the cutting forces calculated

in the previous section.

3.2.1. The Tool Model

Since the idea was to obtain a rough estimate of stresses, a
relatively coarse mesh was used and the tool tip was approximated as
a point. Although it was expected that this approximation will result
in higher theoretical stresses than actual, it simplified the mesh
generation scheme a great deal enabling partial automatic mesh genera-
tion in the preprocessing phase. The element employed in the analysis
was a 3—D isoparametric stress solid element STIF45. It is used for
three dimensional modeling of solid structures and is defined by eight
nodal points having three degrees of freedom at each node (translations
in the nodal x, y and 2 directions). A further advantage of tool nose
approximation as a point lies in the fact that it was possible to
select an eight node volume element rather than a higher order, noné
linear volume element, i.e. 16- or 20-node elements (8). The effect was
a reduction in the degrees of freedom which decreased the solution core
size and the running time. This element is shown in Figure 9. The
resulting tool model consisted of 100 elements and 180 nodes (see Figure

10). In the analysis, the cutting tool was treated as a cantilever beam

24

 

 

 

 

Figure 9. 3-D Isoparametric Element

25

 

 

 

3
..me

 

 

 

 

 

 

 

 

 

 

 

.3 .
fl! .. .
r 9/ ... ....
f. ..x NZ“ WIN
ft /
3m “.7 E.“ f
. ... nu
.. /. .... /. .... .. ...
~ ~ WNW—1.; ar 4
_ ..x ...x an E a. 6%
Mk: .\ .
: ...? ...._ 1..., a l /L a
.. ... a 5% 3 at 9F
__ .... .. «a t. x .t IIIMWIIIIIIJS
__ ._ Ir a... V
.. .... lar \. ...x Va
__ x. .. H a
... ”k1 .... ...... _\ ....—
.. .... I... \- “rw ...
.. .. ...? laws
...... .. ... f... .
, t. v .. . ..
. {waif ...e. N \ IN “ HR
. .... . ....
/ x «v IIII
, .. C , .IIIINIIIIIII .4...
. NM ..V x. . ......
..Qu.‘ . ....
. .. mm ...
...
\
..
K.—
L40

 

we 3
wwmcnm Ho. anHnm mwmamnn Zommw on n50 HOOH

26

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mwmcnm Ho. mHDHnm mwmamnn Zommw om nrm HOOH

27

clamped within the tool holder. Consistent with this situation, all
three degrees of freedom for all the dodes at this end of the tool were
set to zero. The input to the program consisted of the cutting forces

at different times during the tool life.

3.2.2. Stress Distribution in the Tool

The cutting force acting on a tool with a small rake angle
imposes a stress on the rake face which is largely compressive in
character. The mean value of this stress is determined by dividing the
cutting force by the contact area, which in reality is very difficult
to determine accurately. In a lathe, where the tool acts as a canti-
lever, there are also bending stresses giving tension on the upper
surface between the contact area and the tool holder. These stresses
are negligible at the tool tip, because they rise linearly starting at
the tip and reach a maximum at the clamped end of the tool. The tool
model described above was used to calculate the compressive stresses
close to the tool tip by feeding the program with the cutting forces
after cutting times of 1 min., 5 min., 10 min., 15 min., 20 min., 25
min., 28 min., and at the end of the tool life of 30 min. Table 5 lists

these values:

Table 5. Compressive Stresses at the Cutting Edge

Cutting Time (min.) Compressive Stress (psi.)

1 13,650

5 15,750

10 18,900
15 22,200
20 28,450
25 36,300
28 42,350

30 48,350

28

These values, as plotted in Figure 11, are the compressive stress-
es in the first element of the model near the tip and are averaged over
the element. This element, being .1 in. by .1 in., has an area of .01
in2 on the rake face. This area can be, in general, assumed to be lar-.
ger than the actual chip/tool conatact area whose accurate value usu-
ally is not known. But methods have been developed for measuring stress
distribution under certain simplified laboratory conditions (5,9,10).
They provide evidence that the compressive stress is highest near the
cutting edge, diminishing rapidly across the rake face to zero where
the chip breaks contact with the tool. It seems probable that this
stress distribution is common during cutting and that the compressive
stress at the edge is often double the mean stress or even greater.
Figure 12 qualitatively illustrates this distribution for the loads
after a cutting time of 30 min. Under the above assumptions one would
expect stresses to be on the order of 100,000 psi at the cutting edge.
This stress then rapidly decreases along the length of the element
away from the cutting edge. This situation would be comlicated in the
presence of a built-up edge in which case the stresses would probably

be lower.

The very high normal stress levels account for the conditions of
seizure on the rake face, particularly near the cutting edge. The con-
ditions of seizure will be discussed in the next section. If a tool
does not possess adequate strength, it may undergo local plastic
deformation and subsequent cyclic softening under these conditions.

The tool used in these experiments had a yield strength of approximate-

ly 147,000 psi. It is very likely that this tool at least was

29

50.000 ....
T

40.000 ..
30.000 ....
(PSIJ

20.000 ....

10.000 ....

 

J l L l 1A
1 r I I IV

5 10 15 20 25 30
CUTTING TIME WIN.)

“P

Figure 11. Comressive Stress vs Cumulative Cutting Time

30

 

 

 

 

 

 

 

A
100.000
T
80.000..-
R
E
60.000-..
8
8
40.0004.
(PSI)
20.000..-
l\
V
DISTANCE FROM CUTTING EDGE L UN.)

Figure 12. Compressive Stress Distribution on the Rake Face

31

subjected to large elastic deformation locally which means a change

in predefined geometry. This would affect the cutting performance.

The graph of stress vs cutting time reveals similar trends as
indicated by strain vs time and force vs time diagrams. Again the
stress at the cutting edge increases progressively with increasing

cutting time (Figure 11).

4. ELECTRON MICROSCOPIC INVESTIGATIONS

The machining experiments were periodically stopped to assess the
amount and nature of wear at the tool tip. During the course of the 30
min. tool life, the tool was removed several times from its holder and
examined in a scanning electron microscope (SEM) for changes of shape.
Interruptions were more frequent toward the end of the tool life. The
use of an SEM and visual inspection were the only methods employed in
this wear study. Before the discussion of the SEM results, as shown in
Figures 17-29, is given, an overview of the different wear mechanisms

on HSS tools is presented (5,11,12,13,14,15).

4.1. Wear Mechanisms on High Speed Steel Tools

When cutting metals, a tool with the shape of a large-angled
wedge is driven asymmetrically into the work material to remove a thin
layer (the chip) from a thicker body (the workpiece) (see Figure 13).
The chip formation occurs as the work material is sheared in the region
of a plane (the shear plane) extending from the tool edge to the posi-
tion where the upper surface of the chip leaves the work surface
(length OD in Figure 13).In this process, the whole volume of metal
removed is subjected to extensive plastic deformation, as indicated by
the transformation of volume V into V' in Figure 13. The chip thick—

ness t2 is always larger than the feed t1, i.e. the chip thickens as

it leaves the workpiece along the shear plane due to excessive plastic
deformation. The amount of this deformation and therefore of chip

thickening is dependent upon the shear plane angle 0 (see Figure 13).
32

WORKPIECE I

v

...;

CHIP

 

ChiP Format

34

Deformation and thickening are large at small angles, since the chip is
forced to leave the workpiece at a tight curve; they reach an optimum

at ¢ = 45 and grow again as 0 increases.

The wear pattern at the tool/chip interface is significantly de-
termined by the movement of the chip across the rake face and around
the tool edge. In most analyses this has been treated as a classical
friction situation, in which frictional forces tend to restrain move—
ment across the tool surface with a coefficient of friction between the
tool and the work materials as the coupling factor. This approach is
inappropriate to most metal cutting conditions. The concepts of fric-
tion apply when the stresses between surfaces are small compared with
the yield stress of the materials, which is true for many engineering
situations. But for most metal cutting operations the contact between
tool and the work surfaces is so nearly complete over a large part of
the total area of the interface, that sliding at the interface is imq
possible under most cutting conditions. Under these conditions of seia
zed or interlocked surfaces the movement of work material over the
tool surface cannot be adequately described using the terms 'sliding'
and 'friction', because the force parallel to the tool surface is not
independent of the contact area, but on the contrary, the area of con-
tact between tool and workpiece is a very important parameter in metal
cutting. Also, under the conditions of seizure there can be no simple
relationship between the fo-ces normal to and parallel to the tool
surface as is the case under conditions of sliding. Even under seizure
conditions, it is rare that the whole area of contact is seized

together. Some frictional sliding occurs in an intermittent contact

35

area (see Figure 14). The relative movement between chip and tool con-
tinues under conditions of seizure since the contact area is small and
sufficient force is applied (feed force) to shear the work material I .
near the seized interface. Under seizure conditions it can no longer

be assumed that relative movement takes place at the interface as is
teh case with frictional movement, because the force required to over-
come the interlocking and bonding is normally higher than that requ—
ired to shear the adjacent material. Relative motion under seizure

involves bulk shearing in the weaker of the materials.

4.1.1. Plastic Deformation by Shear at High Temperature

The characteristic form of this type of wear is the formation of
a crater, a hollow in the rake face some distance behind the cutting
edge (see Figure 15). The crater is located at the hottest part of the
rake surface from where the hot tool material is sheared since its
yield strength is greately lowered at high temperatures. Another fac-
tor is the increase in the yield strength of chip material which is
subject to high strain rates in the flow zone, thus becoming strong
enough to shear layers of tool material from hot regions. This is a
rapid acting wear mechanism, forming deep craters, which weaken the

cutting edge so that the tool may be fractured.

4.1.2. Plastic Deformation Under Compressive Stress

The compressive stress acting on the rake face is maximum at or

close to the cutting edge and when the stress is very high, the tool

36

AREA OF
SEIZURE CONTACT

...
‘——-

 

 

TOOL

 

Figure 14. Seizure and Intermittent Contact Ares

37

 

  
  

 
  

\
PLASTIC SHEAR AT

HIGH TEMPERATURE DEFORNATION UNDER

CONPRESSIVE STRESS

 

7/

\R

W
\

DIFFUSION WEAR ATTRITION WEAR

   

 

 

SLIDING WEAR

\\\‘

SEIZURE
CONTACT

   

 

\sunme
ABRASIVE WEAR '"EEoRrI‘T'ENT WEAR

 

 

Figure 15. Wear Mechanisms on HSS Tools

 

38

edge may be deformed downward (see Figure 15). This is a deformation
rather than a wear process, since no tool material is removed, but it
results in increased tool forces and brings into play or accelerates
wear processes which reduce the life of the tool. This wear mechanism
limits also the maximum workpiece hardness which can be machined with

HSS tools, since harder materials would cause excessive deformation.

4.1.3. Diffusion Wear

Tools may be worn by metal and carbon atoms from the tool diffu-
sing into and being carried away by the stream of work material flowing
over its surface. Rates of diffusion increase rapidly with temperature.
With HSS tools used in the usual cutting speed range, rates of wear by
diffusion are relatively slow because the interface temperatures are
relatively low. Diffusion accounts for the formation of craters at
speeds below those at which plastic deformation begins. Above these
speeds the damage caused by diffusion wear is obscured by the effect
of plastic deformation, which is a much more rapid wear mechanism. Wear
by diffusion also depends on a rapid flow rate in the work material

very close to the seized surface, to carry away the tool metal atoms.

4.1.4. Attrition Wear

This type of wear is more likely to occur at relatively low
cutting speeds where temperatures are low and wear based on plastic
shear or diffusion does not occur. The flow of metal past the cutting

edge becomes more irregular, less stream-lined or laminar and

39

contact with the tool may be less continuous. Under these conditions
larger fragments of microscopic size may be torn intermittently from
the tool surface. This is usually a slow form of wear, but more rapid
destruction of the tool edge occurs in operations involving interrup-
tions of cut or where vibration is severe due to lack of rigidity in

the machine tool or very uneven work surfaces (see Figures 15,24,28).

4.1.5. Abrasive Wear

Abrasive wear of HSS tools requires the presence in the work
material of particles harder than the martensitic matrix of the tool.
Hard carbides, oxides and nitrides are present in many steels, in cast
iron and in nickel-based alloys, but there is little direct experi—
mental evidence to indicate whether abrasion by these particles does
play an important role in the wear of tools. Where the work material
contains greater concentrations of harder particles, such as pockets
of sand on the surface of castings, rapid wear by abrasion undoubtedly
occurs. But it seems doubtful whether under conditions of seizure,
small, isolated hard particles in the work material can make an impor-

tant conribution to wear (see Figures 15 and 27).

4.1.6. Wear Under Sliding Conditions

At those parts of the interface where sliding occurs, either con-
tinuously or intermittently, other wear mechanisms can come into play.
The parts of the surface particularly affected are those shown as areas

of intermittent contact (see Figures 15,19 and 23). The wear mechanisms

40

operating in these sliding regions are probably those which occur
under more normal engineering conditions at sliding surfaces, invol-
ving both abrasion and metal transfer, and greatly influenced by

chemical reactions with the surrounding atmosphere.

To summarize, the wear and deformation processes which have been
shown to change the shape of the tool and to affect tool life, depend
on many factors: the work material, the machining operation, cutting
conditions, tool geometry, and the use of lubricants. In general the
first three wear mechanisms (4.1.1, 4.1.2, and 4.1.3) are important
at high rates of metal removal where temperatures are high and their
action is accelerated as cutting speed increases. It is these processes
which set the upper limit to the rate of metal removal. At lower
speeds, tool life is more often terminated by one of the last three —

abrasion, attrition or a sliding wear process - or by fracture.

%///

////// CHIP
‘ /

" BU ILT- UP EDGE

TOOL

42

4.2. Results of Electron Microscopic Investigations

To monitor the progress of tool wear as a function of cutting
time, the tool tip was examined in a scanning electron microscope
(SEM). For external examination the SEM is particularly valuable
because of its great depth of focus and it was convenient that the

entire tool fit into the stage of the microscope.

At first the tool tip remained sound and unharmed. Figure 17
shows the rake face of the tool with the left edge being the side
cutting edge. There are no signs of visible wear and the grinding
marks are still visible. This picture was taken after 5 min. of
cutting. At the end of 10 min. of cutting time, one can detect signs
of sliding wear as shown in Figure 18. In this picture of the rake
face, the left edge is the side cutting edge and the right edge is
the end cutting edge. An indentation due to crossing of the chip
across the side cutting edge was forming as shown in the lower left
portion of the picture. The frictional rubbing marks between the tool
and chip just before the chip leaves the tool face are visible in the
lower right quadrant of the picture. These marks also outline the in-
termittent contact area of the tool/chip interface. Some work material,
including a piece of a chip, is shown adhering to the tool along the
edge. This picture was taken after 10 min. of cutting. This wear pat-“
tern remained stable for a large portion of the remaining tool life.
Wear progressed slowly and steadily. Figures 19,20,21 and 22 show
the increase in the size of sliding wear marks. This is very evident

in Figure 19.

43

 

 

Figue 17. Tool Tip After 5 min. of Cutting

 

 

Figure 18. Tool Tip After 10 min. of Cutting

44

 

 

 

 

Figure 20. Tool Tip After 20 min. of Cutting (blow-up of Figure 19)

4S

 

Figure 21. Built—up Edge Adhering to the Rake Face of the Tool

 

Figure 22. Side Cutting Edge After 24 min. of Cutting

 

46

The indentation on the side cutting edge as shown in Figure 19
also extends down the flank face which can be seen in Figure 21. Figure
22 depicts the edge of the tool close to the nose. It is evident that
the edge is rounded and chipped in the lower portion. Although it is
difficult to exactly classify the type of wear occuring, a combination
of abrasive and attrition wear appears likely. These photographs
(Figures 19,20,21,22) which were taken after 20 min. to 25 min. cut-
ting times, also display the phenomenon of seizure between work mate-
rial and the tool. Some degree of metallurgical bonding at the tool/
chip interface might have taken place since the contact remained
intact during the disengaging of the tool from the workpiece at the
end of the cutting pass. There is, however, a considerable variation
in the strength of bond generated, since not all pictures display the
cutting tool with.work material adhering to it. Figures 19,20 and 22
show the ductile tensile fracture of the work material adhering to the
tool face. This occured when the tool was seperated from the chip in
a tensile fracture mode when the tool was disengaged from the work-
piece. As pointed out earlier in the discussion of cutting force mag-
nitudes, the radial force against the nose of the tool did not have a
significant effect as is visible in Figures 19 and 20. The nose is
still sharp and intact. This is evidenced by the fact that the grin—
ding marks on the flank and end relief face can be followed almost up
to the nose (Figure 21). Another important feature of metal cutting
operations, the built-up edge, can also be observed in Figure 21
(see also Figure 16). The built-up edge forms under conditions of
seizure when work hardened material, adhering around the cutting

edge and along the rake face, accumulates to displace the chip from

47

immediate direct contact with the tool. The built-up edge is not a
seperate body of metal but forms a continuous body of metal with the
chip. Its presence transfers the flow zone, where the relative motion
between the chip and tool occurs, to the top of itself. The built-up
edge is sheared off and is not observable at higher cutting speeds (5,

6,13,15).

Due to sliding wear the tool material is weakened progressively
causing abrasive and attrition wear to take over. Figures 23 and 24
show this effect. The weakening of the nose area in particular is
very evident. The rake face is no more smooth, but chipped and cracked
due to attrition and abrasion. These pictures were taken after 26 min.
of cutting time. Shortly after these pictures were taken the nose of
the cutting tool collapsed under machining. Figures 25 and 26 show
the tip of the tool with the nose broken off. The effects of attrition
on the tool can be seen in Figures 27,28 and 29. Large fragments are
being pulled away, leaving a very uneven worn surface. The tool edge

is essentially being 'nibbled' away.

48

 

 

 

Figure 23. Rake Face After 26 min. of Cutting

 

 

Figure 24. Rake Face After 26 min. of Cutting (blow-up of Figure 23)

49

 

 

Figure 25. Rake Face After 28 min. of Cutting

 

 

Figure 26. Rake Face After 28 min. of Cutting (blow—up of Figure 25)

50

 

 

Figure 27. Rake Face After 28 min. of Cutting (blow—up of Figure 26)

 

 

Figure 28. Side Cutting Edge After 29 min. of Cutting

51

 

 

Figure 29. Rake Face After 30 min. of Cutting

5. CORRELATION OF DATA AND SUMMARY OF RESULTS

The results of this study demonstrate that the stress is normally
at a maximum near the cutting edge. Even though no exact numerical
values for the compressive stress acting within the tool/chip contact
area has been determined, the evidence obtained through the finite
element analysis illustrates that they are at least high enough to sig-
nificantly deform the tool tip and side cutting edge. The extent of
this deformation is very likely to be larger with the softening effect
the temperatures on the tool material. Although classified as a sliding
wear pattern, the cratering effect in Figure 23 can very well be
caused by the combined effect of sliding, high compressive stresses
and fatigue on the rake surface. High temperatures would only accele-
rate this mechanism. Considering the interrupted nature of turning ex-
periments, a low cycle fatigue loading of the tool might have taken
place. The collapse of the tool tip towards the end of the tool life,
the crumbling effect seen in Figure 28 and the crack in the destroyed
tool in Figure 29 have, besides the discussed mechanisms, very likely

been caused by the continuously increasing stresses and fatigue.

The difficulty inherent to stress calculations at the tool tip
is in the determination of the tool/chip contact area. The stress is
obtained by dividing the tangential cutting force by this area. This
calculation gives only very rough results owing to the fact that the
area of the chip/tool interface is inherently impossible to determine.
The size of this interface is dependent on what goes on in the flow
zone which is not well understood yet, since it cannot be observed

52

53

directly and is governed by several parameters. These experiments,
through electron microscopic examination of tool wear, also showed
that the seizure between the work material and tool is the normal

condition to be expected.

The dependence of tool wear on cutting time and cutting speed
gas been one of the major areas of research starting with Taylor's
tool life tests around the turn of the century. It is well established
that tool life decreases with increasing cutting time and cutting
velocity in a progrssive exponential manner. The results of the study

are in line with this existing knowledge.

Although the effects of temperature on wear have been excluded
from consideration, it was expected that they would be limited under
the cutting conditions selected. This assumption was confirmed by the
types of wear observed, i.e. abrasion, attrition, wear under sliding

conditions, which characteristically occur at low temperatures.

6.DISCUSSION

This study set out to investigate the feasibilty of correlating
tool wear to cutting forces and stresses. The idea was to understand
the underlying difficulties as well as to discover hidden potentials
with a long term goal of on—line monitoring of tool wear on the pro-
duction level. Much effort has been and is being spent to increase the
productivity on the shop floor. The recent developments in Numerical
Control (NC), Direct Numerical Control (DNC), and Computer Numerical
Control (CNC), although enormous in their positive impact on pro-
ductivity, can only reduce the non-productive time in manufacturing,
i.e. workpiece handling, setup of the job, lead times, tool changes
and operator delays. Although NC has a significant effect on downtime,
it can do relatively little to reduce the in-process time compared to
a conventional machine tool. The most promising answer in reducing the
in-process time lies in the use of Adaptive Control (AC), which deter-
mines the proper speeds and/or feeds during machining as a function
of variations in such factors as workpiece hardness, cutting depth,
air gaps in part geometry, tool wear, and so on. This is a control
system that measures certain output process variables such as spindle
deflection, force, torque, temperature, vibration amplitude, horse

power and tool wear (5,16).

Measurement of forces and wear was the objective of this research.
Although this was accomplished by indirect methods, it gave much

insight into the problems that lay ahead. Much remains to be learned

54

55

about the flow zone at the tool/chip interface, the behavior of
materials involved at high strain rates and temperatures, the nature
and size of the contact area between the tool and the chip and the

accurate modeling of the tool tip.

10.

11.

12.

13.

14.

15.

LIST OF REFERENCES

. Oberg, E., Jones, D. F., Horton, H., L., Machinery's Handbook,

 

Industrial Press, Inc., let ed., 1980.

. Metal Cutting Tool Handbook, Metal Cutting Tool Institute, 1954,

 

Hildreth Press.

. Machining Data Handbook, Metcut Research Associates, Inc.,

 

Cincinnati, Ohio, 2nd ed., 1972.

. "Life Tests on Single-Point Tools", American Standards Association,

ASA 85.19-1946, ASA B5-13-1939.

. Trent, E., M., Metal Cutting, Butterworth & Co., 1977.

 

. Childs, T. H., C., Richings, D., Wilcox, A., B., "Metal Cutting:

Mechanics, Surface Physics and Metallurgy", International Journal
of Mechanical Sciences, Vol. 14, 1972, pg. 359/375.

 

 

. Usui, E., Hirota, A., Masuko, M., "Analytical Prediction of 3-D

Cutting Process", Transactions of the ASME, Vol. 100, 1978, pg. 222.

 

. ANSYS, Engineering Analysis System, User's Manual, Swanson Analysis

 

Systems, Inc., Houston, Pennsylvania.

. Zorev, N., N., International Research in Production Engineering,

 

42, 1963, Pittsburgh.

Amini, E., Journal of Strain Analysis, 3, 206, 1968.

 

Zakaria, A., A., ElGomayel, J., I., "On the Reliability of Cutting
Temperature for Monitoring Tool Wear", International Journal of
Machine Tool Desing,Vol. 15, 1975, pg. 195/208.

 

 

Tseng, M., M., Noujaim, R., A., "On the Measurement and Propagation
of Flank Wear in Cutting Tools", Journal of Engineeringgfor Industry

 

1979, vol. 101, pg. 109/115.

Wright, P., K., McCormick, S., P., Miller, T., R., "Effect of Rake
Face Design on Cutting Tool Temperature Distributions", Journal of
Engineering for Industry, 1980, Vol. 102, pg. 123/128.

 

 

Bhattacharyya, A., Ham, 1., "Analysis of Tool Wear, Part 1:
Theoretical Models of Flank Wear", Journal of Engineering for Indus—
try, 1969, pg. 790/798.

 

Opitz, H., Konig, H., "On the Wear of Cutting Tools", Advances in
Machine Tool Design and Research, 1967, pg. 173/200.

 

 

56

57

16. Groover, M., P., Automation, Production Systems and Computer-Aided

 

Manufacturing, Prentice Hall, Inc., 1980.

 

“III‘IIILIIIIIIIIIIIIIIIIIIIIIII