Astakhov V. Tribology of Metal Cutting

Подождите немного. Документ загружается.

Tribology of the Tool–Chip and Tool–Workpiece Interfaces 167

200 120 40 0

(b)

(a)

20 0 60 80

(W/mK)

(°C)

Fig. 3.31. Inﬂuence of mean contact temperature on the CCR for different tool materials: 1 –

beryllium copper UNSC17000, 2 – tungsten, 3 – high-speed steel M42, 4 – high manganese steel

and 5 – Ti Grade 1; work material – lead.

contact temperature towards the cutting edge. As such, the average contact temperature

decreases which, in turn, leads to the corresponding increase in the tool–chip contact

length. The inﬂuence of these two factors lowers the tool life when machining the steel

materials, i.e. when the contact stress and temperature are high. In the author’s opinion,

this is the major cause of very poor tool life of diamond tool in the machining of steels,

although practically all the known literature sources on metal cutting (for example, see

Ref. [37]) state that the prime cause for this is the transformation of diamond to a graphite

form and/or interaction between diamond and iron and the atmosphere. The latter sug-

gestion cannot explain very poor tool life of these tools in the machining of nickel alloys,

practically having no carbon and iron.

Mutual adhesion properties of the work and tool materials. Analyzing a great body of

research works done in the ﬁeld of the adhesion phenomenon that occurred in the contact

of two materials, Poletica concluded [69] that their results can hardly be applicable in

metal cutting because most of these were obtained using conditions that are not partic-

ularly for metal cutting. For example, the results obtained using pin-on-disc machines

(discussed in Chapter 6) cannot be applicable in metal cutting at all (see Chapter 6).

Poletica also concluded [69] that solubility of the work and tool materials is the major

contributing factor in their adhesion interactions. Therefore, the only proper way to assess

168 Tribology of Metal Cutting

Table 3.1. Properties of copper in terms of its ability to form solid-state solutions.

Material

(component)

Mutual solubility Formation of

intermetallic

compounds

Liquid state In solid state

In copper In material

(

wt.%/at.%

)(

wt.%/at.%

)

Aluminum Unlimited 9.4/19.6 5.7/2.5 Cu

and others

Tungsten Unlimited 9.4/19.6 Insoluble No

Iron Unlimited 9.4/19.6 8.5/7.5 No

Cobalt Unlimited 12.8/42.0 4.1/3.4 No

Manganese Unlimited 31/25 Unlimited No

Molybdenum Insoluble Insoluble Insoluble No

Nickel Unlimited Unlimited Unlimited No

Titanium Insoluble 4.3/5.6 2.1/1.6 Ti

Cu, TiCu

Chromium Limited Insoluble Insoluble No

Silicon Unlimited 4.5/10 Insoluble Cu

Si and others

Carbon Insoluble Insoluble Insoluble No

adhesion interactions between work and tool materials is to test a single work material

against several tool materials purposefully selected as having distinctive properties in

terms of their solubility of solid phases.

Copper was selected as the work material. This is because copper forms solid solution

with a number of relatively hard materials. In machining copper, the selection of tool

materials is not very restricted (in terms of their hot hardness and allowable temperature).

Copper has low hardness and high thermoconductivity. Table 3.1 illustrates the properties

of copper in terms of its ability to form solid-sate solutions.

Technically pure tungsten, titanium alloy Ti Grade 1 and beryllium copper were selected

as tool materials. The ﬁrst one was selected because it is totally neutral to copper.

Ti Grade 1 contains 90% of Ti, which has very limited solubility in copper and can form

chemical compounds with copper. Beryllium copper contains about 98% of copper and

thus should have the maximum afﬁnity. Single-point cutting tools made of Ti Grade 1

and beryllium copper were heat-treated to hardness HB360 and ground with rake angle

γ = 25

◦

. Tungsten tools were ground with γ = 10

◦

because of the high brittleness of

tungsten.

The test results shown in Figs. 3.32 and 3.33 indicate that the adhesion properties of the

tool material with respect to the work material have a marked inﬂuence on the tool–chip

contact length and CCR. As seen, the contact length increases 5.5 times and CCR –

almost 4 times when the beryllium copper tool is used when compared to the tungsten

tool.

Figures 3.34(a) and (b) show signiﬁcant inﬂuence of adhesion properties of the tool

material on the mean normal contact stress and on the ratio of the mean shear and

normal contact stresses. There are no data for the tungsten tool in these ﬁgures because

this tool has a different rake angle having its own inﬂuence on these parameters. As seen,

the mean shear stress in machining with the Ti Grade 1 tool is three times higher than

Tribology of the Tool–Chip and Tool–Workpiece Interfaces 169

1.4 2.8

n (m/s)

Beryllium copper

Ti Grade 1

Tungsten

(mm)

Fig. 3.32. The tool–chip contact length in machining copper with three selected tool materials

having considerably different adhesion properties. The depth of cut d

= 4 mm and feed f =

0.17 mm/rev.

40 80 120 n (m/min)

Tungsten

Ti Grade 1

Beryllium copper

Fig. 3.33. CCR in machining copper with three selected tool materials having considerably dif-

ferent adhesion properties. The depth of cut d

= 3 mm, feed f = 0.17 mm/rev, tool cutting

edge angle κ

= 70

◦

and inclination angle λ

= 0

◦

that of machining with the tool made of beryllium copper. As such, the ratio of the mean

shear and normal contact stresses is 2–2.5 times higher.

Although the contact length, mean normal contact stress and CCR in machining with

different tool materials vary signiﬁcantly as seen in Figs. 3.32–3.34, the mean shear

stress remains practically the same as seen in Fig. 3.35. As before (see explanations to

Fig. 3.23), it is explained by the fact that this stress depends only on the properties of

the work material.

When machining the work material with a tool having different adhesion properties, the

inﬂuence of tool material would be different. Figure 3.36 shows the results of a test

170 Tribology of Metal Cutting

100

(MPa)

(a)

160 n (m/min)80

Beryllium copper

Ti Grade 1

(b)

Fig. 3.34. Inﬂuence of cutting speed in machining copper: (a) on the mean normal contact stress

at the tool–chip interface, (b) on the ratio of the mean shear to the mean normal contact stresses.

Rake angle γ = 25

◦

, depth of cut d

= 3 mm, feed f = (0.07–0.3) mm/rev, tool cutting edge

angle κ

= 70

◦

and inclination angle λ

= 0

◦

160 80

100

Beryllium copper

- s = 0.07 mm/rev

Ti Grade 1

s = 0.17 mm/rev

- s = 0.30 mm/rev

- s = 0.07 mm/rev

- s = 0.17 mm/rev

- s = 0.30 mm/rev

200

300

n (m/min)

(MPa)

Fig. 3.35. Inﬂuence of cutting speed on the mean shear stress in machining copper.

where CCR for aluminum were determined as a function of the cutting speed using

the beryllium copper and steel tools. Iron is much more chemically active to aluminum

than copper. As a result, higher CCRs were obtained for the steel tool compared to the

beryllium copper tool although the thermoconductivity of the beryllium copper tool is

much higher than that of the steel tool.

The above-discussed test results illustrate the mutual action of thermoconductivity and

adhesion properties. To separate these two issues, the tools used in these experiments

were plated with a very thin layer of chromium. This layer could not change the

Tribology of the Tool–Chip and Tool–Workpiece Interfaces 171

160 320 n(m/min)

Steel AISI 07

Beryllium copper

Fig. 3.36. Inﬂuence of cutting speed on the chip compression ratio in machining aluminum for

two different tool materials. Rake angle γ = 20

◦

, depth of cut d

= 2 mm, feed f = 0.21 mm/rev

and tool cutting edge angle κ

= 45

◦

thermoconductivity signiﬁcantly while it signiﬁcantly affected the adhesion properties

at the tool–chip contact. The test results are shown in Fig. 3.37. As seen, CCR for the

plated beryllium copper tool is much smaller than that for the non-plated one. In the case

of titanium tool, the difference is rather small. For the tungsten tool, the result was the

opposite and the difference is signiﬁcant.

Inﬂuence of mechanical properties of the work material. It is much easier to study the

inﬂuence of mechanical properties of work material on the tribological conditions at the

tool–chip interface than that of the chemical composition of the work and tool materials.

The mechanical properties of the work material can be altered in a wide range by its

heat treatment. Poletica proposed [69] the use of beryllium copper as the work material.

As mentioned above, it is an excellent test material because its mechanical properties

can be changed over a wide range by heat treatment while the phase composition and

microstructural parameters remain practically unchanged.

The test results with beryllium copper are shown in Figs. 3.18, 3.19(b), 3.23(b), 3.27

and 3.29. Figure 3.38 shows the contact characteristics obtained under invariable average

contact temperature as functions of the true stress at the fracture of the work material

represented by the hardness of the work material [69,81].

As it follows from Fig. 3.38, the mean normal stress, σ

increases with the hardness of

the work material. The mean shear stress τ

also increases in proportion to the hardness.

The latter result, obtained under special test conditions, is part of a more general depen-

dence shown in Fig. 3.25 where this stress is correlated with the strength of various work

materials. Therefore, the results shown in Figs. 3.38 and 3.25 complement each other.

172 Tribology of Metal Cutting

80 160 n (m/min)

2 - Ti Grade 1

1 - Beryllium copper

3 - Tungsten

Non-plated

Cr-plated

Fig. 3.37. Inﬂuence of cutting speed on CCR in machining copper with the non-plated and

Cr-plated tools. Rake angle γ = 20

◦

, depth of cut d

= 2 mm, feed f = 0.15 mm/rev and

tool cutting edge angle κ

= 45

◦

800 1000 1200 1400

(MPa)

Contact

Stress

(MPa)

1200

800

400

Fig. 3.38. Contact characteristics in machining beryllium copper as functions or its hardness

represented by the true stress at fracture. Rake angle γ = 10

◦

, tool cutting edge angle κ

= 45

◦

feed f = 0.15 mm/rev and cutting speed ν = 1.67 m/s.

Tribology of the Tool–Chip and Tool–Workpiece Interfaces 173

80 160

n(m/min)

Fig. 3.39. Inﬂuence of cutting speed on CCR in machining of various materials: 1 – steel AISI

1010, 2 – copper, 3 – aluminum 2014, 4 – lead, 5 – Armco iron, 6 – forging brass UNS C37700,

7 – steel ASTM A514 (0.18% C, 1.1% Cr, 1.3% Ni), 8 – steel AISI 07, 9 –Ti Grade 1 and 10 –

cadmium.

It follows from the results shown in Fig. 3.38 that the apparent friction coefﬁcient at

the tool–chip interface calculated as the ratio of the shear and normal contact stresses,

depends primarily on the hardness of the work material and, secondarily, on other prop-

erties of the work and tool materials and the machining regime. Figure 3.39 shows

the experimental results for various work materials. These materials differ not only in

mechanical properties but also in many other physical and chemical properties. As such,

only the mean shear stress would depend on the strength of the work material while

other tribological characteristics at the tool–chip interface can vary in wide ranges. As

such, CCR would depend only on the mutual adhesion properties of the work and tool

materials. As seen in Fig 3.39, CCR is practically the same in machining work materials

having considerably different mechanical properties such as cadmium and titanium alloy.

At the same time, CCRs for steel AISI 07 and Ti Grade 1 are different although these

work materials have approximately the same mechanical properties.

Inﬂuence of surrounding medium. The cutting process can take place in different media.

The most common media are gases and liquids. Although inﬂuence of medium is often

treated as the inﬂuence of cutting ﬂuid (as discussed in Chapter 6), one important aspect

of the inﬂuence of the surrounding medium on the tribological conditions at the tool–chip

interface, namely the inﬂuence of atmospheric air is discussed here. While cutting in the

presence of air, one might expect the occurrence of certain chemical reactions of oxygen

and nitrogen contained in the air with freshly formed surfaces of the work and tool

materials. It is a strong belief among the metal cutting specialists that the formation of

174 Tribology of Metal Cutting

oxidized, nitrogenized and oxi-nitrogenized ﬁlms and water vapor on the contact surfaces

are the results of such reactions.

To clarify this long-standing issue, a series of tests were carried out in air and under inert

gas using a specially designed pressure chamber [69] installed on a lathe. The chamber

was ﬁlled with argon with a slightly higher (by 0.05 MPa) pressure than the atmospheric

pressure. The actual pressure of argon in the chamber was controlled by a manometer.

Two work materials were selected for the test, namely Armco iron and copper since they

were having very high sensitivity to the properties of the medium in cutting. Moreover,

copper has an extremely high sensitivity to oxygen that should be very helpful in detecting

any inﬂuence of oxygen in the air on the cutting process.

The test results are shown in Fig. 3.40. No difference in CCR and thus in the tool–chip

contact length was found. The only difference that was observed is the absence of temper

colors on the chip obtained in cutting with argon.

Although it is incorrect to extend the obtained result for all kinds of work materials, par-

ticularly in the case of machining some titanium alloys as pointed out by Poletica [69],

it allows the conclusion that disproves the above-mentioned common belief on the

inﬂuence of oxygen in metal cutting.

60 100

n(m/min)

Air

Argon

Copper

Armco iron

Fig. 3.40. Inﬂuence of the medium on CCR. Cutting feed f = 0.10 mm/rev and depth of cut,

= 3 mm.

Tribology of the Tool–Chip and Tool–Workpiece Interfaces 175

3.2.5 Generalizations

The results of the theoretical and experimental studies on the tribological conditions at

the tool–chip interface can be summarized as follows:

• The concept of the coefﬁcient of friction is inadequate to characterize the sliding

between the chip and tool and thus should not be used in metal cutting studies.

The known modeling of metal cutting where a constant friction coefﬁcient is used

(practically all commercial software packages) cannot be considered as adequate to

any real cutting process. Instead, contact stresses at the tool–chip interface and their

dependence on the parameters of the cutting process should be considered.

• There is a great discrepancy in the reported theoretical and experimental results

on the distribution of the normal and shear stresses over the tool–chip interface.

The most recently reported uniform distribution of the normal and shear contact

stresses is in direct contradiction with the known experimental results and everyday

practice of metal cutting.

• The tool–chip interface can be conditionally divided into two approximately equal

distinctive parts: plastic and elastic.

• Because the chip formation process is of a cyclic nature, the normal and shear

stresses as well as the shape of their distributions vary within each cycle of the chip

formation. Therefore, the mean normal and shear stresses at the tool–chip interface

are of prime interest in metal cutting. These parameters should be considered in the

process analysis and optimization.

• The length of the tool–chip interface depends both on the uncut chip thickness

and rake angle. The dependence of the contact length on the cutting speed directly

resembles the dependence of CCR on the cutting speed.

• The contact length is directly proportional to the uncut chip thickness for different

work materials. One of the most important tribological characteristics at the tool–

chip interface is the ratio of the contact length to the uncut chip thickness referred

as the Po-criterion. This criterion remains invariant to changes in the mechanical

and physical properties of the work material.

• There is a direct correlation between the Po-criterion and CCR. This correlation is

expressed by Eq. (3.42). Moreover, it represents the condition of static equilibrium

of the chip.

• The normal and shear stresses at the restricted rake face is always higher than those

with the natural contact length. The discussed increase can be readily determined

by the condition of static equilibrium.

• The mean shear stress at the tool–chip interface is a function of properties of the

work material. It does not depend on the tool geometry, tool material and cutting

feed. Among other factors, the most surprising, and seemingly paradoxical, is

the independence of this stress on the mean contact temperature at the tool–chip

interface. The mean shear stress is well correlated with the ultimate tensile

strength of the work material. The listed unique properties of the mean shear

176 Tribology of Metal Cutting

stress should be considered the second (next to CCR) stable important tribological

characteristic of the tool–chip interface. Knowing CCR and the mean shear stress

at the tool–chip interface, one can easily calculate 80–90% of the energy spent in

the cutting system. Moreover, because the mean shear contact stress determines to

a large extent the temperature at the tool–chip contact, it can be stated that this

temperature is solely a function of the cutting speed and the work material.

• The mean normal stress at the tool–chip interface is very sensitive to many parame-

ters of the cutting process. Being work material speciﬁc, this stress depends mainly

on the cutting speed, cutting feed and cutting tool rake angle. This stress increases

with the cutting speed for a wide range of metallic work materials and decreases

with the rake angle. The mean contact stress was found to be a function (and a char-

acteristic) of the state of stress in a contact zone. It depends on the Po-criterion in the

same way as this criterion affects the state of stress in the deformation zone. More-

over, for a wide range of work materials, the mean normal stress at the tool–chip

interface is uniquely related to the Po-criterion.

• Among many properties of the tool material, the greatest inﬂuence on the con-

tact conditions at the tool–chip interface is its thermoconductivity and adhesion

properties while the inﬂuence of the elastic constants of the tool material are small.

• The inﬂuence of thermoconductivity on the tribological conditions at the tool–chip

interface manifests in two ways. First, it affects the mean contact temperature, and

second, – it affects the temperature distribution over this interface.

• The CCR and thus the amount of work of plastic deformation in metal cutting

increases when the thermoconductivity of the tool material increases. When the

thermoconductivity of the tool material increases, the average temperature at the

tool–chip interface increases. Therefore, it can be suggested that the relationships

shown in Figs. 3.30(a) and (b) represent the dependence of CCR on the average

contact temperature. The greater the thermoconductivity of the tool material, the

higher is the CCR.

• Any change in the average contact temperature leads to the corresponding change in

the contact pressure at the tool–chip interface. Moreover, this pressure also changes

when the distribution of the contact temperature changes. The thermoconductivity

of the tool materials has the greatest inﬂuence on the distribution of the contact

temperature.

• The adhesion conditions at the tool–chip interface affect the cutting process in terms

of energy spent much more than considered today in the literature on metal cutting.

As such, the solubility of the work and tool materials is the major contributing factor

in their adhesion interactions.

• The mutual adhesion properties of the work and tool materials should be carefully

considered in the selection of the tool material for a given application. A small

alternation in the chemical composition of the work material (often the case in

the automotive industry) can change these properties signiﬁcantly resulting in a

signiﬁcant variation in the energy spent in cutting and thus in tool life.

• The same coating material applied for a different substrate can cause signiﬁcant

decrease or increase not only in the energy spent at the tool–chip interface, but also