Smith G.T. Cutting Tool Technology: Industrial Handbook

Подождите немного. Документ загружается.

sizes of 25 µm. is range includes four-ute square-

ended cutters (i.e. similar to the one depicted in Fig.

55 – bottom le), together with ball-nose end mills,

plus some customised micro-mill tooling.

Complementary to these micro-milling cutters, are

their respective toolholders

, which must hold and

securely contain the tool’s shank in an accurate and

precise manner. Any form of tool runout when held

in the toolholder must be kept to absolute minimum.

e toolholder’s importance in the micro-milling op-

eration is oen overlooked, at the user’s peril! By way

of illustrating this fact, if one has a two-ute φ0

.5 mm

end mill then the chip-load will be <0.010 mm tooth

–1

So, if the micro-mill has a runout of 5 µm, the cutter

is only utilising one of its utes – by a factor of 100%.

is micro-mill’s runout condition, leads to cutter in-

stability coupled with a poor milled surface situation,

with the potential for either reduced tool life, or break-

age. Micro-milling tool deection and its subsequent

breakage when utilised for micro-machining opera-

tions, are principally caused by three main factors,

these are:

1. M

icro-machining creates a substantial increase in

the specic energy, as the chip thickness decreases

– meaning that here, as the chip gets thinner with

smaller D

’s, the micro-mill is subject to greater

resistance, when compared to that of ‘macro-mill-

ing’. Moreover, it is almost as if the workpiece mate-

rial becomes harder during micro-machining. is

resistance force to machining here, is strong enough

to exceed the bending strength of tool – even prior

to any wear occurring, leading to tool breakage.

A method of minimising tool breakage and prevent-

ing its occurrence, is to ensure that the chip thick-

ness is smaller than the radius of the tool’s edge,

2. D

uring micro-milling, a sharp rise in the cutting

forces and stresses resulting from chip-clogging

may cause tool breakage – when say, utilising a two-

ute cutter, each cutting edge removes chips from

the machining vicinity by only a half rotation of the

tool. Likewise, if chip-clogging occurs – within a

few micro-tool revolutions, the cutting forces and

66 ‘Micro-milling/-drilling toolholders’ , are typically manufac-

tured with taper-/face-tments of the : ISO 15 to 30, or HSK-

E25 to 32 types. A typical range of micro-toolholders, might

cover tool shank sizes in a range from: φ0.5 mm to φ2 mm, in

0.01 mm increments. With ‘matching micro-tooling’ , to typi-

cal tolerances of: <+0/-4 µm, with these micro-tools typically

being, for example, rated @ 40,000 rev min

–1

, having a runout

of <3 µm @ 4xD.

bending stresses increase beyond the limit of the

tool’s bending strength, thus causing it to break. A

possible solution to this problem, is to utilise alter-

native tooling materials such as micro-grained M-2

HSS, as they are more exible and as such, they can

tolerate any likelihood of chip-clogging in a more

compliant manner than their cemented carbide

counterparts,

3. W

hile micro-machining very ductile workpieces,

micro-mills can lose cutting eciency as a result of

BUE – this results in increased lateral pressure (i.e.

as feeding occurs) on the micro-mill, causing it to

slightly deect. is increasing tool deection due

to the presence of BUE, increases the stress gener-

ated with every cutter rotation, quickly causing the

micro-tool to break. is well-known BUE phe-

nomena is termed: ‘extensive stress-related break-

age’ ,

which could possibly be minimised by adopt-

ing a somewhat more ecient and benecial cutting

uid lubrication strategy.

NB D

ue to these (above) micro-tooling related

phenomena, many of the latest micro-milling ma-

chines are equipped with sensors to dynamically-

measure and monitor the cutting forces acting on

the micro-mills.

Micro-Boring Tools and Internal

Machining Operations

e production problems of drilling small holes in

workpieces is a big challenge, but this is nothing com-

pared to that of the technological complexity of bor-

ing very minute holes and other internal features in

components. Some tooling manufacturers oer insert-

style tools that are specically-designed to bore-out

small hole diameters, even down to just ≈φ0

.3 mm.

A typical micro-boring tool is illustrated in Fig. 255c

(i.e. ≈φ0

.3 mm), where the main features of the tool-

ing are explained. In the enlarged diagram of a mi-

cro-boring insert geometry schematically illustrated

n Fig. 255 – middle right (i.e. for a ≥φ0

.7 mm boring

insert). is particular micro-boring insert geometry,

is designed for both boring and proling operations

into holes of ≥φ0

.7 mm, although the clearance (‘t

max

’)

will only cope with shallow prole-depth features, be-

fore fouling on the tool’s shank. Micro-tool stiness is

quite high and occurs due to the enlarged shank di-

ameter, allowing reasonably large L/D ratios of >10:1

to be bored – which is remarkable, considering the

minute size of these insert’s. Other micro-machin-

Machining and Monitoring Strategies 523

ing

operations that can be undertaken include (Fig.

255 –

middle-right): grooving; threading; face-groov-

ing; back-boring (i.e. not depicted).

Some important micro-machining factors need to be

addressed, prior to boring-out previously drilled holes

with these micro-boring tools (Fig. 255c), such as:

•

Setting the micro-boring tool at the hole’s centre

height (

Fig. 255d – top) – this is the most impor-

tant preliminary step when about to commence a

micro-boring operation. A micro-boring tool that

is incorrectly set below the hole’s centre-height (Fig.

55d –

bottom) – adversely aects its performance

in several ways. It could foul on the curvature at the

bottom of the pre-drilled hole, through a reduced

edge clearance angle (primary relief). Moreover,

the ‘tool fouling’ causes the insert to rub against

the hole impeding the cutting action, which in

turn, creates vibration causing the insert’s tip to be

‘driven-down’ still further below centre. As a conse-

quence, the tip is forced deeper into the workpiece

material – due to the radial sweep of the bore. us,

as the top rake angle is increased – relative to that of

the workpiece, the clearance angle is reduced. is

geometric change in the micro-boring tool’s geom-

etry, causes the insert to ‘snatch, or grab’ material

rather than cut it, which then increases vibration,

so tool breakage will shortly ensue,

NB D

ue to the minute dimensions of these micro-

tools

, it is very dicult to set the tool exactly on

67 ‘Micro-machining’ , has been dened according to a dierent

approach, namely, concerning the actual workpiece’s volume,

as follows: ‘It is the [workpiece] size in which the work envelope

is smaller than 490 cm

’. (Source: Destefani, 2005)

68 ‘Meso-scale machining’*, this term has been coined and has

been dened as: ‘Millimetre-sized parts, with micron-sized fea-

tures’. For example, minute component features that can have

icron-sized tolerances (<1 µm). Recently in the USA, these

‘meso-machining technologies’ have included both micro-

milling and -turning, with in the former case, utilising micro-

end mills of ≈φ20 µm, while in the latter case, using micro-

turning tools of 10 µm in width. Hence, these technologies can

machine part features in the 25 µm range.(Source: Kennedy,

2006)

*e term: ‘meso-’ meaning: ‘middle’ , or ‘intermediate’.

(Source: Concise Oxford Dictionary)

NB In this case, the so-called ‘meso-machining operations’ ,

refer to machining in between the:nano- and micro-machin-

ing ranges.

centre-height – the ‘ideal positioning’. erefore,

the tool can be set marginally higher than ‘true’ cen-

tre-height, which increases the angular clearance –

relative to the hole, thereby allowing a freer cutting

condition. Further, if any potential vibrations occur

the micro-tool is both deected downward toward

the centre and (radially) slightly out-of-cut, some-

what reducing this ‘grabbing tendency’.

•

Choosing the right speed and feed – when the bore

is <φ6 m

m, then ‘standard’ speeds and feeds cannot

be used. For example, when boring say, a φ1 m

hole, one would expect to utilise a cutting speed of

erhaps ≈140 m min

–1

, which equates to a spindle

speed of ≈44,600 rev min

–1

– which is totally un-

realistic for most types of turning machine tool. If

any vibration occurred, then the micro-boring tool

would be immediately destroyed. So pragmatically,

f we limit the spindle speed to 6,000 rev min

–1

which would signicantly drop the cutting speed to

≈

19 m min

–1

, we would also need to complemen-

tarily reduce the micro-boring insert’s pressure by

reducing the D

to <0.1 mm which – to minimise

tool deection and potential breakage. Feedrates

or ‘macro-boring operations’ , are normally dic-

tated by the ‘close-relationship’ between its tool

nose radius and the bored-hole surface texture re-

quirements. Conversely, for micro-boring these are

not the controlling factors anymore. Here, minimi-

sation of cutting forces is vitally important by se-

lecting a feedrate t

hat should not exceed 0.125 mm

rev

–1

– which automatically overcomes any micro-

bored surface texture issues,

•

Ensuring adequate chip evacuation – is a real dif-

culty with such small bored holes, as little in the

way of unlled volumetric capacity exists with the

micro-boring tool situated inside the hole. So, with

the micro-boring tool inside the hole – accounting

for 60% of the available volumetric space, how can

the chips escape? By utilising a ‘through-insert-

coolant’ micro-boring insert (Fig 255c and d) with

the coolant under pressure, it can reach the cutting

edge. is coolant aids in both forcing and ush-

ing chips out of the bore’s mouth, which minimises

any of these work-hardened chips, creating a chip-

packing tendency in the bore, with the potential of

causing tool breakage,

•

Providing adequate tool stability and location –

these are important factors for micro-boring tooling.

Micro-boring inserts and their respective toolhold-

ers are designed for quick, simple and repeatable

524 Chapter 9

setups, with some tooling manufacturers design-

ing the tooling assembly to avoid inserts twisting

in-cut. ese design innovations range from: in-

sert-clamping ats; inserts having angle-ground

back-ends; to that of ‘teardrop-shaped inserts’ (i.e

see Fig. 255d) – this latter type rmly ‘wedging’ the

insert as it attempts twist – under the torque and

bending moments while boring.

If these micro-tooling factors are adhered to, then the

problems that are likely to be encountered when mi-

cro-machining are signicantly reduced, which means

that any form of micro-machining activities can be

achieved – with due diligence. So that a considerable

amount of micro-machining can be undertaken, ma-

chine tool companies have been developing a range of

specialised machines to cater for this market. Hence,

with the expansion of micro-machining activities this

being a somewhat ‘growth industry’

, let us briey

consider these specialised micro-machine tools and

the technical challenges they had to overcome in order

to cope with such minute cutters and invariably minis-

cule workpiece volumetric dimensions.

9.10.2 Micro-Machine Tools

CNC machine tools designed to machine parts, or

moulds that have small dimensional size, typically

with a linear dimension of ≤ 10 mm, or having detailed

part features of ≤ 0.1 mm, require some signicant ac-

curate and precision enhancements, if they are to cope

with the micro-machining demands of late. A typical

micro-machining machine tool will be mentioned,

‘high-lighting’ some of its important design features,

so that one can gain an insight into the careful at-

tention to detailing necessary for minute component

manufacturing.

One such machine tool produced by the Japanese

company Makino (not shown), has a ‘footprint’ of

≈ 1.8 m × 2.4 m, with a signicant weight of ≈ 5 tonnes,

with a worktable size of ≈ 300mm × 200 mm, having

three axis travel of: 200 mm in X-axis; 150 mm in Y-

69 As a ‘total aside and not related to the main topic’ and, for you

‘acionados’ of English language – isn’t this above statement

(i.e. in italics), the basis for a: ‘double’ – oxymoron?

** Oxymoron: expression with contradictory words – e.g.

‘Wise fool’ or, ‘Legal murder’.

axis; and 150 mm in Z-axis. Obviously for such minute

micropart features to be machined, the positional ac-

curacy and repeatability of the slideways are of crucial

importance, as such, the machine’s positional accuracy

is ± 0.3 µm, with a repeatability of ± 0.2 µm. is ma-

chine features unique workholding equipment, such

as a direct-chucking spindle, which has been designed

to eliminate the toolholder-induced variables, en-

abling miniature components to be produced, such as:

medical instruments; semi-conductor devices; optical

products; etc.

A major factor with any micro-machining machine

tool like the one mentioned above, is its machining

environment and more specically, its ‘thermal condi-

tions’

. As the minute part’s temperature changes along

with that of the machine’s spindle – during machining,

any dimensional modications on say, a ‘macro-scaled

part’ could normally be considered as negligible, but

on micro-sized workpieces, these linear variations be-

come signicant dimensional issues. In order to vir-

tually eliminate spindle growth the machine tool has

to be designed for a stable environment, when one is

attempting to hold ‘micrometre-accuracies’. Machine

tool features necessary in reducing this spindle/ma-

chine expansion, include, an automatic spindle lu-

bricant temperature controller – to reduce spindle

growth, coupled with the machine’s granite base – as

granite has only 10 to 20% of the thermal conductivity

70 ‘ermal conditions and eects’ , in particular will aect either

the machine tool’s linear expansion/contraction – depending

upon whether there is a temperature rise, or fall, respectively.

Simplistically and in this instance, ignoring any uncertainty

factors, then , ‘Coecient of thermal expansion’ which is nor-

mally denoted by the symbol ‘α’ , can be dened as: ‘A measure

of the change in length of a material subjected to a change in

temperature’. us:

α =

Change in length



(�T)

strain

(�T)

Where: L

= original length (mm), ∆T = change in temperature

(°C). Or, this thermal expansion equation can be alternatively

written, as follows: ∆L = (L) (Cα) (∆T) Where: ∆L = change

in length – by thermal expansion (mm), L = original length

(mm), Cα = coecient of thermal expansion, ∆T = change in

temperature (°C).

NB e uncertainty contributions here, may be combined, in

the following manner: u

= √(u

)

+ (u

)

Where: u

= ‘design-

stage’ uncertainty, u

= interpolation uncertainty, u

= instru-

ment/equipment uncertainty. (Source: Figliola and Beasley et

al., 2000)

Machining and Monitoring Strategies 525

of an equivalent cast iron structure, thus structurally-

minimising the eects of ambient temperature changes.

Moreover, for any form of extremely critical and preci-

sion miniature part manufacture, the entire machine

tool can be situated in a ‘thermal chamber’ ,

this in ef-

fect acts as a ‘controlled-temperature environment’.

Micro-tools of for example, φ5

0 µm – as has already

be mentioned, are almost impossible to see, let alone

attempting to set them to length. In order to facilitate

this minute tool setting operation – to sub-µm accu-

racy, the machine tool company developed a hybrid

automatic tool-length measuring system. is system

of tool length measurement is achieved by combining

a static low-pressure contact sensor in conjunction

with non-contacting sensing

, this being performed

while the tool is rotating at speed. Together, these two

sensing techniques permit sub-µm tool positioning

accuracy, during machining operations.

Obtaining the most advantageous micro-machine

ool for a particular type of minute ‘workpiece group’ ,

should be taken by considering the part’s features – in-

cluding its geometric conguration, together with the

level of accuracy and precision required and whether

these parts are produced as ‘one-os’ (i.e. as custom-

ised-specials), or in various batch sizes.

9.10.3 Nano-Machining

and Machine Tools

When attempting to machine components to toler-

ances of nanometric dimensions, the actual problems

considerably exacerbate, even when compared to that

of machining in the sub-micrometre range. Usually,

conventional cutting edges that have been honed, can-

not hope to cope with miniscule D

conditions, as

the edge is just simply not sharp enough and will tend

71 ‘Non-contacting tool sensing’ , is ahieved as follows: while

the spindle is rotating the tool’s tip position is measured via a

non-contact electro-magnetic sensor. is tool measurement

takes into account the thermal displacement of the tip, caused

by its rotation. us, the system’s contolller merges these two

measurements.

NB e tool’s length measurement operation occurs by mea-

suring the spindle growth and waiting until it stabilises – within

specic user-dened limits, such as for example, within2 µm.

Once the spindle is ‘stable’ , it can machine the workpiece at

the desired level of accuracy and precision.

to ‘plough’ , instead of cut. In order to machine such

components, oen on either very ductile workpiece

materials, or glasses, monolithic diamond tooling is

invariably utilised with the tool orientated so avoiding

its natural fracture planes. It is normal practice when

machining components to a few billionth parts of a

etre (i.e. 1 nm = 10

–9

m), that a wide range of ‘pro-

cess, environmental and machine tool inuences’ are

acknowledged and then subsequently minimised. In

fact, the Japanese Professor Nakazawa (1994) consid-

ered the machine tool’s inuences when machining at

high precision and, stated they could be broken down

into the three following requirements, mentioning

that the:

. Machine tool’s built-in reference must not vary,

2. Machine tool’s must follow this kinematic reference

at its highest precision,

. Machine’s movement must be accurately trans-

ferred to the workpiece.

Moreover, Nakazawa presented a useful table of the ma-

jor factors that disturb the relative tool-to-workpiece po-

sitions in ‘forced machining’ ,

as presented in Table 17.

In fact these ‘requirements’ are usually adequate

for the level of machining into the sub-micrometre

range, but once one proceeds to ‘ultra-high precision

machining’ operations within the nano-range, then

many other factors contribute to the overall success

of the cutting process (Fig. 256). As previously men-

Table 17. Factors that disturb relative tool-to-workpiece cut-

ting positions

Factors: Internal/

External:

Conditions:

Heat source Internal Machining energy at machining

point motor, ball leadscrew, bea-

ring, guide, hydraulics

External Convection (air conditioning)

Radiation (lighting, body heat)

Vibrations Internal Vibrations created by the ma-

chining mechanism motor, ball

leadscrew, joints, bearings, guides

External Other machines, moving vehicles,

people

Dust Internal Matter from the workpiece, or tool

External Dust in surrounding air, particles

in cuting uid

[Source: Nakazawa, 1994]

526 Chapter 9

Figure 256. The error sources from the processes; environment and machine tool; on workpiece accuracy. [Source: Breuck-

mann & Langenbeck, 1989]

Machining and Monitoring Strategies 527

tioned, ultra-precision machining does not usually en-

tail machining very diminutive components, but it is

normally concerned with holding exceptionally tight

tolerances on macro-sized parts (Fig. 257b). In fact, a

diamond turning machine tool that can manufacture

components to nano-tolerances is depicted in Fig.

257a. is ultra-precision CNC lathe is considered to

be the most accurate and precise machine of its type

currently available. erefore, it is worth discussing

the machine and its environment, as its installation

and operation encapsulates all of the error sources

shown in Fig. 256.

Nano-Machine Tool and its Facilities

In Fig. 257a, is shown a very expensive and highly

sophisticated machine tool, it was delivered to the

Atomic Weapons Establishment (Aldermaston, UK)

from the manufacturers Craneld Precision Engi-

neering (UK). Its intended purpose was to machine

Perspex optics and large thin-walled aspheric shells

aving form errors of <5 µm, together with wafer-thin

laser targets having micro-machined features, but held

o tolerances of ± 10 nm, by single-crystal (monolithic)

diamond tools – these diamond cutting edges being

orientated along their correct crystallographic plane,

allowing turned surface texture values (Ra)

of <1 nm

to be achieved.

If an ultra-precision machine tool is required to

work at nanometric resolution, then if it needs to be

located within a manufacturing plant, the machine

tool must have a special-purpose facility designed,

constructed and built to exceptionally-stringent

and specied requirements. Prior to discussing this

facility, it is worth describing some features of this a

machine tool, so that one can comprehend the signi-

cant technical problems that had to be overcome, in

order for it to achieve a ‘true’ nano-machining capabil-

ity.

e machine tool was constructed on a polymer

concrete base, that consisted of 8 tonnes of synthetic

granite, giving the desirable properties of: excellent

thermal stability; high stiness; plus good vibration

damping characteristics. Strangely for a diamond

turning machine, the headstock was equipped with a

hybrid hydraulic/air spindle rather than an air-bear-

ing design, because this spindle’s specially-designed

construction enabled an increased load-carrying ca-

pacity (57 kg), coupled to superior stiness. Spindle

peed range was from: 200 to 5,000 rev min

–1

. ree-

axes were tted, two linear axes – running on fully-

constrained hydrostatic dovetail bearings with linear

otor drives and one rotational axis. ese axes kine-

matics were: X-axis (520 mm); Z-axis (220 mm); B-

axis (360°) – as illustrated in Fig. 256-middle le. A la-

ser-positioning system was tted to the axes, having a

resolution of 1.25 nano-metres, incorporating a wave-

length tracker, to compensate for any environmental

changes to the air: temperature; pressure; humidity;

hydrocarbon content; etc. e CNC controller utilised

high resolution (1 nm) fast feed-forward operation,

thereby reducing servo-following errors, combined

with real-time axis compensation together with a

two-dimensional error compensation capability – for

straightness and orthogonality.

Tool-setting errors on the monolithic diamond

tooling, were minimised by a unique probe and optical

etting technique (Fig. 256 – middle-right), reducing

form errors, when utilising the rotational B-axis for

spherical/aspherical turned component geometry

e nanocentre machine tool, was housed in a tem-

perature-controlled environmental enclosure, consist-

ng of the temperature being held at: 20°C ± 0.5°C, en-

losed within 100 mm thick high density polyurethane

foam panelling – providing high thermal resistance,

with an air-tight seal. e positive-pressure air-supply

unit was situated outside – in an adjoining plant room,

ducted into the enclosure by dra-free ducts, with ten

platinum resistance temperature probes (i.e. resolu-

tion 0.001°C) strategically situated within the volume

space, continually monitoring the enclosure’s temper-

ature. Lighting from the uorescent lights had their

chokes removed from the enclosure, thereby reducing

the localised air temperature ‘stratication eects’ by

72 ‘Form errors’ , the principal causes of form error when spheri-

cal/aspherical diamond turning components, can be sum-

marised in the following manner:

Conical error – due to spindle misalignment,

Chevron and Ogive errors* – due to rst-order tool wear

and centring errors,Waviness error – resulting from sub-

strate vibration and tool prole error,

Astigmatism – created by the component xturing arrange-

ments and material stiness.

*Chevron error occurs when turning a convex spherical formed

component – due to the tool ‘over-shooting’ , while, an Ogive

error results from the tool stopping short – this latter error

creating an ‘ogival arch eect’ , with both these prole errors

being due to an incorrect tool centre height setup.(Sources:

Myler and Page, 1988; Wheeler, 2001)

–

528 Chapter 9

Figure 257. Nanometric machining to ultra-precise dimensional and surface texture characteristics. [Source:

Lamb & Gull, 1999]

Machining and Monitoring Strategies 529

over 1°C. e hydraulic and electronic cabinets were

temperature – controlled to ± 1°C. e ‘refractive index

of the air’ had to be corrected – based upon a modi-

cation to the Edlen (1966) equation, as the laser path

positional monitoring system would otherwise be af-

fected, with a correction factor being entered into the

CNC controller.

e ‘T-shaped base’ of the Nanocentre was sup-

ported on three pneumatic mounts that were ‘tuned’ to

liminate oor-borne vibrations of ≥ 2.5 Hz. Two types

of vibrational sources occur, namely forced- and self-

excited, with the forced vibrations originating from

external sources – through the foundations, while the

self-excited vibrations normally being the result of in-

ternal sources. A ‘oor vibration audit’ on the vibra-

tional inuences was conducted, to establish whether

the overall enclosure was suciently vibration absor-

bent. e oor vibration spectra gave typical vibra-

ional readings of 1 nm (rms) at frequencies of 25 Hz,

during the tests, with the external air compressors

mitting a oor borne 25 Hz frequency component,

which had to be subsequently nullied. Further testing

procedures were undertaken, including ‘modal analy-

sis’ and ‘thermal imaging’ of the machine’s structure,

together with a full calibration of the machine tool’s

kinematics.

Once all of these tests and various others had been

completed and compensated for, then a machining

testing program could then be undertaken. A typi-

cal test piece is illustrated in Fig. 257b, where an al-

uminium 6061-T6 part was heat treated and then

stabilised, of φ2

50 mm copper-plated (200 µm depth

coating) part. ese testpieces were faced-o with a

monolithic diamond tool – taking very shallow D

’s

of just a few micrometres, producing for example, a

ace-turned surface texture averaging ≈ 2.8 nm Ra.

Later, proling tests were also conducted, prior to -

nal operational acceptance by AWE, from the machine

tool builder.

Prior to completing this summation of the just

some of the rigorous testing procedures carried out to

ensure that the Nanocentre machine tool could oper-

ate within the nanometric range of ultra-high machin-

ing operation, it is worth making an unusual point

concerning human intervention at this exacting-level

of machining. It was found that when several person-

nel were within the machine tool enclosure while

machining took place, then the thermal output from

these people, inuenced the part’s dimensional size

– without actually contacting the machine, by simply

acting as a heat-emitting source

. Moreover, it was

also found that when diamond-turning

by facing-o

a very ductile testpiece similar to that depicted in Fig.

257b, when these people were in conversation during

the nanometric cutting of the part, their ‘voice signa-

tures’ – in the form of air-borne vibrations were ‘ma-

chined’ into the surface – in a similar manner to that of

an acrylic recording of a record in the past! erefore,

in order to ensure that both the human thermal eects

and the vibrational perturbations (i.e. by air-borne vi-

brations – talking), the personnel had to be removed

while any ultra-precision machining operations were

in progress.

Ultra-precision machining at these nano-metric

levels of operation, severely stretches today’s levels

of technological challenges for: machine tools, me-

trology, plant and equipment, as we approach that of

atomic-levels of precisional uncertainties

. It is not

just the case of purchasing an extremely accurate and

73 ‘Human body – as a heat source’. e average body – in a

‘sedentary state’ , will emit ≈100 W of heat. So, here in this

case, when there are two people present in the machine tool

enclosure, they will radiate ≈200 W of heat – inuencing

the machine’s and hence, the workpiece’s thermal expansion

– when machining at nanometric levels of accuracy and preci-

sion. (Internet source: Burruss, R.A.P., Virtual People, 2005)

74 ‘Monolithic diamond’ ,

has some of the following characteris-

tics: hardness of ≈8,000 Hv; Density (ρ) of 3,515 kg m

–3

; com-

pressive strength of 7,000 MPa; and a Young’s modulus (E) of

930 GPa.(Source: Cardarelli et al., 2000)

75 ‘Atomic radius’ ,

for example, for some typical elements, ranges

from that of: carbon, having an atomic radius of ≈0.071 nm

(i.e. its atomic ≈φ0.142 nm)*, iron’s atomic radius is ≈0.124 nm

(i.e. ≈φ0.248 nm, or ≈¼ nm)*, Aluminium’s atomic radius is

≈0.143 nm (i.e ≈φ0.286 nm, or >¼ nm), Cesium’s atomic radius

is ≈0.265 nm (i.e ≈φ0.530 nm, or >½ nm).(Source: Callister, Jr.

et al., 2003)*Slight digression here, may help explain why these

atomic radii are important, when certain elements are alloyed

together, such as iron and carbon, these being the main con-

stituents of plain carbon steel.When an allotropic change oc-

curs to the iron’s atomic lattice structure (i.e. BCC→FCC @

≈910°C), then the carbon being somewhat smaller, can t into

these (now) larger interstitial sites – voids – within the FCC

iron lattice – distorting the adjacent iron atoms. Upon rapid

cooling (e.g. by water quenching), some of the carbon is

‘trapped’ and severely distorts the structure as it attempts to

transform back to the original BCC form. Hence, this dis-

torted structure of iron-carbon – termed martensite, is both

a very hard, but brittle structure, requiring tempering: if it is

to act as a hardened and tempered workpiece material. is is

the basis (i.e. somewhat simplied), behind this iron-carbon

heat-treatment process.

530 Chapter 9

precise machine tool (Fig. 257a), and hoping to utilise

it to machine when approaching nano-metric resolu-

tion levels. ere are many oen interrelated factors

that have to be considered and then dealt with, if one

is to successfully operate at this ultra-precision level of

machining operations.

9.11 Machine Tool

Monitoring Techniques

Introduction

One of the most fundamental requirements for increas-

ing productivity of CNC machine tools, is the ability

to operate them ideally, in an untended manner, but

at the very least, minimally-manned – whether they

are ‘stand-alone’ machines, or part of a exible manu-

facturing cell, or system (FMC/S). So, if an untended

operation has been decided upon, then the absence

of an operator will create a considerable number of

problems that must be overcome, if the machining op-

erations are to be satisfactory. ese problems arise in

performing the monitoring and service functions that

are usually seen by the operator, who would normally

undertake the: monitoring of the cutting tool’s condi-

tion and its performance; replacing worn, or defective

tooling by interrupting the cutting cycle; assessing the

workpiece quality during machining; changing speeds

and feeds – if required; plus responding to unusual

conditions that are either seen, or heard, during the

cutting operation. While, in an unmanned machin-

ing environment, the associated monitoring systems

must provide the ‘articial intelligence’ (AI), necessary

to ‘mirror’ the experience gained by a fully-skilled op-

erator and their instinctive reactions and, to provide

the type of expertise usually associated with human

involvement. To cope with these every-day human-

intelligence activities and their subsequent interven-

tion during any machining operations, a considerable

number of monitoring systems have been developed.

In general, monitoring systems can be classied as:

process-monitoring; workpiece-monitoring; machine

tool monitoring; and tool-monitoring systems. Typical

applications of these monitoring systems for untended

operation on machine tools, include:

•

Monitoring the correct loading of the workpiece,

correcting any set-up misalignments, or datum o-

sets, while checking the quality of the workpiece,

•

Checking that the correct tools are available, by

identifying both the tools and their setting osets,

monitoring for tool wear and breakage and, initiat-

ing tool replacements – as necessary,

•

Adjusting speed and feed as appropriate and, com-

pensating for such eects as tool wear, thermal de-

formation and chip congestion,

•

Monitoring of machine elements, including the

CNC controller and taking any necessary correc-

tive action in response to: program failure; diag-

nostic error messages; etc.

Whatever the function that is to be monitored, there

is a need for some form of sensor to be incorporated

into the system – to detect any problems as they arise,

so that action can be appropriately taken, if necessary.

us, a sensor’s output – triggered by an error mes-

sage, must be processed to obtain the correct informa-

tion, allowing decision(s ) to be made. e machine’s

control unit, will then receive this ‘sensed’ result and

initiate controlled actions to either correct, or re-

cover the situation. Various types of monitoring and

sensing systems are currently available for machine

tools. Although, because this subject matter is so vast

and sophisticated, only several of these monitoring

techniques and sensing systems will now be consid-

ered.

9.11.1 Cutting Tool Condition

Monitoring

Introduction

Whenever an operator is present during a machining

operation, one of their major functions is to monitor

the tool’s condition while the cutting continues, where

they continually assure themselves that a tool in-cut

is performing productively. e tool-related monitor-

ing functions performed by an operator during any

component’s manufacture, may be classied into four

groups, briey these are:

1. Tool identication – this ensures that the correct

tool will be used for a specic operation, with a va-

riety of techniques being employed to achieve this

crucial tooling activity. Techniques include the use

of: touch-trigger probes (Fig. 133); non-contacting

probing methods (Fig. 134); ‘tagged’ tooling of the

contact (Figs. 116 and 117), or non-contact variet-

ies (not depicted),

Machining and Monitoring Strategies 531

2. Tool-oset measurement – of the cutting edge’s po-

sition is necessary, in relation to that of the part’s

datum point. is can be accomplished by the

‘probing-techniques’ and tool identication meth-

ods mentioned above,

. Tool life monitoring – is necessary to estimate the

extent of a worn tool’s condition, which must be

replaced prior to tool failure. e are a range of

sensing devices available and they can be classied

into two main groups: ‘Direct sensing’ – include:

radioactive techniques; measurement of electrical

resistance; optical observation of the wear zone;

easurement of workpiece dimensional changes;

or the distance between the workpiece and the

tool post, ‘Indirect sensing’ – based upon either:

temperature; sound; vibration; acoustic emission

and force. is latter method can be measured

and monitored either directly, by dynamometry

(Figs. 178–180, 237 and 244), or indirectly via mea-

surements of power, current, or torque – some of

these techniques will shortly be discussed,

. Tool breakage detection – can be monitored to en-

sure that the cutting edge does not fail in-cut, as

damage to both the tooling assembly and the work-

piece may occur as a result. Once again, a variety

of commercially-available techniques based upon

force-related signals are available, including: those

methods that use a dynamometer, either situated

on the tool block, or in say, a turning operation be-

low the tooling turret (Fig. 179); thrust-/feed-force

sensors (Fig. 258); spindle-bearing/motor-current

monitoring (not shown); power-/torque monitor-

ing (Fig. 259a). is latter technique (Fig. 259a),

is oen known as: ‘Torque-controlled machining’

(TCM).

NB I

n order to fully appreciate the complexity and

sophistication of any tool- condition monitoring,

on CNC machine tools, the following section has

been included.

Tool-Condition Monitoring –

With Feed-Force Sensors

Modern microprocessor-designed tool-monitoring

systems can be utilised for a variety of reasons, for

example, to monitor the tool’s condition, or to reduce

machining value-added costs. e advantages of using

monitoring detection, are

•

Tool wear is monitored and tool changes initiated

when necessary, so avoiding damage to the ma-

chine, or workpiece,

•

If breakage occurs, a signal will be immediately

produced to stop the machine tool – usually within

milliseconds,

•

e system detects if a tool, or workpiece is miss-

ing, thus eliminating wasted machine time and the

likelihood of unpredictable crashes.

While the cost advantages of using tool monitoring

are:

•

Tool life can be optimised, meaning that the tools

need to be only changed when they are worn – to a

specied amount (Figs. 174 and 176) and so reduce

the tool costs (Fig. 177e),

•

Down-time (i.e. here, it is normally associated with

unanticipated wear rates, or tool crashes) is re-

duced, which increases the machine’s output and as

a result, improving cycle-time and costs per part,

•

Repairs to the machine tool and cutting tools may

be reduced to a minimum, so the maintenance costs

are lower,

•

e machining operation is automatically moni-

tored, limiting any costly labour rates by subse-

quent operator involvement.

e above listed advantages for tool-condition moni-

toring are quite an impressive recommendation, but

how does it achieve consistent and accurate tool moni-

toring, while simultaneously controlling the cutting

process? ese questions will now be considered, deal-

ing in the rst instance, with how the system monitors

the tool’s performance during machining.

well known fact is that a tool will produce rela-

tively high loads during a cutting operation, as it

begins to wear. For eective ‘process monitoring’ it

is important that the signal utilised should vary in a

progressive manner as the tool wears and, not just at

the time that it actually breaks. It has been shown (Fig.

258b) that during a machining operation, the axial

force component (F

) provides a better indication of

the cutting edge’s condition as a function of tool wear,

than the torque value (M). us, the increase in the

axial force is more clearly dened – in both cases, from

hat of a sharp tool (Fig. 258b – le) to that of a worn

tool (Fig. 258b – right). is change in the force gener-

ated whilst cutting is instantly detected by a feed force

sensor (Fig. 258a). e sensor transforms the force

change into an electrical signal which is transmitted

532 Chapter 9