Mechanical Engineers Reference Book 12 Part 15 ppsx

Acoustic noise %/I39 adopted takes these factors into account and relates the measured sound to a reference level.. In this case the equation is dB sound intensity level = 10 loglo, - I

Vibration isolation and limits 15/137

Figure 15.172 Human sensitivity: Rezher-Meister scale (vertical vibration)

Octave pass band

Qne-third octave band centre frequency

IS0 vibration criteria for a person in a vertical position

At relatively low vibration levels cracks can occur in plaster

(particularly around windows) At higher levels, structural

members may crack and ultimately fail These two types of

damage may be easily attributed to mechanical vibration

Another type of damage may result from building settle-

ment caused by ground-borne vibrations compacting the

ground differentially beneath buildings This type of damage is

indistinguishable from settlement caused by other occur-

rences Specifications for maximum permissible vibrations

may be found in DIN standards which are given in terms of

maximum velocity (in mm s-l) which is allowable for different

classes of buildings from ruins and historical buildings up to

reinforced concrete structures More accurate criteria may be

found in the technical press and HMSO publications One

such type of drawing is shown in Figure 15.174

15.9 Acoustic noise

15.9.1 Introduction - basic acoustics

Sound can be defined as the sensation in the ear caused by

pressure variations in the air For a pressure variation to be

known as sound it must occur much more rapidly than

barometric pressure variations The degree of variation is

much less than atmospheric pressure

Audible sound has a frequency range of approximately 20 Hz

to 20 kHz and the pressure ranges from 20 X N m-2 to

200 N m-’ A pure tone produces the simplest type of wave

form, that of a sine wave (Figure 15.175) The average

pressure fluctuation is zero Measurements are thus made in

Figure 15.175 Sine wave

terms of the root mean square of the pressure variation

(abbreviated to RMS) For the sine wave the RMS is 0.707

times the peak value

Since RMS pressure variations have to be measured in the range 20 x N m-2 to 200 N m-* (a range of 10’) it can

be seen that an inconveniently large scale would have to be used if linear measurements were adopted Additionally, it has been found that the ear responds to the intensity of a sound (q2) in a logarithmic fashion The unit that has been

Acoustic noise %/I39

adopted takes these factors into account and relates the

measured sound to a reference level For convenience, this is

taken as the minimum audible sound (Le 20 X PO-6 N m-')

at 1 kHz

The logarithm (to the base 10) of the ratio of the perceived

pressure l(squared) to the reference pressure (squared) is

known as the Bell, i.e

Since this would give an inconveniently small scale (it would

range from approximately 0 to 14 for human response), the

Bell is divided numerically by 10 to give the decibel The

equation Itherefore becomes:

15.9.2 Sound intensity

Sound intensity, I , is a measure of energy and its units are

watts per metre Intensity is proportional to the square of

pressure Sound intensity Ievel (SIL) is defined in a similar

manner to sound pressure level In this case the equation is

dB (sound intensity level) = 10 loglo, - I

{ref

15.9.3 Sound power

Similarly, the power of a source (measured in watts) can be

expressed in terms of decibels (in this case called the sound

power level (SWL))

dB (sound power level) = 10 loglo, -

Wref is taken as IO-''

It can thus be seen that it is important not only to express

the unit but also to state sound pressure level (SPL), sound

intensity ]level (SIL) or sound power level (SWL)

W

Wref

15.9.4 Addition and subiraction of decibels

For coherent sound waves addition of values is possible It will

be apparent that as the scale is logarithmic, values cannot

merely be added to one another Intensities can, however, be

added anld thus the equation becomes

Figure 15.176 Noise-level addition graph

The square of individual pressures must be added and thus the

equation in this case must utilize P(tota1) = V(P: + P;)

15.9.5 Addition of decibels: graph method

It is possible to use a graph to calculate the addition of decibels, even in the case of multiple additions (Figure 15.176) The graph is used in the following way:

In the case of the addition of two levels - the difference

between the higher and lower levels is plotted on the lower scale of the graph The correction is then read from the vertical scale by projecting a horizontal line across to this scale from the point on the graph The correction is added

to the highest original level to give the total level

In the case of subtraction of levels - the difference be- tween the total sound level and the one to be subtracted is plotted onto the graph and the correction obtained as above In this case the correction is subtracted from the total level to give the remaining sound level

In the case of multiple additions - if there are more levels

to be added the first two levels are added using the graph and then the third is added to the resultant using the same method

15.9.6 The relationship between SPL, SIL and SWL

The total acoustic power of a source can be related to the sound pressure level at a distance r by the following equation (assuming spherical propagation):

w = P2/(pc47Tr?) where p = density of the medium and c = velocity of sound in that medium By substituting this back into the SPL equation

we obtain SPL = SWL - 20 loglor - 11 (spherical propagation) situations, i e

Point source on a hard reflecting plane Line source radiating into space Line source on a hard reflecting plane These equations are:

SPL = SWL - 20 loglor - 8 (hemispherical propagation)

SPL = SWL - 10 loglor - 8 (line source in space)

It is also possible to derive equations for other common

W 1 4 0 Plant engineering

SPL = SWL - 10 loglor - 5 (line source radiating on a

These equations are useful for calculating distance attenuation

effects

If the sound pressure level at distance ro is known it is

possible to calculate the sound pressure level at position rl

If rl is double ro it will be seen that the SPL will be

approximately equal to 6 dB (2OlO log 2) This gives us the

principle of a decrease in level by 6 dB per doubling of

distance (inverse square law) For the line source the same

calculation produces a difference of only 3 dB per doubling of

distance

15.9.7 Frequency weighting and the human response to

sound

In practice, noises are not composed of one single pure tone

but are usually very complex in nature It is essential that more

than the overall noise level (in dB) is known in order to

appreciate the loudness of a noise, as the ear does not respond

uniformly to all frequencies

As previously stated, the ear can respond from 20 Hz to

20 kHz and the response can be demonstrated by equal-

loudness contours (Figure 15.177) It can be seen in Figure

15.177 that there is a loss in sensitivity (compared to 2 kHz) of

approximately 60 dB at the low-frequency end of the chart It

will also be seen that all the curves are approximately parallel,

but there is a tendency to linearity at the higher noise levels

In order to produce meaningful readings it is therefore

important to state the sound pressure level in dB and the

frequency of the noise A weighting can be imposed on noise

readings which corresponds to the inverse of the equal-

loudness contours If this weighting is used all readings which

are numerically equal will sound equally loud, regardless of

Figure 15.178 Weighting networks

Originally, three networks were proposed (A, B and C) and

it was suggested that these be used for low, medium and high noise levels, respectively It was proved, in practice, that this introduced numerous difficulties particularly with rapidly changing noise levels when a change of filter network was necessary It was also found that the ‘A’-weighting network corresponded very well to annoyance levels at all noise levels (Figure 15.178) It was therefore decided that the ‘A’ weight- ing would be used as the norm for noise readings concerning human response There is another weighting network (the ‘D’

network) that is used for aircraft noise measurement

If it is necessary for engineering purposes to know the tonal make-up of a noise, several approaches are possible The noise can be processed by a bandpass filter The most common filters are octave band filters and the agreed centre frequen- cies are as follows

31 63 125 250 500 l k 2k 4k 8k 16k (Hz)

If further resolution is necessary, one-third octave filters can

be used, but the number of measurements that are required to

be taken is most unwieldy It may be necessary to record the noise onto tape loops for the repeated re-analysis that is necessary One-third octave filters are commonly used for building acoustics

Narrow band real-time analysis can be employed This is the fastest of the methods and most suitable for transient noise

Narrow band analysis uses a visual display screen to show the graphical results of the fast Fourier transform (FFT) and can also provide octave or one-third octave bar-graph displays

15.9.8 Noise indices

All the previous discussions have concerned steady-state noise It will, however, be apparent that most noises change in level with time It may therefore be necessary to derive indices which describe how noise changes with time The commonest

of these are the percentiles and equivalent continuous noise levels

Percentiles are expressed as the percentage of time (for the stated period) during which the stated noise level was ex-

ceeded, i.e a 5-minute L90 of 80 dB(A) means that for the

Acoustic noise 15/141

5-minute period of measurement for 90% of the time the noise

level exceeded 80 dB(A) Therefore LO is the maximum noise

level during any period and Ll00 is the minimum

The variation of noise levels within a discrete period of time

can best Ibe described by a set of Ln results (the more results

available, the greater the representation of the noise event)

Sound-level meters commonly measure Ln’s at seven points

(commonly, L1, L2, L10, L50, L90, L95, L99) More so-

phisticated modern machines are capable of being adaptea by

the user ,and non-standard Ln’s are available

Leq (the equivalent continuous noise level) is defined as the

continuoils steady noise level which would contain the same

total acoustic energy as the actual fluctuating noise, measured

over the same period of time This concept may be understood

by considering electrical power consumption If a machine

the total usage of power is 15 kW h-’ The equivalent power

for the &hour period would be 1.875 kW

If two events are to be added together and the Leq derived

we must first convert to intensity units Addition may then

take place directly using the equation:

Y

Zeq = - (Zltl + 12r2 + 13i3 .)

where 7 = total time, I1 = intensity for the first event,

tl = time for the first event, I2 = intensity for the second

event; t2 = time for the second event, etc The total intensity

is then converted back to decibel units by

T

I

dB(A) = 10 loglo-

Zref

where Iraf = reference intensity However we usually know

the levels in terms of dB(A) rather than intensities, therefore

by substitution

where L l = level 1 in decibels, etc As noise is often measured

on the ‘A,’-weighted scale Leq is usually expressed in this way

In this case the nomenclature becomes LAeq

A further derivation of equivalent continuous level is the

single-event level (SEL), also known as sound-exposure level

or Lax This a special type of Leq used for transient events

such as the passage of aircraft, gunshots, etc The SEL is a

one-second Leq and can be defined as the steady level which

over one second would contain the same ‘A’-weighted energy

as the actual event (regardless of its duration) Thus

1

T

Leq = IO log - (tl x 1 0 ~ 1 ’ “ + r2 x 1 0 ~ 2 ” ” + , , etc.)

SEL = 10 log (tl x 10~1’” + t2 x 1oL2’l0 + etc.)

where tl + t2 etc are the durations of levels L1, Lz, etc in

seconds

15.9.9 Noise-rating curves

These are a set of graphs that are commonly used as a

specification for noise from machinery They are similar to

Noise Criteria Curves (used in the USA to specify noise from

ventilation systems) The rating of a noise under investigation

15.9.10 Community noise units

Noise has been defined as unwanted sound To quantify noise

is therefore much more complicated than to quantify sound

itself (which is what we have previously considered) Units have to be derived from these purely acoustic measurements

by assessment of experimental psycho-acoustic data It has been found that the response to different types of aural stimulation cannot be described by one single measurement, and hence a number of different noise measures are used We now have three distinct classes of measurement:

1 Noise Units - these are the basic physical measurements

of sound (i.e decibel)

2 Noise Scales - these are composed of a combination of physical measurements (usually sound level, time, etc.) (i.e Ln’s, Leq, SEL)

3 Noise Indices - here other factors are used to modify the noise scales in order to more closely relate the noise scale

to other factors (annoyance for instance)

A criterion is a noise index value which is used to describe the reaction of a given percentage of the population

15.9.11 Road traffic

Road traffic is assessed by an 18-hour L10 This is not the percentile for 18 hours but rather the arithmetic average of the

15/142 Plant engineering

18 one-hour LlO’s between 6 a.m and midnight on a normal

working day

15.9.12 Air traffic

It has been found that annoyance caused by airdraft flyovers is

related to the average value of the maximum perceived noise

levels and the number of events The index is known as NNI

(noise and number index) and is obtained from:

NNI = Lpn(max) + 15 log(lON)-80

where Lpn(max) is the logarithmic average of the maxima of

the flyovers and N is the number of flyovers

15.9.13 Railway noise

Railway noise is assessed in Leq units 65 dB(A) Leq is the

usual criterion at which double-glazing is fitted where new

housing is built near to railway lines

15.9.14 Noise from demolition and construction sites

Hourly Leq is used as the index

15.9.15 Noise from industrial premises

British Standard 4142: 1990 is described in detail in Section

15.9.25 and is derived from the noise measured in Leq

compared to a background level measured in Ln

15.9.16 Measurement of noise

The simplest sound-level meter consists of a microphone, an

amplifier and a meter of some type Sound-level meters are

graded according to British and international standards For

most precision work a Type 1 (precision) sound-level meter is

used This has an accuracy of approximately *1 dB(A) Type

0 meters (laboratory) grade are rarely encountered Type 2

(industrial) grade sound meters may be suitable for some

initial survey work but may not be sufficiently accurate to

comply with legislative requirements at all frequencies In

particular, the lower grade of instruments have poor perfor-

mance above 10 kHz (the human ear responds to noise at least

up to 16 kHz)

15.9.17 Microphones

The microphone is a device for converting pressure fluctua-

tions in the air into an electrical signal For precision work two

types may be chosen The polarized condensor microphone

consists of a very thin metal diaphragm stretched in close

proximity to a back plate This diaphragm is charged to a

polarization voltage of 200 V (some are lower) The diaph-

ragm thus forms a condensor with the back plate Sound

causes the diaphragm to move in relation to the back plate,

thus changing the charge on the condensor This can be sensed

electrically and used to measure the sound

The pre-polarized (or electret) microphone is a develop-

ment of the polarized microphone, the main difference being

that the charge across the diaphragm is permanent (or almost)

and no polarization is needed (which simplifies the electronics

of the pre-amplifiers) The disadvantage of the polarized

microphone is that it is very moisture sensitive Condensation

on the diaphragm may result in electrical breakdown which

causes sparks These damage the diaphragm, thus ruining the

microphone The pre-polarized microphone has the disadvant-

age of slightly reduced long-term stability (although this has

now been largely overcome) Other types of microphone have

been used - notably the piezoelectric type - but these are not suitable for anything more than the most basic noise ‘survey’ meters

Microphones should be capable of measuring the pressure changes in the air without altering the pressure waves they are trying to measure This may seem to be a fairly fundamental point but, unfortunately, this is not physically possible The diagphragm must have sufficient frontal area in order to capture the pressure wave and hence produce a reasonably sensitive output Some reflections,will occur at the diaphragm and hence produce addition and/or cancellation effects with incoming pressure waves This effect will differ depending upon the angle of incidence of the sound on the diaphragm and the frequency of the pressure fluctuations

In the past it was necessary to have 25 mm diameter

diaphragms in order to get a sensitive response and reflection errors were a significant problem It is now common to employ

12 mm diameter microphones and these problems are now reduced There are, however, still many specialized micro- phones produced but they fall broadly into three types:

1 Pressure microphones - used for measuring sound in ducts, etc.;

2 Free field - used for measuring sound (usually out of

doors) in which the angle of incidence is at 0” to the centre line of the microphone; and

3 Random incidence - used for measuring sound (usually

indoors) in a reverberant field where the angle of inci- dence is more random

Note that most precision sound-level meters are fitted with a switch which can change electronically the response between free field and random response

For infra-sound (sound below the normal audible range) measurement special microphones may have to be used Although some ordinary microphones are capable of operat- ing at low frequency, great care has to be exercised in impedance matching if low-frequency cut-off is to be avoided

15.9.18 The sound-level meter

The precision sound-level meter incorporates the pre- amplifier in the nose of the meter (usually in the stem that the microphone fixes on to) The main amplifier is contained within the body of the meter and may either be auto ranging or

may have one or more user-adjusted ranges In older instru- ments the range had to be adjusted in 10 dB steps (which was very awkward to use with rapidly changing noise levels) Simple sound-level meters merely display the output of this amplifier onto an analogue meter (Figure 15.180)

Modern sound-level meters are equipped with internal filters and intergrating circuits and can produce outputs in terms of percentiles, Leq and frequency spectra Some sound-

level meters have a computer-controlled circuitry that is

addressable from a ROM cartridge which is inserted to load a program and then removed These sound-level meters can then perform many functions as several cartridges are avail- able The sound-level meter thereby becomes dedicated to one particular type of task

Memory power of sound-level meters is increasing daily and

it is now common to hold many sets of data (for instance, percentiles) in the sound-level meter memory and download later (perhaps in a kinder environment) either to a printer directly or to a personal computer If the PC option is chosen the data can be introduced to a graphics program and results displayed in a chosen graphics format which can produce elegant displays

Digital outputs are available on most sound-level me’ters which will enable connection to portable computers if much

Acoustic noise W 1 4 3

These units are now available in laptop computers They are not, at present, being produced by the major instrumentation companies, who continue with their dedicated machinery It has to be said, however, that the add-on units are not as fully developed as they might be

Current developments include the provision of amplifiers and power supplies to enable microphone connection directly, and if these prove successful the end of the dedicated sound- level meter may be in sight

Figure 15.1180 Schematic diagram of a sound-level meter

greater memory is required (or if on-site processing is chosen)

Sound-level meters are also equipped with a.c or d.c outputs

which will enable the connection of tape recorders, etc

Ruggedized sound-level meters are available which are

designed for leaving out of doors These devices (often

referred lo as environmental noise analysers) are fitted into

steel wealhertight cases and have a large battery capacity (and

the provision for external battery connection) They are fitted

with their own printers Battery and paper life is in the order

of six days Longer life may be obtained by the use of external

batteries and minimizing the amount of data being printed to

the paper roll

15.9.19 Digital signal analysis

While analogue filtering of signals may be of some use, as

previously described, if detailed information is needed inevi-

tably digital processing is called for The principle of frequen-

cy analysis is known as Fourier Analysis The Fourier series

states that any complex signal can be represented as a series of

sine waves of various frequencies, magnitudes and relative

phase angles

An example of this is the square ‘wave This signal may he

represented by the series of sine waves composed of the

fundamental frequency - a sine wave at three times the funda-

mental and one-third of the amplitude, a sine wave at five

tiirnes the frequency and one-fifth the amplitude, etc., with the

progression carrying on to infinity Electrically, this process is

known as, FFT (Fast Fourier Transform) analysis The narrow

band FFT analyser displays this signal graphically (as a display

with frequency on the x-axis and amplitude on the y-axis)

Octave OF one-third octave analysers usually employ digital

filters which are arranged such that real-time analysis is

possible ((where the whole of a signal is analysed rather than

merely a snapshot) The sophistication of the machine and the

required upper frequency will determine whether real-time

operation is possible or not Bothi types of analyser have

digital outputs which will enable downloading to larger com-

puters for further manipulation or to allow long-term storage

It is now possible to obtain add-on hardware and software

systems for existing personal computers which will enable

them to be used both as statistical (Ln and Leq, etc.) and

frequency analysers (both narrow and octave band, etc.)

15.9.20 Noise control

Noise is capable of causing psychological, physiological and pathological reactions as well as physical damage to plant, machinery and building structures The need for the control of noise is recognized in many statutes for the protection of both workers and members of the public in their homes

15.9.21 Noise nuisance

Section 80 of the Environmental Protection Act 1990 gives local authorities the power to serve a notice where certain classes of nuisance have occurred or may occur The expres- sion ‘nuisance’ is not defined in the Act or indeed in any other The use of the expression ‘nuisance’ can be traced back to

legal action as far as the thirteenth century and its meaning is

now well understood

Nuisance describes anti-social un-neighbourly behaviour, and has been taken to mean the interference with one’s neighbours in their day-to-day-activities Noise nuisance can therefore be a statutory nuisance (by virtue of the Environ- mental Protection Act), a private nuisance (actionable at common law as a tort) or a public nuisance (a crime) FOF a noise to be a statutory nuisance it must also be a common law nuisance and hence a private or public nuisance

The concept of private nuisance is now well developed Private nuisance is a land owner’s tort and is a complaint that the use or enjoyment of his or her land has been interfered with The nuisance only applies to the occupier of the land and not his or her family or sub-tenants

There are two types of private nuisance The first concerns

rights attached to land (for instance, right of way) and the

second to enjoyment of the land (which does have relevance to noise control) This class of nuisance is described as ‘where a person is unlawfully annoyed, prejudiced or disturbed in rhe enjoyment of land or with his health, comfort or convenience

as an occupier’ The interference must be substantial and the duration, nature and level of the noise must be considered A single event may not therefore constitute a nuisance The area affected by the nuisance must therefore be consi- dered One often-quoted remark is taken from the case of

Sturgess v Bridgam (1879), in which Theiseger, L J., said

‘What would be a nuisance in Belgrave Square would not necessarily be so in Bermondsey’ However, care must be

taken if it is to be assumed that because an area is already noisy extra noise will not constitute a nuisance In one case

another printing press in Fleet Street proved to be a nuisance (1907)

Two other legal precedents should be considered at this

stage The first concerns sensitivity In the case of Walrer v

Selfe (1851) the expression ‘ought this inconvenience to be considered - not merely according to elegant or dainty modes

of habit or living, but according to plain and sober and simple notions amongst the English people’ was quoted This forms a cornerstone of nuisance law and gives rise to the question of reasonableness of a nuisance Special sensitivities are not therefore to be considered when the question of nuisance arises This may have relevance to shift workers, for instance,

1 5 / 1 4 Plant engineering

who while they might expect their daytime sleep to be

protected by law, may be disappointed to find that the law will

only protect their property against noise that would affect the

enjoyment of the average person (i.e one who is not sleeping

during the day) The second precedent concerns the case of

the aggrieved person who moves next to a noise source and

hence suffers a nuisance

The law of prescription concerns private nuisances (but not

public) and states that if things are done which affect your

neighbour (with his or her knowledge) and continue for 20

years, you obtain the right to continue However, this does not

translate well to noise nuisance If, for example, the noise has

continued for more than 20 years but no one has been affected

by it, there has been no noise nuisance and hence there can be

no prescriptive right

This can be illustrated by the case of Sturgess v Bridgrnan

(1879) The plaintiff was a doctor who built a consulting room

at the bottom of his garden against a neighbouring property

and was affected by the noise of machinery from that proper-

ty The judge ruled that as the doctor had not known about the

noise until he built his consulting room no prescriptive right

accrued Therefore in the common case of a complainant

moving next door to a factory the normal rules of nuisance will

apply, despite the factory occupier’s insistence that ‘they were

there first’

15.9.22 Health effects

Exposure to noise has been shown (in clinical experiments) to

cause nausea, headache, irritability, instability, argumenative-

ness, reduction in sexual drive, anxiety, nervousness, insom-

nia, abnormal somnolence, and loss of appetite, as well as the

more well-known hearing loss Generally these health effects

were shown to occur at noise levels greater than 85 dB(A)

In the case of hearing damage, numerous experiments have

been conducted with the aim of arriving at a safe exposure to

noise It has been found that some individuals are much more

susceptible to hearing damage than others Some people may

suffer permanent damage over a few months’ exposure while

others may take years to develop the same damage (at the

same noise levels)

Physical injury occurs at sound pressure levels in excess of

140 dB (at this level there is a risk of rupture of the tympanic

membrane) while levels greater than 130 dB result in acute

pain Statistical studies on workers exposed to noise levels

between 75 dB(A) and 9.5 dB(A) lead to the following conclu-

sions:

1 For a 40-year working life a daily Leq of less than

75 dB(A) will lead to negligible risk

2 The experimental data would indicate that for higher noise

levels, and corresponding shorter time periods, the risk to

hearing damage is the same For example, 78 dB(A) for an

8-hour period is the same as 81 dB(A) for a 4-hour period

3 Above 7.5 dB(A) 8-hour Leq the risk of hearing damage

increases proportionately with the rise in levels

4 Most countries have legislation which restricts noise levels

to 85 dB(A) k 5 dB(A) with a tendency to reducing

acceptable levels It should be noted that at the UK’s limit

of 90 dB(A) there is some risk of hearing damage

5 Infra-sound (sound below the normal human audible

range) is capable of causing health effects More recent

research indicates an effect similar to excess alcohol con-

sumption and indeed a synergistic effect with alcohol has

been noted It may be that in certain cases infra-sound is

capable of causing an increase in accident rates High

infra-sound levels are noted in the foundry industry and in

drivers’ cabs in large vehicles

15.9.23 Damage to plant/machinery/building structures

Noise can lead to damage in two ways:

1 Directly - as a result of induced vibrations

2 Indirectly - as a result of interference with the operative’s normal function

Direct damage includes vibration fractures of electrical com- ponents (particularly switch contacts), structural panels, etc Damage to buildings occurs particularly around windows (infra-sound is particularly troublesome in this effect) Indirect damage is probably the greatest effect of noise levels Operator performance is affected by fatigue and also the inability to hear potential problems with the machine (that might ordinarily be attended to with no significant damage resulting) In addition, the inability to hear shouted warnings may result in accidents and further plant damage

15.9.24 Legislation concerning the control of noise

15.9.24.1 Environmental Protection Act 1990, Section 80

A notice may be served where a nuisance has occurred or the Local Authority think a nuisance may occur Noise nuisance is not defined as such, but includes vibration The notice may not be specific and may merely require the abatement of the nuisance A notice may, however, require the carrying out of works or specify permissible noise levels The time period for compliance is not specified in the Act, but must be reasonable Appeals against a Section SO notice must be made to the magistrate’s court within 21 days of the serving of the notice The grounds of appeal are given in the Statutory Nuisance (Appeals) Regulations 1990 and are as follows:

1 That the notice is not justified by the terms of Section 80 The most common reason for this defence is that the nuisance had not already occurred, and that the Local Authority did not have reasonable grounds to believe that the nuisance was likely to occur

2 That there had been some informality, defect or error in,

or in connection with, the notice It may be that the notice was addressed to the wrong person or contained other faulty wording

3 That the Authority have refused unreasonably to accept compliance with alternative requirements, or that the requirements of the notice are otherwise unreasonable in character or extent, or are unnecessary This defence is self-explanatory

The Local Authority are only permitted to ask for works that will abate the noise nuisance Other works (perhaps to comply with other legislation) should not be specified in the notice They may, however, be contained in a letter separate from the notice An example of this would be where food hygiene requirements were breached by the fitting of acoustic enclosures to food-manufacturing machines Readily cleanable enclosures may be a require- ment of the Food Hygiene Regulations, but it should not

be contained in a Section 80 Environmental Protection Act notice

4 That the time (or, where more than one time is specified, any of the times) within which the requirements of the notice are to be complied with is not reasonably sufficient for the purpose

5 Where the noise to which the notice relates is that caused

by carrying out a trade or business, that the best practic- able means have been used for preventing or for counter- acting the effects of the noise ‘Best practicable means’

Acoustic noise 1 5 / 1 4

incorporates both technical and financial possibility The

latter may be related to the turnover of a company

Theredxe a solution that may be the best practicable

means for one company may not be so for another

6 That the requirements imposed by the notice are more

onerous than those for the time being in force in relation to

the noise to which the notice relates of

(a) Any notice under Sections 60 or 66 of the Control of

(b) Any consent given under Sections 61 or 65, or

(c) Any determination made under Section 67

Section 60 relates to a construction site notice Section 61

is a consent for construction works Sections 65-67 relate

to noise-abatement zones (see below)

7 That the notice might lawfully have been served on some

person instead of the appellant, being the person respon-

sible for the noise

8 That the notice might lawfully have been served on some

person instead of, or in addition to, the appellant, being

the owner or occupier of the premises from which the

noise is emitted or would be emitted, and that it would

have been equitable for it to have been so served

9 That tlhe notice might lawfully have been served on some

person in addition to the appellant, being a person also

responisible for the noise, and that it would have been

equitable for it to have been so served

Pollution Act 1974, or

15.9.25 British Standard 4142: 1990

This British Standard is a revision of a standard first published

in 1967 and was revised in 1975, 1980, 1982 and 1990 The

standard purports to rate noises of an industrial nature affect-

ing persons living in the vicinity It gives a method of

determining a noise level, together with procedures for assess-

ing whether the noise in question is likely to give rise to

complaints It does make the point that while there is a

correlation between the incidence of complaints and general

community annoyance, quantitive assessment of the latter is

beyond the scope of the document, as is the assessment of

nuisance

The previous document has been used extensively as a guide

to the assessment of nuisance in various circumstances (cer-

tainly outside the scope of the document) and has gained a

status that outweighs its original intention Unfortunately, the

early document was very flawed in its methodology (as is the

current one) and resulted in numerous difficult legal decisions

when it was produced in court as the definitive guide to noise

nuisance In particular, the old BS 4142 had a method for

obtaining a ‘notional background level’ where the actual

background level (i.e that level which exists when the noise in

question was suppressed) could not be measured, which was

widely discredited as being grossly inaccurate

The new BS 4142 rates noise in terms of Leq over a

measured time interval (one hour in the daytime and 5

minutes at night) and compares this level with a background

measured in terms of the L90 of the ambient If a noise has a

duration shorter than the measurement period in question, an

‘on-time’ correction is applied by the use of the following

equation:

Ton

LAeq Tr = LAeq T, -k 10 loglo -

Tr where

LAeq T, = Leq for reference period

LAeq T,,, = measured Leq for the event

Ton = time on

T, = reference time period (5 or 60 minutes)

Table 15.33 Corrections to noise level readings

Noise level reading LAeq T

minus background LA90, T

Correction subtract from

noise level reading

to source and back-calculate theoretical noise ievel in isolation from background

A further correction may need te be applied if the specific

noise does not exceed the background by more than 10 dB A

simplified correction table is used (Table 15.33)

Finally a correction is applied dependent upon the nature of the noise If the noise contains a distinguishable, discrete continuous note (whine, hiss, screech, hum, etc.) or if there are distinct impulses in the noise (bangs, clicks clatters or

thumps), or if the noise is irregular in character enough to

attract attention, add 5 dB to the specific noise level to obtain the rating level The assessment for complaint purposes is then made by comparing this rating level with the background noise level If the rating level exceeds the background by 5 dB the standard states that the result is marginal, and if the rated level exceeds the background by 10 dB or more, complaints are

‘likely’

This background noise level is one of the main criticisms of the document as it is intended to include any existing noise sources in the area The new noise source is therefore com- pared against the existing noise climate, even if most of this is produced by the same factory The example given in the British Standard further reinforces this point by considering premises which produce 40 dB(A) at the nearest house when operating normally and yet the ambient fails to 29 dB(A) during a factory shutdown Thus the existing contribution is already 11 dB A new source is assessed which adds 4 dB to the existing (40 dB) ambient, and the result is determined to

be marginal! If this situation were to continue the background sound level (as defined in the standard) would ‘creep’ upwards

- obviously an undesirable situation and one that is addressed

in a planning circular (Circular 10/73) that deals with planning and noise This circular particularly addresses creeping am- bients and states that ‘the introduction of a new noise source into an area is liable to result in a creeping growth of ambient noise level, and consequent deterioration in the quality of the environment, even though each of the new sources, consi- dered separately, would not be liable to give rise to com- plaints’ This point alone is sufficient for the method to be discredited by Environmental Health Officers when investi- gating nuisances, and the standard is unlikely to be used by them as a definitive guide Consequently, operators of indus- trial premises should not use the information given in this British Standard as evidence when arguing (in legal situations) that they are not causing a statutory nuisance

Further, this British Standard takes very poor account of the effects of discrete frequency components It is quite possible for a narrow band component to cause a serious nuisance while being almost unmeasurable on an ‘A’-weighted scale Consequently, more detailed narrow band analysis would be necessary and it is essential to compare the actual noise with the background noise within that narrow band (usually octave

or one-third octave) The British Standard makes no mention

of such a situation

15/146 Plant engineering

15.9.26 Noise-abatement zones

Local Authorities are empowered by the Control of Pollution

Act 1974 to designate areas as noise-abatement zones Within

these areas noise levels are measured and entered onto a

register It is an offence to increase noise levels beyond

register levels unless a consent is obtained If the Local

Authority are of the opinion that existing noise levels are too

high, noise-reduction notices can be served

In the case of new premises the Local Authority will

determine noise levels which it considers acceptable, and these

will be entered into the noise level register Appeals against

notices or decisions are made to the Secretary of State

15.9.27 Planning application conditions

Local Authorities are empowered to impose conditions on

planning applications to protect environmental amenities of

neighbours Noise is commonly controlled by conditions

Local Authorities may ask for more onerous controls on

planning conditions than the mere avoidance of nuisance

Planning conditions are designed to avoid reduction in amen-

ity of neighbours This may mean that a process has to be

almost inaudible (particularly in the case of Light Industrial

Consents) Appeals against planning conditions are made to

the Secretary of State

15.9.28 The Health and Safety at Work etc Act 1974

Section 2 of this Act imposes a general duty on employers to

ensure, so far as is reasonably practicable, the health, safety

and welfare at work of all his or her employees This general

section will include the acoustic environment It should be

noted that there are specific regulations made under this Act

(The Noise at Work Regulations 1989 - dealt with later) that

control noise (primarily to protect hearing) but they do not

completely satisfy the overall requirements of Section 2 For

instance, if high noise levels mask an audible alarm such that a

risk of injury is caused, a breach of Section 2 would be likely

(despite the fact that the noise levels in The Noise at Work

Regulations have not been exceeded) Section 3 of the Act

deals with an employer’s duties to nsn-employees and again

imposes a duty of care

The Noise at Work Regulations have no relevance to

members of the public as they apply only to persons at work

Section 3 of the Act would, therefore, control noise exposure

to non-employees who would be likely to suffer risks to health

and safety Noise nuisance is controlled by other legislation

(The Environmental Protection Act 1990 and others)

Section 4 of the Act imposes the general duty on employers

to care for the health and safety of persons, who not being his

or her employees, nevertheless have resort to premises under

his or her control This duty is designed to protect subcon-

tractors, etc

Section 6 concerns articles manufactured for use at work

and will control the acoustic output of machines, etc The

control of the manufacture of noisy machines is carried further

by The Noise at Work Regulations

15.9.29 The Noise at Work Regulations 1989

These Regulations came into force on 1 January 1990 and

control the exposure to noise of persons at work They

establish three noise levels known as the first, second and peak

action levels Different regulations are applicable as each

action level is exceeded The unit of measurement is known as

equivalent continuous sound level and may be defined as ‘that

notional continuous steady level which would have the same

‘A’-weighted acoustic energy as the real fluctuating noise measured over the same period of time’ For the purposes of the Regulations an 8-hour time period is used and the 8-hour equivalent continuous sound level is abbreviated to L E P , d: The first action level is 85 dB(A) LEp.d

The second action level is 90 dB(A) LEP,d The peak action level is 200 Pascals (equivalent to 140 dB) Damage to the hair cells in the inner ear is proportional to the noise energy received This is a dose concept comprising the product of noise level and exposure duration It follows, therefore, that the same amount of deafness will follow from the exposure to a very intense sound for a short period as to a lower level for a proportionally longer period

It has been shown that the exposure time has to be halved for each 3 dB(A) increase in the noise levels 3 dB(A) repre- sents a doubling of sound energy, hence this rule has become known as the equal energy damage risk criterion It follows that 93 dB(A) for 4 hours is also 100% of the permitted exposure for a day; similarly, 2 hours at 93 dB(A) would be

50% of the permitted exposure Where an employee is likely

to be exposed to above the first action level the employer shall ensure that a competent person makes an assessment of the noise levels which is adequate for the purposes:

1 Of identifying which of his or her employees are so

exposed, and

2 Of providing him or her with such information with regard

to the noise to which those employees may be exposed as

will facilitate compliance with his or her duties under the Regulations, specifically:

(a) Reduction of noise exposure (b) Ear protection

(c) Ear protection zones (d) Provision of information to employees

15.9.29.1 The requirements of the Regulations

An employer must:

1 Carry out an assessment when an LEp,d of 85 dB(A) is

likely to be exceeded

2 Review the assessment if changes necessitate this

3 Record the exposure and keep records

4 Reduce the risk of damage to hearing to the lowest level reasonably practicable

5 Every employer shall, when any of his or her employees is likely to be exposed to the second action level or above or

to the peak action level or above, reduce, so far as is reasonably practicable (other than by the provision of

personal ear protectors) the exposure to noise of that employee

6 If an employee is exposed to greater than the first action level and less than the second action level the employer shall provide hearing protection if so required by the employee

7 If an employee is exposed to greater than the second action level or greater than the peak action level the employer shall provide hearing protection which, when properly worn, will reduce the risk of hearing damage to below that arising from exposure to the second action level

or, as the case may be, to the peak action level

8 Ear-protection zones (i.e areas where the second action level is likely to be exceeded) shall be established Employees must wear ear protection in this zone The employer shall erect suitable signs

9 Information, instruction and training shall be provided for employees where exposure is likely to exceed the first

action level or the peak action level This information shall

include:

(a) The risk of damage to an employee’s hearing that such

exposure may cause;

(b) What steps an employee can take to minimize that

risk;

(c) Tine steps that an employee must take in order to

obtain the personal ear protectors which the employer

must provide

(d) The employer’s obligation under the Regulations

15.9.30 Noise control engineering

Be€ore attempting noise control it is important to consider the

nature of the problem The first (and usually the most

cost-effective) approach is to silence the noise at source In

order to aippreciate likely noise sources and the methods used

to reduce their emission, we will consider the ordinary recipro-

cating piston engine

The first source is the crankcase wall ‘ringing’ under the

reciprocating forces of the combustion Excess noise will be

produced if the frequency of the combustion pulses is at the

resonant frequency of part of the engine The solution is to

de-tune the block by stiffening (which may have an added

mass-law effect) or by the addition of damping materials

The next source to consider is the crankshaft and bearings

Most shafts will be out of balance to some degree and will have

a resonant point A well-designed engine should not run at this

resonant frequency Bearings are two main types - sliding (or

plain) be:uings and roller Excess clearance can give rise to

bearing knock and the solution here is to replace bearings

Poorly designed systems can give rise to bearing knock if the

shaft has a bending mode within the engine’s operating range

Plain (oilite) type bearings can produce screech on start-up

from cold (these bearings are used in electric motors, starter

motors and alternators, etc.) Replacement may make matters

better

Roller bearings are generally quieter than plain bearings but

can produce considerable noise if damaged Frequency ana-

lysis of the noise can be used to assess the source but there are

many modes involved in a bearing ‘click’ Again the solution is

replacement, with care being taken to locate the new bearing

without causing damage (by only inserting the bearing with

force on the outer race) and by careful checks of the dimen-

sions of the bearing housing

Gearboxes give rise to noise as teeth contact each other If

excess clearance (and/or poor lubrication) is present, noise

will be exacerbated Again, frequency analysis can help to

locate the source of trouble by comparing the dominant

frequency of the noise with the gear teeth meshing frequen-

cies It should also be noted that gears themselves can ring

(particularly if they are free and not cast as part of a layshaft,

etc.) The dominant frequency may be at the damaged gear’s

resonant frequency

Finally, hydraulics can give rise to noise due to the intermit-

tent force pulses produced by pumps The solution would be

to introduce some flexibility into the receiver system to damp

out the very high intermittent pressures produced by the

incompressibility of the fluid

Other examples of machines which lend themselves to noise

reduction at source are:

1 Presses-is the degree of impact necessary? Can it be

adjusted? Can the press operate by pressure alone?

2 Air discharge - use of air tools and nozzles The turbu-

lence in the boundary layer of air between the rapidly

moving airstream and the atmosphere is heard as noise

Can the airstream be diffused (silencers fitted to the

Acoustic noise 15/147 exhaust)? Nozzles used for cleaning can have devices fitted which give a gradual transition from the rapidly moving air

to atmosphere by the use of an annular ring of small nozzles round the central nozzle These silence with very little loss in efficiency

Reciprocating compressors - these give rise to very high noise levels at low frequency (below 250 Hz typically) These low-frequency noises are very difficult to attenuate The most popular solution is to use rotary (vane type) compressors instead These are inherently quieter and have the further advantage that the noise they generate is

at high frequency (typically above 1 kHz) and is, there- fore, easy tQ attenuate

Cutting machines - modifications to the method of res- training material being cut to reduce ‘ringing’ Reducing free length of material

reduction of noise at source is not possible (or does not provide sufficient reduction) the transmission pathway must

15.9.31.1 Insulation

The simplest insulator is a sheet of material placed in the sound-transmission pathways Sound energy reaches the sur- face in the form of a pressure wave Some energy passes into the partition and the rest is reflected

Energy that passes into a partition may be partially ab- sorbed and transformed into heat This is likely to be very small in a plain partition The remainder of the energy will then pass through the partition by displacement of molecules and pass as sound in the same way that sound travels in air This can then pass to the edge of the partition and be

re-radiated as sound from other elements of the structure

- this is known as flanking transmission By far the greatest amount of energy, in a thin partition, will pass through the partition by actually causing the partition to vibrate in sympa- thy with the incident sound and, hence, re-radiating the sound

on the opposite side The amount of sound transmitted through a partition is represented by the ratio of the incident energy to the transmitted energy This factor when expressed

as decibels is known as the sound-reduction index:

1 Transmission coefficient The movement of the panel (and hence its resistance to the passage of sound) is controlled by a number of factors: The surface mass affects the inertia of the panel Greater mass causes a corresponding greater inertia and hence more resistance to movement At high frequencies this becomes even more significant The mass law can be expressed:

where rn is the superficial weight (kg m-*) and f is the

frequency (Hz)

Stiffness - at very low frequencies the movement of the panel will be controlled by the stiffness, as inertia is a dynamic force and cannot come into effect until the panel has measurable velocity Stiffness controls the perfor- SRI = 20 lOgl& - 43 dB

Plant engineering

Frequency (Hz)

Figure 15.181 Typical insulation characteristics of a partition

mance of the panel at low freqencies until resonance

occurs As the driving frequency increases, the resonance

zone is passed and we enter the mass-controlled area The

increase in sound-reduction index with frequency is app-

roximately linear at this point and can be represented by

Figure 15.181

3 Coincidence - a panel will have a bending mode when a

wave travels along the length of the sheet of material The

frequency of this bending mode is known as the critical

frequency This mode of bending will be introduced by

sound incident at angles greater than 0” At the critical

frequency coincidence will only occur for a sound wave

with a grazing incidence (90”) At greater frequencies the

partition will still be driven, but in this case by progress-

ively lower angles of incidence The coincidence dip is not,

therefore, a single dip but will result in a loss of sound-

reduction index at progressively higher frequencies The

desirable insulation panel will, therefore, be massive but

will not be stiff

15.9.32 Absorbers

15.9.32.1 Porous absorbers

As sound passes through a porous material, energy is lost by

friction within the material The material is usually employed

by fixing it to the surfaces in a room The absorber will have

the highest efficiency when positioned where the air molecules

are moving the fastest (and hence more energy is absorbed)

At the wall surface the molecules are stationary If we plot a

single-frequency graph we find that the maximum particulate

velocity occurs at A14 (one-quarter wavelength) from the

surface In practice, incident sound is rarely of single frequen-

cy But the principle can be observed that the absorber must

be one-quarter of the wavelength away from the wall (for the

frequency of the sound to be absorbed) This can be arranged

either by having a thickness greater than h/4 for the lowest

frequency to be absorbed or, alternatively, to mount the

absorber on a frame some distance away from the wall such

that the centre of the absorber is at h/4 for the frequency to be

absorbed

15.9.32.2 Resonant absorbers

The simplest resonant absorber is known as the Helmholtz

resonator This device consists of a chamber connected to the

duct (or whatever area is to be controlled) by a narrow neck The volume of air in the chamber will resonate at a frequency,

F,,,, determined by the volume of the chamber, the length of the neck and the cross-sectional area of the neck

S

IV F,,, = 55 - (Hz)

where S = cross-sectional area (m’), I = length of neck (m)

and V = volume of enclosure (m3) As the chamber reso- nates, air is forced through the narrow neck and hence energy

is absorbed in overcoming the resistance

The degree of attenuation at the critical frequency can be very large, but this type of silencer has a very narrow bandwidth This device may be suitable when the machine being dealt with emits sound predominantly of a single wave- length The absorption bandwidth of a Helmholtz resonator can be expanded by lining the chamber with absorbers but this has the effect of reducing the efficiency

The perforated absorber which forms the basis of many acoustic enclosures and silencers is a development of the resonator principle As stated previously, the bandwidth may

be broadened by packing the chamber with an absorber, but this lowers efficiency This may be overcome by using multiple absorbers in the sound path It can be arranged by placing a perforated sheet some distance away from the rigid outer wall

of the enclosure and filling the cavity with absorber It is not necessary to use cross-walls between the ‘chambers’ so

formed In this case the equation becomes:

15.9.33 Vibration isolation

Vibration in machinery or plant can be induced in a number of ways, including:

Out-of-balance forces on shafts

0 Magnetic forces in electrical apparatus Frictional forces in sliding objects

The first course of action in vibration isolation is the reduction

at source This may be achieved by balancer shafts in engines, stiffer coils in electrical apparatus or better lubrication be- tween adjacent sliding surfaces

When all possible vibration reduction has been achieved the machine must be isolated from the structure by some form of spring mounting Spring mounts have a resonant frequency depending upon the stiffness of the spring and the weight of the object placed upon it It will be apparent that the static deflection of the spring will also be proportional to the resonant frequency

As the driving force of the mass/spring increases from zero

up to the resonant frequency the amount of transmission of the vibration increases until resonance is reached and the

Acoustic noise acting as a spring If the two elements of the wall were in rigid connection the insulation would be 3 dB more than the single element alone (mass law) and if totally separated it would be the sum of the figures In practice, a cavity wall with ties and a

50 mm cavity gives approximately 10 dB more reduction compared to a single-skin wall of half the surface mass

Double-glazed windows work on the same principle It is

important to avoid the coincidence of the resonant frequencies

of the two elements and hence it is usual to arrange for the glazing panels to have different thicknesses (and hence a different resonant frequency) This is not necessary if one element is subdivided by glazing bars to give different size panes from its opposing element The reveals of a double- glazed window should be lined with acoustically absorbent material to damp the sound within the cavity The width of the cavity should not be less than 150 mm

If insulation panels are not of uniform construction, as in the case of a wall containing a window, the average sound- insulation value has to be derived for use in calculations The total transmission coefficient for the composite panel will equal the sum of the individual coefficients multiplied by their respective areas and divided by the total area Thus:

t," = and the SRI of the total panel derived from

Figure 15.182 Performance of anti-vibration rnountirg

transmission becomes infinitely large As the resonant point is

passed, the transmission begins to reduce until at some point

the transmissibility falls below one (see Figure 15.182), i.e

isolation occurs

In practice, however, spring systems have some inbuilt

damping and this will have the effect of reducing the ampli-

tude of the resonance below infinity This is very necessary in

real systerns to avoid excessive excumions of mounted machin-

ery A damped mounting will follow the second curve in

Figure 15 182 and it will be noted that the vibration isolation

at high-fiequency ratios is less than that for undamped

systems It is important, therefore, to use the lowest degree of

damping that is necessary

15.9.34 ]Practical applications

15.9.34.1 Acoustic enclosures

Panels of multi-resonator material are made from perforated

plate sandwiched with solid plate with an intermediate

absorber layer These panels can be built up into enclosures

taking care to seal all junctions adequately Typically, these

enclosures are made to surround small machines (e.g com-

pressors) They may be fitted together with spring catches to

aliow for dismantling for maintenance purposes

Ventilation may be a problem but can be dealt with in

several ways:

I Acoustic louvres - louvres are constructed of the absor-

bent panel material (suitable for small degree of noise

reduction only);

2 Silencer fitted to ventilation duct (see later);

3 Baffled enclosures to ventilation duct

15.9.34.2 Building insulation

Single-panel insulators have been described earlier In build-

ing insulation it is usual to provide double insulation In

theory, if the insulation panels have no interconnection it

should be possible to arithmetically aldd the sound reduction of

the two elements of the structure In practice, it will be found

that there is bridging, either by the structure, wall ties or

flanking iransmission or by the air between the two elements

SRI = 10 loglo (L) (dB)

tav

15.9.34.3 Control of noise in ducts

Fans produce the least noise when operating at their maximum efficiency It is, therefore, important to select the correct fan for the airflow and pressure characteristics required It is also important to remember the noise generated within the system (as opposed to at the fan) depends on the air velocity, and, hence, for a required airflow rate, a larger cross-section duct (with a correspondingly lower velocity) will give quieter results It will also give other advantages when it comes to providing extra noise attenuation and fitting of silencers

It is most important in the design of systems to eliminate as much turbulence as possible To achieve this, the fans should

be mounted some distance away f1om bends (at least one and

a half duct diameters) Junctions between pipes and con- nectors should present a smooth internal profile and inlets to systems should be tapered and not plain Outlet grilles should

be of larger diameter than ducts and have aerodynamically smooth profiles where possible

If it is necessary to add extra attenuation to a duct it is essential to decide on the required amount If only a relatively small degree of absorption is required the first course of action

is to line part of the duct with absorber The length of duct to

be lined will be determined by the degree of attenuation required and the thickness will be determined by the frequen-

cy of the noise The data for these factors are available from many sources and are usually published as tables by manufac- turers For further attenuation it is necessary to provide a centre-pod type attenuator (Figure 15.183) This increases the area of the absorber and also aids low-frequency attenuation For further low-frequency attenuation an in-line splitter si- lencer is employed (Figure 15.184)

These are capable of providing a high degree of attenuation, depending on the width between the elements The smaller the gap, the higher the attenuation Again, tables of perfor- mance are published by the major manufacturers It is necess- ary, in order to decide on the Sesign of the silencer to be

Plant engineering

Figure 15.183 Centre-pod silencer

Figure 15.184 Splitter silencer

installed, to know the required attenuation (and something

about the frequency/noise level profile) and the permitted

pressure loss in the system

Manufacturers’ data can then be consulted Splitter si-

lencers are also available, made into bent shapes, and these

can provide even higher degrees of attenuation as well as

aiding installation Silencers should ideally be fitted in systems

as near to the noise source as possible to avoid noise break-out

from the duct Other obstructions in the duct must be consi-

dered, however, as they may generate further aerodynamic

noise which, if it occurs after the silencer, will not be atten-

uated

15.9.34.4 Anti-vibration machinery mounts - in practice

Again, the characteristics of the system need to be considered

The weight of the machine and the frequency will determine

the static and dynamic deflections of the mounts and hence the

material of which the mount is to be constructed At very high

frequencies, mats may be placed under machinery These may

consist of rubber, cork or foam At middle frequencies it is

usual to use rubber in-shear mounts At low frequencies metal

spring mounts are used

15.9.34.5 Mats

Anti-vibration mats are very useful for frequencies about

25 Hz They have the disadvantage of being liable to attack by

oils and if they become saturated or deteriorated they will

compress and lose their efficiency

15.9.34.6 Rubber mounts

Although these are loosely termed ‘rubber’ mounts, they are often composed of synthetic rubbers which are not readily attacked by oils and can operate over a much wider tempera- ture range Typical maximum static deflections are 12.5 mm

15.9.34.7 Steel spring mounts

Steel springs have the disadvantage of transmitting the high frequencies along the length of the spring It is usual to mount the spring with a rubber or neoprene washer under its base Steel spring mounts are also most vulnerable to resonance problems, and the solution is to build in a damper device This has the disadvantage of reducing the isolator’s efficiency

Machinery must be positioned so that all mounts are equally loaded, and failure to do so will result in the possibility of a rocking motion developing This may require mounting the machine on a subframe If this is not possible the load should

be assessed at each mounting point and mounts of different stiffness used

15.9.34.9 Installation

Mounts should be installed so that the whole machine is isolated from the structure Services (e.g power, hydraulics, etc.), should be mounted flexibly Bridging is the most com- mon fault when providing vibration isolation to machines and building structures, and should be carefully avoided Services should be designed to withstand the degree of movement permitted by the anti-vibration mounts without suffering damage

Further information on acoustic noise and its control can be found in references 68-70

8 Balje, 0 E., Turbomachines, John Wiley, Chichester (1981)

9 Csanady, G T., Theory of Turbomachines, McGraw-Hill, New York (1964)

10 API 610, Centrifugal Pumps for Refinery Service, 7th edn (1989)

11 IS0 5199, Centrifugal Pumps: Class I1 (1986)

12 BS 5316, Acceptance Tests for Centrifugal Mixed Flow and Axial Pumps Part 1 Class C Tests

13 American Hydraulic Institute Standards (1983)

14 BS 848, Fans for General Purposes Part 1: Methods of Testing Performance (1980): Part 2: Methods of Noise Testing (1985)

15 BS 2009, Code for Acceptance Tests for Turbotype Compressors and Exhausters (1953)

16 BS 1571, Testing of Positive Displacement Compressors and Exhausters Methods for Acceptance Testing (1987): Part 2:

Klay, H R and Reich, B., ‘Gas compressors - a brief

survey’, Sulzer Technical Review, 2/1988

Final Report of the Advisory Committee on Asbestos (Vols 1

and 2) October 1979, HMSO

Asbestos Regulations 1969 Statutory Instrument No 690,

HMSO

Health and Safety Executive Guidance Note: Environmental

Hygiene: 10 Asbestos

Department of the Environment, Waste Management Papers:

Number 18: Asbestos Wastes

Number 23: Special Wastes (Chapter 4), HMSO

‘Material Health & Safety Data Sheets’, Asbestos Information

Centre

A Mechanical Seal Guide io API 610 Standard, 7th edn, John

Crane Inc., New York (1990)

Summers-Smith, J D (ed.), Mechanical Seal Practice for

Improved Performance, Mechanical Engineering Publications,

London (1988)

Flitney, R K., Nau, B S and Reddy, D , The Seal User’s

Handbook, BHRA, Cranfield (1984)

Fern, A G and Nau, B S , Seals, Engineering Design Guide

15, Oxford University Press/Design Council, Oxford (1976)

Merry, S L and Thew, M T , ‘Comparison between a

hydra’dynamic disc seal and neckrings for a small process

pump running in water and in mercury’, 9th B H R R

International Conference on Fluid Sealing, 1981, Paper H2,

Neale, M J (ed.), Tribology Handbook, Butterworths,

Londmon (1973)

p 333

Warriing,‘R k, Seals and Packings, Trade & Technical Press

Limited, London (1967)

CIBSE Guide, Volume A , Design Data (1986)

CIBSE Guide, Volume B , Installation and Equipment Data

(1986)

ASHRAE Handbook, Fundamentals (1989)

ASHRAE Handbook, H V A C S y s t e m and Applications (1987)

Threshold Limit Values and BioLogical Exposure Indices f o r

19884989 American Conference of Governmental Industrial

Hygienists

BRE Digest 119, Assessment of Wind Loads (July 1970)

BRE Digest 210, Principles of Natural Ventilation (B B

Ductwork Specification DW142, HVCA (Heating and

Ventilation Contractors Association)

CIBSE Technical Memorandum TM8, Design Notes for

Ductwork Industrial Ventilation, A Manual of Recommended

Practice, 20th edn, American Conference of Governmental

Induijtrial Hygienists (1984)

I S 0 7730: 1984, Modern thermal environments -

determination of the PMV and PPD indices and specification

of the conditions for thermal comfort

Fire Paper 7, ‘Investigations into the flow of hot gases in roof

venting’, HMSO (now available from Colt International Ltd)

(1963)

Fire Paper 10, ‘Design of roof venting systems for single

storey buildings’, HMSO (now available from Colt

International Ltd) (1964)

BS 7346, Components for smoke and heat control systems

Part 1: Specification for naturai smoke and heat exhaust

ventilators

BS 7346, Components for smoke and heat control systems

Part 2: Specification for powered smoke and heat exhaust

3 Economic use of fired space heaters for industry and

4 Compressed air and energy use

5 *

6 *

7 Degree days

8 The economic thickness of insulation for hot pipes

9 Economic use of electricity

10 Controls and energy savings

11 The economic use of refrigeration plant

12 Energy management and good lighting practices

13 The recovery of waste heat from industrial processes

14 Economic use of oil-fired boiler plant

15 Economic use of gas-fired boiler plant

16 Economic thickness of insulation for existing industrial

17 Economic use of coal-fired boiler plant

18

19 Process plant insulation and fuel efficiency

20 Energy efficiency in road transport

commerce

buildings

*New titles in preparation

49 Energy Audit Series, No 18

50 Payne, G A The Energy Managers’ Handbook, Westbury

51 NIFES, Energy Managers’ Handbook, Graham and Trotman,

52 Energy Manager’s Workbook, Energy Publications, Cambridge

53 Energy Manager, Maclaren Publishing, Croydon (monthly)

54 Energy Management, Department of Energy (monthly)

55 CIBS Guide, Chartered Institution of Building Services,

The Engineering Industries, Department of Energy (1984)

House, London (1980) London (1985) (1982)

London A3 Thermal Properties of Building Structures

A4 Thermal Response of Buildings

56 Murphy, W R and McKay, G Energy Management,

Butterworths, London (1982)

57 CIBS Code for Interior Lighting, Chartered Institution of

Building Services, London

58 Lyons, S L Handbook of Industrial Lighting, Butterworths,

63 Angela, M., Vibration Monitoring of Machines, Technical

Review No 1, 1987, Bruel & Kjaer, Denmark

64 Broch, J T., Mechanical Vibration and Shock Measurement, Bruel & Kjaer, Denmark (1976)

65 International Standards Organization, I S 0 237213, Vibration Severity Standards

66 International Standards Organization, I S 0 3945, Vibration Severity Standards

67 Thomson, W T and Rankin, D., ‘Case histories of on-line rotor cage fault diagnosis’, Conf on Condition Monitoring

1987 McGraw-Hill, New York (1971)

68 Beronek, L L (ed.), Noise and Vibration Control,

69 Burns W., Noise and Man, John Murray, London (1968)

70 Kerse, C S , Noise, Oyez Publishing, London (1975)

Yroductrvity and welding e

16.3.2 _ - Soldering and brazi I

rg 1oiuu

Large-chip metal removal 16/3

streamer itself presents disposal problems, frequently wrapp- ing itself round the workpiece, the cutter or parts of the machine tool, creating a hazard to both the process and the operator

Chip breakers are extensively used to induce continuous chips to break into short lengths which are relatively safe and can be easily disposed of These push against the underside of

the chip and cause it to curl into a tight spiral, the free end of which strikes against the tool, and the resulting bending stress causes fracture The earliest form of chip breaker, still extens- ively used with flat-top tools, consists of a hard wedge-shaped block of sintered carbide clamped to the rake face of the tool about 2 or 3 mm from the cutting edge (Figure 16.2(a)) The introduction of disposable sintered carbide inserts has allowed more complicated rake-face geometries to be used which act

as built-in chip breakers (Figure 16.2(b)) Effective chip breaking is largely a matter of trial and error, being influenced

by the feed, tool bluntness and cutting speed as well as by the material being machined

The development of new and improved cutting materials has enabled a hundredfold reduction in cutting time to be achieved since the beginning of the twentieth century Unfor- tunately, the reduction in idle time, caused by the need for tool adjustment, and in the tool approach and retraction times before and after cutting has not been of a comparable order

In achieving lower production costs the emphasis has now

rightly moved away from furthe; reducing cutting time to

attacking the disproportionately large amount of non-cutting time

Disposable sintered cutting inserts are made to a high level

of precision which allows them to be indexed or replaced in tool holders in a few seconds, usually without the need for

sizing cuts When all the cutting edges are worn the inserts are discarded, obviating the need for time-consuming regrinding

which is common practice when using high-speed steel tools

When resetting lathes between work batches a substantial time saving can now be achieved by using preset tools mounted in holders which can be replaced as cartridges in the tool post

No doubt further improvements will be achieved in the development of cutting tool materials but reduction of manu- facturing time in future will be determined mainly by reducing idle time, both by better tool changing mechanisms and by improved machine tool design to facilitate chip disposal and to reduce the tool approach and retraction times

The shaping group of machine tools produce chips by a relative linear motion between the cutting tool and the work

It includes shaping machines, planing machines and slotting machines, all of which are used mainly for tool manufacture or maintenance work and have little application in modern production They operate on a reciprocating principle, cutting

16.1 Large-chip metal removal

16.1.1 Large-chip processes

A11 the large-chip processes use cutting tools of defined

geometry which are applied in a controlled manner to remove

metal at a predetermined rate The processes could be classi-

fied in many ways, but it is convenient to consider them in

terms of the kinematics of the inachine tools With this in

mind, they have been separatcd into four main machine

Turning machines embrace the wide variety of lathes and

vertical boring machines which can be controlled manually or

automaticially Automatic control can be achieved using cams,

sequential controllers, hydraulic copying devices or numerical

programming All machines in this group are capable of

performing six basic operations as shown in Figure 16.1 In

addition, copying lathes and numerically controlled lathes can

generate mon-parallel forms by traversing the tool simulta-

neously in two planes

Most turning processes use tools with a single cutting edge

where the cutting action is characterized by a relatively

uniform section of material being presented to the cutting

zone resulting in a continuous chip when cutting ductile

materials or a repetitive form of short discontinuous chips

when cutting brittle materials Although the production of

continuous chips indicates an efficient cutting action the chip

Shaping (reciprocating tool or work)

Drilling and boring (rotating tool)

Chip

Figure Basic lathe operations Figure 16.2 Chip breakers: (a) clamped, (b) built-in

on the forward stroke and idling on the return stroke Al-

though they have quick-return mechanisms the cutting time is

only in the order of half the reciprocating cycle time Swarf

disposal is usually no great problem due to the intermittent

nature of the cut Figure 16.3 shows typical configurations of

these three machine types, which have changed little in recent

years

Other machines in the shaping group are gear shapers and

gear planers, outlines of which are shown in Figure 16.4 In

gear shaping the cutter resembles a side-relieved spur gear,

the involute profile being generated by rapid reciprocation of

the cutter while slowly revolving the cutter and gear blank in

synchronism Helical gears can be generated using a cutter

with helical teeth and applying an appropriate helical motion

to the spindles Gear shaping is used for producing gears when

hobbing would be impossible due, for instance, to a turned

shoulder close to the involute profile Gear planers have little

modern use, particularly in a production environment The

cutter is in the form of a straight-tooth rack, suitably relieved,

and the gear is generated by reciprocating the cutter and

moving the gear blank and the cutter at a constant speed To

enable a short rack to be used it can be removed from the cut

and indexed back at intervals

Broaching machines also belong to the shaping group, but

these produce the required form in a single pass Internal

broaching is for opening circular holes to produce non-circular

forms The cutter is a broach which has a number of cutting

edges along its length and which is usually drawn (but some-

t o paper

Figure 16.3 Outlines of reciprocating machines: (a) shaping, (b)

planing, (c) slotting

Saddle Cutter

Gear blank Face plate

0.05-0.08 mm, so the number of cutting teeth is determined

by the form to be produced

Push broaching is limited to broaches with a small number

of teeth which have a low lengthkross-section ratio and which would not buckle under compression Surface broaching is of more recent introduction, and is used as an alternative to milling for the production of external surfaces Surface broaches are rigidly clamped to a machine slide and traversed against the component being machined, producing a surface in

a shorter time than is required for milling, and usually giving a superior finish Whereas internal broaching is usually the only feasible method for producing the desired shapes, surface broaching is an alternative to milling and is usually justified only if the quantities required are sufficient to absorb the high equipment and tooling costs

The milling group comprises a large range of manually operated or numerically controlled machines, many of which can perform operations such as drilling, reaming and boring as well as the accepted milling operations Milling cutters gene- rate surfaces either by means of cutting edges on the periphery

or the face of the cutter Peripheral milling is now seldom used for generating large plane surfaces, its main use being for machining slots or profiles Although peripheral cutters can be fitted with carbide cutting edges the majority are of high-speed steel and, except when used for machining the more exotic hard materials, will probably continue to be so in the foresee- able future Frequently they also have shallow teeth on the cutter face, although these teeth usually contribute little to the

total metal removed A range of typical peripheral cutters is

shown in Figure 16.5

Face milling cutters are essentially for generating plane surfaces They are fastened to the end of stub arbors in the machine spindle and their configuration makes them suited to the use of specially designed carbide inserts (Figure 16.6) The cutting edges of both peripheral and face milling cutters are in contact with the uncut part of the component for, at most, half

Figure 16.5 Peripheral cutters: (A) high radial rake cutter, (B) helical cutter, (C) side and face cutter, (D) end milling cutter, (E) slot drill

Face milling cutters: (A) zero corner angle, (B) corner angle

16/6 Manufacturing methods

a revolution Since the chip length is in the order of one third

of the length of the uncut surface the chips have a maximum

length approximating to the cutter radius, so chip breaking

poses few problems

Gear hobbing machines also belong to the milling group

Hobs are in the form of a screw with a straight side rack-form

thread, gashed to give cutting edges and relieved to provide

cutting clearance Gears are generated by rotating the blank

and the hob in synchronism and the hob is fed parallel to the

axis of the arbor on which the blank is mounted When a

number of identical gears are required, several blanks can be

fitted to the same arbor and machined at a single pass

The machines in the drilling and boring group can be

sub-divided into drill presses, radial drilling machines, jig-

boring machines and horizontal boring machines They have

rotating spindles which hold drills, stepped cutters, taps,

reamers or single-point boring tools, and the cut is applied by

feeding either the spindle or the work table Typical configura-

tions within this group are shown in Figure 16.7

Increased metal-removal rates, made possible by the devel-

opment of new cutting materials, have forced machine-tool

manufacturers to design new machines capable of large ranges

Column

1 Spindle head

Column

e D Movement perpendicular (d 1

Flgure 16.7 Machines in the drilling and boring group: (a)

single-spindle drill press, (b) radial drilling machine, (c) gantry-type

of spindle speeds and feeds This has necessitated more powerful motors and structures having high rigidity to resist the increased cutting forces and to reduce the likelihood of self-induced vibrations, giving rise to chatter

With increased rates of metal removal the problem of swarf disposal has become more acute The magnitude of this problem can be visualized when it is realized that swarf occupies about one hundred times the volume of the metal removed, so a 10 kW motor running at full power can generate about 1 m3 of mild steel swarf or about 3 m3 of aluminium swarf per hour

Modern high-production metal-cutting machines commonly cost in excess of 5100 000 If they are amortized over a period

of 5 years the depreciation cost of such a machine when used

continuously on a double-shift basis is more than E5 per hour

It follows that such plant requires high utilization, efficient programming to produce at optimal metal-removal rates, an effective system of tool management to reduce non-cutting time and intelligent application of terotechnology to minimize lost time due to maintenance

Where large-scale production justifies continuous or large- batch manufacture the achievement of these objectives be- comes a feasible possibility Unfortunately, few products are marketed in such large quantities and a large-batch approach usually results in uneconomically high stocks British manufac- turing industry in the 1970s became notorious for its inflexibil- ity and for the disproportionately high stocks which pursuit of large-batch policies entailed Present policies are directed towards small batches, and this manufacturing philosophy has highlighted the need for rapid change-overs and for manufac- turing systems accommodating large numbers of tools which can be called into use in response to the demands of small batches

on the end of a thin-walled tube

In practice, the cutting tools usually approach the work obliquely and have rake angles in both directions on the rake face together with a nose radius at the end of the cutting edge The direction in which the chip flows across the tool surface is determined by this complicated geometry British Standard 1296: 1972 defines the angles on single-point cutting tools in terms of the normal rake system (Figure 16.9), based on two coordinate rake angles.' The back rake or cutting edge

inclination As is measured parallel to the cutting edge in the vertical plane and the normal rake yD is measured in a plane at

right angles to the cutting edge and perpendicular to the rake

face & is the tool approach angle and K: is the horizontal clearance angle, or the tool minor cutting edge angle In

addition, the tool is relieved to give vertical clearance angles

of about 5'

Other systems of tool nomeclature relate the rake angles to the coordinate axes of the tool shank, or to the cutting edge, measuring the angles in each case in the vertical plane Although these systems are conceptually simpler, they are of little use in deducing the direction of chip flow The British Standard relates to single-point tools but it can also be applied

to multi-point tools and is generally preferable to the other systems

Large-chip metal removal 16/7 16.1.3 Cutting-tool materials

The variety of cutting-tool materials has increased rapidly in recent years due to the development of more difficult-to- machine materials and to the insistent demand for higher productivity Any attempt to summarize these materials is unlikely to be completely successful due to the rate at which

improvements and innovations are occurring However, this is

no excuse for ignoring the published state of the art at the time

of writing

When selecting a tool material for a particular application it

is necessary to measure its rating against the following list of properties, some of which are mutually opposed For instance,

in most cases hardness and impact resistance of competing tools tend to be inversely related The essential properties are:

I

2 High compressive strength

3 Adequate impact resistance

4

5

6 Low interface friction

7 Good abrasion resistance Temperatures at the chipltool interface can be about 1000°C

when machining steel, and considerably higher when machin- ing some of the more exotic materials, particularly if heat is applied to help soften the work material Since cutting temp- erature is largely dependent on cutting speed, it follows that pursuit of higher productivity creates increasing demands on the high-temperature properties of the tool materials

High hardness at elevated temperatures

Insusceptibility to violent local temperature variation Chemical inertness at working temperatures

Figure 16.8 Orthogonal machining

Elevation on X-X

16.1.3.1 High-speed steel

High-speed steels (HSS) are likely to continue to be used in the foreseeable future for many applications such as drilling, reaming, tapping and dieing, forming, broaching and milling due to the ease with which they can be shaped in both the soft and hardened state Typically, they consist of carbon steel alloyed with tungsten or molybdenum, together with percent- ages of chromium, vanadium and cobalt The alloying ele- ments raise the temperature at which tempering occurs, allowing HSS to be used at temperatures up to about 650°C Their hardness is limited to 750 HV adequate for machining most of the common metals, including alloy steels in their unhardened forms

Cutting speeds are necessarily limited to prevent excessive rise in temperature When machining mild steel, cuzting speeds of about 1.5 m/s are possible if a plentiful supply of

coolant is provided A recent development is the coating of

HSS drills with a deposit of about 3-5 pm of titanium nitride which allows rotational speeds to be increased, resulting in a

50% increase in penetration rate and longer tool life With a few exceptions of this sort it is unlikely that HSS will ever again pose a threat to the supremacy of sintered carbides for heavy-metal removal

Despite their relatively low hardness and susceptibility to softening at high temperatures, high-speed steel tools are tougher than most of the competing materials, enabling them

to be used for interrupted cuts without fear of fracture They can also be reground, giving a number of cutting lives before they must be finally discarded

- I

I

X

Plan

Angles in the Normal rake system

These alloys, consisting of cobalt, chromium, tungsten and carbon, although less versatile than high-speed steels, enable cutting to be performed at higher temperatures Their main use is for drilling, where their superior hardness at elevated temperatures is an advantage when the application of fluids is

Manufacturing methods

frequently a problem Their cutting performance is generally

superior to high-speed steel but inferior to sintered carbides,

so they are unlikely to grow in popularity

16.1.3.3 Sintered carbides

The introduction of sintered carbides for cutting has been the

most important single contribution to increased productivity

during the past SO years They are essentially cermets, which

consist of hard carbide ceramic particles embedded in a metal

matrix Early carbide tools were usually of tungsten carbide

and cobalt Their brittleness encouraged the use of negative

rakes to promote compressive stresses and restricted their use

to continuous cutting Subsequent improvements in the sint-

ering process, and the introduction of alternative ceramic and

metallic components, has enabled a range of carbide tools to

be produced which withstand the rigours of thermal and

mechanical shock, making them suitable for interrupted cutt-

ing in turning and milling

The first generation of carbide tools consisted of sintered

tips brazed to steel shanks This provided a fairly rigid cutting

system and the tools could be reground when blunt More

recently, the brazed tip has been almost completely super-

seded by disposable tips which are mechanically clamped into

steel tool holders These tips are polygonal, with three or

more cutting edges which can be indexed when worn to expose

new edges The negative rake varieties can be inverted to

double the number of edges They are a throwaway concept,

regrinding being uneconomic Their introduction has forced a

reappraisal of metal-cutting economics, making tool lives in

the order of 15 min a desirable objective Practical cutting

speeds are about three times as great as could be achieved with

high-speed steel, S m/s being a typical maximum when

machining mild steel

16.1.3.4 Ceramics

Sintered ceramic tools based on aluminium oxide (A1203)

have been available for more than 30 years Their brittleness

and poor thermal shock resistance can be improved by addi-

tions of zirconium oxide and titanium carbide, but they are

generally unsuited to interrupted cutting They are suitable for

machining hardened steels and chilled cast irons, and are

similarly suitable for heat-assisted machining of hard materials

such as the nimonic alloys and Stellite However, their brittle

behaviour has proved a severe limitation for general-purpose

machining of steels, an area in which their high metal-removal

potential would have been an advantage

Mixed ceramics, based on carboxides with dispersed tita-

nium carbide, have achieved better impact resistance without

significant loss of hardness They can be used at high cutting

speeds in the order of 15-20 m/s when operated at low feeds,

making them suitable for finish turning and finish milling hard

materials

A recent addition to the range of commercially available

ceramic cutting tools uses silicon nitride (Si3N4) with differing

levels of aluminium and oxygen substitution The silicon

nitride ceramics have good resistance to thermal and mech-

anical shock, enabling them to be used for discontinuous

cutting They can be operated at higher cutting speeds than

carbides together with higher feeds than other types of cera-

mics Their main applications to date are for rough turning

and rough milling grey cast iron and for turning nickel-alloy

steels Due to their high thermal shock resistance they can be

used to cut dry or with coolant, the latter method being

inadvisable with other ceramics Chemical action occurs at

cutting temperatures, causing rapid wear when machining

most steels, so at present there is little likelihood of silicon

nitride supplanting carbides in this important area of manufac- ture

16.1.3.5 Cubic boron nitride (CBN}

Polycrystalline cubic boron nitride is another comparatively recently introduced addition to the range of metal-cutting materials It has a hardness considerably in excess of ceramic tools and retains this hardness at temperatures well in excess

of 1000°C The main application is for machining hard ferrous materials at very high cutting speeds, giving a surface finish comparable to grinding It has relatively good impact res- istance, allowing it to be used for interrupted cutting It is also

of use for hot machining of refractory metals such as Stellite, where a 90% reduction of machining time compared with carbide has been claimed Due to the high cost of CBN tools, the manufacturers do not claim great cost savings, but the time saving is very significant

16.1.3.6 Diamonds

Diamond is the hardest material known to humans and it has found a limited cutting application where this is an important attribute Natural diamonds, brazed to steel holders, have been used for many years for producing fine finishes on copper and aluminium Being monocrystalline, natural diamonds have planes of weakness which render them unsuitable for anything but fine finishing cuts

The production of polycrystalline synthetic diamonds

(PCD) has extended the usefulness of diamonds by improving their impact reistance PCD cutting tools are now extensively used to machine abrasive aluminium-silicon alloys, fused silica, and reinforced plastics They are chemically reactive at high temperatures, so they are of little use for machining ferrous materials

16.1.3.7 Limitations imposed by machine tools

Industry has, until recently, been very reluctant to replace machine tools while they continue to perform the function for which they were purchased Machine-tool manufacturers market machines which adequately utilize the cutting tools available at the time of purchase Inevitably, with the rapid development of new cutting materials, the existing machine tools cease to provide a service which uses the cutting tools in

an economic manner

Self-induced vibration, giving rise to chatter, is undesirable

in any cutting operation It is particularly undesirable when using brittle cutting tools where catastrophic failure can be- come a very real possibility The problem of self-induced vibration becomes more acute as metal-removal rates increase

at high cutting speeds Ideally, resonant frequencies should be

as high as possible, but this requires high structural stiffness and low mass Unfortunately, dynamic stiffness tends to be directly related to mass, so a simple scale-related solution does little to reduce chatter The solution, if it exists, lies in structural redesign to enhance stiffness without a proportional increase in mass

Optimal cutting speeds using modern cutting tools require a large increase in the rotational speed of spindles and a corresponding increase in input power Manufacturers of machine tools recognize this need, which is reflected in their latest designs, together with improved provision for handling the greater volumes of swarf which are produced

16.1.4 Cutting fluids

Cutting fluids are used for three main purposes: as a lubricant

at low cutting speeds, to cool the tool and work, and to assist

Large-chip metal removal 16/9

in clearing the swarf At cutting speeds in excess of about

0.7 m/s thiere is little noticeable lubricating effect Below this

speed extreme pressure (EP) mineral oils containing sulphur

or chlorine additives can be used to reduce friction in the latter

stages of chip/tool contact due to the formation of low-shear-

strength sulphides or chlorides The balance of the cutting

forces is affected, giving rise to a larger shear angle and

reduced contact length, and encouraging the thinner chip to

curl, making for more effective chip breaking High compress-

ive stresses near the tool point prevent lubricant penetrating in

this area, so the lubricating effect is limited to the latter part of

the chipitool interface where sliding friction occurs The

chemical reaction giving rise to low-strength compounds is

both temperature and time dependent Hence, at higher

cutting speeds the lubricant rapidly loses its effectiveness

High-speed steel cutting tools start to soften at temperatures

above about 650°C When using these materials the cooling

effect of cutting fluids enables the tools to be operated at

higher cutting speeds than would be possible when cutting dry

Water-soluble oils, having high specific heat and good metal-

wetting properties are better coolants than the mineral oils

used as lubricants With carbides and ceramics the poor

resistance to thermal shock makes the use of cutting fluids

inadvisable except in special circumstances Fortunately, these

materials can be used satisfactorily at high temperatures, and

coolants are therefore not usually required

The purely mechanical function of using cutting fluid to

assist swarf disposal is sometimes of prime importance An

example of this application is in deep-hole drilling, where

cutting fluid is pumped to the cutting edges at high pressures

of about 6 Nimm'

16.1.5 Forces and power in metal cutting

Most lathes and milling machines lack the power to exploit the

cutting tmools in an economic manner This shortcoming is

usually aggravated by a natural trepidation on the part of

operators to run machine tools near their power limits for fear

of stalling the drive motor It is surprising that few machine

tools are fitted with wattmeters, so operators usually have no

idea how near they are to causing an overload

Among the more sophisticated numerically controlled

machines, very few are fitted with adaptive control devices

which cause feed or cutting speed to respond to excessive

power demands The vast majority operate from a predeter-

mined program which has been based on safety considerations

where the power requirements are well within the rated output

of the motors

There is an ill-founded belief that the cutting forces, and

hence the power required, increase significantly as the tools

wear, increases of 40% sometimes being quoted In fact,

cutting power seldom increases by more than 10% over the life

of the tools

Cutter

rotation

At rated power, transmission losses usually account for

about 30% of the input power, with a correspondingly greater percentage loss when operating at lower energy levels Tran- smission losses are higher when the machine tool is cold, and drop significantly over the first half-hour of operation It is desirable, therefore, to record the transmission power over the full range of cutting speeds and feeds on a machine tool in both the cold and warmed-up conditions Only then is it possible to know the available power which can be used for cutting

Although a knowledge of cutting forces is desirable to prevent excessive structural loads, the main reason for wishing to know is as a basis for estimating power The power in watts is simply the product of the peripheral speed in metres per second and the tangential cutting force measured unewtons '

Due to the formation of a built-up edge when cutting steel at low speeds, forces on the tool vary in an unpredictable manner, but above about 2.5 mis become relatively constant when built-up edge ceases to have a significant effect At cutting speeds below about 0.7 mis the lubricating effect of

cutting fluids can do much to inhibit build-up, and on the rare occasions when such low speeds are used the cutting force can often be reduced by this means When using sintered tools it is usually possible to operate at cutting speeds high enough for the forces to be considered constant

Peripheral milling removes metal by means of teeth on the circumference of the cutter It is seldom used to produce large flat surfaces, which are more effectively generated by face milling Mostly, peripheral milling cutters are used for end milling slots or for producing slots or stepped surfaces by using one or more horizontaliy mounted side and face cutters or

helical slab milling cutters

End milling seldom requires the rated power of the drive motor The limiting factor is usually the maximum recom- mended feed per cutting tooth which will prevent damage to the cutter Horizontal peripheral milling, however, can be limited by the power of the drive motor, and it is useful to consider the way in which the cutting parameters affect the power required

Peripheral cutting can be performed in either the upcut or

climb mode (Figure 16.10) In upcut milling the cutting edge must penetrate the previously cut surface before chip genera- tion commences This causes a high radial force at the commencement of the cut which does not happen with climb milling Cutters with large radial rakes have a weak tooth form which results in rapid wear when subjected to the high radial forces associated with upcut milling Radial forces experi-

Cutter rotation

\ \

and

16/10 Manufacturing methods

enced in climb milling are much lower and wear is usually not 16.1.7 Tool-life assessment

a severe problem

of

specific power can be achieved when climb milling with a high

Summarizing, the most economic performance in It is fortunate that, with few exceptions, tool wear occurs in a

predictable manner Although it takes different forms, each is radial rake cutter operated at large feeds It is preferable to

operate at high feeds rather than high cutting speeds since the

associated with a known

lar ranges Of feed and

which happens particu-

are

’peed’ The most

associated with crumbling of the cutting edge, cratering of tungsten carbide, plastic deformation of the tool, thermal and mechanical shock, or attritive wear on the clearance face

index of feed in the cutting power equation is less than unity

whereas power is directly proportional to cutting speed

16.1.5.3 Hot machining

Some work materials pose machining problems which cannot

readily be solved by conventional methods These include the

nimonic alloys and Stellite cast alloys To give some idea of the

problems which are encountered, some of the nimonic alloys

when machined with carbide tools on a 100 mm diameter bar

necessitate cutting speeds as low as 0.15 m/s and tool failure

commonly occurs after machining a 100 mm length of the bar

When the surface of the bar is preheated with a gas tungsten

arc or a transferred plasma arc struck between the electrode

and the work surface and using ceramic or CBN tools, cutting

speeds of about 2.5 mls are possible, and the tools remain

serviceable after machining a considerable length of bar

This technique is not one which would be advised if

alternatives are possible, but with some of the more refractory

metals now in use the hot-machining process is frequently the

only practical solution The surface preheat temperature is

about 6OO0C, giving such high interface temperatures that

carbide tools cannot be used

16.1.6 Surface-finish considerations

Built-up edge is one of the main factors contributing to poor

surface finish When machining most materials this can be

reduced (if not eliminated) by operating at high cutting

speeds Where finish cuts are required, the uncut chip area is

relatively small so the cutting power is never likely to be an

important consideration, even at very high cutting speeds If

ceramic tools are used the speed limitation is usually that

imposed by the available spindle speeds, but when using

carbides or high-speed steels the speed constraint is usually

that imposed by tool wear

The theoretical surface roughness in turning is determined

by tool plan geometry, a pointed tool operated at a given feed

producing a rougher surface than one having a nose radius

The surface generated by a peripheral milling cutter is

directional in property In the direction of feed the theoretical

surface is geometrically similar to that for a turned surface, the

cusps having a radius equal to that of the cutter, and the pitch

between cusps being equal to the feed per cutting tooth Due

to the almost inevitable lack of straightness of the arbor on a

horizontal milling machine, the contour generated by the

cutter teeth varies as some teeth take a greater depth of cut

than others In severe cases one tooth may take such a

disproportionately deep cut that the surface generated has a

periodicity corresponding to the feed per revolution rather

than the feed per tooth, and the cusps are correspondingly

deeper

Face milling usually produces a finish superior to that

generated by peripheral cutters The geometry of the cutter is

specially designed so that the combination of corner angle,

end cutting-edge angle and nose radius produce very flat

cusps In addition, due to the use of carbide cutters, cutting

speeds are much higher than those achievable using HSS

peripheral cutters

16.1.8 Economics of metal cutting

The main financial objective of a manufacturing company is maximization of return on capital This implies a knowledge of

profitability which, in turn, requires a knowledge of cost and selling price Production engineers are concerned with the processes needed to make components which are eventually assembled into finished products Selling price, therefore, is not usually a very useful statistic for the process planner He

or she must settle for sub-objectives such as minimum cost or, sometimes, maximum output which, although not synony- mous with profitability, at least contribute to its achievement When discussing manufacturing economics these are the ob- jectives to which we must address ourselves

Metal cutting is an intrinsically wasteful operation, involv- ing the removal of large quantities of material Although there are no reliable figures to support this contention it is probable that only about 70% of the material purchased is contained in

finished parts, the balance being expensively converted to swarf which has a very low resale value Intelligent design can

do much to increase material utilization but material wastage will always be a significant proportion of the total component cost In spite of the attractiveness of contending production options such as metal forming, it is inevitable that cutting processes will continue to be extensively used The subsequent analysis assumes that due cognizance has been taken at the design stage of the importance of material utilization, and the cost factors include only the direct cost of manufacture and its associated overhead The operating cost, taking account of direct labour, machine depreciation and factory overhead, may well be in the order of $220 per hour

The cost of manufacture per component, K , can be divided into five parts:

1 Set-up and idle time cost per component, K1

2 Machining cost per component, K 2

3 Tool-changing cost per component, K 3

4 Tool-depreciation cost per component, K 4

5 Tool regrinding cost per component, K S

where K = K 1 + K2 + K3 + K4 + Kg Assuming that dis-

posable inserts are used, there is no regrinding cost, so K5 can

be ignored

The setting cost can be substantially reduced by using preset tooling However, with the current trend towards small batch sizes, the setting cost ascribed to each component will increase proportionately The idle time per cutting cycle is composed of loading and unloading time in addition to the tool approach and tool retraction times before and after machining has taken place Set-up and idle time can therefore contribute signifi- cantly to production cost, and its reduction is frequently the largest single factor in cost minimization

Machining cost is directly related to cutting time which, in turn, is dependent on cutting speed and feed The use of large feeds and high cutting speeds reduces machining cost but decreases tool life and, consequently, increases both the unit

Large-chip metal removal 1611 4

process parameters, but a typical boundary is shown in Figure

16.12, where the maximum width of cut under stable machin-

ing conditions decreases as the spindle speed increases A

large tool approach angle and nose radius increase the effect- ive length of the cutting edge, which is tantamount to increas- ing the width of cut and decreasing the uncut chip thickness If

chatter is likely to occur it is obvious that tool approach angle and nose radius should be kept as small as possible, a recommendation which conflicts with recommendations as regards tool life

The influence of feed on the stability boundary is more significant but this does not appear to have received the attention it deserves Figure 16.13 shows how the critical width

of cut increases as feed increases, making the case for using large feeds to oppose the onset of chatter.’

tool depreciation and unit tool changing costs It is therefore

the minimization of the total of these three costs which

determines minimum production cmt for any given set-up

Figure 16.11 illustrates by way of a carpet plot how K

typically varies with both feed, f, and cutting speed, v The

main point worth noting is that within the broad range of feeds

and cutting speeds selected the minimum cost OCCUKS inside the

speed range and at the maximum value of feed

A great deal of research has been published on the stability

boundary between width of cut and spindle speed.8 The

boundary envelopes vary with the machine tool and the

f2.5

Figure 115.1 1 Production cost related to cutting speed and feed

Spindle speed (rev/s)

Sure 16.12 Variation of stability threshold with spindle speed

Feed, f

Figure 16.13 Variation of stability threshold with feed

16/12 Manufacturing methods

In conclusion it is pertinent to stress again the importance

of large feeds in achieving not only chatter-free performance

but also low specific cutting power and minimum-cost machin-

ing

16.2 Metal forming

16.2.1 Introduction

Metal forming, i.e changing the shape of the material without

actually removing any part of it was practised at least 3000

years ago in Egypt, where hammer forging to produce gold

sheet, cut subsequently to make wire, is recorded in the Bible

to have taken place Rolling in wooden mills was employed to

manufacture papyrus Manual swaging and wire drawing were

well established in the Middle Ages but naturally, were

limited in scope by the power available It was only with the

advent of the Industrial Revolution that progress was made

and processes like extrusion and cross and longitudinal rolling

became available But even here, the restrictions imposed by

the low quality of tool materials, lubrication problems and the

lack of understanding of the basic precepts of plasticity

impeded progress until, in some cases, well into the twentieth

century

T h e ever-increasing demand for high quality products-

often of sophisticated shape in difficult to process materials

economically produced, fabricated or semi-fabricated, com-

bined with the rising cost of metallic engineering alloys has

focused attention on metal-forming processes and techniques

T h e emphasis here lies on the ‘chipless’ approach to shap-

ing This provides an economical direct means of converting a

cast ingot to slab, plate, billet or bloom and then-in another

chipless operation-of changing these basic shapes into pro-

filed finished or semi-finished products T h e avoidance of the

removal of the material during a forming operation enhances

the economics of the process by reducing wastage associated

with the swarf-producing machining Whereas the latter has,

of course, a very considerable and necessary role to play in the

range of manufacturing activities, its indiscriminate use (a

feature of the early years of plentiful supply of cheap labour

and materials) is no longer acceptable when high tonnage of

accurately manufactured product can be obtained at a much

lower cost

In the most simplistic terms, the desired change in shape is

effected either in the cold, warm or hot state (the latter below

the melting point of the material) by the application of

external forces, pressures or torques of sufficient magnitude to

induce plastic flow and thus a permanent set, of the material

through the forming pass Depending on the operation, the

material is forced to flow between driven rolls, through (or

into) open or closed dies, or between sets of dies and rolls

Solid or hollow sections are thus produced from the initially

solid blocks of metal

T h e standard basic operations are:

1 rolling (flat, oblique or longitudinal),

2 extrusion (axisymmetrical or asymmetrical)

3 drawing (solid or hollow components),

4 sheet forming (deep drawing, bending, pressing or bulg-

ing),

5 forging (solid and hollow sections) and

6 cropping (shearing and piercing)

Within the compass of any of these operations, a number of

variants exists which reflects not only a variety of manufactur-

ing routes and subroutes, but also the nature properties and

characteristic responses of the processed materials Modern

metal-forming technology makes use of solid and semi-solid (‘mashy’ state), and superplastic, as well as explosively pre- welded metallic composites and dynamically compacted parti- culate matter Mixtures of metallic and/or ceramic and poly- meric materials are formed to manufacture composites of very specific properties T h e problem of forming these into desir- able shapes presents the engineer with new and often difficult situations t o solve Selection of the appropriate forming process, the tool design, the effects of the pass geometry on the final physical and mechanical properties of the product, the dimensional accuracy, and the achievement of the as near

as possible final shape in the minimum of operational stages have to be faced

The apparently simple sequence of ingot-slab-semifabri- cate-finished product becomes complex unless there is good understanding of the basic characteristics of the individual processes and an appreciation of the principles of the theory of plasticity, as well as that of the concepts of tool and process design T h e bases for and fundamentals of the major processes and technological developments are discussed in the following sections, but detailed treatment of the individual topics is only indicated by reference to the appropriate literature

16.2.2 Classification of processes

For a given application, the selection of the correct process necessitates the introduction of a criterion of process classifi- cation Since hot working homogenizes and refines the crys- tallographic structure of the material and thus, ultimately, improves its strength and toughness, whereas cold working increases strength, hardness, dimensional tolerances and im- proves surface finish, these temperature-induced effects are often used to differentiate between the various manufacturing methods

Important as the processing temperature is, in some cir- cumstances other criteria of classifying metal-forming pro- cesses may well be more appropriate From a purely manufac- turing point of view, quantity and shape may have to b e considered, while the likely response of the processed material

to the level and/or rate of stressing, as well as the manner of application of the forming load system, may offer a better clue

to the desirability or otherwise of using a particular technique

or operation

T h e parameters that characterize forming operations give rise to the following possible classification systems:

1 operational temperature (hot, warm or cold forming),

2 shape effect (bulk or sheet forming),

3 operational stress system,

4 operational strain rate,

5 starting material (ingot slab, billet, bloom, slurry, or

powder)

16.2.2 I Operatiotial-temperature criterion

The idea behind the subdivision into hot, warm and cold processing of materials is not only to indicate the nature of the operation, but also to draw attention to the plant and ancillary equipment needed, to the level of force parameters required, and to the likely metallurgical response of the processed material

A n outline of this classification scheme, including only the basic operations, is given in Table 16.1

Starting with a cast ingot, the primary hot operations of flat, billet and slab rolling, and slab forging will produce the starting stage for the secondary, further processing of the slab into plate, billet or a large forging These, in turn, will form the first step in the manufacturing route of a more sophisti-

Metal forming 18/13 Table 16.1 Classification of dynamic regimes

Primary load High or moderate Slow deformation Rapid loading or

Usual method Constant load or Conventionai Fast-acting

of loading constant stress hydraulic or hydraulic or

plastometers, low impact devices Dynamic Strain versus time Constant-strain-rate Machine stiffness

testing machine

High velocity impact or loading High velocity impact devices, expanding-ring technique, high- speed metal cutting Elastic-plastic wave propagation

Very high velocity

or hypervelocity impact Light gas gun or explosively accelerated plate

or projectile impact

Shock wave propagation, fluid-like behaviour

cated, profiled product Hot operations are carried out at

elevated temperatures exceeding annealing and normalizing

ranges and, consequently, yield a hot-finished product show-

ing a relatively Bow level of flow stress However, the force

parameters required match the mechanicai properties of the

material and are also relatively low It follows that the rate of

wear of the tooling can be kept at an economical level

especially if the lubrication problems are well under control

To improve the mechanical properties of the product, while

at the same time keeping the loading at a moderate level,

warm processing is used Here, the temperatures are well

above ambient but, equally, well below the hot-processing

range, ansd usually slightly less than for recrystallization The

increased material ductility is sufficient to reduce the power

requirement of the plant Cold-working conditions are con-

fined to ambient temperature and are characterized by a high

energy requirement-necessitated by large operational forces

and/or torques-but result in very high quality final product

displaying both good dimensional tolerances and mechanical

properties

A rouglh guide to the temperature ranges can be obtained by

considering the operational ternpera.ture/melting point ratio

On this scale, hot working takes place when the ratio is >0.6,

warm working when the ratio is 0 3 4 5 (the latter corresponds

to recrystallization conditions], and cold when the ratio is

<0.3

The effect of shape reflects the geometry of both the initial

and final component and, consequently, the nature of the

change imposed on it by the forming operation

A process in which a component of a relatively small initial

surface area/thickness ratio is deformed in such a way that the

ratio is increased, is often classed as a ‘bulk deformation

operation’ On the other hand, the component of an initially

high surface aredthickness ratio, shaped in a process which

does not impose any change in the thickness but effects shape

changes only, is said to be ‘sheet formed’ Any change in the

thickness of such a component can easily lead to tensile plastic instability and incipient localized yielding

Bulk processes are those of rolling, extrusion, forging and solid- and/or hollow-section drawing Bending, pressing, deep drawing, spinning and shearing are the main sheet-forming operations

16.2.2.3 Operational-stress system

Because of the inherent severity of many forming processes, particularly the rotary ones, a consideration of the type and property of the induced stress field is of primary importance The success of the operation may well depend on its compati- bility with the properties of the processed material

The presence of tensile and compressive stress fields results

in the appearance of shearing stresses which, in turn, lead to the sliding of molecular planes and, eventually, to the yielding and plastic flow of the metal Stress systems containing these components are most likely to give rise to plastic flow which, if

it is controlled, will produce the desired amount of deforma- tion

Purely compressive or tensile systems create conditions of hydrostatic pressure in a triaxial field (absence QE shear), or

produce shearing stresses in uni- and bi-axial conditions Clearly, since it is the configuration of the individual stress system that is indicative of the type of deformation which can

be expected, its assessment prior to choosing a forming system

is imperative These various possibilities are illustrated, diagrammatically in Figure 16.14

As an indication of the incidence of any of the stress systems, the following, non-exhaustive, list can be considered:

Figure 16.14 Process classification system based on operational temperature

Uniaxial tensionhniaxial compression:

1 between the rolls in roll forming, and

2 in the flange in deep drawing

Uniaxial tensionlbiaxial compression:

1 in the drawing die

Compressive stress systems

in the closed forging die,

near the die throat in extrusion of bar, and

under the punch in tube extrusion

between the rolls of a longitudinal rolling mill with no

front and/or back tension, and

in the upsetting, open dies

16.2.2.4 Operational strain rate

A number of engineering alloys and even some practically

'pure' materials, e.g commercially pure aluminium, are sus-

ceptible to the changes in the rate of straining Modern

technological techniques have either 'speeded up' conven-

tional processes-for instance, wire can be drawn at some

120 m min-l -or have introduced new ones that operate in truly dynamic conditions Impact extrusion, explosive form- ing, welding and compaction, and mechanically and elec- trically induced discharges of energy producing high strain rates, have all combined to introduce an entirely new field of

high-energy rate fabrication, known commonly as HERF

The range of possibilities arising in this context are listed in Table 16.2 which provides a detailed insight into the effect of different strain rates and the means of producing them in an industrial environment

Table 16.2 Mass-velocity relationship contributing to total kinetic energy

Forming system Mass contribution Velocity Vdocityimass

contribution

(m s-') (kg)

16.2.2.5 Starting material

Since some modern processes do not require bulk solids as starting materials, but utilize particular matter and semi-solid substances, a classification based on the initial physical state of the material offers an interesting alternative to the more conventional approach

Typical examples of unconventional starting materials are: 'mashy' state processing, leading to conventional rolling of

Metal compositse sandwich components; the Conform-type extru-

sion, starting with a powder or gr,anulated material, or an

exp!osive compaction of powders

16.2.3 Characteristics of the basic groups of processes

Of the major processes listed in Section 16.2.1, forging is the

most diverse and cannot therefore be described in more

general terms For this reason, the basic characteristics of only

four groups of processes are indicated here and those of

forging, sheet forming, cropping, etc., are discussed later

All rolling processes rely on the forces transmitted through

the roils to the material to effect deformation and on the

rigidity of the roll system for the dimensional accuracy of the

product

Sheet and plate are initially obtained from a slab by rolling

the slab in a relatively simple system (Figure 16.15) Driven

rol!s introduce the material into the roll gap, or working zone

of the pass, and reduce the thickness The success of any

further processing to obtain strip rather than a sheet or large

area of pllate, depends on the ability of the system to maintain

a constarit width of the processed metal and on reduction of

the thickness (this being equivalent to the reduction in the

cross-sectional area) These requirements call for a plane

strain operation which is possible only if the lubrication of the

pass is very efficient Processing in this mode can proceed in

either cold or hot conditions

A much more complex rolling system is that of longitudinal

rolling, which is employed in the production of axisymmetrical

billets, bars and hollows (Figure 16.16) A train of suitably

shaped rolls mounted on stands (either in pairs or in three-roll

configura.tions) inclined at right angles (between the success-

ive stands) is used, as shown diagrammatically in Figure

15.16(a) A gradual reduction in the cross-sectional area of the

material takes place (Figure 16.16(b)) as the specimen moves

axially forward through the sets of driven rolls While fully

engaged in the train, the processed material experiences,

additionadly, axiai tensions resulting from a differential distri-

bution of‘ successive stand velocities The ovality of the early

passes is slowly reduced along the train until the last stand is

reached Here, the final, circular cross-section is expected to

be achieved

An alternative to longitudinal rolling is offered by the

oblique-rolling system in which a single set of two or three

driven rolls produces tractive, frictional forces which propel

the specimen axially while, at the same time, causing it to

Figure 16.15 The principle of sheet and plate roiling

rotate The motion of an element of the worked material is thus forward, but helical

Figure 16.17 illustrates, using an example of tube rolling, the basic principle involved In this case, three profiled, driven rolls, disposed at 120” to each other, and inclined at an angle a (the feed angle) to the horizontal ail1 axis, and an angle p (the cone angle) in the vertical piane, introduce the bloom (sup- ported internally in the bore by a mandrel) into the forming pass The bloom is ‘sunk’ onto the mandrel in the zone AB and has its wall thickness reduced on the roll ‘hump’ BC Slight elastic recovery takes place along DE The bloom is thus elongated and its wall is thinned The amount of deforma- tion imposed depends on the size of the inter-roll opening or

the ‘gorge’

These basic characteristics of oblique rolling operations (the variants of which are discussed later) are common to all operations, as indicated, for instance, in Figure 16.18 This shows, diagrammatically, the operation of the so-called

‘secondary piercing’, or ‘oblique plug rolling’ of a tube-a process in which a long cylindrical mandrel is replaced by a short profiled plug

On the other hand, processes of profiling by rolling can take various forms, one of which is indicated in Figure 16.19 where

a stepped shaft, required to acquire a series of specific profiles, can be manufactured by oblique rolling in a single three-roll stand An operation in which the billet is rotated and fed through a system of driven rolls produces this effect

In another variant of oblique rolling, a two-roll system of helically ribbed rolls (Figure 16.20) will produce metal balls out of a solid cylindrical billet

These few examples illustrate the versatility of rolling operations, a more detailed discussion of which is given in Section 16.2.4

When the initial shape of the work piece has been imposed

on it by one of the processes described above, there often arises the problem of how to achieve a degree of further deformation leading, possibly, to the final product Drawing processes answer this need by providing a means of producing either solid (bar, rod or wire) or holiow tubular sections, either circular or non-circular in shape The drawing operation

is carried out in a d i e - o r a set of consecutive dies forming a tandem drawing system-into which the work piece, with a swaged leading end, is introduced (Figure 16.21) An axial force is applied through a gripping device (as indicated by the arrow in Figure 16.21) and the work piece is pulled through the die In the case of a solid specimen, the outer dimension only is reduced, whereas with a hollow section there is also a change in the wall thickness Lubrication of the working zone

of the pass (the part of the die surface along which the deformation is effected) is of importance from the point of view of the magnitude of both the drawing load and the induced drawing stress, and in view of the surface finish Similar results can be obtained in extrusion, a process in which the starting billet (sometimes referred to as the ‘slug’) is inserted into a cylindrical container and is then pushed mecha- nically through a suitably profiled die (Figure 16.22) There is

a number of variants of this process (see Section 16.2.6); but the two basic operations are those of forward (or direct) and inverse (or backward) extrusion In the forward extrusion a solid moving ram is brought into direct contact with the billet and activates the latter by moving it axially forward through the die In inverse extrusion a hollow ram is in contact with a movable die which bears onto the billet, firmly held in the container When the pressure exerted by the tooling is suffi- ciently high to exceed the yield stress of the material, plastic flow is initiated and backward extrusion into and through the hollow ram takes place

Considerable control over the dimensional accuracy can be exercised in such systems but, again, solution of the lubrica-

Roll axis

-

Metal forming 16/17

Figure 16.18 Oblique, tube rolling on a plug (secondary piercing)

Figure 16.19 Three-roll shaft shaping

Figure 16.22 Direct and inverse extrusion

tion problem is of importance In this latter context, hydrosta-

tic extrusion (to be described later) provides an important

alternative to the conventional arrangements indicated here

A large group of ‘unorthodox’, dynamic processes in-

troduces a number of new elements and opens new opera-

tional possibilities of using materials which are sometimes

difficult to process and of reducing manufacturing costs by

dispensing with heavy plant and equipment

The high-energy-rate processes stem essentially from the

usually overlooked fact that the working of metal requires

energy and not merely the application of force, and that, in

addition, the rate of dissipation of energy is of importance A

simple consideration of the basic equation for kinetic energy

shows that a comparatively small change in the velocity of a body will have a more pronounced effect than will a change in its mass A typical conventional system approaching the conditions of high-energy forming, i.e drop-hammer forming,

is limited in its usefulness by the necessity of using large masses and, therefore, unwieldy and costly equipment The sources of energy used in the high-velocity systems are chemical explosives, electrostatic and magnetic fields, and pneumatic-mechanical devices The basic processes are those

of forming (shaping), welding and powderiparticulate-matter

compaction A variety of forming systems exists, each display-

ing specific characteristics associated with either sheet or tube forming, for which it is intended

16.2.4 Rolling processes and products

Traditionally, both routes start with a cast ingot (Figure 16.23)

which is then rolled down to slabs (route 1) by cogging In route 2, cogging again leads to the production of a bloom (a product of over 10 cm2, or equivalent, in cross-section), and then to either a variety of small flats or large rounds or,

through a billet mill, to a billet (a product of cross-sectional area less than 10 cm2) However, very satisfactory develop- ments in the area of continuous casting have led to the introduction of casting machines into these cycles In the new, fully automated and computer-controlled, high-productivity works, continuous casting of slabs has to a great extent eliminated the cast ingot

In route 2, in a modern mill the stress is on the use of continuous billet casters (in preference to bloom casters), thus eliminating one stage of the production line Where blooms are still required, normal practice is to employ two or three strands of material which are then rolled in two or three passes

to produce blooms With smaller sizes of billet, up to six strands can be cast

It is clear from Figure 16.23 that the manufacture of a wide range of either semifabricates or finished products calls for a

variety of mills and plant settings A very brief review of these

is provided here but, again, detailed information can only be obtained from the Further Reading at the end of this chapter Basically, the process, whether hot or cold, begins with the preparation of stock such as an ingot (in older plant) or continuously cast bloom or billet In hot operations this is followed by heating in a strictly controlled atmosphere and temperature, and then rolling proper Finishing of the work piece includes a number of operations such as cutting, cooling and, very often, straightening In cold operations, which are used to enhance the mechanical properties of the material and improve dimensional accuracy, the ancillary equipment con- sists of furnaces for heat treatment and plant for surface finishing Whereas modern plant comprise not only the rolling mill(s) proper, but also a number of pieces of ancillary equipment concerned with the preparation of the material prior to and post rolling, interest centres mainly on the actual mill since the dimensional quality of the product will depend mainly on its performance

According to their actual functions, rolling mills are sub- divided into the following classes:

1 cogging mills (production of blooms, billets and slabs from ingots, where these are still used);

production of plate and strip; or production of billets, bars, rod, sheet, tube or sections