BRITISH STANDARD BS 1134-2:

1990

Assessment of surface
texture
Part 2: Guidance and general
inforrnation

UDC 62 r.9.015:620. 179. 1 18(083.93).00 r.8

ffi^rffir&
ItL-E
K@-WW
No copyrNc wrrf,our Bsr pE3l[ssroN ExcEpT As pERMTTTED By copygrcnT LAw Mn#qdnF ilX
BS 1134-2:1990

Committees responsible for this

British Standard
The preparation of this British Standard was entrusted by the General
Mechanical Engineering Standards Policy Committee (GME/-) to Technical
Committee GME/10, upon which the following bodies were represented:

Department of Trade and Industry (National Engineering Laboratory)

Department of Trade and Industry (National Physical Laboratory)
GAMBTCA (BEAMA Ltd.)
Gauge and Tool Makers'Association
Institution of Production Engineers
Ministry of Defence
University of Warwick

This British Standard, having

been prepared under the
dilection of the General
Mechanical Engineering
Standards Policy Committee,
was published under the
authorit5r of the Boald of
BSI and comes
into effect on
31 July 1990

o BSI 11-1999

BS 1 134 first publiehed

December 1950
Second edition April 1961 Arnendrnents issued since publication
First published as BS 1134-2
August 1972
Second edition July 1990

The following BSI references

relate to the work on this
standard:
Committee reference GME/10
Draft for comment 88/71611 DC

ISBN 0 580 17819 6

 BS 1134-2:1990

Contents

Page
Committees responsible Inside front cover
Foreword ii
0 Introduction 1
1 Scope I
2 Geometry of surfaces 1
3 Measurement 4
4 Control methods 5
5 Results obtained from common production processes and materials 12
6 Guidance on suitable cut-off values L2
7 The need for control ofsurface texture L2
8 Costs 13
I Procedure for the measurement of surface roughness parameters 13
10 Calibration and adjustment of stylus instruments 15
Appendix A Procedure for surface roughness inspection 17
Figure 1 Influence of direction of measurement on the effective
spacing of- profrle crests
Figure 2 A surface texture representing the combined effects
-
of several causes
Figure 3 textures having the same maximum height
- Surface
but differing in geometrical profrle
Figure 4 Analysis of surface texture chart showing the
- of longer spacing errors
elimination
Figure 5 Relationship of stylus point to the actual profile
-
ofthe surface
Figure 6 Behaviour of the stylus when traversing ridges
-
and grooves 8
Figure 7 Relationship of skid to wavelength of texture 8
-
Figure 8 Effect of different horizontal magnifications on
- profile
the surface 10
Figure 9 Types of surface profile having the same height of
departure- from the nominal profile 10
Figure 10 Types of surface profile having the same spacing and
same average- height 11
Table 1 Sampling lengths for the measurement of .Bu of
- profrles (e.g. ground profrles)
non-periodic L2
Table 2 lengths for the measurement of R, of
- Sampling
non-periodic profrles (e.g. ground profrles) t2
Table 3 Sampling lengths for the measurement of .Eu and B, of
periodic -profiIes (e.9. turned profrIes) 12
Table 4 Surface roughness values produced by common
- processes and materials
production t4
Publications referred to Inside back cover

oBSI 11-1999
BS 1134-2:1990

Foreword
This Part of BS 1 134 has been prepared under the direction of the General
Mechanical Engineering Standards Policy Committee and replaces
BS 1134-2:1972 which is withdrawn. BS 1134 was first issued in 1950 and
revised in 1961 and 1972.
BS 1134 consists of two Parts. Part l covers methods for the assessment of
surface texture parameters including instrumentation and Pafi 2 provides
guidance and general information.
It is hoped that Parts 1 and 2 of BS 1134 will provide a general level of
understanding of the subject.
BS 6393 refers to calibration specimens for use with stylus instruments used for
the measurement of surface texture and gives detailed information for practical
assessment.
A British Standard. does not purport to include all the netessary provisions ofa
contract. Users of British Standards are responsible for their correct application.
Cornpliance with a British Standard does not of itself confer imrnunity
frorn le gal obli gations.

Sumrnary of pages
This document comprises a front cover, an inside front cover, pages i and ii,
pages 1 to 18, an inside back cover and a back cover.
This standard has been updated (see copyright date) and may have had
amendments incorporated. This will be indicated in the amendment table on the
inside front cover.

o BSI 11-1999
 BS 1134-2:1990

0Introduction The complete texture of any surface is, therefore, a

combination of irregularities of various kinds and
In practice, all manufactured surfaces depart to magnitudes ar"ising from different causes.
some extent from absolute perfection and it has long
been recognized that careful finishing of It is convenient to recognize four principal orders of
components can lead to longer life, improved geometrical ilregularity, resulting from the
efficiency, functional interchangeability and other following:
benefits. This is reflected throughout all fields of a) machine tool inaccuracies, deformation of the
industry. workpiece due to cutting forces, effect of
At one time the terms "rough machine", "medium workpiece mass, etc.;
machine" and "fine machine", or equivalent symbols b) vibrations.of any kind, e.g. chatter marks;
were used on engineering drawings, leaving the c) the inherent action ofthe particular production
surface to be controlled by limitations of the process;
machining process involved and the arbitrary
d) rupture of the material during chip removal.
opinions of operator and inspector which, all too
often, did not coincide. Problems resulted which It follows that an ideally complete assessment
became increasingly acute as the demand for a more would involve the measurement of each and every
comprehensive specification increased to keep pace departure from the ideal surface plus an assessment
with technological development. of the effect of the combined texture on the
functioning of the component surface.
Most of the uncertainties of specifying surface
requirements have been eliminated by development Measuring techniques and knowledge of surfaces
of instruments for the measurement of surface are constantly improving but, in state-of-the-art
texture on a numerical basis and by the issue of measuring, a simple method of approach, as given in
various national standards. However, the standards Part 1, is realistic and likely to remain in use for
embraced much that was arbitrary and in need of some years to come.
wider agreement and for some years the whole By implementing Part 1, the complex problems of
question has been under study within the assessment which are three-dimensional in
International Organization for Standardization character are reduced to two-dimensional
(ISO). Some very important areas of international metrology. This is done by confining individual
agreement have now emerged and have been measurements to profrles of plane sections taken
incorporated in Part 1 of this standard. through the surface. On surfaces where the texture
is multi-directional in character, the plane chosen
1 Scope for the purpose of measurement is usually
This Part of BS 1134 gives general information and immaterial, but on turned, ground and other
guidance on the methods of assessing surface surfaces having definite directional scratches or
toolmarks forming the texture, totally different
texture given in BS 1134-1.
readings could be obtained from measurements
NOTE The titles ofthe publications teferred to in this standard made in different planes (see Figure 1). The term
are listed on the inside back cover.
"lay" is applied to the directional marks constituting
2 Geometry of surfaces the topography of the surface and a further piece of
simplification (see 2.1.4 of BS 1134-1:1988) is to
The imperfections of any surface take the form of a confi.ne measurement to a plane approximately at
series of peaks and valleys which may vary both in right angles to the predominant direction of lay.
height and spacing and result in a texture which, in There are exceptions to this rule but they are
feel or appearance, and in properties generally, is comparatively rare.
often characteristic of the process employed in its
production. Surfaces produced by cutting tools tend
to be of uniform spacing with well-defined direction
of tool marks whilst those produced by grinding
processes, though usually directional in character,
are generally ofirregular spacing. Surfaces
produced by the reciprocating action of abrasives,
such as honing or lapping, tend to be less directional
in character.

o BSI 1 1- 1999
BS L134-2:1990

(a) Various directions of measurement on a surface

,l.!Lf \ /L/ -,\-,\ \-/

F---{ F---{ F---{ F---{
ABCD E

(b) Profiles of the surfaces shown in (a)

Figure I Influence of direction of measurement on the effective spacing of profile crests

-
Having conveniently reduced the process of When specifiiing a surface to be achieved,
assessment to an analysis of the form of the proflle specification of the sampling length relative to
as revealed by a plane cross section through the numerical height value limits the maximum spacing
surface, Part I seeks to controlthe spacingofprofile of irregularities and indicates the meter cut-off to be
irregularities to be taken into account in any one used in assessment of the product specified. In
assessment. The ir:regularities on any surface effect, the meter cut-off causes the instrument to
commonly represent the joint effects of roughness filter out and thus omit from an average reading
arising from the inherent action of the cutting (taken over a number of sampling lengths) errors
process [see item c) of this clause] and waviness having a spacing greater than envisaged. Thus, if
attributable to vibration, machine deflection and the interest lies primarily in roughness, a short
similar, phenomena [see items a) and b) of this sampling length would be quoted with similar,
clause]. These may be superimposed on departures attendant, instrument cut-off length but detection
from true geometrical form and thus the spacing and measurement of any waviness present (on
and, hence, the sampling length chosen to make an which roughness might be superimposed) would be
assessment of surface texture is of vital importance. revealed by selection of a greater instrument cut-off.
Indeed, careful study of Figure 2 shows that a In speaking ofthese features, it is not only
misleading height value could be obtained for the convenient but precise to associate the term
profrle should the length of sur{ace (sampling "sampling length' with the actual surface (or its
length) selected for evaluation of the sudace be representation on a drawing), leaving the term
inadequate. "cut-off'to apply to the instrument only.
Control is achieved by laying down a standard series
of sampling lengths which are now internationally
agreed (see clause 3 of BS 1134-1:1988). Various
instruments for the measurement of surface texture
have meter cut-offs which correspond with the
standard sampling lengths.

o BSI 11-1999
 BS 1l-34-2:1990

Waviness

Emoc of form

Figure 2 A surface texture representing the cornbined effects of several-causes

-
A further piece of simplification can now be effected. It has thus far been established what should be
Experience has shown that the majority of surfaces revealed by an ideally complete assessment of
subject to control can conveniently be assessed over surface characteristics and what will be revealed by
a sampling length of 0.8 mm when roughness only is the simplified approach of BS 1 134- 1. That Part also
to be taken into account. This value is, therefore, gives the procedures which have become
that most popularly used as instrument cut-off for internationally accepted for evaluating surface
assessing the finer surfaces. It is not possible, characteristics. As a basis, BS 1134-1 gives the
however, to assign any particular spacing as being methods of assessment from a graphical recordingof
the demarcation between roughness and waviness the surface profrle but, in practice, modern
because the scale ofeach ofthese features depends instruments provide a metered reading
upon the class of work and the manufacturing automatically of one or more of the parameter
processes employed. values relative to a defined reference line within the
Iimitations of the selected instrument cut-off. The
more sophisticated instruments provide a chart
record of the surface profrIe in addition to a metered
reading of the selected parameter. To ensure
uniformity of instrument performance it is therefore
necessary to state the basic requirements with
which the various designs of instruments shall
comply. These features are the subject of
international agreement and are given in section 3
ofBS 1134-1:1988.

o BSI 11-1999
BS 1134-2:1990

3 Measurement Thus, various values can be obtained for the ,Ry

value ofthe surface according to the length of
Surface textures commonly met with in practice are
surface selected for measurement. It might seem
complete in type and often represent combined
that the length of surface which most fairly
effects due to several contributing causes. For
expresses the quality ofthe surface texture is the
instance, see Figure 2 which shows greatest, i.e. 13. That this may not be so is
diagrammatically a surface containing roughness, demonstrated in Figure 3 in which profiIe A
waviness and error of form. If the measurement is represents, diagrammatically, a finely finished
confined to a short sampling length of the surface, 11,
surface having a large error of form, while profrIe B
the value obtained for the maximum height will represents a more coarsely finished surface having
be .Ry1 (maximum height being considered here for
rro error of form. The maximum heights are the
demonstration purposes). This is a measure of the same, but the surfaces are obviously far from
roughness neglecting the occasional deep scratches;
identical in quality.
it is nominally the same for all parts of the surface
and it neglects the irregularities of greater spacing.
As the sampling length is increased, however, the
height will eventually increase until, for a length 12,
it reaches a new value By2, which takes into account
the waviness, but which still ignores errors of
geometrical form. Finally, for the whole surface, if
such a measurement is practicable, sample
length'/3, would give a value, Ry3, including all
irregularities and errors.

Figure 3 Surface textures having the sarne maxirnurn height but differing
- in geometrical profile

oBSI 11-1999
 BS 1134-2:1990

An obvious method of measuring the roughness 4 Control methods

regardless of the other ir"regularities in Figure 2 is
4.1 Instruments
to limit the measurement to a sufficiently short
Iength ofthe surface. In the case ofthe finer 4.1.1General
surfaces, however, this length becomes very small Instruments are available for the assessment of
(of the order 0.8 mm) and the difficulty is then surface texture which operate on mechanical,
encountered that, even within a relatively small mechanical-electrical, pneumatic, or optical
area of the surface, the measured value over a principles. At present, ISO agreements and national
sampling length may vary considerably from point standards throughout the world are based on
to point. Such variations should not be confused instruments of mechanical and electrical design
with true variations of texture in different parts of since these are host commonly used.
the surface, as they are merely the incidental result
The proper application ofthe technique of surface
of the method of analysis and have to be smoothed
texture measurement requires a knowledge of the
out. This can be done, as shown in Figure 4 by
mode of operation and the limitations of the
taking, as the true value, the mean of a number of
instrrrments employed. The importance of making
observations. The observations may conveniently be
sure that electrical instruments are not misused by
taken in a row along a short length ofthe surface.
applying them to surfaces for which their meter
Thus, ifthe roughness does not exceed the given
cut-offs are inadequate will be apparent. When
spacing, e.g.0.8 mm, the graph may be divided into
nothing is known about the specimen, the first
successive sections 11, 12, ls, etc., each 0.8 mm long
safeguard against this misuse is to have a good look
and the average height ofeach section found
at the surface before taking a reading; dominant
separately. Taking the mean value from a few
(e.g. five) consecutive sections will usually suffice to spacing can often be seen and measured directly
eliminate the effect of variations between the with a scale. Traces can then be taken, using an
appropriate device for generating the datum. Some
individual sections. The effect ofthe procedure is to
electrical instruments are provided with means for
eliminate from profile A the more widely spaced
selecting the meter cut-off values, arld with these an
components of the texture and to measure only
extra safeguard lies in taking a second reading
those components the spacing of which is less
using a longer cut-off value. If there is no material
than 0.8 mm, as shown in the modified profile B,
increase in the reading, it can be assumed that no
which is obtained by rearranging the sections with
significant irregularities just beyond the range of
their individual axes in line. Each section will
the frrst off-cut are present on the surface.
constitute one sampling length to which the meter
cut-off is equivalent. 4.1.2 The pich-up
It is clear, therefore, that in order to obtain Most instrrrments in general use for the
consistent results from surface texture measurement of surface texture are designed to
measurements, the meter cut-off has to be specified, respond to the irregularities ofthe surface through
as well as the ft value. It is for this reason that the the agency of a stylus, which rests on the surface
use of standard meter cut-off is essential. The meter and is traversed across it. Movement is measured
cut-off selected for this purpose should be sufficient relative to a selected datum. The ideal datum would
to give a satisfactory average ofthose components of have the nominal shape of the surface, but the
the texture which it is desired to control, but not so datum most commonly used is provided by a skid
great as to include other components of wider which has a relatively large radius of curvature in
spacing. The meter cut-off at 0.8 mm is considered the direction ofthe traverse. This also rests on the
to be suitable for a large proportion ofordinary surface but follows its general contour, riding over
engineering work (see Table 2). the crests of smaller irregularities without
responding to them individually. The movements of
the stylus normal to the surface, measured relative
to a datum corresponding to the path followed by the
skid, are recorded by the instrument.

o BSI 1 1- 1999
BS 1134-2:1990

Figure 4 Analysis of surface texture chart showing the elimination of

- longer spacing errors
There are two limitations on the response achieved The second limitation is that the error introduced by
by the stylus and skid device. First, the stylus the skid will depend on its radius in relation to the
cannot have a mathematically sharp point, but has height and spacing ofthe texture. Ifthe radius ofthe
to be finished to a small but definite size. The profile skid is too small it will tend to ride up and down over
of a stylus having a rounded tip is shown by the the peaks and valleys representing the components
dotted outline at A in Figure 5. It is evident that of wide spacing in the texture instead of correcfly
such a stylus would just reach to the bottom of the following a path parallel to the general contour of
valley in the profiIe immediately to the right of it. the surface, see Figure 7. Ifthe distance A to A
Had the actual valley in the surface been deeper the represents the skid on a surface which includes a
stylus could not have penetrated to its full depth. In series of openly spaced ridges, it will rise and fall
fact the record itselfaffords no evidence ofthe through a distance d as it traverses the surface, and
non-existence offine deep scratches in the actual the record may be falsified to this extent. The
surface. The behaviour ofa rounded stylus in magnitide of the error introduced into the record
passing over a surface with sharp grooves and also depends on the position of the stylus relative to
ridges is illustrated in Figure 6. The full depth of the the skid. If the stylus is located so that it touches a
groove is reduced by an amount, b, in the record valley when the skid is at its lowest point, the depth
profrIe and the trace is also smoothed out over the of the record waveform will be reduced by an
sharp corners and crests. Apart from the reduction amount equal to d, but if the stylus is located at a
in the total depth, however, these defects of point half a spacing (or any odd multiple of half a
reproduction would be practically unobservable in spacing) ahead of or behind the lowest point of the
the ordinary record with the horizontal scale skid, as indicated by the distance S' to S in Figure 7,
compressed relatively to the vertical scale in then, as can be seen from the diagram, the depth
Figure 8. Further, in an average reading, the loss in recorded will be increased by an equal amount. Any
the valley may be offset by the gain resulting from intermediate condition may, of course, occur with
the apparent rounding of the crests. various differences in the actual form ofthe record
Fortunately, exp'erience shows that errors due to the produced.
finite radius of the stylus are not so serious as might Some instruments have two skids, touching the
at first sight appear and that the irregularities on surface with the stylus acting midway between
most surfaces tend to be relatively shallow, them. The chances of meeting extreme phase
compared with their spacing, so that the records conditions are then reduced and in practice such
obtained from them are not usually subject to instruments operate satisfactorily with skids
serious limitation from this cause. It is desirable, having a smaller radius than is needed for the single
none the less, that the stylus should be as sharp as skid (see Figure 4 of BS 1134-1:1988).
is practicable. A nominal radius of.Z pm is specified
for profrle recording instruments, while a nominal
value of 10 pm radius is acceptable for instruments
slving average readings only.

o BSI 1r-1999
 BS 1134-2:1990

For closely spaced textures, errors due to the skid It is clear that ifthe pick-up is traversed across the
can generally be neglected. The risk ofinaccurate surface at constant speed, s, each value ofthe
recording of more widely spaced features increases, wavelength,.l, if recurrent, gives rise to a definite
however, as the square ofthe spacing, and sets an frequency, /, which equals the speed divided by the
effective limit on this form of datum device. For spacing, i.e. f = s/,1,. Thus, at a speed of 2.5 mm/s, a
spacings too great for the skid, two other devices are spacing of 2.5 mm produces fundamentally an
possible. One is to use a guide in the form of a shoe alternating current of 1 Hzl) while spacings of,
with a highly polished face shaped to conform to the e.g. 0.8 mm and 0.25 mm produce currents of 3 Hz
general contour ofthe surface. The other is to use a and 10 Hz respectively. If the respective waveform
guiding mechanism, providing a datum for the is other than sinusoidal, in addition to the
stylus movement, with a shape conforming to the fundamental frequency, alternating components of
nominal profile of the surface, independently of any higher frequency than the fundamental appear.
actual contact with it. These componentp can be accounted for by a
4.1.3 Electrical measuring instrufitents mathematical proposition known as Fourier's
theorum which shows that the form of any recurrent
Many electrical measuring instruments can provide profile can be expressed as the sum ofa series of
graphs and meter readings but some are destined pure sinusoidal components having appropriate
mainly for the workshop and provide meter amplitudes and phase relations. The fundamental
readings only. wavelength is that of the profile, and the
To ensure a high degree of fidelity a recording wavelengths of the other components are successive
instrument should have the sharpest stylus that is simple fractions, e.g.t/2, Yz, t/e, Lll. The integrating
practicable, together with adequate means for instmment, in effect, resolves the form of the profile
providing the datum and a t5pe of pick-up that is under examination into its sinusoidal components,
responsive to sustained displacement of the stylus, sifts these components by means of the wave filter
so that everything found by the stylus is properly and recombines those which pass through it in such
reproduced. a manner as to produce automatically an
Instruments having generator type pick-ups tend to alternating current representing the irregularities
be velocity sensitive and respond to amplitude only coming within the meter cut-off, this curuent being
over a limited pass band with the result that long finally integrated by the meter.
waves are suppressed and steps are distorted. Thus, if the whole profile of Figure 2 is examined by
Meter instruments are designed to exclude the an instrument having a meter cut-off equal to lr, the
wider spacings and transmit to the meter only the alternating current emerging from the filter and
spacings coming within the pass band terminated passing to the meter would look substantially like
by the meter cut-off. This is generally accomplished the curve in this figure representing the roughness
by making use of the properties of generator t5pe alone. If the meter cut-off is equal to 12, tLle current
pick-ups and/or of electric wave filters which passing to the meter would have the shape of the
transmit or reject alternating currents according to curves representing roughness and waviness
their frequency. together.
The wave frlter consisting of a group of electronic
components serving no other purpose than to form
the filter or having some other part to play
(e.g. coupling circuits or even the pick-up itself in
the case of generator transducers) can be so
proportioned as to attenuate below the cut-off
frequency. If the instrument is to have a meter
cut-off, e.g. 0.8 mm, the filter circuits are designed
to transmit to the meter all frequencies higher than
those produced by 0.8 mm spacing at the intended
speed oftraverse and to attenuate all frequencies
corresponding to a wider spacing.

1)
I Hz = I cycle per second.

o BSI r1-1999
BS 1134-2:199O

r.<d

Figure 5 Relationship of stylus point to the actual profile of the surface

-

'-\1--",
/\,/ --j-l-
/\ l:Ji
i r-
/\
1---1)--l -'

Figure 6 Behaviour of the stylus when traversing ridges and grooves

-

Figure Relationship of skid to wavelength of texture

7
-
Although it is usual to speak of the cut-off as though It is convenient to express the cut-off in terms of
it occurs abruptly, in practice the transmission falls spacing rather than frequency because it is the
off gradually. Standard transmission limits are spacing that is a basic characteristic ofthe surface,
given in Table 5 of BS 1134-1:1988. the resulting frequency depending on the speed of
traverse which may be anything the instrument
designer chooses as being suitable for the pick-up.
It is clearly important to operate pick-ups at the
speed the instrument designer chooses. In the case
of motor driven instruments this will be ensured
automatically but in the case of manually operated
pick-ups the proper speed must be stated and
observed.

o BSI 11-1999
 BS 1134-2:1990

4.1.4 Profile record.ings It is notsuggested that all or any ofthese

An example of a trace of the cross section of a short diagrammatic forms are likely to occur exactly in
length of ground surface as obtained with a stylus practice, but approximations to most of them, as
instrument is shown in Figure 8. well as many other more complicated forms
undoubtedly do. The functional properties ofsuch
Since the irregularities of the surface which it is differing surfaces may vary greatly and hence arises
desired to examine are minute in depth, it is the impossibility of expressing the complete
necessary for graphical representations of them to roughness characteristic of a surface by means of
be shown to a considerable degree of magnification, any single number. It is for this reason that the R
often up to x 100 000. It will be appreciated that to value which serves to define its quality or grade
the same scale of magnifrcation, a sample of should be supplemented, e.g. by referring in
adequate length, even if only 0.25 mm long, would sufficient detail to the process of manufacture.
require a graph the length of which would be quite
unmanageable. Fortunately, the spacing of 4.1.5 Measurernetut of surfaces hauing greater
successive irregularities is usually much greater spacing of imegularities
than their height and it is therefore customary for While the majority of surfaces that have to be
instruments used for obtaining graphical measured in normal engineering practice have
representations of surface profiles to have different ir:regularities of less than 2.5 mm spacing, surfaces
magnifications for the vertical and horizontal with greater spacing can be found especially on
scales, the former being usually milled and planed surfaces and in the domain of
from x 10 to x 500 the latter. The extent to which waviness.
this alters the appearance ofthe trace is seen by When the surface is smooth and sufficiently
comparing the true trace of a short length of surface reflective (in which case the search will be mainly
taken at the same vertical and horizontal for waviness) direct interference methods can be
magnification, as shown in the upper trace of used, but in other cases stylus methods are Iikely to
Figure 8, with its condensed version shown in the be the most convenient.
Iower trace.
For satisfactory measurement, the stylus has to
Two principal quantities that can be noted on the traverse a sufficient length of the sur{ace relative to
graphs are the height of the irregularities a datum, the errors of which are small compared
(which may be measured in various ways, with those of the surface, and this can be
e.g. maximum height values or average values) accomplished by using an optical flat as a reference
and the separation oftheir peaks. surface.
A variety of differing surfaces may exhibit the same Sometimes the more closely spaced irregularities
height of departure from the nominal profrle. are of no functional relevance and can be removed or
See Figure 9. Figure 9(a) represents the profile of a frltered either by using a sufficiently large radius for
perfectly smooth flat sur{ace of, e.g. 150 mm length. the tip of the stylus or (in electrical instruments) by
Figure 9(b) shows the same length of surface, means of a wave frIter.
concave to the extent of 2.5 pm, and Figure 9(c) to
The relative magnitude of maximum height and B"
Figure 9(f) indicate waves
height, the former from about three to seven times
of 150 mm, 50 mm, 7.5 mm and 0.075 mm pitch
respectively, but all except Figure 9(a) have the the latter for the kind ofprofrles considered here
should be remembered when contrasting waviness
same height of.2.5 pm.
assessed in the one way and roughness assessed in
The surfaces in Figure 9(b) and Figure 9(c) have the other.
errors of geometric form but would appear smooth.
As the spacing is reduced the quality of the surface
changes although the same height of wave has been
maintained in each of the diagrams. This would
become noticeable to the eye. The errors in
Figure 9@) to Figure 9(d) are practically
undetectable in surface texture measurements and
would need to be determined by ordinary
metrological methods, whilst those of Figure 9(e)
and Figure 9(fl are revealed by a waviness recorder
and by nominal surface texture measurements
respectively. Consider also the difference between
the various forms shown in Figure 10 aII of which
have the same spacing, and the same average height
but vary extremely in character.

o BsI 11-1999
BS 1134-2:1990

l*
x5000 / 5000

Air

Metat

Air

Metal

xs000 /100
| ---
I
Figure 8 Effect of different horizontal magnifications on the surface profile
-

V.l4el al'./ / / / / / / /,/,/ // / //,/ /,/ / /,/ / / /,/ / / A (o)

{b)

.1,=7.5mm

Izspm
vryylv:-- -:
: :::::: : ::: ::--:::==72222 ( f I

'-ll-.-,1.= 0.075 mm lnot in proporiion)

Figure e Types of surface profile height of departure

- HH?||;i;*"

t0 o BSi 11-1999
 BS 1134-2:1990

Figure 10 Types of surface profile having the same spacing and sarne average height
-
4.1.6 The use of replicas The replicas will have an inverted profile. The
recorded trace from the replica may thus differ from
Surfaces inaccessible to the measuring instrument
that of the master surface, if, e.g. the master surface
can be measured indirectly via a replica. The choice
contains fine holes inaccessible to the stylus but
of replication method will depend on the answers to
which become accessible when inverted on the
the following questions.
replica.
a) Is very high frdelity required?
4.2 Roughness comparison specimens
b) Does the replica have to be flexible to be
released from the specimen surface? For roughness comparisons, see BS 2634.
Roughness comparison specimens are used to give
c) Is coating the specimen with release agent
engineers, draughtsmen and machine operators a
permissible?
comparison ofthe feel and appearance ofthe
d) Is the surface susceptible to damage or common surfaces with which they are dealing and to
modification? control production on a comparative basis. The
Whatever the method chosen, it should be assessed operator is given a specimen, or preferably two
for fidelity and general acceptability on a specimens, one having the assigned maximum value
representative test surface and results from the test and the other perhaps haUthat value. The product
surface and its replica shsuld be compared before is compared with these specimens by feel and
applyrng it to an important specimen. appearance or with the aid of an instrument
Examples of flexible replicas exhibiting reasonably remembering that in the latter case the life of the
good fidelity and generally not requiring the use of specimen may be much reduced.
release agents are cellulose acetate sheet moistened While greater accuracy is likely to be obtained by
and softened with a solvent such as acetone and comparing the product with a carefully checked
curable elastomers such as silicone rubber. Some specimen of the actual part, standard specimens
with convenience in storekeeping and distribution
long-range fidelity is lost because of the flexibility of
thereplica. Low instrument stylus pressures should are available.
be used. Where a drawing indicates a single roughness value
For rigid, high-fidetity replication, thermosetting this is considered as an upper limit. The details of
resins such as epoxies and polyesters are the most the process employed should be planned so as to
widely used and are available under a number of achieve a surface which is well within the limit in
trade names. These almost always require the use of order to provide adequate allowance for process
a release agent. For high-fidelity work, the replica variations.
system should be evaluated on the smoothest If an upper limit or a maximum value is stated the
sur{ace available not just on a test specimen of the surface is considered acceptable if not more
quality of the intended specimens. than 16 o/o of a\7 measured values exceed that
specified (see 8.2 and 8.3 of BS 1134'1:1988).

o BSI 11-1999 1t
BS 1134-2:1990

When instruments are used for comparison the Table 2 Sampling lengths for the
important requirement is that if the instrument rneasurement- of R"of non-periodic profiles
reading is in error it should be equally so both for (e.g. ground profiles)
specimen and product. But since the magnitude of R,
the error may depend partly on the characteristics up to and Sampling length I
Over
ofthe surface under test the question ofadequate including
similarity is still present. For example, in the case of pm l/m
stylus comparison instruments a blunt stylus may 0.10 0.08
read low on some surfaces but high on others while,
0.10 0.50 0.25
if the crest spacing of the surfaces being compared is
not the same, uncertainties in speed of traverse 0.50 10.0 0.8
(affecting the instrument's meter cut-off) and also 10.0 50.0 2.5
skid effects may upset the comparison. 50.0 200.0 8.0
200.0 25.O
5 Resultsobtained from conunon
production processes and materials Table 3 Sampling lengths for the
- of .R" and R" of periodic
measurement
Table 4 shows results in terms of -Ru values that can
profiles (e.g. turned profiles)
be expected from various common production
processes and materials. These are to be regarded as s_
a guide to designers and others who are urged to Up to and Sampling length I
Over
note that the true criterion in specifiiing actual including
requirements is successful and economic production
practice.
0.032 0.08
6 Guidance on suitable cut-off values 0.032 0.1 0.25
Table 1, Table 2 and Table 3 grve values of the 0.1 0.32 0.8
sampling lengths suggested as suitable for the 0.32 1.0 2.5
measurement of the B, and fi, parameters of
periodic and non-periodic profiles in cases where the 1.0 3.2 8.0
sample length to be used is not stated on the 3.2 25_O
drawing or in the manufacturing specification. NOTE Although the S- palameter in Table 3 is in essence S-
Table 1 Sarnpling lengths for the as defined in2.2.4 of BS 1134-1:1988 for the purpose of Table 3
-
measurement of R. of non-periodic it only needs to be estimated either directly fi'om the sur{ace or
from a graphical record of the profile.
profiles (e.g. ground profiles)
Ra
up to anct Sampling length J
7 The need for eontrol of surface
Over including texture
pm pm Control is generally instituted not with a view to
o.o2 0.08 seeking superlative finishing of components, but to
0.02 0.1 0.25 secure a surface texture ofknown type and
roughness value which experience has proved to be
0.1 2.O 0.8
that most suitable to grve long life, fatigue
2.O 10.0 2.5 resistance, maximum effrciency and functional
10.0 80.0 8.0 interchangeability at lowest cost; together with
attendant benefrts such as reduction ofvibration,
wear and power consumption.
There are two schools of thought on whether or not
all surfaces should be specified and controlled either
of which can be justified in a particular
circumstance.

T2
oBSI 11-1999
 BS 1134-2:1990

Some organizations seek to control the For each ofthe production processes there is a value
requirements ofevery surface on the grounds that of surface finish above which increases in
each has an essential surface pattern and order of permissible roughness have only a slight influence
roughness for function and/or economic reasons. on the finishing cost but below which the cost rises
Their motives are to secure all the functional steeply. Roughness values should not be placed on
benefits and at the same time to ensure that close drawings as the arbitrary opinion of the designer
control is only applied to those components which but should result from investigation taking into
merit any higher finishing costs involved. consideration all the factors including load,
In contrast, other companies apply control only to lubrication, materials, relative movement, speeds
essential parts leaving the remainder to be taken and temperature, etc. In the absence of full
care of by good machining and finishing practice information, experience with similar designs and
usually commensurate with the achievement of the processes is often a useful starting point. The use of
required dimensional accuracy. roughness compalison specimens as described
in 5.2 can also be helpful, particularly where
It is not a function of this British Standard to roughness requirements are specified for purely
advocate either approach; as stated, both have merit
cosmetic reasons, when it is often found that
and are valid under the right circumstances. For
reduction in surface finish demands of one or even
instance, it would be quite unmerited to institute
two grades have little adverse effect on the
the full control and inspection procedures of the
appearance of the surfaces produced.
aerospace industry in some areas of general
engineering.
9 Procedure for the measurement of
8 Costs surface roughness parameters
It is appropriate to mention here this all-important In order to decide whether or not a particular
consideration. It has already been mentioned that manufacturing process gives the required surface
the whole object is to prescribe and to obtain the finish, it is necessary to compare the value of the
surface topography and roughness most suited to roughness parameters of the workpigce surface with
the application. Failure to recognize and appreciate the requirements specified on the drawings or in
this philosophy has perhaps been the most common technical documents.
factor in delaying the widespread adoption of The surface roughness of the workpiece being
surface texture control as a useful tool to aid inspected can appear to be homogeneous or to be
production. Consequently, an enormous amount of quite different over various areas. This can be seen
production time has been wasted and continues to by visual examination of the surface. In cases when
be wasted on seeking too good a finish regardless of the surface roughness appears to be homogeneous,
cost or of functional need. roughness parameter values determined over the
whole surface should be used for comparison with
the special requirements.
If there are separate areas with obviously different
d e grees of surface roughness, surface roughness
parameter values which are determined on separate
areas should be used for comparison with the
specified requirements.

o BSI 1r-1999 13
BS 1134-2:1990

Table 4 Surface roughness values produced by cornmon productiorl

- processes and rrraterials

K.y, Iilll'lalllllllaveraseapplication V;'z-/77771 lesa frequent appli€tion

Bougboes values (pm ll")

Prces
50 25 12.5 63 1.6 OB 0.4 02 0.05 0.025 0.0125

Flame atting
Snagging

Sawirg

Planing, shaping

Invstment @ting

Ertrodiag

Cold rclling, drawing

Dic csting

NOTE. The rangs shou above are t)rpical of the preaeee liated, Higher or bwer vahree may be obtained uder epecial conditionE

t4 @BSI 1r-1999
 BS 1134-2:1990

For requirements specified by the upper limit of the The reliability of the decision as to whether or not
surface roughness parameter, those separate areas the surface being inspected meets the specification
of the surface shall be used which appear to have the and the precision of the mean value obtained for the
maximum roughness, i.e. the maximum values of surface roughness parameter of the same surface
the surface roughness parameter. For requirements depends upon the number of sampling lengths
specified by the lower limit of the surface roughness within the evaluation length over which the single
parameter, those separate areas ofthe surface value of the sur{ace roughness parameter is
which appear to have the minimum roughness obtained and also on the number of evaluation
should be used. Ifthe roughness ofthe surface being Iengths, i.e. the number of measurements over the
inspected or of some of its areas does not meet the surface. The minimum evaluation length is equal to
requirements of the drawing or specification, then the sampling length. The evaluation length
the method of visual evaluation or the method of including five consecutive sampling lengths
comparison with roughness comparison specimens (cut-offs) is taken..as the standard.
according to BS 2634-1, BS 2634-2 and BS 2634-3 The greater the number of measurements taken
can be used for the surface roughness inspection. In over the surface and the longer the evaluation
other cases methods of measurement using length, the greater is the reliability ofthe decision
instruments should be used. as to whether the surface being inspected meets the
For requirements specified by the upper limit of the specification and the higher is the precision of
surface roughness parameter, the surface is determination of the mean value of the roughness
considered to be acceptable if not more than 16 o/o of parameter. However, an increase in the number
all the measured values of the surface roughness of measurements Ieads to an increase in the
parameter exceed the specified value. In cases measurement time and the cost of measurement.
where the lower limit is specified, the surface is Therefore the inspection procedure should reflect a
considered to be acceptable if not more than LG % of compromise between reliability and cost. A
all the measured values of the surface roughness description ofone ofthe possible procedures for the
parameter can be exceeded by the specified value. inspection of the workpiece surface roughness is
For values specified by the maximum value of the given in Appendix A.
surface roughness parameter, none of the measured
values ofthe surface roughness parameter bfthe 10 Calibration and adjustrnent of
whole surface being inspected should exceed the stylus instruments
specification value. T\vo forms of calibration procedure are recognized.
NOTE In cases when the values of the roughness parameter of One is basic and complete but costly and involves
the surface being inspected are distributed according to the
normal law, the upper limit conforms with the limit determined
the direct evaluation of the magnification of
by the value p + o, where p is the alithmetic mean value of the recorded profrles, followed by accurate assessment
sudace roughness palametel, and o is the standard deviation of oftheir parameters from the records. With the aid of
these values. digital recording and computer techniques this form
To decide whether or not a workpiece surface is in has now been raised to a very high level. The other
accordance with the specification a set of single procedure, suitable for workshops and inspection
values ofthe surface roughness parameter, each departments, is based on the use of instrument
determined from a number of consecutive sample calibration specimens (see BS 6393). These
lengths constituting one evaluation length, should specimens in the range of values ideally required
be used. are still in a state ofevolution. It is for this reason
Using these measured values the mean value of the that the range of workshop and inspection
surface roughness parameter can be determined. department testing has so far been restricted to
those features which are under the direct control of
the user and which require to be checked and if
necessaly re-adjusted by the user after setting up
the instrument and as often thereafter as may be
necessary to maintain the standard of performance.

oBSI rr-1999 15
BS 1134-2:1990

Calibration should begin by ensuring that the stylus Calibrations should be made under the best possible
is in good order. Pending the evolution of conditions. Difficulties may sometimes arise,
instrument calibration specimens suitable for e.g. when there is too much floor vibration to allow
checking the sharpest tips, suffrcient inspection to for the use of an instrument with a skidless pick-up
ensure that the tip is not damaged can generally be at the highest magnification or when the skid
carried out with a microscope or projector. The tip normally fitted has too short a radius to give good
can then be viewed in side and end elevation at the results on a calibration specimen. Care should be
highest magnification available. A magnifrcation taken to reduce the danger of damage to the stylus
of x 100 to x 300 will generally suffice. or calibration specimen when lowering the pick-up
Calibration procedures have to be suited to the onto the surface of the specimen or on to any
features of the instrument being tested, an4 will contacting surface'
usually be defined in the maker's handbook.
Adjustment of the overall amplification, the
recording section of the instrument and of the
parameter meters, is effected by means of
potentiometers which have to be set in conjunction
with calibrated test specimens supplied with the
instrument by the maker. These adjustments are
needed on a pretty regular basis and when the
highest standard ofaccuracy is required this can
best be secured by calibrating the instrument
immediately before use for the next part of its
working range.

16 o BSI 11-1999
 BS 1134-2:1990

Appendix A Procedure for surface A.4.1 Where the indicated parameter symbol does
roughness inspection not contain the suffix "max." initially, the surface
will be accepted and the test procedure stopped if:
A.l General
a) the fi.rst measured value does not exceed 70 %
The following example illustrates one of a number of of the specified value indicated on the drawing;
methods of workpiece surface roughness
inspections. b) the fi.rst three measured values do not exceed
the specifred value;
A.2 Visual test
c) not more than one of the first six measured
Visually inspect workpieces to select those where it values exceed the specified value;
is obvious that inspection by more precise methods
is unnecessary, e.g.because the roughness is d) not more than two of the first 12 measured
obviously better or obviously worse than that values exceed the specifred value.
specified, or because a surface defect which Sometimes, e.g. before rejecting high valued
substantially influences the function of the surface workpieces, more than 12 measurements may be
is present. taken, e.g.25 measurements with up to four
A.3 Comparison test exceeding the specifred value.
Ifthe visual test does not allow a decision, tactile A.4.2 Where the indicated parameter symbol does
not contain the suffix "max." usually at least three
and visual comparison with roughness comparison
measurements are taken from that part of the
specimens should be carried out,
surface from which the highest values are expected
see BS 2634-1,2634-2 and BS 2634-3.
e.g. where a particular deep groove is visible or
A.4 Measuring inspection equally spaced if the surface gives the impression of
NOTE 1 The most reliable results of sudace roughness homogeneity.
inspection arr achieved with the help of measuring instruments.
Thelefore, the lules and procedures for inspection of the most
important details should be followed with the use of measuring
instruments fi'om the very beginning.
NOTE 2 Ifthe comparison test does not allow a decision to be
taken, measurements should be made on that part ofthe surface
on which the critical values can be expected according to visual
examination as follows-

o BSI r 1- 1999 t7
 BS 1134-2:1990

Publications referred to
BS 1134, Assessment of aerface textu,re.
BS 1134-1, Methods antd htstru;nentation^
BS 2634, Specification, for roughness contparisort specimens.
BS 2634-1, Specification. for tu,m,ed, ground, bored, milled, shaped and planed specirnens.
BS 2634-2, Specification. for sparlz eroded, shot blasted, grit blasted and polished specintens,
BS 2634-3, Cast su,rfaces.
BS 6393, Specification, for calibration of stylu,s irlstruments.

o BSI 11-1999
BS 1134-2:
1990
BSI British Standards Institution
-
BSI is the independent national body responsible for preparing
British Standards. It presents the UK view on standards in Europe and at the
international level. It is incorporated by Royal Charter.
Revisions
British Standards are updated by amendment or revision. Users of
British Standards should make sure that they possess the latest amendments or
editions.

It is the constant aim of BSI to improve the quality of our products and services.
We would be grateful if anyone finding an inaccuracy or ambiguity while using
this British Standard would inform the Secretary of the technical committee
responsible, the identity of which can be found on t\e inside front cover.
TeI: 020 8996 9000. Fax: 020 8996 7400.

BSI offers members an individual updating service called PLUS which ensures
that subscribers automatically receive the latest editions of standards.
Buying standards
Orders for all BSI, international and foreign standards publications should be
addressed to Customer Services. Tel: 020 8996 9001. Fax: 020 8996 7001.

In response to orders for international standards, it is BSI policy to supply the

BSI implementation of those that have been published as British Standards,
unless otherwise requested.

Information on standards
BSI provides a wide range of information on national, European and
international standards through its Library and its Technical Help to Exporters
Service. Various BSI electronic information services are also available which give
details on all its products and services. Contact the Information Centre.
TeI: 020 8996 7111. Fax: 020 8996 7048.

Subscribing members of BSI are kept up to date with standards developments

and receive substantial discounts on the purchase price ofstandards. For details
of these and other benefits contact Membership Administration.
TeL 020 8996 7002. Fax: 020 8996 7001.

Copyright
Copyright subsists in all BSI publications. BSI also holds the copyright, in the
UK, of the publications of the international standardization bodies. Except as
permitted under the Copyright, Designs and Patents Act 1988 no extract may be
reproduced, stored in a retrieval system or transmitted in any form or by any
means - electronic, photocopying, recording or otherwise - without prior written
permission from BSI.

This does not preclude the free use, in the course of implementing the standard,
ofnecessary details such as symbols, and size, type or grade designations. Ifthese
details are to be used for any other pulpose than implementation then the prior
written permission of BSI must be obtained.
If permission is granted, the terms may include royalty payments or a licensing
agreement. Details and advice can be obtained from the Copyright Manager.
BSI Tel: 020 8996 7070.
389 Chiswick High Road
London
W4 4AL