Assessment of Surface Roughness as per Indian Standards (Metrology)

There are various methods of evaluating the surface roughness but the most prominent and commonly used methods are the M [Centre Line Average method (CLA) or the mean line] system and the E (Envelope system) system. TheM system expresses the arithmetical average departure of the actual surface both above and below a mean line, within a specified sampling length, and in a plane substantially normal to the direction of surface. A similar system also
based on the mean line expresses the departure of the actual surface as a root mean squared value. CLA system is more popular than RMS value method. To eliminate the effects of secondary texture the profile is split into a number of sampling lengths and a separate mean line computed for each length. In this way, the mean lines of adjacent sampling length may not coincide even. The E system expresses the arithmetical departure of a surface both above and below a ‘mean’ curve. This mean curve is developed from a contacting envelope by displacing it to a position, where the areas enclosed by the profile above and below the mean curve are equal. The contacting envelope referred to above is obtained by rolling across the surface a sphere of radius r = 25 mm and displacing the locus of centre for this circle towards the surface by an amount equivalent to r.
The merits and demerits of the two systems (i.e., M and E) are given below. The M system is very widely used and all the instruments for measuring surface roughness available in the country are designed to measure in this system. It is more useful and a satisfactory means of controlling, at the point of production, the consistency of results from a process when the production parameters have been established.
This system has the limitation that it is unable to control the functional qualification of a surface whence associated with a machine process.
The E system, though, at present in its early stage of development is going to be widely accepted in future because of its better definition of peak measure and the fact that it is more easily applied to the surface finish instruments based on the interference principle.
M system. The surface roughness in this system is expressed by Ra value which is very useful for comparison of various surfaces obtained by similar operation ; of course, it does not indicate the limits of irregularity Rmax value (measure of peak value) which is perceived by looking at the profile and by touching the surface is probably the most direct method of expressing the surface roughness value. This value is also very useful in identifying the surfaces for comparison and record purposes but there is no direct relationship between Rmax and Ra value and as such Rmax value alone can’t be specified in drawings.

Information to be given in the Statements of Surface Roughness.

The various roughness grade numbers N 1 to N12 in 5 groups are specified as under by BIS. The relationship between the roughness grade numbers and the commonly used and generally accepted system of indicating surface roughness by symbol is given below.

Roughness grade number Roughness value Ra )im Roughness symbol
N12 50 -
Nil 25
N10 12.5 V
N9 6.3
N8 3.2 w
N7 1.6
N6 0.8
N5 0.4 vvv
N4 0.2
N3 0.1
N2 0.05 WW
Nl 0.0025


The values of the surface roughness expected from various manufacturing processes are indicated below.

Manufacturing Process Ra value in \un
Casting Sand casting
Permanent mould casting
High pressure casting
5 to 50 0.8 to 6.3 0.8 to 3.2 0.32 to 2
Hot working Hot rolling
Flame cutting, sawing and chipping
2.5 to 50
1.6 to 25 0.16 to 5 6.3 to 100
Machining Operation Radial cut-off sawing
Hand grinding
Disc grinding
Turning and milling
1 to 6.3 3.6 to 25 1.6 to 25 0.25 to 25 1.6 to 50 1.6 to 25 1.6 to 20 0.32 to 25 0.4 to 6.3 0.4 to 3.2 0.4 to 3.2 0.4 to 3.2
Grinding and
Surface grinding
Cylindrical grinding
0.063 to 5 0.063 to 5 0.025 to 0.4 0.012 to 0.16 0.04 to 0.16 0.04 to 0.8 0.16 to 0.32



This is the direction of the predominant surface pattern and is determined by the production method used. The surface roughness is generally measured across the direction of the lay.
The various directions of lays depending upon the production process can be as shown in Fig. 11.17.

Preferred Values of Ra andRz.

Preferred values for arithmetical mean deviation Ra in ^m are selected from :

Fig. 11.17. Lay.

Selection of Sampling Length.

For measuring surface roughness, the value of the sampling length I is selected from the following series depending upon the process of manufacturing:
0.08, 0.25, 0.8, 2.5, 8, 10 and 25 mm.

Process Suitable Sampling Length
Milling 0.8 2.5 8 10 _
Boring 0.8 2.5 8 10
Turning 0.8 2.5 .—
Grinding 0.25 0.8 2.5
Planing 2.5 8 10 25
Reaming - 0.8 2.5 — .
Broaching 0.8
Machined Diamond boring 0.25 0.8
Surfaces Diamond turning 0.25 0.8 ,
Honing 0.08 0.25 0.8
Lapping 0.08 0.25 0.8 —■
Super-finishing 0.08 0.25 0.8 ■ —
Buffing 0.8 2.5
Polishing 0.8 2.5
Shaping 0.8 2.5 8 10
Spark machining 0.8
Non- Burnishing 0.8 2.5 _ _
machined Drawing — - 0.8 2.5
surfaces Extrusion 0.8 2.5
Moulding 0.8 2.5
Electron-polishing 0.8 2.5

It is usual practice to choose shorter value of the finer and larger value for the coaser grade of a given process when more than one values are given.


It may be emphasised here again that the surface texture represents the combined effects of several causes, of which the roughness, waviness and the error of form are more predominant. If only the total height (i.e., the difference between the highest and lowest points on the plot) is considered then idea of above three characteristics can be had by considering different sample lengths. If we are interested only in the measurement of roughness of the surface then the sampling length should be neither too big as to include the waviness, nor too small as to ignore the occasional deep scratches. The best thing is, therefore, to limit the measurement to a sufficiently short length of the surface and in order to take into consideration the considerable variation of roughness from place to place, several readings at various places are taken. When the surface roughness has a directional quality, the readings should be taken in a direction perpendicular to the direction of lay as this way the effective spacing of the crests will be at short distances.. The effective spacing of crests goes on decreasing as the direction of measurement is reduced from 90° to direction of lay and in the extreme case when measurement is made parallel to the lay, a minimum result of roughness is obtained.

Determination of Ra Value.

The Ra value can be determined either by graphical methods or by direct reading instruments. Whenever the surface texture has a directional quality, the direction in which the measurement is made is approximately at right angles to the ‘lay’. It is always preferable to compute thei?a value by taking mean results from the measurement of several sampling lengths taken consecutively along the profile so that it gives better indication and is the correct representation of the whole of surface roughness.
Graphical Methods. In this method, it is presumed that the surface is nominally flat and that the record is produced in rectilinear co-ordinates in which a truly flat surface is represented as a straight line. First of all mean line of successive sampling lengths is determined. In case profile is not nominally straight, then mean line of appropriate curvature could be assumed ; for the record could be developed such that base line is straightened. For such cases, the profile records are made on plastic replicas by taking from curved surfaces and flattening out; or such instruments may be used which operating on a curved surface under appropriate conditions produced records in which the nominally curved base line is automatically straightened.
The mean line is determined by first drawing any straight line parallel to the general course of the record over the sampling length. Then the sum of areas above this line and below it are computed and their difference divided by the sampling length gives the distance by which the above drawn line is to be shifted in order to obtain the central line. Then the arithmetical sum of all the areas above and below the central line divided by the sampling length and the vertical magnification of the record gives the Ra value in mm, if areas are computed in mm2 and sampling length in mm. This value multiplied by 1000 gives Ra value in microns.

Some Notes on Stylus and Skid Method.

It is obvious that the stylus which rests on the surface and traverses across it responds to the irregularities of the surface relative to a selected datum provided by a skid having a large radius of curvature in the direction of the traverse. The skid simply follows the general contour, riding over the crests of smaller irregularities without responding to them individually. This method is subjected to some limitations in giving exact response corresponding to the shape of the irregularities. It is firstly due to the impossibility of having a mathematically sharp point for the stylus, which at the most can be finished with a small but finite radius. The effect of this is that the full dcpth’of the groove is reduced, the trace is smoothed out over the sharp corner and crests, and the trace offers no evidence of fine deep scratches in the actual surface. However, it may be appreciated that for an average reading the loss in the valley is fully offset by the gain resulting from the apparent rounding of the crests. Further detailed studies indicated that the surface irregularities being relatively shallow compared with their spacing, the errors due to finite radius of the stylus are not very serious and that a nominal radius of 0.003 mm for the tip of the stylus gives quite satisfactory results. Secondly, the error is also introduced due to the displaced position of the stylus in relation to the skid, because it is just possible that the stylus is touching a valley when the skid is at its lowest point and vice-versa. In order to reduce the chances of meeting extreme phase conditions, it is common practice to have two skids having a small radius and both touching the surface with the stylus acting midway between them.
Keeping the above in view, the stylus type instruments, employed for determining the surface roughness should comply with the following requirements :
Stylus. For graphical recording, nominal radius of the tip of stylus should be 0.003 mm. For determination of Ra value by direct reading instruments the nominal radius of the tip of stylus should be 0.01 mm. The variation in the tip radius of the stylus should be within 30% of its nominal dimension or 0.003 mm whichever is greater.
The shank may be either conical or pyramidal, and in either case the included angle of cone should not exceed 90°.
The maximum force between a stylus and the surface under test should not be more than 0.016 x (tip radius in microns)2 grams.
Skid. If a single skid is employed its radius in the direction of traverse should not be less than 50 times the sampling length. In case of two skids, the radii-should not be less than 8 times the sampling length.

Next post:

Previous post: