Surface texture has always been evaluated with the assistance of profile measurement, moving a stylus tip along a straight line. The result obtained describes the altitude in z as a function of the position in x: z = f(x). It is called a 2D scan and one talks of 2D profilometry.

Profile measurement, z = f(x)

Topography systems made their first appearance only twenty years ago. Their operating principle is to measure parallel profiles, with regular spacing, covering a rectangular surface. A height measurement z is obtained as a function of the x position on the profile, and the y position of the profile in the surface. It is called a 3D scan, and one talks of 3D topography or sometimes of surfometry.

Surface measurement, z = f(x,y)

Since the amount of 3D data is much larger, it was necessary to wait for sufficiently powerful computers to become available before topography systems caught on. Measurement time can be fairly long because it involves several hundred profile scans. In addition to "lateral scanning" systems, there are systems that use a camera to record the surface texture directly on an x,y matrix, for example white light interferential microscopes, and confocal microscopes. These systems are faster but they have a limited field of measurement.

When a profile is measured, in 2D, the sensor only records the relief along its track. Hence the measurement is only representative of the surface itself under certain conditions. These conditions depend upon the isotropy of the surface.

A surface is said to be isotropic when it presents the same characteristics regardless of the measurement direction. For example, surfaces with a random texture without any distinctive mark or feature, are isotropic.

Unfortunately this type of surface is fairly uncommon and most of the surfaces found in industry have a directed texture (turned, ground, brushed, etc.) or a periodic structure (EBT impacts, grained plastics, etc.). In these cases, the surface is said to be anisotropic.

surface with an isotropic texture	surface with an anisotropic texture (double turning)

In order to work around the problem of anisotropic surfaces when using 2D profilometry, metrologists and manufacturers have defined rules for handling isotropic cases, and for making do with a profilometric measurement. The ISO 4288 standard specifies that profiles must be measured in a direction which has the greatest amplitude, which in general corresponds to a direction perpendicular to the lay of the texture.

A brushed surface presents parallel grooves which can be completely analysed by measuring a profile perpendicular to the brush marks. Whatever profile is measured in these conditions, the surface texture is representative of the whole surface.

A turned surface presenting machined grooves in an arc of a circle can be characterised by making a radial measurement (along a radius) perpendicular to the grooves. Whatever direction is measured, provided that the profile is radial, the structure and spacing of the machined grooves will be correctly recorded.

	profile extracted from a turned surface

The anisotropy of a surface is related to the scale. At a large scale, the surface of the plastic lid of a notebook computer appears to be perfectly isotropic and homogenous. At a small scale, it can be shown that in reality the surface is made up of multiple small grains which give the surface a matte appearance. At an even smaller scale, for example with a scanning probe microscope, one can only see elevated mountains and a chaotic landscape.

This property can be exploited to characterise a surface that is moderately anisotropic, for example a fine grained plastic, with the help of profile measurements. In this case one proceeds with numerous measurements in all directions and takes the average value of the parameters that are obtained. The more significant the statistical sample, the more the characterisation is reliable. One compensates for the slight anisotropy of the surface by the number of evaluations.

Nevertheless, some surface textures pose irremediable problems for 2D analysis. When profilometric analysis fails, this manifests itself in less than striking correlations between surface texture parameters and a production characteristic or quality factor. In such cases, 3D surface measurement and the analysis of surface texture using 3D parameters significantly improve the correlations, and provide information that it is impossible to obtain in 2D. They provide a far better evaluation of the quality of the surface.

A flat surface containing uniformly spaced pores provides profilometric measurements with numerous deep holes.

A flat surface including narrow deep grooves presents the same type of profile when measured perpendicularly to the grooves. It is not possible to say, with respect to a profile, if a valley is due to a groove or a hole. This example shows that the same profile can originate from two different surface types, with different functional properties. Only 3D analysis makes it possible to record the surface texture in a reliable manner.

The analysis of a surface defect is an analysis, in essence, concentrating on a detail. It cannot be carried out using statistical methods and therefore excludes the use of classical surface texture parameters. The ISO 8785 standard lists all of the types of imperfection that one can meet on mechanical parts:

Buckle	Craters
Flaking	Blowhole

The isolated character of these imperfections and their specific nature means that surface measurement is much more pertinent and reliable than profilometric measurement. Once the surface has been measured, tools like the measurement of volume, perimeter and maximum depth can be used to characterise the imperfection. Also it is important not to forget the power of images which can sometimes give an engineer a much better idea of the cause of a problem than a series of numbers.

These operations are currently carried out in laboratories working on the analysis of fractures (accident assessment), on the optimisation of production tools (foundry, fast machining), surface deposits (scratch resistance, adherence), materials analysis (resistance to shock, temperature, chemical compounds), etc.

When motifs in surface texture are too large to be analysed statistically, as described above, it is necessary to make a surface measurement and to analyse the morphology of the motifs and their functional role in the part.

Macroscopic texture measured on plastic imitation lizard skin (10 mm x 10 mm)

The classic 2D and 3D parameters based on amplitude statistics (such as Ra, Sq, Rsk...) can't be used reliably here. It is preferable to use parameters relating to volume, form factor, summit density, etc Tools like valley vectorisation prove to be essential for certain applications.

3D measurement also makes it possible to analyse matter on different levels separately, for example ink deposits on paper, or textured or grained plastic.

The example below shows two surfaces measured on a dashboard covering. The texture is transferred by rolling in order to achieve large aesthetic patterns (grains) on a fine texture background with a silk or matte finish.

Using a worn roller, the background texture is lost.	Using a new roller, the background texture is clear.

The surface grains are eliminated in order to quantify the roughness of the background texture and make a reliable distinction between good parts and parts made using a worn roller.

Thanks to the spread of topographical measurement systems in industry, there are new characterisation methods based on algorithms that are intrinsically 3D, and not merely 3D extrapolations of 2D parameters.

representation of dales and pits.

One of these new methods is the segmentation of motifs on the surface. This method is used to isolate the significant motifs in a texture and to find representative peaks and valleys. These specific points facilitate understanding of the role that the peaks and valleys play in overall part function. The study of the relations between adjacent dales makes it possible, for example, to predict flow phenomena. It is not only possible to study static lubrication (as with classical methods), it is equally possible to understand the dynamics: how will fluid behave when it is compressed after one part has made contact with another?

In recent years, numerous publications cite studies showing the superiority of 3D characterisation when it comes to discriminating between similar surfaces based on functional criteria. 3D tests make it possible to increase product life cycles, to optimize production tools, to make economies by improving lubrication and reducing leaks, etc. It is probable that within a decade, industry will make routine use of 3D control, even in the workshop.

Nevertheless industry needs time to accumulate a knowledge base on 3D surface states that is as substantial as the existing 2D knowledge base, built up over decades of practice. The tools are available now, what remains is to refine their applications.

Our everyday world is three dimensional. The visible world is three dimensional. In the nature of things, objects, their size, their surfaces and their texture extend into three dimensions. Likewise, the interactions between objects (contact, wear, sliding, etc.) take place in three dimensional space.

It is therefore logical to carry out analysis in all of these dimensions, so long as the methods of control are powerful enough. The profilometer is a simplification that was essential during a certain period of time for purely material reasons.

For forty years, the ISO standards have been based on profilometry alone. The ambition of current work is to redefine these concepts and tools in 3D, in the most general manner possible. The initial 3D standards are about to be published, and others will follow in a couple of years' time. As soon as this corpus is published, the goal will be to reform the collection of 3D standards and make them follow from the 3D standards as special cases.

Old habits will give way to new 3D concepts, to the benefit of both industry and consumers alike!

Whatever method you choose, to explore the advantages of analysing 3D surface textures, or to capitalise on years of experience with profilometry, Digital Surf's products and technologies will provide you with suitable high performance tools.

Pioneer in the field of 3D surface textures ever since 1990, Digital Surf has acquired a certain stature in metrology circles worldwide, supplying its technologies to numerous leading manufacturers. Digital Surf participates in ISO committees responsible for drawing up future 3D surface texture standards, and its MountainsMap® software includes 3D tools that have been validated and proven, even before standards are published.

The inescapable movement towards non-contact measurement and 3D measurement places Digital Surf's technologies at the heart of your present and future concerns.

Other relevant pages:

  Metrology guide
  Mountains® platform
  Standard packages