Study of several roughness measurement techniques for the characterisation of mass standard

Project Description

Roughness measurements on the plane and cylindrical surfaces of several metallic materials to make mass standards of the OIML shape and cylindrical shape are made with two optical methods (ARS and TIS) as well as mechanical profilometers (Talysurf, Talystep, ...)

The results will be compared and discussed with respect to the spatial frequency range of each device.

In order to compare polishing techniques of mass standards, the validity limits of roughness parameters of each measurement method will be defined.

The correlation between mass stability and surface roughness of mass standards could be studied.

Final Report 2003-02-26

The goal of this project is to compare and discuss the advantages and limits of several roughness measurement techniques to characterize mass standard surface quality. This is in order to choose the best polishing methods, to study the correlation between mass stability and surface roughness and to propose recommendations for testing the surface quality of mass standards.
The measurement techniques used were four optical methods, complemented by one mechanical method, as follows:

• two instruments based on the angle-resolved-scattering method (ARS) (BNM-INM and NAWC)
• one device based on the total-integrated-scattering method (TIS) (SP)
• one instrument based on near field microscopy (LPUB)
• one instrument based on phase shifting interferometry (NIST)
• one mechanical profilometer (NAWC)
   

Four mass standards have been studied with these techniques, which were chosen, not for their pristine surface condition, but rather as being representative of weights in use:

• 50 g and 100 g in the cylindrical shaped masses, the first one made from stainless steel, the second from Alacrite
• 500 g and 2 kg weights of the OIML shape both made from stainless steel
   

There is no one roughness value for a surface. Apart from actual local variations in surface roughness, the present study emphasises for the weighing community the well-known fact that different methods cover different spectral ranges of surface spatial wavelengths. This study showed that near field optics, phase shifting interferometry, light scattering techniques and mechanical profiler give different values of rms roughness (Rq) because the instruments measure over different surface spatial wavelength ranges. The main difficulty lies in the interpretation and comparison of the results obtained using these instruments.

It is thus necessary to give further information together with values of the surface parameters provided by a roughness measurement technique as following:

• instrument characteristics (spatial frequency range, illuminated surface area for light scattering method, profile length for profiler, …)
• measurement conditions (studied zone position on the sample, micrographs obtained with a Nomarski microscope, measurement direction with regard to principal grooves direction, …)
   

The instruments using light scattering and the mathematical model used to determine the surface parameters must be  adapted to study the cylindrical surface of mass standards.
The results of this project should support future studies of relevance to mass metrology:

• to assess which ranges of surface parameters are most relevant for mass stability
• to assess the efficiency of cleaning method by differential analysis of roughness measurements obtained before and after cleaning
• to improve the manufacturing process by measuring surface roughness for each polishing step
   

The report concerning this study will be available in April 2003 and two articles should be submitted in March 2003 for publication in Metrologia.