Finding Common Ground For Measurement Disagreement

When we think about making a simple measurement such as an outside diameter, many methods can be used. These would include calipers, micrometers, mechanical snap gages, air snaps, CMMs and special fixture gages, just to mention a few.

The problem with all these methods is that while each is good in its own right, there may be a bias seen among them. Each method has slight variations that can cause differences in result. Possible variations are gaging force, different process methods (such as the difference between contact and air gaging). Different measuring techniques include a CMM doing average diameters as opposed to a caliper using two point measurements.

With all of these techniques and potential pitfalls, how do you account for multiple measurements made by different personnel, in either the same or different locations, and then agree upon the results? The bottom line is, you don’t. The only correct way to do this is to determine an agreed-upon method for making the measurement and ensuring that each location follows the same process.

The good news is that these methods have already been fairly well defined in ASME B46.1-2002 for Surface Texture (Surface Roughness, Waviness and Lay). When there is a disagreement about the interpretation of texture measurements, this standard provides common ground for making the measurement. This includes defining the type of instrument and its parameter settings. With these defined, all users are at a common starting point.

The first recommendation of the ASME Standard is for the type of instrument to be used. The choice in this case—as seen in Section 2.3, Classification Schemes—is a Type I Skidless Contact profiling instrument. This is the most common analytical tool for surface measurement that includes the ability to measure smooth and rough surfaces; measure roughness and waviness compared to an outside surface; and use a selection of filters and parameters for data analysis.

Once the particular instrument is chosen, the next step is to agree on the same measurement characteristics. Section 9 of the Standard specifies that the instrument must have Gaussian (50 percent) filtering. These are specified because they tend to separate roughness and waviness more effectively than the 2RC filters often found on older systems. Also, the recommended bandwidth, stylus tip and radius and sampling interval are determined using Table 9.2 in Section 9 based on the desired roughness cutoff (lc).

The last setting to select is the stylus force. This can be found in Section 3, paragraph 3.3.5.2 based on the desired radius tip. The chosen stylus force is sufficient to maintain top-surface contact but not so large that the stylus will cause damage to the surface. The maximum recommended values are always determined by the stylus radius.

These are all fairly common settings selected for the skidless measuring system. Those familiar with these systems are no doubt proficient with making this setup and would be satisfied with choosing this common format.

However, there may be some unusual measuring conditions that require further definitions. These may occur, for example, if the surface structure being assessed requires a short wavelength cutoff smaller than the 2.5 micrometers specified in Section 9.2. In this case, Section 7 in the Specification deals with measuring Nano Surface Texture and Step Height measurements with Skidless Profiling Systems. It talks about the measurement of nanometer sized features on surfaces that may have irregular roughness and waviness features or a depth measurement such as a groove. Following the recommendations in this section will reduce the uncertainty of the measurement of very small features. It will improve the comparison results from different laboratories.

Finally, there are some particular cases where the above recommendations may not apply. These usually involve soft surfaces where significant damage can occur to the surface using a contact stylus system. In these cases, the parties involved have to devise and jointly agree on a measuring process and its characteristics.

In the end, it’s always satisfying to have various measurements made in various locations arrive at nearly the same results. The only effective way of doing this is to agree on a standard measuring technique and adhere to it.