What is the most accurate way to check if a measuring tool works within its specifications? Guillaume Bull, product manager at Creaform, explains in this article.
When replacing old measuring equipment, it is common to validate that both the old device and the new device measure the same data and provide quality control (QC) with the same results. To do this, correlation tests are performed.
To facilitate and speed up the work, it is tempting to test a regularly manufactured part. After all, its specifications are well known. However, this choice of part may lead to a false diagnosis and an incorrect conclusion regarding the accuracy of the new measuring device.
Therefore, the most accurate way to check if a measuring tool works within its specifications is to use a calibrated artefact for which measurements have been previously validated and the data is traceable.
Using a common artefact for the old device and the new device helps to minimize the variables that can influence the correlation tests. Among these variables, which will induce measurement differences, are the extraction methods that are different from one technology to another, the alignment methods that are rarely the same, software that does not process or calculate data in the same way, the setups that are generally different depending on the technologies, and the environment that, if not maintained exactly the same, will greatly influence the measurements.
Using a calibrated and traceable artefact enables operators to validate that both devices work within their specifications. As a result, if the measurements taken on this calibrated artefact give the right value, we will know for sure that the measuring devices work properly.
A manufacturing company working in the automotive industry wants to replace its CMM with a 3D scanner. In order to validate the new equipment, a correlation test is performed between the two devices—the old and the new. When the two measurements are compared, there is a difference; the instruments do not correlate with each other. Why? Should we not get the same measurement on both instruments? What is causing this difference? Since we know that the old equipment has been accurate historically, should we conclude that the new equipment has an accuracy issue?
When testing for correlations between two types of equipment (i.e., comparing the measurements obtained on the same part with two instruments), there are many variables that can induce errors in the measurements. These variables include extraction and alignment methods, software calculation, setup, and environment.
We measure the same part, but we do not extract the same points with one measuring tool as we do with the other tool. The consequence is a difference in measurement due to the imperfection of the geometry of the part. Indeed, when we probe a surface plan by taking a point at the four corners, this method does not consider the surface defaults of the plan. Conversely, if we scan this plan, we measure the entire surface and get the flatness. Therefore, if the surface has a slight curve, the scanned plan might be misaligned compared to the probed plan. Thus, there will be a difference in measurement between the two methods.
We measure the same part, but we use two different methods of alignment. The consequence is a slight difference in the alignment method, which can lead, due to leverage, to large deviations at the other end of the part. Even if the same method of alignment is used, as mentioned above, a difference in the extraction method of the features used in the alignment can lead to a misalignment of the part. The positioning values are based on the alignment, which must not differ from one instrument to another, neither in the construction method, nor in the way it is measured.
We measure the same part, but we use different software that does not use the same algorithms for data processing. The consequence is a difference in the calculation of a feature from the software, even though the measured data is the same. The more complex the construction of the measurement is, the more likely it is to have deviations between calculations.
We measure the same part, but we do not have the same setup on both instruments. The consequence is different measurements of this same part. For example, a part of large dimensions is measured on a CMM. The marble on which the part is placed has an excellent flatness (30 microns). The same part is then measured with a 3D scanning system. But the surface on which the part is put has a different flatness (800 microns). As a result, the part twists and deforms slightly when placed on the second marble. Although the same part is measured, the two setups give different measurements because the support surfaces have different degrees of flatness.
We measure the same part but under different conditions. The consequence is a difference in the measurements. Indeed, if we measure an aluminium part of one meter on a CMM at an ambient temperature of 20 deg C and we measure the exact same part at 25 deg C, then the difference in temperature will result in a lengthening of the part by 115 microns at 25 deg C.
It is crucial for quality control to minimize these different variables that could lead to correlation errors. The easiest way is to use, on both instruments, a common artefact for which measurements have been previously validated and the data is traceable.
Artefacts have the distinguishing characteristics of being calibrated and traceable. All features have been previously measured and verified in a laboratory, eliminating any doubt and uncertainty regarding measurements.
A value commonly obtained with a traditional measuring instrument is not a reference value that can be relied upon 100%. The reason for this is that equipment is not an artefact. There is always uncertainty associated with any measuring instrument. Therefore, the verification, validation, or qualification of a measuring instrument cannot be done with any part for which dimensions have not been previously validated.
The only way to certify that a measuring tool works within its specifications is to compare it with an artefact whose dimensions are calibrated in a known laboratory. Only an artefact makes it possible to correlate measurements between equipment because only an artefact can subtract all the variables that could interfere with the measurement. Thanks to an artefact, there is no doubt; the equipment measures accurately.
If two devices get the same measurement with an artefact, but do not correlate on a specific part, then the difference is not attributable to the instruments. Rather, it will result from measurement processes that will need to be checked and scrutinized further to obtain the desired measurement.
Check these articles out:
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!