Theodore D. Doiron

Uncertainty and Dimensional Calibrations

The calculation of uncertainty for a measurement is an effort to set reasonable bounds for the measurement result according to standardized rules. Since every measurement produces only an estimate of the answer, the primary requisite of an uncertainty statement is to inform the reader of how sure the writer is that the answer is in a certain range. This report explains how we have implemented these rules for dimensional calibrations of nine different types of gages: gage blocks, gage wires, ring gages, gage balls, roundness standards, optical flats indexing tables, angle blocks, and sieves.

A Careful Consideration of the Calibration Concept

This paper presents a detailed discussion of the technical aspects of the calibration process with emphasis on the definition of the measurand, the conditions under which the calibration results are valid, and the subsequent use of the calibration results in measurement uncertainty statements. The concepts of measurement uncertainty, error, systematic error, and reproducibility are also addressed as they pertain to the calibration process.

20 °C-A Short History of the Standard Reference Temperature for Industrial Dimensional Measurements.

One of the basic principles of dimensional metrology is that a part dimension changes with temperature because of thermal expansion. Since 1931 industrial lengths have been defined as the size at 20 °C. This paper discusses the variety of standard temperatures that were in use before that date, the efforts of C.E. Johansson to meet these variations, and the effort by the National Bureau of Standards to bring the United States to the eventual world standard.

Dimensional metrology of bipolar fuel cell plates using laser spot triangulation probes

As in any engineering component, manufacturing a bipolar fuel cell plate for a polymer electrolyte membrane (PEM) hydrogen fuel cell power stack to within its stated design tolerances is critical in achieving the intended function. In a bipolar fuel cell plate, the dimensional features of interest include channel width, channel height, channel parallelism, side wall taper, straightness of the bottom or side walls, plate parallelism, etc. Such measurements can be performed on coordinate measuring machines (CMMs) with micro-probes that can access the narrow channels. While CMM measurements provide high accuracy (less than 1 µm), they are often very slow (taking several hours to measure a single plate) and unsuitable for the manufacturing environment. In this context, we describe a system for rapid dimensional measurement of bipolar fuel cell plates using two laser spot triangulation probes that can achieve comparable accuracies to those of a touch probe CMM, while offering manufacturers the possibility for 100% part inspection. We discuss the design of the system, present our approach to calibrating system parameters, present validation data, compare bipolar fuel cell plate measurement results with those obtained using a Mitutoyo UMAP (see footnote 1) fiber probe CMM, and finally describe the uncertainty in channel height and width measurements.

Measurements of large silicon spheres using the NIST M48 coordinate measuring machine

The NIST M48 coordinate measuring machine (CMM) was used to measure the average diameter of two precision, silicon spheres of nominal diameter near 93.6mm. A measurement technique was devised that took advantage of the specific strengths of the machine and the artifacts while restricting the influences derived from the machines few weaknesses. This effort resulted in measurements with unprecedented accuracy and uncertainty levels for CMM style instruments. The results were confirmed through a blind comparison with another national measurement institute (NMI) that used special apparatus specifically designed for the measurement of these silicon spheres and employed very different measurement techniques. The standard uncertainty of the average diameter measurements was less than 20 nanometers. This paper will describe the measurement techniques along with the decision-making processes used to develop these specific methods. The measurement uncertainty of the measurements will also be rigorously examined.

Accuracy and versatility of the NIST M48 coordinate measuring machine

The NIST Is continuing to develop the ability to perform accurate, traceable measurements on a wide range of artifacts using a very precise, error-mapped coordinate measuring machine (CMM). The NIST M48 CMM has promised accuracy and versatility for many ears. Recently, these promises have been realized in a reliable, reproducible way for many types of 1D, 2D, and 3D engineering metrology artifacts. The versatility of the machine has permitted state-of-the-art, accurate measurements of one meter step gages and precision ball plates as well as 500 micrometer holes and small precision parts made of aluminum or glass. To accomplish this wide range of measurements the CMM has required extensive assessment of machine positioning and straightness errors, probe response, machine motion control and speed, environmental stability, and measurement procedures. The CMM has been used as an absolute instrument and as a very complicated comparator. The data collection techniques have been designed to acquire statistical information on the machine and probe performance and to evaluate and remove any potential thermal drift in the machine coordinate system during operation. This paper will present the data collection and measurement techniques used by NIST to achieve excellent measurement results for gage blocks, long end standards, step gages, ring and plug gages, small holes, ball plates, and angular artifacts. Comparison data with existing independent primary measuring instruments will also be presented to show agreement and correlation with those historical methods. Current plans for incorporating the CMM into existing measurement services, such as plain ring gages, large plug gages, and long end standards, will be presented along with other proposed development of this CMM.

Minimizing errors in phase change correction measurements for gauge blocks using a spherical contact technique

One of the most elusive measurement elements in gage block interferometry is the correction for the phase change on reflection. Techniques used to quantify this correction have improved over the year, but the measurement uncertainty has remained relatively constant because some error sources have proven historically difficult to reduce. The precision engineering division at the National Institute of Standards and Technology has recently developed a measurement technique that can quantify the phase change on reflection correction directly for individual gage blocks and eliminates some of the fundamental problems with historical measurement methods. Since only the top surface of the gage block is used in the measurement, wringing film inconsistencies are eliminated with this technique thereby drastically reducing the measurement uncertainty for the correction. However, block geometry and thermal issues still exist. This paper will describe the methods used to minimize the measurement uncertainty of the phase change on reflection evaluation using a spherical contact technique. The work focuses on gage block surface topography and drift eliminating algorithms for the data collection. The extrapolation of the data to an undeformed condition and the failure of these curves to follow theoretical estimates are also discussed. The wavelength dependence of the correction was directly measured for different gage block materials and manufacturers and the data will be presented.

High-speed high-resolution fine wire diameter measurement system

A fine wire diameter measurement system, for on-line monitoring, has been proposed by using a Machine Vision System and a visible diode laser. The system uses the Fraunhofer diffraction principle. The diffraction pattern, generated by a small wire exposed to a collimated laser beam, is acquired by a CCD industrial camera that is connected to a processing board inside a PC computer. Two different methods of measuring the diameters, static and dynamic, have been proposed in order to get high precision and high measurement rate. Wires with diameter from 10 to 350 micrometers have been measured by this system with 0.06% resolution. The accuracy is less than +/- 0.5% over a range of 90 - 350 micrometers diameter. For thinner wires, the measurement system should be calibrated to eliminate the systematic errors. The estimate random errors are +/- 0.25%. The instrument can measure the wire diameter at a 1000 Hz rate and allows it to move laterally in a 1 mm square window, maintaining the above accuracy. The system is compact and there are no moving parts.

Dimensional Review of Scales for Forensic Photography

Scales for photography provide a geometrical reference in the photographic documentation of a crime scene, pattern, or item of evidence. The ABFO No. 2 Standard Reference Scale (1) is used by the forensic science community as an accurate reference scale. We investigated the overall accuracy of the major centimeter graduations, internal/external diameters of the circles, error in placement of the circle centers, and leg perpendicularity. Four vendors were selected for the scales, and the features were measured on a vision‐based coordinate measurement system. The scales were well within the specified tolerance for the length graduations. After 4 years, the same scales were measured to determine what change could be measured. The scales demonstrated acceptable stability in the scale length and center‐to‐center measurements; however, the perpendicularity exhibited change. The study results indicate that scale quality checks using certified metal rulers are good practice.

Probing of Two-Dimensional Grid Patterns by Means of Camera Based Image Processing

Camera based probes and machine vision have found increased use in coordinate measuring machines over the past years and the calibration of artifacts for these probes has become an important task for NIST. Until recently these artifacts have been calibrated using one or two dimensional measuring machines with electro-optic microscopes or scanning devices as probes. These sensors evaluate only a small section of the edge of a grid mark, and irregularities in this particular spot from local deformations or contamination influence the measurement result. Since these measurements result in a single number based on the entire field of view, the influence of small irregularities are not easily detected. Since different probes scan different parts of the grid mark edge they may give systematically different positions of the mark. The conversion to video based sensors has allowed more flexibility it edge detection, although most instruments still use least squares fits as the substitute geometry of straight edges. This method is very susceptible to noise and edge irregularities. We present some experiments for finding the sub-pixel edge point locations and fitting the set of edge points to a line using a fairly simple least sum of absolute deviations fit. Data from a high accuracy 2D measuring machine is used to show the strengths of the algorithms.