Logo of the Physikalisch-Technische Bundesanstalt
Production sequence of Si-spheres and interferometrical determination of the sphere volume

Gear calibrations using scanning mode


Gears are three-dimensional machine elements whose geometrical deviations are often determined on coordinate measuring machines (CMMs). While the pitch characteristics and the diametrical two-ball dimension are recorded as point distance measurements, the characteristics for profile and flank line are measured in lines according to DIN ISO 1328-1:2018-03 with at least 150 points along the involute and helix, respectively.

For the traceability of the latter two gear tooth parameters, the measurement and evaluation of the slope, form and total deviation are specified in nationally and internationally applicable standards. The measurement result for each gear parameter is determined from 24 measurements in different positions in the measuring volume of the CMM. This measurement strategy, which is nevertheless relatively time-consuming, has the advantage that the geometric errors of the measuring devices are largely compensated. However, the smallest measurement uncertainties that can be specified could previously only be achieved if the measurements were performed tactilely in single-point probing mode.

For some years now, measuring devices have been continuously improved in terms of their sensor technology and adapted to the industry's demand for shorter measuring times. This goal has been achieved primarily by scanning measurement methods. However, the shorter measurement time has been at the expense of accuracy. One of the challenges of scanning is to find the optimum scanning parameters for the measurement process. In the interaction of measuring speed, probing force and filtering, undesirable dynamic effects can lead to existing shape deviations not being reliably detected, for example. In contrast, filtering should minimize noise effects. Furthermore, the materials of the stylui, usually balls, can also play a role. Here, diamond-coated styli offer much lower wear than the more common ruby balls.

The introduction of the scanning measurement method was tested on 29 gear calibrations (a combination of 23 profiles and 26 flank lines with helix angles from 0 ° to 45 °) on different gear standards, optimized and verified by comparative measurements in proven single-point probing. The measurement results show that almost equal standard deviations of the individual measurement parameters can be achieved with the scanning method for both profile and flank line calibrations. The differences are in the range of ± 0.2 µm compared to the measurement results of the single point measurements and are thus in the same order of magnitude as the reproducibilities of the CMMs. The En value calculation (measure of the agreement of measured values considering the specified measurement uncertainties) confirms the good measurement results for all gear parameters (see Figure 1). Only in the case of the form deviations (f) must it be noted that with the scanning method, depending on the quality of the flank and here in particular the straightness, filtering according to ISO 1328 1 can cause a significant change in the measurement result (see Figure 2). In the calibrations analyzed, the shape deviations show differences between the single-point measurements (ffα,EP) and the scanning measurements (ffα,scan) of up to 1.3 µm, and thus also have En values > 1. The measurement method and the parameters of the filtering are clearly described in the calibration certificates, so that the prerequisites for a reproducible and reliable transfer of the measurement results are given.

The time saved by the scanning method means that it has been possible to almost compensate for the 41% increase in fees for the division of dimensional measurement quantities that came into effect at PTB on February 1, 2019.

Evaluation of the calibrations of scanning and single point measurements
Figure 1: En-values of gear calibrations for verification of scanning and single point measurements.

Comparison of a profile form deviation measurement in single point and scanning mode
Figure 2: Profile measurement with form deviations for single point measurement mode (ffα,EP) and scanning measurement mode (ffα,Scan).



[1] M. Stein, A. Wedmann, S. Jantzen, K. Hierse, K. Kniel, Involute gear calibration using tactile CMMs in scanning mode, MST 31 (2020), 7, 1 – 12



Physikalisch-Technische Bundesanstalt
Bundesallee 100
38116 Braunschweig