WinWerth® measurement software

In practice, a few sizes of production parts often need to be determined "quickly". This task is also carried out by employees who are not permanently involved in the operation of coordinate measuring machines. To enable effective work in this environment, operation is limited to the bare essentials. The "intelligence" of the WinWerth® measurement software then takes over e.g. B. the exact determination of the object area to be captured, the selection of the geometric element to be measured (e.g. B. straightness, circle, corner point) as well as the linking algorithms for determining geometrical characteristics such as distances, angles and diameters.

For more complicated measurement tasks, the operator can take over parts of the automatic processes (setting windows, selecting features) themselves and gradually familiarize themselves with the more detailed control of the measurement sequences. For support, measurement points or scan lines are automatically distributed on the geometry elements to be measured, e.g. as circles, cylinder surface lines, stars or spirals, taking into account the necessary traverse paths. The complete measurement sequence, including evaluation, is first created offline using the CAD model or online with the minimum number of points for the respective geometry element. Measurement points and scan lines can be subsequently moved, deleted or added with the mouse or via a dialog. Measurement sequences specified in this way can be saved and called up as an automatic sequence in the event of repetition.

Another benefit of the CAD module integrated in WinWerth® is that CAD information can be used to position the coordinate measuring machine. Werth was probably the first manufacturer of coordinate measuring machines to introduce this technology back in the mid-1990s under the name CAD-Online®. The entire measurement sequence can be controlled by selecting the geometric features on the CAD model. The measuring machine automatically moves to the generated measuring positions and measures with the selected sensors.
In this way, for example B. automatically capture measurement points as point clouds using stylus or measure larger areas with the Werth 3D Patch or confocal sensors by automatically stitching the individual measurements together in high resolution. Technology parameters such as the illumination setting for the image processing sensor can be set directly on the measuring machine, taking into account the interaction between the illumination, measuring object and imaging system. Collisions are avoided by automatically modifying the movement sequences based on the workpiece and machine or sensor geometry.

In contour image processing, the image within an evaluation window is viewed as a planar whole. Contours are extracted from this image using suitable mathematical algorithms (operators). Each pixel of a contour corresponds to a measurement point. The measurement points are strung together like a string of pearls. This makes it possible to detect and filter out interfering contours caused by surface structures, breakouts and dirt during measurement (contour filters) without changing the shape of the contours. It is important for practical use that several contours can be differentiated within a capture area and the desired one can be selected. This allows reliable edge detection even with large tolerances and rigid scanning in transmitted light. In a further step, modern systems interpolate the coordinates of the measurement points within the pixel grid and thus allow higher accuracies.

Contours larger than the field of view of the respective lenses can be captured as a whole using automatic contour tracking in conjunction with the CNC axes of the coordinate measuring machine (contour scanning). This scanning method is well suited to checking a few relatively large contours, e.g. on punching tools. 
Another method for capturing larger areas of the workpiece is "raster scanning HD" (patent). Here, the image processing sensor captures images of the workpiece at high frequency during movement. These are resampled and superimposed to create an overall image with a resolution of up to 20,000 megapixels. In this way, for example, 100,000 small bores on large fiber couplers are measured in just 35 minutes instead of 7 hours. Accuracy is also increased by measuring even large areas at high magnification and averaging over several images, which improves the signal-to-noise ratio. The process can be adapted to the requirements of the measuring task.
With Raster Scanning HD P, image acquisition only on areas of interest using a preset path results in a further reduction in measuring time and data volume compared to rectangular raster scanning of the entire workpiece with Raster Scanning HD N. On rotary axis devices, Raster Scanning HD ROTARY enables image acquisition during rotation with measurements on the "unrolled" overall image of the lateral surface of rotationally symmetrical workpieces.

Eccentric Tomography allows the workpiece to be positioned anywhere on the rotary table (patented). This eliminates the need for complex and time-consuming alignment of the workpiece and increases ease of use. Sectional tomography or ROI tomography (ROI: Region of Interest) is used to measure parts of the measuring object with high resolution without having to capture the entire measuring object, e.g. with Raster Tomography, which is time-consuming and memory-intensive.  Multi-ROI tomography offers a combination of the benefits of eccentric and sectional tomography. Several high-resolution parts can also be selected at any position in the measuring object.

Eccentric Tomography allows the workpiece to be positioned anywhere on the rotary table (patented). This eliminates the need for complex and time-consuming alignment of the workpiece and increases ease of use. Sectional tomography or ROI tomography (ROI: Region of Interest) is used to measure parts of the measuring object with high resolution without having to capture the entire measuring object, e.g. e.g. with Raster Tomography, which is time-consuming and memory-intensive. Multi-ROI tomography offers a combination of the benefits of eccentric and sectional tomography. Several high-resolution parts can also be selected at any position in the measuring object.

In the X-ray tomographic measurement of metal-plastic components such as feeding connectors, the metal pins often cause artifacts due to beam hardening and scattered radiation, which make measurements on the plastic housing more difficult. In Dual-Spectra Tomography, the measurement software combines two CT measurements at different cathode voltages into one volume. The radiation spectra are matched to the two materials. The corresponding reduction of artifacts in the volume reduces the measurement uncertainty when determining sizes between the different materials. To this end, the WinWerth® MultiMaterialScan uses the patented subvoxeling process to automatically calculate separate STL point clouds for each material from the CT volume data, even for several different metal components.

Depending on the task, the measurement results are either calculated or displayed in a reference coordinate system that has been measured beforehand (e.g. vehicle coordinates in automotive engineering) or in a coordinate system that has been generated by optimally fitting selected surface areas relative to the CAD model.

The two fitting strategies WinWerth® BestFit and ToleranceFit® can be well illustrated using the example of a 2D cross section. In the first case, the position of the measured points is optimized by minimizing the distances to the target points. As tolerances of different object areas are not taken into account during fitting, tolerance overruns may be detected, although the tolerance could be maintained by shifting the coordinate system. This method is therefore only suitable for quality control to a limited extent.

The optimization criterion for WinWerth® ToleranceFit® is to keep the distance between the measurement point and the tolerance limit as large as possible or, if the measurement point is outside the tolerance limit, to keep the tolerance overrun as small as possible. Objects recognized as faulty according to the BestFit method (red areas present), but not actually faulty, can be classified as functional according to the ToleranceFit® method. The contour is checked as with a gauge.

To illustrate the deviation of the workpiece geometry from the target values, a comparison to the CAD data with a color-coded display of the deviations in WinWerth® is suitable. This procedure is essential for checking free-form surfaces. For measurement, the areas of interest of the object are scanned or captured as a point cloud. WinWerth® then compares the measured values with the CAD model. The result is documented by vectorial or color-coded deviation to the CAD model. This evaluation can be carried out as part of the measurement sequence on the machine or in offline mode at a separate evaluation station. The colors of the measurement points illustrate the deviation between the target and actual values. To include the part tolerances in the display, they are divided into four basic classes:

• positive within tolerance
• negative within tolerance
• positive outside tolerance
• negative outside tolerance

The amount of deviation is displayed using color coding. Alternatively, the user can configure color coding according to their wishes.

In order to incorporate the measured or calculated deviations into the manufacturing process, the specification data can be modified largely automatically with WinWerth® FormCorrect. For this purpose, the deviations between the original CAD model and the measured data of a sample workpiece are determined and mirrored on the model. From this, the measurement software generates a corrected CAD model that can be used to compensate for systematic manufacturing deviations in the plastic injection molding process and 3D printing. In contrast to conventional reverse engineering, the application is considerably simplified. Due to the high precision, often only one correction loop is required, meaning that the costs of the development process can be significantly reduced. For high-resolution corrections and for the modification of internal surfaces, the use of coordinate measuring machines with X-ray computed tomography is recommended. A similar procedure is possible with the 2D BestFit software. Mold correction can be used both for running in new cutting tools (profile grinding, form milling) and for correcting positioning deviations during wire EDM.

If no CAD data is available, the measurement points can be selected interactively by the operator. In WinWerth®, both direct selection with the mouse and automatic decomposition into standard geometry elements are possible. Starting from a starting point, further points are automatically added all around until the form error of the selected feature (e.g. e.g. cylinder) increases noticeably. This signals that the limits of the feature have been reached and the process is completed.

It is more effective to define the measurement sequences using 3D CAD data. Simply selecting CAD features automatically selects the necessary measurement points (patent). Based on the selection of CAD patches, all measurement points of the measured object that can be geometrically assigned to this patch are selected, taking into account the specified edge distances. This results in a complete capture of the mold of the corresponding feature with the maximum number of points.

In practice, it is common to define drawing dimensions in 2D views and cross sections. This must also be taken into account when evaluating tomographically generated measured data. For this purpose, planes can be defined in the workpiece coordinate system and intersected with both the CAD target data and the actual point cloud. WinWerth® automatically extracts contours representing the target data and the actual contours. The same software functions are used to evaluate the 2D sizes in cutting contours created in this way as are available for evaluating contours scanned using image processing or a stylus.