Surface inspection

While producing turned parts, milled parts or plastic parts surfaces are increasingly exposed to higher requirements on matters of material and quality. Either being used in a visible area or maybe subsequent steps in the line of production require non-disturbed surfaces, customers demand 100% quality. During production there are many reasons for flaws such like chip casts, scratches, blow holes or damaged edges.

We offer modules and complete inspection systems that might be used stand-alone or integrated into a production line to inspect, evaluate and sort surfaces of turned parts, milled parts or plastic parts. As a decade-long system integrator we connect automated feeding stations and transfer systems with testing machines we develop and design. Our specialty is the automated placement and rotation of products-to-be-tested in front of a camera and the subsequent unwinding of its surface for testing and evaluation.

With the help of lighting which is adapted to the tested part we can even check mirroring surfaces or polished metal as well as dull materials. To do so, we have developed a software which compares captured images to reference images with defined flaws. The size of the flaws can be adjusted dynamically depending if the error is within or outside of a given tolerance and needs to be sorted out.

PENTACON surface inspection systems are used with automotive suppliers and system integrators who usually require 100% quality control at high quantities. As a certified automobile supplier, we use our very own inspecting machines with our production to reach the highest quality demands.

Mode of operation

The part-to-be-tested will be placed either manually or automated by a robot or a similar feeding system. After put in place a standardized digital camera with an high-resolution Schneider lens captures the surface that a computer unwinds afterward. The part is either rotated or captured during free fall by several cameras placed in a 90 degree angle. The captured image is subsequently compared to a reference image containing flaws. The errors are grouped according to error types and may be adjusted individually depending on sizes and tolerances. After the testing the system sorts out the parts automatically.

The software we use for comparison is based on the machine vision libraries Matrox or Halcon whereas we combined the advantages of both libraries into one power-full application. Errors may be highlighted in color for visualization and can be saved for further post-processing steps. Our software solution uses integrated statistics and logging features which can easily be adjusted to customer demand.