While 3D printing has been widely embraced as a means of speeding up the product development process X-Ray 3D scanning has been somewhat less utilized, despite its tremendous usefulness. Part of the reason for this may be that while a 3D print can be held and touched the data files from CT scans are enormous and difficult to process for anyone but the company making the scan. This post will walk through the analysis and reverse-engineering capabilities that CATIA can offer, and hopefully provide some ideas as to how you can put CT scanning to use in your development process.
The image on the left is from the raw data file that was output when a minifig was scanned with an X-Ray CT scanner. It has been loaded into CATIA, where the points have been plotted so that the internals become apparent. CATIA with the Digitized Shape Editor workbench contains a suite of tools for importing, segmenting and editing stl files like this one. CATIA lets you trim any noisy data away, and section down to the parts of it you’re interested in through sweeping, flooding and 3D box-selecting. Once you have trimmed down to the region of interest, then you can use the Quick Surface Reconstruction workbench to create surfaces from the points and to compare the mesh to nominals.
On the right, you can now see the meshed minifigure after a few cleaning extraneous points that came in from the scan. With CATIA you can take measurements directly from the part by fitting primitives to sub features, for instance the peg on the top of the head has been fit to a cylinder in order to determine its diameters. What’s more, you are able to generate a 3D color map of deviations from the mesh to the nominal shape, as in the second image below where the hand grip is compared to a perfect cylinder.
As you can see, the grip is very nearly a perfect cylinder, with almost the entire area within 0.008mm of nominal. The only real deviation is at the bottom where the hand has started to slope away from the circular grip. This image says quite a lot about the quality of the injection molding process that produced the hand.
Continuing the analysis, you see in this image that deviation plots have been added on the top peg and the cylindrical regions of the hip, and a thickness map has been added at the chest. The thickness plot has the effect of colorizing the embossed graphic on the figure’s chest, which was only barely apparent in the raw data.
If you wanted to characterize the critical features of this part you would likely look at the peg and the holes, since these will drive the interactions with other blocks. In the image below the diameters of both the peg on the top of the head and the hole in the bottom of the leg are measured by taking the diameter of a best-fit cylinder of each. As you can see the dimensions are the same right down to the micron, but interestingly there are two small flats on the hole that cause the interference that locks the parts together. The fit will be tight since it’s line-to-line everywhere else, and the flats will insure that the mating parts will stay attached even as they wear through play. This is very impressive molding capability for such a small and inexpensive toy.
Finally, the image below shows some of the internal structure. This technology makes it possible not just to see how the figure is put together, but to quantify fits and dimensional values inside it in a totally non-destructive manner. Hopefully this will give you some ideas about how CATIA and CT scanning can be used in your development process; all dimensions from a typical part can be taken for first article inspection, and beyond that surface textures and figure can be very precisely quantified in a way that would be impossible with any other approach. In subsequent posts we will show how CATIA can be used to compensate for warpage after molding and how to reverse-engineer designs from scanned objects.
If you have any questions or would like a demonstration of the capabilities of CATIA to manage and analyze point cloud data by all means please reach out to us.