If you have a question not covered, please email me and I will post the answer here if appropriate.
How accurate is scanning?
Accuracy depends on many factors, including object size, geometry, surface finish, colourisation and specific geometry. Typically the larger an item, the less accurate the scan because the point cloud is less dense. Parts with sharp edges and compromised visual access present challenges and need to be assessed on a case by case basis, but a solution can usually be found. If a surface requires preparation before scanning to make it visible, it is the surface preparation that gets scanned, not the object. A thin layer of paint, for example , adds around 30 microns. All these factors and many others contribute to the overall accuracy achieved. As a rough guide, however, small components (< 300mm) should be within 0.05mm or better, and large components (500mm – 1m) within 0.25mm or better, but each project is different. To ensure the accuracy of the scan data, components are checked where possible with conventional inspection equipment.
Does scanning an object produce a CAD Model or IGES file?
A common misunderstanding of the scanning process is that the direct output is a CAD model that can imported into any CAD system. Sadly this is not the case. The direct output of all optical scanners is point cloud data, which is like STL data. A CAD modelling stage is needed if a geometric solid model is required.
What can you do with Point Cloud Data?
If you have the right software, point cloud data tells you everything you need to know in order to reconstruct a part in CAD. Unfortunately most mainstream CAD packages are not particularly good at working with this type of data. Specialist packages such as Rapidform XOR, Geomagics and Polyworks to name but a few are excellent at converting point cloud data into CAD models.
Are there limits on the size of object that can be scanned?
The lighting conditions are one of the biggest limitations to scanning large objects. Small objects can be scanned indoors, so lighting conditions can be easily controlled. Large objects such cars and boats can be scanned in a large garage or outside when dark (and dry). The pattern that the light projector produces must be easily seen by the cameras. The brighter the ambient light level, the more difficult this pattern is to see. A bus could easily be scanned indoors with low ambient light, but not outside during the day.
What do you model when reverse engineering?
When scanning an object all surface geometry is captured, including imperfections caused by the manufacturing process and any damage the part may have suffered. Typically, unless expressly asked not to do so, the part will be remodelled capturing the design intent and disregarding imperfections. There are some good reasons for this. First, modelling in every single defect would be time consuming and therefore expensive. Secondly, one of the main reasons for reverse engineering is to remake the part. Therefore the requirement is to create a ‘perfect’ part representing true design intent. Sometimes the true design intent is difficult to determine due to defects in the part. In these situations, the part will be modelled on a best effort basis, unless otherwise instructed.
What resolution can FlexScan Produce?
Unlike fixed focal length scanners, such as Laser scanners, the resolution is variable. With FlexScan the resolution is determined by the resolution of the cameras and the distance from the object. The nearer the scanner is to the object, the higher the resolution. This is a very useful characteristic because scanning a boat hull does not require points every 0.02mm and would be virtually impossible to process because of the volume of data produced. If however scanning an intricate item like a coin or piece of jewelry, a point every 0.005mm would be desirable. A typical setup with 5MP cameras will produce an approximate point to point distance of 0.2mm when capturing a field of view width of 500mm.
Can models be provided in Solidworks format?
Yes, however models are not always easy to modify afterwards. When reverse engineering a part, the emphasis has to be on representing the original geometry and design intent. As every experienced solid modeller knows, there are many ways to approach a model. Modelling techniques vary and therefore modifying someone else’s model is possible, but not necessary straightforward. Complex models are more difficult than simple ones, and reverse engineered parts are very often complex. No liability is accepted for modelling technique and the ease with which a Solidworks model can be modified. However, assistance can be provided, if required.