Two files are similar looking (one is the reference , the other is what I need to analyze). I need to check how much accurate the one I have is and what is the level of error?
The softwares that I have used are meshlab, cloud compare and netfabb.
They give some results but I need something more reliable.
The way I usually compare two similar models is to overlay them one on top of the other in an assembly, make them different colors, and look for the differences.
Since the only tool at my disposal is Solidworks, I was able to import the files as parts and place them side by side in an assembly, but it's more or less a visual comparison.
The files don't have a common 0,0,0 and they are quite irregular, geometry-wise, so it's really just an eyeball approximation. If they shared a common origin it would be easier to place them atop one another and measure the differences.
Here's a 3d PDF and step file you can look at, perhaps they will provide you with some useful information.
Good luck, maybe somebody else will come along with more thorough methods and advice.
Mesh R4 repaired and Rscan4 Repaired files.
Thanks for the information. I do have files with same coordinate system which overlap. I was not able to align them on Solidworks. I am attaching those files here. Is there a way of comparing them now? Like could I get X Y Z coordinates of both the meshes and compare them on excel for error percentage.
Comparison software exists, but it usually costs a small fortune. I'll take a look at the models tomorrow morning in Geomagics Control, and see what they look like.
The software designed for this process is handy, because it will help align the meshes with each other. But, if there are no defined datums (i.e. holes, or planes), the best alignment in that case really comes down to an average (unless the two meshes are exactly the same).
Otherwise, the process Robert described is pretty common, but in this case it would be very difficult to align the meshes by hand. "Common" parts with flat features and holes are a better candidate for a manual alignment method.
Some models and images are attached.
The images show the alignment. Grey spots show places where there was not enough mesh/overlap to determine deviation.
The two stl files are saved with new coordinate systems so you can view them in your own stl/cad software.
Angad, The first two images I attached above show the coloring of a deviation map. The units are in inches, and show the part is huge, so you'd want to divide those values by 25.4 if your original units are mm.
This was helpful, I have been able to do the same on my STL software. I have a question regarding how to quantify the difference or error percentage? I uploaded new files with same coordinate system which overlap, does that help?
Does geomagic control have an option of error propagation between two models?