I played a bit with your files, but I couldn’t find a proper solution.
Things I found:
- It returns the origin only, if the X value is positive. That’s why two of them shows up correctly, the other ones don’t. If I move the failing masses to the other side of the origin they show up correctly. I tried with spheres in a generic model family, the result is the same. At some positions it returns a null value.
- I replaced the spheres with cubes and it showed the correct centroids, so the main bug is that they are spheres, and centroid calculation fails on them for some reason.
- I could also reproduce a similar bug in a new empty file, spheres sometimes add null values even with mass or generic model spheres. Cubes never fail.
I guess it’s a bug in the centroid calculation of dynamo.
A workaround would be to use the Element.Getlocation node, it doesn’t fail, and extracts the same coordinates, if you have to use spheres.