Point "On Center" Of A Surface

[dplachinski]

Give the user the ability to create a datum point at the center (centroid) of any surface (in core modeling). Similar to creating a point "on center" of a circular edge.

Creo at this time does not have the ability where NX and SolidWorks both do.

I have seen many "work-arounds" where you have to thicken a surface, extract the center of gravity dimensions (X,Y,Z) from a coordinate system and place a point manually by typing in the extracted dimensions.

https://support.ptc.com/appserver/cs/view/solution.jsp?n=CS73023

https://support.ptc.com/appserver/cs/view/solution.jsp?n=CS40552

I created a random surface in Creo with curvature in both X and Y directions.

I then brought the surface into NX to create a point on the center with ease. In NX you give it's location as a .5 ratio of the U and V Parameter of the face.

I then brought the surface into SolidWorks to create a point on the center with ease. In SolidWorks, you simply select the option "Center of Face.

[MartinHanak]

It would be nice to explain why do you need this functionality.

[dplachinski]

There would be many uses to know where the center of a surface is and have a point placed there.

One that comes to mind would be; where to place a component in an assembly to encompass a certain surface or geometry.

Another would be, where to place a cross section datum through the center of an object that is of random shape.

[mender]

Out of curiosity, how exactly would this be defined?  I can envision an analysis that gets the centroid point for a surface, and a datum point feature that is driven by this analysis, but in saying 'on the surface' instead of 'of the surface', there is an implicit projection of this point onto the surface, and it isn't clear what direction of projection would be suitable (in the case of an arbitrary surface shape).  Is this just semantic hairsplitting on my part, and in the case of a dome (say) you want the centroid, which here is not on the surface?  Or are you envisioning a particular projection that would be best built into the functionality?

[dplachinski]

I believe you are correct Matthew. The title should be "of" the surface. I will see if I can change it. In the two cases I showed; NX puts the point "on" the surface and uses the U/V parameters. Solidworks puts the point on the centroid "of" the surface. I guess I started this idea with a flat surface in mind, but didn't want to limit it to only flat surfaces. I guess if it is a flat surface, it would be in the same plane anyway. If it is at the centroid, the user could then determine in which direction to project it onto the surface if needed.

Thanks,

Dan

[mender]

That makes sense.  I could see U/V parameters as well, but that would be useful for fewer classes of surface, and many of those would be doable with a datum point on the surface, placed by dimensions driven by relation to read to halfway along the surface.  So the centroid is probably the more interesting case.

Caveat: I don't do the relevant sort of development, so can't say how easy/difficult this would be to do.

[dplachinski]

What is the solution Steven?

I will upload the file as a reply to this discussion.

Thanks,

Dan

[dplachinski]

It was suggested that I change the description to include this functionality in the core product. Apparently this function is available if you have a license for ISDX.

[MikkoHinkkanen]

There are various use cases for this. Especially when you "connect the dots" for BMX analysis.