Skip to main content
Oct 1, 2018 12:25 PM UTC
Delivered

Bounding Box as a Feature

Related products:Creo Parametric

Bounding Box as a Feature

Please make a feature in Creo Parametric representing the smallest area or volume in which a design can fit, bounding box.

Please spend a few minutes looking at this video from SolidWorks, it is very simple and easy to understand.

Bunding Box

 

Chees Erik Sallingboe

50 replies

Dec 16, 2021 1:42 PM UTC

In addition to all the ideas above, I've created a fake screenshot with a few more requirements (optionally including quilts and use of subsets in assembly mode)

Hopefully the picture motivates to switch the state of this idea to "In Implementation" 😉

 

product idea: enclosure volume as full featureproduct idea: enclosure volume as full feature

Dec 16, 2021 1:52 PM UTC

@rmueller,

Nice picture!  By subset, do you mean only certain components in the assembly?  If so, could a simplified rep be referenced instead?

Dec 16, 2021 2:05 PM UTC
Yeah, subset should be technically a simplified rep. Here I had the function/button from the shrinkwrap feature in assembly mode in my mind.
Dec 16, 2021 3:49 PM UTC

As an added bonus, there should also be added an option to consider a bounding cylinder instead of a box, to measure the stock of cylindrical parts (no assembly mode here). The internal algorithm could infer the axis (but this is computationally slower), or ask the user to specify the axis or a coordsys axis. The algorithm would process all revolved or cilindrically extruded surfaces (arcs or circles), project all arcs/circles in a plane normal to the axis (and/or centered on the axis, if all cilindrical surfaces are revolved around the same axis), and report the biggest circle in one dimension, and the surfaces length along the cilindrical axis.

Dec 16, 2021 3:58 PM UTC

In the bounding cylinder case, for the algorithm to be more robust, we should also consider the possibility that there are additional flat or nurbs surfaces outside any of cilindrical ones. This would be the case when for instance we select a cilindrical steel stock of a bigger diameter, but machine a "square" inside a circle to have flat faces at one side, and a smaller diameter turn machined at the other end.

Feb 3, 2022 7:29 PM UTC

Same basic idea was posted in 2014 yet PTC still has not implemented idea. 

Can someone from PTC tell us what we have to do to get ideas implemented?

 

https://community.ptc.com/t5/3D-Part-Assembly-Design/Create-a-analysis-feature-that-will-give-you-the-bounding-box/m-p/465666

Feb 3, 2022 7:42 PM UTC

The workarounds suggested using modelcheck in the video do not make sense from a security point of view. We need the geometry around (as a body, as a quilt or wireframe), so that we can quickly visually inspect if the calculated dimensions and orientation make sense for a particular part. Getting the dimensions as a list only can lead to expensive errors when requesting steel or other expensive materials from just a list of values that we would need to confirm manually anyway, measuring each part with a paintstainkingly slow process. With geometry present, in a few seconds we can confirm that the measurments make sense.

Feb 3, 2022 7:57 PM UTC
Should be able to trust that the bounding box feature produces the correct results. If you have to spend the time to double check everything that PTC does you probably should be using a different software that does not do anything automatically for you. I do not want the extra geometry in my models.
Feb 3, 2022 8:06 PM UTC

@WayneSchrock Maybe that's why this much needed feature is not yet implemented. Remember how the A380 was much delayed and had serious cost increases due to the way that cables were automatically calculated? Maybe engineers trusted too much in the software (Catia), and did not bother to check those virtual measurments. Probably there was a lawsuit. Maybe PTC wants to potentially avoid such mess. The way I suggest it, we obtain the speed improvements, but have a way to check if the stock sizes of the parts are correctly calculated, and that a potential bug does not wreak havoc if the user has the opportunity to do a sanity check.

Feb 3, 2022 8:10 PM UTC

A bounding box taking into account only the solids that measures paralell to a provided CSsys should not be difficult to implement, and not be prone to bugs. But people here have also requested for the software to "search" for the correct or optimized orientation, and this takes much more time to calculate, and demands a more robust algorithm... and much more testing to avoid possible bugs. I don't mind that PTC provides only a user specified CSys orientation. Much faster ro regenerate, and it's much better than having nothing, or doing it all manually.