Community Tip - Want the oppurtunity to discuss enhancements to PTC products? Join a working group! X
What is better way to make a model (casting) keeping manufacturing process in mind?
Please also let me know the priority of feature command (Extrude,Revolve,sweep..etc) that i need to follow during modeling a cast model...
Thanks in advance
There really is not anything specific in the software that must be considered that does not need to be considered in casting design to begin with.
Drafts, undercuts, and slide features are all just features. You address each on their own merit. Creo really doesn't care on -when- you create a feature. it acts on the geometry available at the time of the feature creation.
I consider a model in how I can sustain it. If you build a feature on a feature on a feature on a feature... you know you cannot delete the first one without re-routing or deleting the children of that feature. This is -technique- more than -design-.
Specific to any mold design is draft techniques; This is probably the only feature you want to decide if you build it into the original feature or as a separate feature. Many designers want to add it afterward so they can finish the design and then make it ready for the mold. I am divided on this. I like to look at it on a case by case and feature by feature basis. Creo has the techniques to make drafts and parting lines after the fact, but I find them difficult to use since I don't use them often. If this is what you do on a regular basis, it is worth understanding how these tools are used properly.
Sorry I can't write a better response right now but I'd add one thing to Antonius' post: inheritance.
Look up Inheritance models for us in castings. This might give you what you're after.
It seems there are two schools of thought in casting / machining designs.
One is to design the primary model in the "as cast" state and then use cut features to create the final "as machined" model. The advantage here is that you are designing as the part would be built - cast first, machined second. The downside is that the primary model developed isn't the desired final part, the final design becomes secondary if you will.
Two is to design the final "as machined" part first and then add features to create the "as cast" model. The pros & cons are reversed. The good thing is that you are designing exactly what you want to end up with, the bad is that it's a bit harder to add geometry get to an "as cast" state.
Personally, I like the second approach as it is a more direct route to my desired final design. You have to think about the casting along the way anyway.
As far as what kind of features to use, well, that depends a lot on what you're making. A transmission case is going to have a bit of everything, including perhaps some surface features. A pipe fitting is going to be a lot of revolves & some extrudes.
Doug,
Have you ever done this as a family part where the features that are machined are suppressed?
Thanks, Dale
I've not done very many of these, to be honest. In our business we frequently design the end product (as machined) and let the casting vendor determine what material needs to be added to achieve the desired end result. Of course, there has to be an understanding of what areas must be machined and coordination with the vendor during the design process so that the cast shape is appropriate for what they will need to add. We take the approach that we don't really care how much or what shape of material they need to add to get the final result and we don't want to be in a position of dictating how they do their business.
That said, a family table approach is one method and it would seem to work better with the first method I describe above. It would seem to be easier to add the machining cuts to the family table.
The inheritance feature method that Brian mentions is another and is likely the approach I'd take. It would seem to be easier to manage as the casting and machined parts would be separate but related. All the secondary features, be they add ons for the casting or subtractions for machining, would live in that model instead of in the generic.
I've found doing it as a family table worked bst for me. No external refs. You put enough datum curves and surfaces in it to be it's own internal "skeleton", then use that to drive everything, like all the machining.
Thanks to every one for your kind information