The above graphic is a convenient cheat sheet for what to do and what NOT to do when you’re building a parametric model. Stick to those rules and you’ll be a lot happier at work – and your co-workers will be a lot happier with you.
Now that you have that list, let’s take a step back and take a closer look at how a 3D parametric modeler works. Who cares, you say, just as long as it works? Well, that’s exactly how your CAD system should be. You should be able to leave the inner workings of modeling software to us, so you can focus on doing your best design work.
However, it's worth having a little knowledge about the inner workings just so you can appreciate why things can sometimes go wrong. Horribly wrong. You may think you’re making a seemingly innocent change early on in your feature history when all of a sudden, your Feature List lights up like a Christmas tree and your entire model blows up. This has to be the most frustrating aspect of parametric modeling, so it’s worth a little effort knowing how to avoid it.
Behind the scenes, every piece of geometry created by a parametric modeling system has a unique ID number. This number is used by subsequent sketches, features, assembly mates, and drawings, to work out where things should be placed relative to the rest of the model. So if that ID number no longer exists, guess what? That's right, the feature does not know what to do and fails. If a design change makes an edge or face disappear or changes the topology of your model dramatically, it is likely that some downstream features may fail.
As you are probably aware, the Feature List down the left of the screen is the order in which each feature is evaluated. Another way to think of it is the recipe that was used to build the model. As your design progresses, you can add new features that build upon the features already present to add more detail. This is what makes parametric modeling systems so powerful, letting you build in design intent so your models will update consistently and predictably.
But it’s not the features themselves that are the problem, it’s the references you make between them. References are your biggest ally, but also potentially your biggest enemy. A badly considered set of feature references can make your models very fragile.
Here is a simple example. You create a sketch on the face of a solid model and dimension to the edges of the model. Perfectly okay. Later you decide to fillet the edges that the sketch was dimensioned from. If you create the fillet after the sketch, then there's no problem. At the time the sketch is evaluated, the edge ID exists. Now if you reorder that fillet before the sketch, essentially changing the recipe for the model, the fillet removes the edge and the ID number no longer exists. The sketch fails.
What is clever though, is that the ID is not lost forever. It is simply masked by the fillet. If you suppress or delete the fillet, the edge that was there reappears and assumes the same ID as it had before. Your sketch rebuilds once more.
If you had dimensioned to Sketch 1 (that created the original extrude), the ID would never disappear and your model would be bulletproof. With this in mind, consider which references you are using when building features. If your model fails, then now you know why.
If you know things are not likely to change, for example, if you are building a 3D model from a supplier’s 2D drawing, then you can be a little gung-ho with your references. But who can predict what will happen during a design project?
So stick to the simple rules of thumb listed at the top of this post and your models will be much more robust – as well as faster and easier to edit!
Interested in more modeling tips? Register for our Masterclass on 3D modeling today!