How optimal is your CAD model?
I don’t know a designer or a CAD modeler who hasn’t been through a phase where he/she was speaking about his/her work in artistic terms rather than technical ones. I also don’t know anyone who hasn’t had to face someone criticising a design through an artsy tirade leaving them clueless as to what is exactly wrong with their model or how to improve it. Don’t beat yourself up about it. Just keep two things in mind: 1) you have had this attitude at some point and 2) therefore should be understanding and patient. And in order to assess your CAD model, you should quantify or ask for it!
“Your CAD model doesn’t “breathe”, it just doesn’t give, like, a positive vibe about its whole structure...”
“…I don’t know, it just doesn’t feel right. Your CAD model can be improved, shapewise, I think…you see what I mean.”
No, I don’t see what you mean, and unless we’re facing a white CAD model on a white background that I’m meant to display in an art gallery, don’t go on talking about how optimal my design is without giving a single technical criterion. To quantify is to give a clear, precise appreciation of work based on parameters that we, as CAD modelers and designers, should know enough about.
As much as I want to speak about my CAD model in terms of graceful shapes and sober aesthetics relaying magnificent simplicity and innate beauty, I tend to rather say that my main intent was to make it as aerodynamic as possible, or that I’m softening every edge pretty much everywhere.
Criteria are too numerous to assess the model
If you are ever tangled in a discussion over technical CAD but weirdly filled with adjectives, don’t play the blame game over who is supposed to bring up the technical aspects or be accurate. After all, maybe you asked for insight from a colleague passing by and he/she spoke off the top of his/her head. Maybe you didn’t present your CAD model during the meeting with quantitive attributes depending on the audience.
There are many possibilities for a CAD discussion to turn into an artistic discussion or simply one based on preference. You, however, should just cut the losses and be more specific with your questions, with what you want your critics to focus on, or even what aspects they are exactly talking about. A CAD model has numerous parameters to determine how optimal it is and it’s crucial to not only be aware of that, but to also have an understanding of the basic ones that you will need.
The assessment parameters will depend on your audience and your requirements. We will make sure to speak thoroughly about them in the future.
But for now bear in mind the basics:
A matter of price
Your CAD model is optimal when it costs the least. This means cheap manufacturing, cheap generic parts and cheap assembling. Therefore, if this is your main area of trouble, narrow down the question to the shapes you deem a bit difficult to render, or materials a bit more expensive than the usual average of your company/workshop. If you are asking someone about your CAD, tell them simply: “do you think my CAD features won’t exhaust the budget? What do you think of the mechanical system in term of assembly cost?”
A matter of manufacturing
Price aside, even if you can afford a top notch 3D printer, a convulted shape with weird holes would still be a challenge for it. Ask people versed in manufacturing with the pieces of equipment you have, most specifically the workers that use it: “Are my models too hard to manufacture?”
A matter of space
This is one of the trickiest matters you might encounter while designing a CAD model. Are you designing a part, an assembly, or a whole machine? People see and feel there is something wrong but can’t verbalize it correctly and might inquire about weight or volume or cluttering. I’m giving you the simplest way to put it : “is there enough void within the CAD?”. As cluttered as assemblies get nowadays, to leave room within is crucial and often underrated. People in the field since the eighties might point out with infamous expressions such as “ your CAD model doesn’t breathe” or “your assembly is too cluttered.” But all these mean one thing: have you thought about the void ratio within your model? Can lubrification, dilatation, errors of cotations, and equilibrium be fulfilled?
A matter of analysis
This means a wide array of criteria but basically, if you want your CAD model to be optimal, the first thing you need to do is ask your analyst what he/she “doesn’t” want in the model and what possible modifications could be requested. These introductory and rather simple questions are the mere door knobs to a broad gate of requests. Optimizing CAD models for analysis is becoming more of a job in itself and is in fact, when improving a part or looking to assess its functioning, the sole requirement. Therefore, a CAD with removable sharp features or an easily modified sketches is a good start.
Be technical, be quantitive, but mostly be quizzical.
The main point is to be the one initiating the question so as to be as specific as possible when asking. Dare to ask and try to zero in the points that matter to you. If more feedback is coming on aspects you haven’t observed – even better.
But still, make sure to rephrase and ask the person about quantitive details. You’d think engineering a part is mainly about handling the physics and the software, but communication is such an underrated and powerful tool. Be aware of it and use it to your advantage.
Modernize Your Engineering Curriculum
This eBook is aimed at the engineering professor, new or experienced, that is interested in how their peers are thinking about the challenges associated with modernizing.
About the author: (Khadija Ouajjani)
Mechanical design engineer working for the aeronautics industry, obsessed with improving CAD to improve analysis, doesn't accept numbers without a precision scale, bookworm and aspiring writer, obsessed with garlic and Mandelbrot, loves tea with mint and coffee with cinnamon. When not writing for GrabCAD, she's probably drawing fancy propellers and coloring cube trees in a coffeeshop.
All posts by Khadija Ouajjani