Upgrading SolidWorks

Ok, so its that time of year again, where my fingers are crossed and my sphincter is taught, otherwise know as SolidWorks (SW) upgrade time! For those that might not be "in the know" SW is the software program that I use for all my "real job" mechanical design work. It is also a stressful time as sometimes the process doesn't go as smoothly as one would hope. I've taken all the precautions that I could, such as uninstalling the older version first, cleaning out the old registry files, and sprinkling chicken blood all over my cubicle and computer. Luckily I don't work around any vegans.

30-minutes into the process and I am at 18% completion for installing the new version, SW 2011, SP4! The funny thing about upgrading on this particular date is that the beta version of SW 2012 was just released today. That's pretty confident of SW Corporation to do. Why you ask? Its because everyone knows that the world will end in 2012, or more specifically, there will be a major change in the world on or after December 21st, 2012. I'm not a betting man, but my vote goes towards the start of the Zombie Apocalypse on the 22nd.

What is this "Design Intent" I keep hearing about?

Design Intent is not a buzz phrase like "ideation funnel" or "value streaming." It is a real term, with a real purpose. Ok, so those other buzz phrases are real too, but I'm just not talking about them at the moment.

Design intent could be better understood simply by reversing the order of the words. Does "intended design" make any more sense to you? Design intent comes into play when using parametric solid modeling programs, such as SolidWorks, ProENGINEER or Catia. As a refresher, "parametric" refers to a model that has features with a meaningful relationship to other model features or data. It is these meaningful relationships in the CAD world, these parametric relations, that make 3D modeling so powerful to us in the engineering realm.

Back in the days of 2D CAD (way back to the early 1990s), programs could not incorporate design intent into a model. It could display geometry of a design, but the model itself did not contain any information beyond the start and end points of a line, or the diameter of a circle to construct the model. In the mid and late 1990s 3D CAD powered onto the scene, allowing the model to become a virtual representation of a design. Associated data became contained within the confines of the model file, not simply the geometry required to construct the model. These other data covered information such as density, material, and the locations and sizes of features. This data could then be retrieved and modified at any time. When these modifications were changed, due to parametric relationships, the entire model would update itself. Of course as some of us have discovered, the extent and scope of these "automatically updating" features were highly dependent on design intent.

To build a robust parametric model (that is also easily editable), some design intent in the form of forethought on how the model is to be constructed must be given. After all, designs are conceived to fulfill a purpose. Take for example a strip of metal requiring a hole in the center of its width, much like a lawnmower blade. The design mandates a hole to be drilled in the center of the part's width and length. If you model the design by dimensioning a hole half the width from the edge, it is indeed "in the middle" of metal strip, but you have ignored the design intent of the hole. If the width changes in the future your hole would still be a half the original width away from the edge, and no longer in the middle of the part. Had you placed your hole in the mid-plane of the metal strip, no matter how the width changed the hole would remain in the middle of the strip. The design intent of the hole was maintained, and all it took was a little forethought during the modeling process.

This forethought does not stop at the part level however, as parametric relations are not confined internally to part models alone. Parametric relations can span between several models, along with their associated drawings, or be driven by the top level assembly itself. For this reason, design intent is very important and should be considered early on in the design phase of every project.

How do I DO what I do?

I was recently asked, "How do you Do what you do?" Huh? It took a moment for me to grok that question. Was I being asked about my workflow? Was I being asked to train others? Did someone think I was crazy or crazy-like-a-fox? As it turned out, the correct answer was "all of the above." I rambled through some explanation that satisfied my inquisitor, and the more I thought about it, the more I realized that I needed to write something down for my own sake. You get the benefit of reading about here.

When I am faced with a design challenge the first thing I try to do is understand what the design scope is. For example, a customer might comment about a panel being flimsy.

1. They could be saying the panel is too flexible and are worried about drum-skinning.
2. They could be saying that it won't hold up in the user environment and want it more robust.
3. They could be saying that it just looks too cheap and are worried about user perceptions.

All three examples are covered by the "flimsy" comment above, and all three require different approaches in the design. Pick the wrong direction to precede and at a minimum you look like someone that doesn't know what they are doing. In the worst case, you or your employer just lost a customer. Design is a very subjective field, and the sooner a designer knows what the customer truly desires, the sooner a designer can create an item that will satisfy that customer. Of course there is a fine line between what the customer wants and what they need, but I'll save that for another topic. Ask questions, and make sure there are no Yes or No answers. Get your customer to paint a full picture for you.

When I am called to discuss a design I sit back, listen, and take notes. Usually there are others in the room that have spent a great deal on time on a project already, let them talk. They know their product or manufacturing capabilities, and combining the two can only yield one design outcome that they can see. These existing personnel are too close to the design, so close in fact they are married to it and can't see where improvements can be made. This is all right in the short term, at least they are showing some compassion for their work. Its hard to come into a situation like this, as I may be unfamiliar with the industry or the unique challenges that are encountered. The best thing to do is listen to all the arguments being presented, draw on my past experiences, and formulate something that tries to address all the issues. I'm there to provide a new perspective on the problem, collectively being known as "thinking outside the box" and "ideation". Sometimes a successful design might require suggesting a manufacturing method that is not common in the customer's industry. Sometimes I have to literally "turn the world on its ear" and suggest a different assembly process that gets the design back to familiar and manageable territory. You can't get faulted for paying attention and listening, just be sure you stay focused on the task and are ready to provide solutions when its your turn to speak up.

When I am developing the design, I am considering the entire system. How will my widget interface with other items? How will manufacturing produce it? How assembly build it? How will end users interact with it? As I develop my models in SolidWorks, I try to envision what the final assembly will look like. This will lead me to where the assembly origin should be, and in turn tells me which components are the critical driving factors for the design. This also gives me further insight into how each part should be modeled, as in when a part should be a revolved feature or multiple extruded bosses. From the base sketch, should I extrude in only one direction in the plane, or should it be a mid-plane extrude. The proper model construction should be intuitive at higher levels of the model. It should make sense that if I want to grow a particular feature that I should only have to select that feature and modify it, not find myself in a cascading chain of changes.

Age of Computers

Its easy to see that the Age of Computers is well entrenched in our lives. Everyone from kids to seniors are using them today. Some jobs rely on computers more than others, especially engineering. What I find interesting is that even though I have been around and using computers for the last 20 years, there are times when I find that I must still break open my "art kit" to get something accomplished.

My art kit contains a bunch of colored markers, india ink, paint brushes (for the ink), quill pens, mechanical pencils, lead holders, and drawing templates. Although I am well-versed in the use of 3D modeling software and graphics programs, sometimes what gets the job done is a nice hand rendering of an object. I think that's cool.