Bumps in the Road to Expert

By James Woodward - Mechanical Engineer at TriAxial Design and Analysis, Inc.

Certification has never been a particularly important thing in my mind, as I generally look at CAD software as a tool for design in the same way a lathe is a tool for fabrication.  Certification definitely sets you aside for consideration in the rigorous competition of the engineering world, but beyond that, what's the point?

Having coworkers with broader skillsets than mine, however, has changed my mind quite a lot.  I'm surrounded by people that have explored all of the various minutiae of the program itself, led not only by the necessities of problem solving and finding the simplest and most efficient ways to model complex geometry, but also by the more rounded aspects of a standardized testing protocol.

Progression from the CSWA to CSWP and onward to the CSWE, it turns out, is a very good way to flush out the aspects of SolidWorks that you may have overlooked in your daily duties.  

The special case of the CSWE, however, is an interesting one.  Previous tests all function in a general knowledge-base sense, with a cumulative score, but the CSWE requires a passing grade in 4 of 6 subtopics in order to allow taking the complete test; Sheet Metal, Weldments, Surfacing, Mold Tools, and Drawing Tools. 

My particular experience has thoroughly educated me in the ins-and-outs of several of these topics, to the point where a brief review of the topic at hand and a quick overview of the test prep materials available online was sufficient for the Drawings, Surfacing, and Sheet Metal modules.  There is a degree of comfort and competence with the topics relevant to each subtopic which I think is necessary to proceed effectively, and achieving this was a pretty simple matter for everything with which I was mostly familiar.

Surfacing, in general, is a thing most users of parametric modeling have a hazy grasp on, I believe.  It's certainly a different sort of approach - instead of forming extruded and swept and revolved prismatic bodies and sculpting them with increasingly more complex additive and subtractive geometries, it's starting with a desired shape (either in your mind or based on what's in front of you) and manipulating the edges and surfaces that compose it in order to redefine where the material boundaries are.  Working from the outward in, you could say. 

Most often, this really only manifests in the manipulation of existing bodies or to design around an envelope, in a sort of smooth or organic fashion.  The most important element, I would say, is certainly the ability to correctly identify the portions of a surface model which need to be removed, and to remove them cleanly.  Making a working edge from a damaged edge allows the eventual filling of open edges and recreation of a solid body, which comes up regularly.  Besides this, the ability to correctly locate a sketch picture and use it for design guidelines is critical.

Drawings were a curious thing to study - through many a year manipulating and creating drawings, the basic elements are certainly already in place.  Creating a view, adding or altering a dimension, reconnecting a leader line to redefine a dimension, and creating a client-specific standard for BOM or Revision Block were pretty well clear in my mind, as these are things that one does on a day-to-day basis.  But there are definitely other, deeper facets that came as something more of a learning experience than anticipated.

I had limited use, for instance, for the various options available for placing drawing views relative to model features.  It's one of those things that will occasionally come up, and each time it had I'd looked it up for the immediate necessity, but never quite understood fully in terms of operations and potential.  Very much in the same vein, the wide range of potential in any given Bill of Materials is staggering.

If one had to study for anything in the Drawings, section, I'd recommend that.  Manipulating the fields shown in the BOM and understanding how everything is tied to everything else, changing custom properties and configurations, and editing what and how things are displayed via the model properties was the lion's share of the Drawings side of things in my view.

The Mold Tools was also reasonably straightforward, with one very important caveat - as both of my parents used to say (repeatedly and with evident joy) - RTFQ.  Read The...  Question.   This piece of wisdom has cropped up over and over again throughout the years, whenever I'd waste time on an exam trying to solve for a variable that wasn't the one in question, or missing the relevant line in a paragraphs-long essay topic and taking too large or small a bite of the subject.  Over and over, the sound of my parents hammering this one home.  R, T, F, Q.

It seems that, due to a slight peculiarity in the intricacies of the program itself, the correct answer to one aspect of the Mold Tools test can ONLY be correctly answered by selecting the Top Plane as the pull direction as the question outlined.  The topmost surface, despite being parallel to the Top Plane, yielded an answer that was very slightly different from the correct one.  Slight, yes, but definitely enough to be incorrect.  Hours of repeating the operations on the sample I had available and sometimes yielding the correct answer and sometimes not, changing the order of the features in the tree, selecting different options for each one all led me to the inevitable conclusion that when the test called out the Top Plane as the correct pull direction, it meant it.  R, T, F, Q.

And finally, Sheet Metal.  Once again, a perfect application of the 'RTFQ' idea.  If you understand what how the model is supposed to end up, and you understand the basics of K-factors and break styles (and most importantly, where to alter those values within the feature tree) you should largely be prepared.  Assigning materials is quite straightforward, but important.  Mitered edges and sketch bends are commonly encountered (and fortunately for me, had been encountered many a time prior to the test), and knowing the intricacies of how the model is affected by the values used in the bend features will smooth the test considerably.