In Stay Sharp Episode 121, Digital Product Series: Sigmetrix – Why Tolerance Analysis Matters in Digital Engineering, co-hosts Jonathan Scott and Juliann Grant take a deep dive into mechanical variation management and tolerance analysis with James Stoddard, co-founder and research fellow at Sigmetrix. As Jonathan and Juliann learn, Sigmetrix is at the forefront of helping manufacturers understand that variation management isn’t just a single-point activity, but something that should span the entire product lifecycle.
Why This Topic is Important
An important step in the digital engineering process happens just before the transition from design to manufacturing. “All your models are based off of perfect ideal concepts,” James explains. “After that point, you go into producing real parts, and those real parts introduce variation.” No two parts or assemblies are the same, and your models need to capture the variation of the real parts—that’s mechanical variation. Tolerancing is the definition in your part design where you specify the variation that’s allowable for each instance of a part. He says, “You can minimize the variation, but you can never eliminate it. The more you do to minimize it, usually the higher the cost goes. So trying to understand what is allowable from an overall quality standpoint is a key part of the design.”
To help companies be more profitable, Sigmetrix bridges the gap between design and manufacturing. James notes, “If you don’t understand that variation, it’s hard to create a good design.” While variation and tolerances are concepts that apply broadly—even to someone cutting wood in their garage—Sigmetrix is valuable for companies that are building something with a degree of complexity and/or a high cost of failure, such as parts for a pacemaker. Cost savings are also important. “What we’re trying to do is trying to make quality affordable and trying to find what is a sustainable level of quality for any given company in the industry,” James says. “Certain industries require a higher level of quality based on their nature, but every industry has a certain quality threshold they’re trying to achieve.”
Sigmetrix provides software, consulting, training, and support directly to customers—not only for the tools, but also for the subject matter. It’s not that manufacturers haven’t been dealing with mechanical variation and tolerances for hundreds of years, James says, it’s where and how they’ve dealt with it. Until recently, they’ve mostly compensated for them on the manufacturing side. Sigmetrix software can change that. “As computers have evolved, as model-based definition has evolved, we can take a lot of what was done on the manufacturing side and push that understanding upfront to the design side,” he says. “Do that type of iterations and prototyping in a virtual sense, and get that understanding before you have to make significant cost impacting decisions.”
Sigmetrix Product Capabilities
The software’s core functionality is tolerance analysis. Because tolerance definitions exist in CAD models, the software integrates directly with the major CAD systems to access that information and then deliver a prediction on critical measurements or assembly characteristics in the design. Their analysis is fast. James says, “We can solve fairly complex assemblies within minutes. It’s not like you have to run a batch overnight. We feel that that’s important. And some of the smaller things, it’s almost instantaneous.”
From there, the software’s capabilities have expanded organically to geometric dimension and tolerancing (GD&T), also known as geometric product specification (GPS) internationally, because the annotations on a CAD model need to be clear and correct. Sigmetrix offers education and training in the “language” of GD&T, as well as a product, GD&T Advisor, that helps designers apply the product manufacturing information in their CAD models.
The company is continuing to extend their capabilities into the product lifecycle, specifically how to link tolerance analysis to both upstream and downstream stages. Upstream, the goal is to send information back to system requirements, to maintain an understanding of how the actual dimensioned parts relate to them. Downstream, the idea is providing information to manufacturing and inspection to enable better decision-making for manufacturing processes and sequences. “Each of those decisions further impacts the variation that gets introduced into the product,” James says. “Having a living model of this variation of the tolerance analysis that can then live through manufacturing so as they make these decisions, they can adjust the definition and see how that impacts the quality. And if we trace all the way back to system requirements, you’ve got a more digital thread.”
Learn More About Sigmetrix
The full podcast contains more discussion and details on a variety of topics, including what the secret sauce is behind their approach to statistical analysis, how Sigmetrix sits at the intersection of a virtual twin and a digital twin, how proper GD&T annotation should (but doesn’t always) reduce the cost of a part, and more.
Be sure to check out Stay Sharp Episode 121: Digital Product Series: Sigmetrix – Why Tolerance Analysis Matters in Digital Engineering and join us each week for a new podcast.
Follow the Razorleaf Podcast, Stay Sharp in Digital Engineering on:
