Why Do Model-Based Definition (MBD) and Geometric Dimensioning and Tolerancing (GD&T) Go Together?

MBD + GD&T go together like peanut butter + jelly, mac + cheese, or any other delicious duo of your choice. Dan Feighery, expert trainer, discusses why they’re the perfect pairing.

It’s design interpretation on steriods.”

Harmony between MBD and GD&T is a primary focus of your MBD Using Modern GD&T course. What does this look like at a high level?

We spend some time in class getting everyone to a base level understanding of geometric tolerancing. Once everyone’s reset their base understanding of it, we do a demo of how we would utilize GD&T with MBD. I walk the practitioners through my analysis in my head if I were to open a new MBD file and look at it. So when we pair GD&T with MBD and we open a file or design and look at it the first time, I’ve got the ability to dynamically rotate the part around in 3D space and really get a better physical understanding of the part than I could in 2D. That’s the obvious part. But then I start looking at the GD&T, which is just another language to communicate the design intent to us. 

The human brain can process symbols way faster than it can words, so combining GD&T type of language with MBD is shoving all of this intent or these design requirements into a package that then we can consume incredibly quickly. It takes me about three or four minutes to talk through my analysis, and in reality, if I’m just looking at it, it would take 45 seconds. I demonstrate that with MBD and GD&T paired together, I can rotate the part and understand immediately how it’s going to fit into the assembly. There’s no assembly information there. I just have a single component or a single part design, but the way it’s annotated, it’s communicated to me how it fits into the assembly. So that’s first. 

And then I keep rotating around and looking at the various annotations, and with the associativity, I can click on it and not have to do a hunt-and-peck on my own, but instantly recognize that these are the features associated with this annotation. I’m able to see which features are critical based on the tolerances. Again, within seconds, I’m processing this and rotating the part around. And then I can also see holes, or widths, or slots – features of size. I’m able to tell whether it’s a clearance application, or if it’s a press-fit application, or if I’m concerned with how thick the wall thickness is adjacent to that feature. I could never quickly get any of that out of the old 2D direct-dimensioning method with flat drawings. I’d be flipping back and forth between pages. 

So it’s design interpretation on steroids, that’s why it’s such a powerful tool. You can get a partial efficiency just using GD&T. Same thing if you just want model-based definition. But you’d be doing your designs and your colleagues and your suppliers a disservice by just implementing MBD without the geometric tolerancing. Together, they convey all of the design intent up into the assembly, which is why it’s so powerful. And even now I think it takes me five, maybe six minutes to explain all that, when really that’s a 45-second analysis in the person’s head when they’re looking at a design in model-based definition with modern GD&T. 

What are some other benefits of pairing MBD and GD&T? 

The associativity. So with model-based definition, you’ve got annotations in there. The annotations, whether they’re feature-of-size dimensions, or geometric tolerancing, or surface finish requirements, like weld requirements – the fact is that they’re all associated with a model. It’s not intended to make things easier for a human, but visually, when you click on something and it highlights all the features, that really establishes confidence and a better understanding for the consumers.  

But then there’s the fact that depending on the downstream application or computer or machine or software that’s consuming it, nothing has to be reproduced. Now we can take it and add to it and build on it, but we can also reuse. You’re eliminating all this waste from a LEAN standpoint that we’ve traditionally accepted because it’s been status quo. 

If somebody implemented MBD and didn’t do GD&T, where are the problems going to show up? 

I think the biggest problem is going to be that it’s not the right info you need downstream. What’s the feature? What’s important about the feature? What’s the feature giving me from an applications or performance standpoint? In combination, you’re doubling down and getting your power downstream. Otherwise it’s half effort, and there is still a lot of downstream work to do. 

On the other side, what if you just started GD&T and applied it to your two-dimensional drawings? 

You’re making progress, but you’re limiting yourself to the 2D, you’re not getting the benefits they see in class. So yeah, it’s a step forward, but it’s smaller than it could be.  

Pick a saying, right? ‘Go slow to go fast’ or ‘take baby steps,’ but these are two steps that can be taken together, and they’re not going to cost you any significant additional turmoil by combining them and implementing them together. There’s always a learning curve, but the savings you’re going to get out of the two together is better ROI from a time standpoint, if nothing else. That doesn’t even factor cost and lost opportunity, where we know there are additional benefits. 

---

---

---