A simple question with a not so simple answer. Let me start by saying the purpose of this article is not to explain the intricacies of geometric tolerancing. This is more about why you should consider adding it to your company’s toolbox. An often misunderstood or misguided notion that standard tolerances are always the fastest or cheapest way to accomplish the end goal isn’t always the case. Here’s why:
What is it?
GD&T or Geometric Dimensioning and Tolerancing is a pre-defined set of symbols and nomenclature. Collectively, GD&T provides a system for engineers to more precisely convey the intent of a design when working with assemblies of parts. The schema creates additional methods for engineers to define the relationships between bodies (up to and including the intended motion of a system).
Sticking Point
Unfortunately, there is a stigma surrounding GD&T that it is complicated, costly to implement, and all-around unfriendly to the people responsible for manufacturing the parts. This could not be further from the truth, with a few caveats. In many cases, GD&T can clean up “muddied” drawings with lots of notes and leaders, and actually offer a wider envelope of good parts that satisfy all the critical requirements without sacrificing quality or increasing cost.
One hurdle to overcome, as with any new implementation, is a degree of edification required for successful adoption. I’ve found that it is best to start small and monitor how well the changes are received by employees. Adaptation and adoption include everyone involved from the engineer to manufacturing personnel, and through to quality assurance. Let’s take a look at each individual perspective.
Mechanical / Manufacturing Engineer
When designing a new system, it is easy to fall into a routine that I broadly call “myopic part design” or lacking a sense of depth perception. Each part in an isolated context is generally designed to serve a specific function. After all, form follows function. This motor needs:
- the bracket needs to hold X amount of weight
- securely mounted to the frame at location Y
- needs lined up with output shaft Z
Engineers that have the ability to take a step back and see how everything is interconnected are able to visualize the constraints of the machine and account for what could be costly mistakes down the road.
example: a 6” wide mounting face for the motor that tapers in thickness .005” from one end to the other, attached to an 18” long output shaft could cause a .015” mismatch at the far end of the shaft. That might not sound like much in the grand scheme of things but the person assembling the machine won’t visibly notice the difference and will apply the required stress to the shaft to line up the bearings you (the Engineer) placed in the exact spot they need to be on the frame. This of course does not include any positional tolerances on the shaft bearings themselves. This creates undue stress in the shaft, over time will cause bearings to wear out faster than predicted, and could make the machine challenging to assemble. Remember, time is money especially in manufacturing. This could all be mitigated at the design stage with simple flatness and parallel tolerances applied to the mounting surface of the motor and bracket. Don’t look now, but that’s GD&T at work saving you and your customer money, and time (more money) down the road.
As I said before, GD&T doesn’t reside solely on the shoulders of the engineer designing the parts. That design intent outlined in the drawings translates directly to the Manufacturing Engineer who is responsible for designing work-holding solutions and processes to make the parts. If datum A is the critical surface machined during the first operation, and the feature dimensions reference that face for positional tolerancing, then that is most likely the surface you want to use for locating subsequent operations. Creating new geometry from known sizes and locations reduces tolerance stack up and gives you the best chances at correctly machining additional features in the most cost-effective way. For anyone that has ever dealt with sub .001” tolerances knows that stack-up is a very real thing and is unfriendly at the best of times.
Next, let’s head out to the machines where the magic happens. Machinists/Operators are an engineer’s best friend when dealing with manufacturing issues because they can offer different perspectives, insight into the quirks of the machines they operate, and techniques that have proved vital in the past. They also help keep you grounded because every company has limitations when it comes to equipment, metrology, and skill level among other things. Skills especially are honed by years of experience, not awarded through promotions and pay raises.
Machinist / Machine Operator
After the engineer successfully documents their design intent and applies GD&T in a responsible manner, the machine operator needs to understand how to interpret the dimensions for machine setup and performing in-process inspection for the operation they are performing. Metrology is a science in and of itself when it comes to inspecting parts. There is no “magic bullet” so if you are looking for a do-all machine, I’m afraid you will be very disappointed. That being said, metrology comes in many forms such as gauge pins, indicators and calipers, up to more advanced CMM’s, Videoscopes, and 3D scanners. It is the culmination of education and using the tools available that allow the machine operator to know beyond a shadow-of-doubt that they are making good parts. And the ability to make adjustments to continue making good parts (which is an entirely separate topic).
Lastly, every company that wants to have a fighting chance in this economy has some measure of Quality Control or Quality Assurance (pun intended).
Quality Assurance
This person/department takes one last look to check all the boxes and say “this is a good part, it can ship to the customer as-is.” A verification of every dimension and tolerance used to produce that part and to make sure nothing was overlooked. Some QA departments have more advanced measuring capabilities that are not conducive to being out on the shop floor so they may also be responsible for in-process inspection activities. But there still needs to be a proper level of education and training applied to interpret drawing specifications. A part scrapped on the shop floor is unfortunate but sometimes a necessary evil when dialing in the machine. A part scrapped at final inspection has the additional cost and overhead of finishing operations, employee wages, and sometimes rework involved. Again unfortunate, but these things happen. That is a simple fact of manufacturing.
On the flip side, one of the worst scenarios imaginable is scrapping a good part because the tolerances were not correctly evaluated. Sometimes with larger companies, QA might not have direct access to the engineer responsible for the design if a dimension measures out of spec. There could be exceptions made if it will not affect form, fit, or function.
By that same token, maybe the tolerances are unnecessarily tight and a design change is required. This brings me to the concept of continual improvement but I will save that for another article.
Final Thoughts
There are many facets of GD&T that could be explored in-depth and are not well-suited to the discussion as part of this article. The key takeaway here is that GD&T offers additional tools and methods for documenting the design of parts and systems. Every implementation can differ slightly from one company to the next. Start small and see what works and what doesn’t within your resources. You might be pleasantly surprised!
Please feel free to comment and share! And don’t forget to subscribe for future updates.
Disclaimer: The views expressed here are my own and do not necessarily represent that of anyone mentioned in the articles or of my current employer. I am/We are maintaining this blog in an effort to increase awareness and open the door for dialogue in any form about the history of manufacturing, current trends in the industry, and opportunities for the future. Any direct links on this page may or may not be part of an affiliate marketing program and this site may generate revenue from qualifying purchases. Thank you for reading.
Steve Ritter
Mechanical Design Engineer and Project Manager with close to 15 years in a manufacturing setting. I have experience with many CAD software packages over the years (Autodesk (all), Solidworks, Pro/Engineer, UniGraphics), CNC and manual machine operation, CNC programming, and Electro-mechanical prototyping and system design.
This Post Has One Comment
Wow this is a really good. Nice work.
Comments are closed.