Achieving Gearbox Design Excellence

Simplicity and engineering judgment help ensure successful gearbox designs.


In this month’s column, general concepts related to design excellence are discussed. The first one is simplicity in designs. Always seek the simplest and most elegant way to achieve the design goals. Almost any engineer can design and create a complex and complicated device, but it takes skill to reduce a concept to its most simplified and acceptable form. Everything else typically adds weight, cost, size, difficulties in servicing and maintenance, manufacturing, reduced reliability, or other problematic issues.

I like to think of design — and in particular, gearbox and geared system design — and its designer comparable to a sculptor who takes a block of stone and removes material from it until he has achieved the desired result in shape, form, and function. From that perspective, this quote by Antoine de Saint-Exupery is most relevant: “A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away.”

And as Colin Chapman, founder of Lotus Cars, famously said, “Simplify, then add lightness.”

This brings us to the next important concept: knowing when to stop working on a design. When there is nothing left to remove, this tells the creator when the design is finished.

It is easy for us as engineers to ignore this signal and to continue trying to refine and improve the design. Declaring a design freeze by putting a stake in the ground and agreeing to move on to the next phase of the program is a necessary step in good design practice. It is good to remind ourselves to not let the perfect be the enemy of the good, or great.

To a great gearbox designer, experience and ability provides a sense of size, scale, proportion, and relationships between components.

A former colleague of mine believed that, even with CAD computer drawing programs, a great gearbox designer should be able to take a blank sheet of paper and draw a complete gearbox on it with all of the internal components, with only limited drawing tools or by freehand, to the exact scale if possible. And when the gearbox and all of its components were measured and analyzed with their stresses and lives calculated and found to be sized correctly and adequately, the design would be found to be acceptable and would meet the design requirements with few corrections needed. This exercise works much better with pencil on paper rather than with a CAD computer drawing program, because the CAD program does not offer a sense of size and scale that a full-size paper drawing does for most designs. This is still a good training and evaluation exercise.

Another example of this concept of size and scale is the story of an experienced design engineer being asked what size a particular component should be. To indicate this, the engineer held two of his fingers apart for a distance and then had his assistant measure and record the gap between his fingers. That is the correct size that the part should be made to, and it will be correctly sized, within a small margin of error. This is not a guess or an estimate, but rather, true knowledge and understanding.

Finally, let’s further explore the concept of engineering judgment. One trend that I have seen over the years is the increased power and capability of the computer software tools available to us. An example of this is FEA (Finite Element Analysis) programs, which are often used to calculate and display the relative stress and displacements and locations of the stresses in a particular component. The results are typically displayed on the computer screen using a digital model of the part. Stresses are typically depicted using various different colors to indicate the stress or displacement values. Results and accuracy depend on the quality of the input data and assumptions as well as the skill, ability, and proficiency of the FEA analyst that is setting up and creating the FEM (model) and performing the analysis.

The results of this analysis must then be evaluated by a qualified and experienced design engineer to interpret and summarize the results. If the analyst did his or her job well, creating an accurate model, applying the right forces in the right location, etc., then the design engineer can have more confidence in the results.

However, it still requires sound engineering judgment by him in order to use and depend on these results. We shouldn’t blindly accept data like FEA or other similar engineering calculations and simulations and information without thinking about the range of outcomes of this data in terms of size, proportions, and scale, as previously mentioned, and whether, in our experience and knowledge base, these numbers make sense and seem accurate. FEA results can produce pretty pictures with beautiful colors, yet can have incorrect and inaccurate stresses, displacements, and other related data.

It’s helpful in these areas to trust your intuition and judgment. If something doesn’t seem right or feel right, investigate it further and confirm and validate the results.

I have found that there are good engineering analysts, and there are good gearbox and geared system design engineers. The best ones can do both well, but this is not as common as being good at and specializing in one or the other.

Keeping these thoughts and concepts in mind can help to ensure engineering design excellence and great gearbox and geared system design.

Previous articleQ&A with Allen Williston
Next articleRochester Gear Inc.
is president/sole member of Innovative Drive Solutions LLC, which provides gear design engineering services for optimizing gears, gearboxes, and other geared devices. Miller has created over 300 original gearbox system designs. He is a member of AGMA, SAE, and ASME, and he holds three patents. He is vice chair of the AGMA Vehicle Gearing Committee. For more information, visit