One of the scariest things a gear designer can face is a blank sheet of paper. This is why most of our products are “derivative” rather than “revolutionary.” I am very skeptical of anyone who touts their “breakthrough” in gear technology because our field has been “well plowed” by some of the greatest minds in science for several thousand years. Almost every “new” concept gets exposed as having already been studied by one or more of those geniuses. Aside from the obvious advantage of using the Internet to “expose” these claims, we have never been in a better position to learn from the successes and failures of others. Well, maybe not the failures as those tend to get locked down rather quickly by the legal department. Your competitors put an incredible amount of technical information on the Internet and you need to be familiar with it. One of the most frequent questions I am asked is “How big does the gearbox need to be?” A good way to start answering that question is to look at what the competition would offer. A better way is to fully understand your own technology and to have up-to-date computer models of all your products ready for customization. Those of you with “legacy” products will be amazed at the upgrades possible with little change to the overall cost or interface dimensions. Gears have gotten better, bearings have gotten better, and our understanding of stresses has gotten better since those products were first designed. Many designers are surprised to learn just how “scalable” gears are. I make sure the overall proportions are optimized before delving into the specifics of a particular gearbox. By this I mean the reduction ratios at each reduction stage, the face width to pinion pitch diameter ratio, the tooth counts, and the face contact ratio. If you get those factors “right” and run a math model of the set at 1 NDP it is quite easy to arrive at the approximate size of the set needed for your load level. Metric users, please excuse my backwards ways. This proportionality trick will work in the metric system too but I have not bothered to do the math on it. For customary units [feet, inches, horsepower], if you take the cube root of the ratio of the 1 NDP set output torque to your required output torque, you end up with a very close approximation of the NDP needed for your application. The “approximate” is because the dynamic factor varies with the size of the parts. A more thorough exposition on this sizing method can be found in my 2009 AGMA Fall Technical Meeting paper “The Effect of Gearbox Architecture on Wind Turbine Enclosure Size.”