# Pieces of a Puzzle

One of the things I most enjoy about teaching newbies is the need to come up with different ways to explain important concepts. Today’s high school students have had a completely different educational experience than us old codgers, different also from those soon to be graduated engineers or their forty something supervisors.

How different? There is a popular meme making the rounds of social media starring two teenage lads trying to use a rotary dial phone. I am sure there is another one concerning a pay phone, although where you would find a pay phone these days is a mystery.

I refuse to bash people who have been spoiled by the Internet providing instant answers to any question that crosses their mind. We have encouraged them to belief that if you ask enough people and involve enough computers you will get the “right” answer.

But is that a realistic expectation? Stuart Curtis, according to his widow, needed a solid week to design a single gearbox. That was probably in the very early stages of the computer assisted design days and I suspect he had to write a lot of code during that week.

I averaged one custom gearbox a week for twelve years at one point in my career. We had over (100) individual gearboxes in various stages of production during 1999 and it was only possible because of CAD.

My concern with training young engineers and designers is that computers make it very easy to design very poor products. The old joke that a camel is a horse designed by a committee could be applied to a computer designed gearbox without skilled operator input. “Garbage in, garbage out” is still the basic principle of computer programming.

I hope it clicked for this week’s students when I compared the reverse engineering of a gear set to a Sudoku puzzle. All the things you measure or specify on a gear are related to each other. It is a complex web of simultaneous equations, when you start changing pitches or helix angle or rack offset coefficients, it ripples through the other factors.

And the gears are typically the easiest thing to design in a gearbox. We have collectively put a great deal of thought into gear geometry. The same cannot be said for other components like housings, shafts, and lubrication systems.

Perhaps Gear Technology should solicit technical articles on components that do not require teeth.