1. Brock Brown says

"Endless racks and shafts." heh heh

I think we're saying the same thing two different ways: either make a series of approximated cams or set up a fancy cam-driven machine to operate the working head on a lathe.

That would be the fancy method of doing it. Considering the technology of the time, what they probably did was define the 3D shape mathematically using calculus, then set the longitudinal variable to some given value before solving the 2D cam shape. Then glue, putty, sand, etc.

Wiki says that there were cam-based "gun stock copying lathes" in the 1820s and cam automation was advanced by WW1, so some sort of automated machining is possible.

4. Federico Jimbo Smithson says

very thanks

5. christo930 says

Very good, thanks for posting. I hope you can answer this question. I have seen machines that can do amazing drawings. I saw one that draws a very complex ship, probably in the Franklin Institute in Philadelphia (where I live). What are these drawing machines called?

6. iliasasdf says

Fucking awesome.

7. The Serious Account says

Why are these old educational videos so much better than modern ones? They are so explanatory and easy to follow.

@xphreakyphilx There's a few ways of doing it. There are measurement errors associated with all of them, but here's the easiest to understand: Take a bunch of thin sheets. Each one is a solution set for a single range. Stack the thin sheets together and then putty and sand for smoothness, and you've got the 3D cam master. Take a mold of it, then you can cast as many 3D cams as you like.

9. BitUnWise says

how did they manufacture the 3d cam to be mathimaticaly correct?

10. navyreviewer says

@RangerGordon Sorry, accidently pressed post. So you'll always have you're numerator, denominator, and convergent. Putting them linerally is simply to appease human astetics. Where the three parts are in the equation is irrelevant. The beauty of a computer like this is it can always balance the three, and it can continually do it on the fly.

11. navyreviewer says

@RangerGordon In some cases the component chosen as output depended on the application. In many cases which part was the output component could change. That is to say the various parts could "talk back" to the others, changing the equation layout. For example. 2 + 2 = 4. This could also be expressed as 4 = 2 + 2.

12. Brandon Burt says

This is very, very cool. That gear assembly blew my mind! Now, at 6:42, when the narrator reiterates that either the spider shaft, or one of the end gears, might be used as the output (with the remaining two components used as input), it's a matter of algebraic truth (2 + 2 = 4, just as 4 – 2 = 2). But, in practice, were these mechanisms set up so that all three components could be physically driven? Or did the component chosen as output depend upon the application?