[archiver]

Posted by Dodd Stacy at 05/11/2001 10:04 AM EST


Engineer on team #95, Lebanon Robotics Team, from Lebanon High School and CRREL/CREARE.


In Reply to: sticking with double d, thanks...
Posted by Joe Johnson on 05/10/2001 10:53 PM EST:



: Intellectually, I know that a well designed square hole
: and shaft pair can work just fine.

: But...

: Emotionally, I just can't get past the feeling that
: spinning things aught to be roundish.

: Since, for me at least, engineering is primarily an
: emotion driven experience, I am sticking to my beloved
: double d solution.

: Also, I have always been able to make
: them work well for me, so why change now?

: Joe J.

: P.S. I am not kidding when I say that engineering is
: primarily an emotion driven activity for me. It is the
: emotional part of my brain that tells me when the
: direction I am going is a good one or not. I credit my
: emotional thinking with nearly every innovative design
: I have ever come up with.

: Of course, the intellectual part of my brain does the
: detail work much later but typically this is well after
: my emotions are sure of the right path: Either I love
: this new idea of mine and I will stick with it through
: thick and thin or I despise that frumpy tart of an idea
: and I am going to chase after that sexy new number I
: caught a glimps of a few moments ago ;-)

: Many of the creative engineers I admire have similar
: approaches to engineering though perhaps they don't
: express it in quite the same terms as I do. JJ

I can't add a thing to this sentiment.

But for transmitting torque from a shaft to a disc, all of the various spline shapes (square, double D, key, etc) and pin details that have been mentioned involve varying degrees of stress concentration. There is a family of spirals that can be employed in splines with various numbers of lobes that uniformly distributes the stresses in the shaft and disc resulting from torque transmission. If you have EDM capability, it's not difficult to cut this shape. An ellipse with moderately high eccentricity is a reasonable approximation of a 2 lobe implementation of this approach.

Dodd