Rotary to linear translation - an interesting idea.

A long time ago, someone I don't know emailed me the details of his
homemade DRO that used rotary encoders. I've long since lost the
emails, so I can't tell you his name .. too bad.

I was concerned about two issues with rotary encoders and had come up
with no satisfactory solution - his idea purported to solve both.
Unfortunately, the parts are still in a box, awaiting the completion of
about 25 other projects, so I can't give a definitive answer.

( I was REALLY hoping for about 10 feet of snow this winter. Since
Vancouver shuts down at about 4 inches or so, ten feet would give me
lots of time to get everything done. Alas, never more than 2-3" fell
at a time, and always melted before the next snowfall. )

My concerns for linear to rotary conversions were

1. Slippage and lash in the drive.
2. ACCURATELY machining a pulley some multiple of PI in diameter.

The guy's idea was this.

Start with a piece of shim stock, just longer than the travel to be
measured. .008" was recommended thickness. Clamp it at both
ends and stretch it very tight, so that it gives a "twang" when plucked.
Machine your clamps so that the tension is adjustable.

Underneath this piece, place two bearings, about 3/4" OD, about 1" or
so apart, aligned so they JUST touch the shim stock without actually
deflecting it. ( The ASCII drawing below shows too much space )



----------------- shim stock
O O two bearings


Machine a pulley JUST smaller than what you think the size should be,
attach to the shaft of your rotary encoder, and support with bearings as
appropriate. Locate this pulley on an adjustable slide 90 degrees to
the shim stock, above and between the two bearings. The slide allows
the pulley to get closer or farther from the shim stock.

O pulley with rotary encoder
----------------- shim stock
O O two bearings


Now adjust the pulley, along its slide, so that the pulley presses
against the shim stock, causing a slight wow in the stock.


--------------- 0 -------------------
\---/

( magnified view of pulley deflecting shim stock, greatly
exaggerating the deflection )

The friction this creates will cause the pulley to rotate VERY
precisely as a result of linear travel, with no lash and no slippage.

While I have only mocked this up and not built a precise version, the
idea is sound, and seems to absolutely solve my concern #1 above.


Here's the interesting part, that I haven't been able to get my head
around.

The guy claims that by pressing the pulley farther and farther into the
shim stock, causing greater and greater deflections ( adjusting tension
as required ), the pulley diameter is effectively changed - a given
amount of linear travel will cause more or less rotation.

This means that you can "adjust" the system so that a given unit of
travel yields precisely the correct rotation.

At some times of the day, this seems correct, at other times it doesn't.
I just can't seem to convince myself one way or the other, although I'm
certainly leaning towards no adjustability.

Still, I do plan to build up a final version and actually test this.
Even if the "adjustment" claim doesn't pan out, the no-lash/no-slip
feature is very attractive.


Alan

--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta