Design of Wooden Machines

Peter Seddon wrote:

> I am interested in building a CNC milling/routing machine and am
> comfortable with the stepper motor drive and software issues. Not
> being a very strong mechanical engineer I am challenged by the
> design of a suitable milling machine. I would like to build a three
> axis machine capable of machining soft materials (wood, plastic etc)
> up to say 500mm x 300mm x 100mm.

> the idea of building something out of MDF appeals, but is it practical
> and does it provide a workable machine ? What type of slideways should
> I use, what type of lead screws can I use - the idea of using ball
> raced double extension drawer runners and threaded rods appeals but
> again is that workable ?.

> Do you know of any suitable designs or have you built such a machine
> and can you give me any hints and tips or pitfalls to avoid ?

I think the group is still looking for a "great" design. I have
include links to all the homemade XYZ router tables (and table
plan vendors) in my webpages (if I missed anyone, please email me).

I would first suggest looking at CADMAN's machine which is very
similar to what you propose http://24.12.163.105/mill/

My own experience in building wooden machines has led me to
the following concepts (which I am using to make my big XYZ router
tables, keep in mind I am concentrating on inexpensive components,
a fractionally 'better' machine could always be built with precision
motion control rails and ballscrews, but I am intentionally avoiding
those ):

- I didn't like the drawer slides. They are fine in compression,
but offer lots of slop in side-to-side movement. That can be
rectified with two slides being mounted at 90 degrees to each
other (per side), as in:
_ _
| | or /\ /\

these would be the bottom slides in a bridge router (the table
moves, but the bridge holding the Y axis is fixed).

The problem comes with sawdust getting in the bearing works,
so I've decided that is not a good way to build an inexpensive
machine.

- I decided to experiment with hardware store 'readi-rod' which
is long lengths of pre-threaded rod. In fact, I bypassed the
'skinny' 1/4" and 3/8" type rod and went directly to the 1" dia
readi-rod so that I could use it as a structural member (in my
wood designs). If you machine one end of the rod to use a
pair of taper roller bearings against an end plate, you get
a nice arrangement. If you use a coupling nut (3" long vs 1")
then most of the nut backlash can be removed.

After building one machine with one X axis screw down the middle,
there were skew problems, so I replaced it with two X screws (left
and right) and of course the skew was removed. Also, now the table
was completely supported by the 1" rods! (We're talking 30"x30"
table size). I'd recommend this (using two X axis screws).

For the Y axis, I had 4 drawer rails in a box pattern. That was
replaced with four 1" dia threaded rods, which also carried the
weight (although it would be easy enough to add a 'L' bracket with
a plastic strip to support any flex in the middle). Using this
box concept, the rigidity is impressive.

- Currently I have replaced the 1" dia readi-rod with 1" dia precison
rolled acme screws, and instead of using 3" long coupling nuts I
am using 8" long UHMW bearing blocks, which run through 1.5" dia
stainless rods-over-beams to support the weight and keep the
axes running true. When I get this running I'll post pictures
on my website, and later I plan offer kits. The size of these
tables are 5'x10' minimum (5'x9' working area) and should go up
to any (reasonable) length. I think I've seen screws from NOOK which
go up to 38 feet.

Doug Fortune
http://www.cncKITS.com