Hi,

> >Hmmm...  sounds like a perfect use for an old 3.5" floppy head mechanism,
or
> >even the head mechanism from a tape drive.
>
>         More or less that! You can use a screw coupled to the motor, and a
nut to carry the z block up and down. It works!

    It sounds simpler than it is in reality. The 3.5 head mechanism is very
very light and it is spring loaded. That is how they get the precision. The
head mechanism of a streamer drive is a much better option but still hard to
get it to be repeatable enough. The problems is much smaller than doing a
mill or a lathe because you do not have any forces besides the weight of the
mechanism but you still have to get everything to be perfectly square ! Even
at a good CNC lathe it is pretty hard to make 2 parallel holes if you have
to take the part out of the "holding jaws". If you have a cnc mill you still
have to make all the rails and screws adjustable. That is the main advantage
of using Roman's approach and using ready made printer mechanisms but my
experience with the mechanisms is that they have too much backslash when
they change direction and that may not be repeatable. For anyone that wants
to make that for a professional use I still recommend going from the ground
up with real machine parts. Look closely at K. Maxon design of the mill and
you will see that all the parts were put together with allen screws to make
it adjustable. Is is much harder than it looks, specially for milling. The
link I sent on the other messages is the simplest clean gantry design that I
have found. the only problem is that you have really to use good guides
because it is driven from just one side and can stop if it skews too much.
You can not get much simpler than that. Belt drive is definitely a good
option because the acme screw precision is too much for the application.

Best regards,
Alexandre Guimaraes

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.