This is a long post. I have posted in yahoo groups as well. help will be appreciated!
I have been involved in woodworking for many years. My father specialized in
custom wood turnings. We used converted engine lathes for turning wood. one
machine was set up with a clutch and ran about 2600 rpm. The "big" lathe was
capable of 2' diameter and 16 feet long. My dad had a stroke three years ago and
closed his shop. Toward the end he was getting more and more calls for "twisted
turnings" or "rope". We did do some prototype work on mechanically coupled
devises and one production run of parts for a restoration company. We attached a
follower finger to the back end of the cross feed and removed the crossfeed
screw. |In the place of the cross feed screw we installed a lever. this allowed
us to follow a pattern like many of the duplicator set ups on other wood lathes.
That is the basic profiling system of the lathe. the blank is primarily taken to
this shape in the first step.This can be done two ways. If the blank is not to
large in diameter it can be cut with a tool the same way as metal would be in
the lathe provided the surface speed is high enough for a good finish. on larger
turnings they would fly apart if turned at high RPM's. for this a powered cutter
like a router is used to remove material. we used a skill saw or a shaper
cartridge with a cutter.
This is where the fun begins! Once the basic shaping is done we want to add
detailed cuts or nicks to the turning. There are several companies that provide
equipment to do this by mechanically coupling the spindle and the longitudinal
feed just as it would be on a typical engine lathe through gears, sprockets and
or chains. This is great but it has limitations. How many sets of gears and
sprockets does one want to keep around to use or in hopes to use them again. The
range could be from 12 threads per inch on a project where someone wanted a
threaded cover on a jewelery box to one turn in several feet with several leads.
For indexing the work I can make an indexing head for the leads but I think it
would be easy to establish an offset in the drive.
There could be several levels of complexity to this.
The easiest would be a long slow twist with no taper, a straight column. lets
say one half turn in 8 feet. I place the carriage in home position at a stop and
set the cutter depth on the crossfeed. as i hand crank the longitudinal the
spindle is turned by its drive to maintain the ratio I desire. When I finish the
cut I return to the beginning and pick the next lead to make that cut. This
could be done with an angular offset from home for the spindle I would imagine.
One difficulty I can see is that the finer the twist the mor sloppy the
resolution which is why I thought when going to fine pitches I would turn the
headstock by hand and let the step drive on the longitudinal feed keep in time.
This type of turning has its failings ift he column is tapered the flutes or
humps stay the same for the length of the turning and converge as the turning
The next part falls into place quite easily once the easy application is
finnished. Rose engine work.
If you are not familiar with this type of ornamental turning search Youtube for
a video it is interesting to see. Basically in rose engine work a Rosette is
fastened to the spindle. a fixed rubber running against the spindle causes the
spindle housing to rock back and forth moving the center line of the spindle
back and forth closer and further away from the cutter. This is done at slow
spindle speeds! This is typicaly for small parts up to about 12" long with
special adapters. Cant do this with an 8 foot turning. Again mechanical means
can be used to couple the spindle with the cross feed or better yet the compound
but this makes for more sprockets or gears to be stored.
I would like to mount the rosette on the spindle and use a linear encoder in
place of the fixed rubber. Linking this to a drive on the COMPOUND to move it in
and out. The cross feed will still follow the template. A second linear scale
will determine the location of the cross feed slide in relation to the center of
the turning to establish the ratio of travel between the rosette follower and
the action of the compound feed drive. There is a reason for this and it is why
I made the distinction of a straight pillar in the first section. The rosette is
a constant input for angular and amplitude. Lets say I have a rosette running
with 8 lobes 1" tall. as long as the turning is straight the bumps will be
consistent. If I run a taper as the turning gets smaller the bumps get thinner
because the angular relation remains constant, BUT the height remains constant
as well so as the diameter gets smaller the bumps get taller and thinner. They
loose their geometry.
The most difficult is the dynamic combination of these two: turning long spiral
rosette type turnings.
The ELS is much more than I need in one respect and only scratches the surface
so as I mentioned the first course is to drive either axis by hand and have the
other follow on a variable input ratio.
hope this makes sense....I have been thinking about it for a long time.