Re: [CAD_CAM_EDM_DRO] Re: True 4th Axis milling

Brian Pitt wrote:

> Chris
>
> On Sunday 10 June 2001 12:52, you wrote:
> > If I understand correctly, the problem was that the tool offset was
> > accomplished in the rotary axis, so the wall on one side was based on
> > one radial and the one on the other side was based on a different
> > radial, hence they are not parallel, but rather interesect at the
> > center of the radial axis. Sounds like what is needed is to
> > accomplish the offset using the linear axis instead.
> >
> yes exactly
>
> thing is we don't have a CAM system at the shop yet and haven't
> needed one for most of the work we do ,this was the first job in
> about 4 years that we couldn't quickly program by hand at the machine
> so we decided to farm out the programing at $80/hr rather than spend
> big bucks for software that would just collect dust most of the time

Gee, unless this groove was a LOT more complicated than what we're talking
about here, I don't see why someone who knows just a little G-code
couldn't program this directly in the G-code. The linear part of the groove
is just going to be about 8 moves. Plow down the center, cutter comp on,
move over to the edge, cut back up, across to other edge, cut down to end,
move to other edge, cutter comp off. The rotary part should be fairly similar.

>
> the programers (all three of them) just couldnt figure out how to get
> their high end CAM system to handle the Y-axis along with the A-axis
> and went with a 3 axis centerline toolpath (X,Z,A)
> rather than a true 4 axis path (X,Y,Z,A) that would have done what we
> wanted
> probably more of a wetware problem than a software one
> - much harder to debug ;-)

Yes, I can see how a CAM system could get in trouble with that. Also, a
lot of the high end setups assume a 4- or 5- axis machine has the spindle HEAD
moving in arcs, not the workpiece. The lower end machines can do most of the
same work (and some pieces with holes all around even better) than the
traditional 5-axis machines. It would seem that this sort of problem would
have come up already, and they'd have a workbook example to show how to
do it.

Jon