Re: RE: EMC, ISA and other Questions...

> Date: Wed, 22 Jan 2003 17:37:19 -0700
> From: Bob Simon <bsimon@...>
>Subject: RE: EMC, ISA and other Questions...
>

>Here's how it's normally done;
>
>First of all the fourth axis needs it's feed rate supplied in *inverse
>time*. Most controls initiate this with G93. The feed rate number in the
>"F" address needs to be the velocity in inches per minute divided by the
>travel distance or "span length" and 60.

Snipped some explanations

Inverse feed is the reciprical of the time that a given cut is expected to take.
In the example below the cut is the circumference or
25.13 inches cut at 30 ipm=.837 minutes or 50.26 seconds

1/50.26=.02 The inverse feed

>to mill at 30 ipm on a 8 inch diameter part for one full revolution,
>
>G93
>G01 B360 F.020

Bob this is fine for a cut on a cylindrical surface. However if you program a line
that passes from one side at the cylinder surface through or close to the center of
rotation and then back to the other side, this will not make a usable method of
programming feed rates, because of the effects of changing the radius during the cut.

This does permit cutting axis parallel lines which are a problem with some
controllers, as there is no statement of degrees per minute that will allow
movement along the X axis if no rotation occurs about the A.(Degrees per minute F=0, hence no motion)

If you program G01X2.0 while rotary contouring degrees per minute feed rate is in effect, no movement occurs.
However if the controller has enough smarts to recognize a modal linear feed rate no problem.

My advice to controller designers for rotary 4th axis, is to ignore inverse time and do the math correctly.
This is what Art has attempted and hence the user can simply program a
feed in inches per minute, an initial radius reference, and then ignore feed
rate as the controller correctly cuts at the desired feed rate in the inches per minute as desired.
Only one feed rate need be specified at the beginning.

The improvement on this as a rotary axis programming technique is to interpolate
the lines into short enough segments that each move causes a change in feedrate
appropriate for the then current radius. Art could probably add this capability and the
line would actually cut at the programmed feed rate , even though the radius had major
variations throughout the length of the G01 cut (It would be replaced with many shorter G01's).
My guess is that it is not worth the trouble to compute an internal feed adjustment every
few thousandths of an inch. The user can easily (if needed) control the variation in feed rate
by the geometry size in the CAD program

Most any cad-cam program will do this, even those that are limited to wrap cylinder
or axis substitution types of rotary axis programming.

Best Regards, Fred Smith- IMService

Listserve Special discounts and offers are at:
http://www.imsrv.com/grp.html

We are THE source for low cost Cad Cam