Re: Newbie question of the day

On Sun, 04 Sep 2005 19:00:37 GMT, Geoff <geoff@xxxxxxxxxxxxxx> wrote:

>On Sun, 04 Sep 2005 08:12:59 -0400, Cliff <Clhuprich@xxxxxxx> wrote:
>
>> And are named "forwards" from ABC.
>> DEF
>> GHI
>>
>
>Only in a robotic application. In a machining center DEFGHI are
>reserved for other purposes for obvious reasons and cannot serve as
>axis designators in ordinary G-code syntax.

By the time you have that many axes it's not really an "ordinary"
machine.
What I stated seems to be basically correct per RS-267-B
(Axis and motion nomenclature for numericslly controlled machines)
section 8 (additional axes) subsection 8.2, as well as my
recall of 12 axes and other machines.

>In a machining center the
>secondary and tertiary rotary axes are designated A1 B1 C1 A2 B2 C2

Then how would you command A1180.0, etc

>and there are special codes for setting the control mode to bring
>these in and out of context for coordination with the other axes since
>it might be hard to define whether one was referring to A1 80. or
>A180. Modern controls can use A1=180. or the like, any delimiter would
>probably work depending on the control vendor.

And similar could be done following the standard.
OTOH It's also possible to use other code formats for things like
what you are thinking of as "D", F", etc.
It's also possible I expect to load registers in sequence.
F180 F5 might rotate the F axis 10 & use feedrate 5.
Type II & III data are also often used in any case as these
machines often have to be programmed in inverse time
parametrically.

>
>In the Siemens 840D they can have any name the OEM likes. There are
>even X1 Y1 Z1 axes on some machines but these are usually in channels
>other than the primary channel and they serve another purpose from
>actually cutting parts. (e.g., Tool changers.)

Which is part of why we have posts ....

>There are axes and there are axes. Any machine tool manufacturer who
>labels an axis that is not directly involved in the actual machine
>tool cutting geometry an "axis" for purposes of counting axes as in
>"this is a 6-axis machine" is fooling himself and others. To say this
>machine has 6 axis drives but only 4 of them are machining axes would
>be permissible but to call it a 6 axis machine is deception. To be
>credible as a machine tool specification the axes would have to
>programmable in a coordinated path.

Positioning-only axes are usually 1/2 axes.
But, as BB can probably tell us, 1/2 + 1/2 = 1 <G>.

Not that I'd call it that in this case.

>It's credible to call a twin-head
>VTL a 4 axis lathe, but to call it a 5 axis because the spindle/table
>can be positioned would not be permissible since coordinating it with
>the 4 linear slides would not be advantageous.

It would be for things like threading. But that's usually not
visible to the user (but has to be in some cases).

>Simultaneous inside and
>outside threading might be construed to be a 5 axis move since it
>requires coordination of slide motions with spindle rotation but I
>think this is stretching the definition beyond the breaking point.
>
>Twin tilting gantries on a single bed cannot be defined as a 10 axis
>machine since they are almost always controlled from 2 CNC controls
>and each one would only be a 5 axis machine.

There are also "split" controls with an axis in common. Back to
inverse time, usually.
BTW, That sounds like a 9 axes, were it a twin column with
a moving bed.
With a gantry it would probably be a twin spindle 5 axes ....
but triples are often used (we had quite a few at McD ...)

>Gantry axis drives, leader-follower, master-slave or anti-backlash
>drives are properly counted as ONE axis since they control only one
>kind of motion in the machine geometry. The fact that it takes two
>drives to do it is an aspect of the fundamental construction of the
>machine and not a "feature" to be counted as an extra axis. Most of
>the time these drives are controlled by only one controller output
>channel and the follower drive is slaved off the primary and isn't
>"seen" by the control.
>
>K&T made a machine that had a tilt table controlled by twin elevating
>screws driven by two motors commanded by the CNC with separate outputs
>but the drive channels were locked together by the servo system
>internally and commanded in G-code as a single axis (A-axis).

I've probably programmed a few of those.

>The
>feedback was a rotary encoder mounted on the trunion but each motor
>has a resolver for encoding the position of the elevating screws and
>each of these was linearly compensated to maintain the load balance
>between the drives and to maintain geometric accuracy. For you pocket
>protector types, imagine converting linear ball screw motion to rotary
>tilt coordinates from an offset trunion encoding rotary motion. The
>machine was a 5-axis, X Y Z A B machine with a rotary commandable
>spindle but they didn't call it a 6 axis machine. The customer also
>had the option of adding a contouring tool adapter (U-head) that would
>have made it a 6 axis machine.
>
>On the issue of 6 axis machines. It is possible to have a 6 axis and
>they have even been built. A tilting head spindle or even a B and C
>axis rotary table with an A axis tilt head spindle are possible
>geometries. Then you would have XYZ ABC without having to even mention
>the "spindle as axis" idea. G&L has built HBM's with X Y W A B C
>geometries in a tilting, swiveling 6 axis contouring machine for
>aerospace applications.

Yep. IIRC *** Parsons at P&W wrote the OS for those in Pascal.
G&L was running way behind.
I barely escaped the task. Probably would have been more interesting
(in retrospect) than the 12 axes ones I ended up with.

I *really* wanted to violate te standard & make the
tool's axis the Z axis so that the tools could be easily extracted
from holes in MDI mode .....
Oh, well .. it was a good argument<G>.

>There are even machines built with extra linear slides for machining
>purposes that only come into play when the column is in a certain work
>zone. A machine can have X Y Z W1 for main machine motion with Z being
>a quill spindle and W1 being a ram carrying the spindle forward for
>extended travel.

Some are very good for some classes of mold work too.

>This particular machine also had a rotary table
>B-axis carried on a W2 slide. The X axis was also a leader-follower so
>the machine actually had 7 axis drives, not counting the 100 HP
>spindle drive. The spindle could be rotationally positioned too with C
>axis commands but calling the spindle an "axis" is specious. Siemens
>considers axis drive motors to be "spindles" and the digital servo
>systems inside the machine make no distinction between an axis
>"spindle motor" and a spindle motor per se. They are all just drives
>to be configured and commanded. The servos are put into position mode
>or speed mode, linear or rotary, depending on the OEM's application.
>
>The point of this rather long post is that you can have a 5 or 6 axis
>machine pretty easily, even when confining your definition to
>"working" geometric axes that actually apply to coordinated
>interpolated machine tool motion without resorting to calling the
>spindle an axis or calling an axis outside the work zone an axis. Yes,
>they are axes to the controller but they are not necessarily germane
>to the geometric capability of the machining center.

What do you call a 1/2 + 1/2 + 1/2?
--
Cliff
.