Multi Spindle CNC Router control options

[carboncymbal]

I’m interested in having 2spindles each on an independent Z axis on a future router build. I am wonder if it is possible to do this with an acorn 6? I know an approach used in mach3 and Linux CNC is to use a swap axis approach where the Z axis is done by different machines.

Is this possible with Acorn and centroid?

I have considered other approaches to accomplish this if each axis can’t be independently driven.

[cnckeith]

yes there are several approaches to this.

what exactly are you after? not all dual Z machines act the same.

some are Z1 Z2 are slaved and work in unison, others are two separate axes Z and W

do you need rigid tapping on the 2nd Z axis? etc..etc..

[cncsnw]

Keith,

What can you tell us about M333, and whether it could be applied to this type of machine?

[carboncymbal]

I’ll try to explain the machine layout a bit to provide some context, I may not use the correct identifies for each axis. Avid CNC has offered a mechanical setup similar to
What I am hoping to accomplish. I have attached a photo

The machine will be a fixed gantry, moving bed router. A small machine but built very rigidly.

Ideally there would be 2 independent Z axis assemblies (I think Z and W axis), all moving right to left together on the Y axis (fixed gantry). I’d like to control each spindle in the Z orientation with its own servo, but only one at a time. The purpose of this is to get 2 tools for an operation without the need of an ATC setup (reducing cycle times). The system would be programmed using traditional 3 axis cam. Ideally an M6 command to another tool retracts the active spindle, changes the active spindle and associated offset, and starts the cut. This approach would allow me to maintain maximum rigidity on 2 main spindles without the need for pneumatics and an extra set of linear rails. I have seen people do this with wincnc and Mach 3. I think in both cases they essentially swap the z axis between the two as part of a custom m6 macro.

If I can’t do it as described above, I could do it pneumatic ally with custom M6 macros. However this is less ideal because of the additional mechanical systems needed (and extra set of vertically mounted linear guides and a pneumatic actuation system.

My cycle time on this machine is low, possibly as low as 15 seconds. In this case I may actually run both spindles during the entire cut to prevent a ramp up time to 14-18k for each spindle. I expect the spindles to be HSD or Hiteco spindles in the 2-4 hp range.

I do not need rigid tapping on either spindle.

[carboncymbal]

Keith,

What can you tell us about M333, and whether it could be applied to this type of machine?

Thanks for mentioning that, that appears to function similarly to the "swap axis" concept in other controllers. I think this would do what I need provided it is fully functional. In a manual I just looked at it was shown as "experimental" I'm hoping that is just because it would be a fairly specialized use case, not because the feature is incomplete.

One reason why I am interested in a dual servo axis approach is that it would allow me to piggy back additional small spindles as needed pneumatically using simple M6 macros (I'd like to add a light weight spindle for drilling and possibly a high RPM spindle for engraving). It also permits full rigidity and mechanical simplicity of the main 2 spindles.

[carboncymbal]

yes there are several approaches to this.

what exactly are you after? not all dual Z machines act the same.

some are Z1 Z2 are slaved and work in unison, others are two separate axes Z and W

do you need rigid tapping on the 2nd Z axis? etc..etc..

Any thoughts on this CNCKeith? I answered your questions above but failed to use the reply function.

[cnckeith]

thanks, busy day working with Delta Drives and Hickory, just getting around to looking at the forum now.

[cnckeith]

if you are not going to rigid tap on the W axis, independent Z axes can be controlled with G code, you could simply have your post processor put W in the G code for the 2nd Z axis when you want that 2nd Z to move and then the W axis would move not Z

M333 is not sorted and is experimental, not sure if we will continue with this path or not at the moment, i'm looking into it.

we do have some other plans on how to make it easy to arrange and support independent Z axes and simultaneous Z axes (Z1, Z2) with different or same spindle controls. M3.2 M4.2 etc.. will be sometime this year when we get to that. For now editing the post to create g code to do what you want is how others are doing what you are after.

[carboncymbal]

if you are not going to rigid tap on the W axis, independent Z axes can be controlled with G code, you could simply have your post processor put W in the G code for the 2nd Z axis when you want that 2nd Z to move and then the W axis would move not Z

M333 is not sorted and is experimental, not sure if we will continue with this path or not at the moment, i'm looking into it.

we do have some other plans on how to make it easy to arrange and support independent Z axes and simultaneous Z axes (Z1, Z2) with different or same spindle controls. M3.2 M4.2 etc.. will be sometime this year when we get to that. For now editing the post to create g code to do what you want is how others are doing what you are after.

I’m very interested in your future solutions to simplify a dual spindle router as you describe.

In the mean time, your idea of doing it in the post processor makes total sense. I’ll start researching method using post processors in Fusion 360.

Thanks

[carboncymbal]

if you are not going to rigid tap on the W axis, independent Z axes can be controlled with G code, you could simply have your post processor put W in the G code for the 2nd Z axis when you want that 2nd Z to move and then the W axis would move not Z

M333 is not sorted and is experimental, not sure if we will continue with this path or not at the moment, i'm looking into it.

we do have some other plans on how to make it easy to arrange and support independent Z axes and simultaneous Z axes (Z1, Z2) with different or same spindle controls. M3.2 M4.2 etc.. will be sometime this year when we get to that. For now editing the post to create g code to do what you want is how others are doing what you are after.

OK, I have a few questions related to the custom post solution. Customizing a fusion post processor would be an in depth project for me and I'd have to learn a lot. However it occurred to me that for the types of programs I intend to run I will all cuts with spindle 1, then all cuts with spindle 2. If I understand your proposal correctly, I could take the following approach to program spindle 2.

1. Generate traditional 3 axis code via my chosen cam software (in my case Fusion 360), using a setup assigned to G55 WCS
2. Do a mass find/replace on that file to find all Z commands and change them to W.
3. Run the program, the Z axis remains where it is, but the W axis operates just as it would if it were the Z axis.

Is this correct?

Would things like helical interpretation work in this case? For example, this code is copied from a centroid manual for xyx movement.

G17 G90 F30 ; select XY plane and absolute positioning
G00 X3.0 Y2.0 Z1.0 ; rapid to start position X3, Y2, Z1
G02 X2.0 Y1.0 I-1.0 J0.0 Z0.0 ; CW XY arc from X3,Y2 to X2,Y1.
; Center at X2, Y2
; Helical Z move from 1 to 0

would this code, a literal find and replace, accomplish the same thing, but with the W axis instead of Z?

G17 G90 F30 ; select XY plane and absolute positioning
G00 X3.0 Y2.0 W1.0 ; rapid to start position X3, Y2, W1
G02 X2.0 Y1.0 I-1.0 J0.0 W0.0 ; CW XY arc from X3,Y2 to X2,Y1.
; Center at X2, Y2
; Helical W move from 1 to 0

It almost seems to me that G18 and G19 would require some redefinition of being based upon the WX or YW plane. Although maybe not since the WX=ZX plane and YW=YZ plane.

If my interpretation of all of the above is correct, then this would be incredibly simple given the simple structure of my programs. I could write a simple script to automate the edits from standard gcode output from Fusion.