Continually rotating rotary axis

[oddsfellow]

Doing some multiaxis machining in Mastercam and the surface is quite flat but then it has rounded corners. (Think of a cube with radii on the edge). For the vast majority of the motion its a straight linear move then as the part goes around the radius the rotary axis engages and has to accelerate very quickly and decelerate very quickly as the cutter goes around the corner before the machine goes back into a purely linear move. I thought if i drew a round curve around the object and used that to create the path it could then continually rotate the part avoiding the acceleration and deceleration and potentially smoothing out the motion. I cannot seem to get a toolpath that does this at the moment. Is there a way to do this inside mastercam? 

[Colin Gilchrist]

Create a Cylindrical Surface, "inside" the part you are cutting. Use the Cylinder as the Cut Pattern, and the Tool Axis Control method. Then, use "collision control" to specify "what surfaces the cutter is tangent to, as it moves over the Cut Pattern surfaces". This is known as the "clean core" method of 5-Axis path creation.

You can then choose to use "Lead" or "Lag" to push the cutter "forward (lead = positive value)", or "backwards (lag = negative value)". You could even use "side tilt" if you want, or a combination of Lead-Lag/Tilt, to keep your ball cutting on the flute, and not directly at the Tool Tip.

I find at least 4 degrees of Lead, Lag, or Side-Tilt, really improves tool life and surface quality, when cutting with a Ball. If you get get 15-20 degrees, even better, but the best results will be over 45 degrees (from about 50-70 degrees, you've got the "sweet spot" of the ball.)

[crazy^millman]

can you make a sample of what you are doing and post it up? I can think of at least 30 different ways a 5 Axis machine can be built. Knowing your type of machine and kinematic setup along with a sample file will go a long way to help narrow down some good suggestions.

[Aaron Eberhard]

Agreed with Ron, but it sounds like you've got your tool axis control set to be Normal to the Surface (i.e., always be perpendicular to the point that you're cutting).   If you want to force multiaxis motion, try telling it to Rotate Around Axis, that'll treat it like a rotary move, so it's always spinning.   If the feature isn't actually rotating around an axis, you can create your own by drawing a line through it.

 

[crazy^millman]

12 minutes ago, Aaron Eberhard said:

Agreed with Ron, but it sounds like you've got your tool axis control set to be Normal to the Surface (i.e., always be perpendicular to the point that you're cutting).   If you want to force multiaxis motion, try telling it to Rotate Around Axis, that'll treat it like a rotary move, so it's always spinning.   If the feature isn't actually rotating around an axis, you can create your own by drawing a line through it.

 

Quickest way on any surface to draw a line normal to it to use in this situation is Wireframe/Line/Normal to point.

 

Then in the Rotation Angle Drop down there is a the ability to pick a line. Grab that line and good to go.

[oddsfellow]

Thanks for all your help guys. Using a trunion table bolted onto a 3 axis vmc but machining 5 axis simultaneous. i did find this article during searching yesterday which explains exactly what i am trying to do. essentially i am trying to do figure 7. https://www.moldmakingtechnology.com/articles/the-misunderstood-cutter-path . 

Will try what you suggested today and post back the results. Unfortunately it is a confidential part so cannot post pictures sadly. 

[crazy^millman]

3 hours ago, oddsfellow said:

Thanks for all your help guys. Using a trunion table bolted onto a 3 axis vmc but machining 5 axis simultaneous. i did find this article during searching yesterday which explains exactly what i am trying to do. essentially i am trying to do figure 7. https://www.moldmakingtechnology.com/articles/the-misunderstood-cutter-path . 

Will try what you suggested today and post back the results. Unfortunately it is a confidential part so cannot post pictures sadly. 

I would think about 4 vectors in the corners and try driving the toolpath from the 4 vectors using Lines as the Tool Axis Control Process. The Lead/Lag can still be used for a majority of it, but in the corners the tilt lines might limit the axis travel enough to stop the extreme motion. Sometimes mounting a mold at a 10 or 15 degree angle on a trunnion also help create enough difference from a singularity event which is what it sound like is going on here.

[Aaron Eberhard]

One other option is to slow down the lead/lag changes in the Tool Axis Control.  For example, here's a morph toolpath normal to surfaces:

Note: I have backplot Vectors turned on, hit the little "!" and turn on toolpath vectors in the lower right to see the same thing.

On the Tool axis control page, turn "Smoothing"

Then slide the bar to adjust:

 

That'll give you this:

 

 

[crazy^millman]

9 hours ago, Aaron Eberhard said:

One other option is to slow down the lead/lag changes in the Tool Axis Control.  For example, here's a morph toolpath normal to surfaces:

Note: I have backplot Vectors turned on, hit the little "!" and turn on toolpath vectors in the lower right to see the same thing.

On the Tool axis control page, turn "Smoothing"

Then slide the bar to adjust:

 

That'll give you this:

 

 

That is new for 2022 so for those stuck in previous versions this is not available in in tool axis control encase you are wondering. Helps with large axis move also.

[Aaron Eberhard]

1 hour ago, crazy^millman said:

That is new for 2022 so for those stuck in previous versions this is not available in in tool axis control encase you are wondering. Helps with large axis move also.

Good call on that...  I forgot that not everyone is playing with the shiniest toys

[oddsfellow]

Once again thanks for everyone's help. From the assistance given here was able to get a continual rotary axis path which avoided the acceleration and deceleration spikes but it did increase cycle time. Can only go as fast as the slowest axis and the linear axis will be quicker than the rotary but guessing this only really applies if you have long linear moves to make, which i do in this case. There was some visual, almost reversal marks after the original cut which did disappear doing a continuous rotary which is great so it ended up as a trade off between cycle time and surface finish. 

Then tried Aaron's approach of what i would call fanning and this sped up the cycle time and also gave an improved finish. Its a very smooth toolpath and is a blend between what i was doing and what i was trying to do but with better results. The other thing i am getting with this is because i am rolling around the ball rather than fixed at one contact point the tool life is also increasing. That would also apply to the continuous rotary toolpath too.

[Aaron Eberhard]

6 hours ago, oddsfellow said:

Once again thanks for everyone's help. From the assistance given here was able to get a continual rotary axis path which avoided the acceleration and deceleration spikes but it did increase cycle time. Can only go as fast as the slowest axis and the linear axis will be quicker than the rotary but guessing this only really applies if you have long linear moves to make, which i do in this case. There was some visual, almost reversal marks after the original cut which did disappear doing a continuous rotary which is great so it ended up as a trade off between cycle time and surface finish. 

Then tried Aaron's approach of what i would call fanning and this sped up the cycle time and also gave an improved finish. Its a very smooth toolpath and is a blend between what i was doing and what i was trying to do but with better results. The other thing i am getting with this is because i am rolling around the ball rather than fixed at one contact point the tool life is also increasing. That would also apply to the continuous rotary toolpath too.

That's great to hear you're making good progress on it