5 Axis Morph

[david]

Good morning all. With 2019 Mastercam on a Okuma MU-400VA equipped with an OSP-200 control I am trying to get a set of surfaces cut using the "Morph" tool path. My part is a cavity with the centerline running left to right or parallel with the "X" axis. The "A" axis can go from +20 to -110. In order for the tool holder to miss the parting line, 7 degrees is all the tilt needed.

When I post the program out, the "C" axis wants to constantly spin around to make the tool cut with the "A" working from 0 to minus 7 degrees. In misc. values I can make it work with a "A" positive solution. Why would this tool path go from "A-7" to "A7" degrees without spinning the "C" axis constantly? My 5 axis tool path knowledge is very very limited, so this may be a no brainer for an experienced person. Can anyone point in the right direction?

Thank you.

[gcode]

Without seeing your file, I think you are having a singularity issue

When the A axis passes through A0, there are an infinite number of solutions for the C axis thus the C is spinning wildly

with very little actual tool motion

This can sometimes be solved by changing the alignment of your part on the machine so that the tool is not passing through A0

[huskermcdoogle]

Post a sample file.  We are flying blind.  Depending on the post, there is a decent chance are you can get it to do what you want, but you will have to make sure not to contour through the singularity.  I am guessing you just want it to tilt back and forth from A-7 to A7 during the linking moves?  I don't know if the generic 5 ax post can be forced to work this, but I think it can.

Post a sample file, simplified geo if need be, I think we can get what you need.

[huskermcdoogle]

Thinking about it, I have never been able to get my head head post to do this, but I can get my head table post to do it, but it is always a mile away from a singularity as a singularity would equal a crash 99.9% of the time...

As gcode said, Can you mount the part on an angle plate or something and attack it at tipped say a minimum of 15 degrees (A needs to be tipped more than 14 if you want a lead angle of 7 without passing through table vertical singularity).

[huskermcdoogle]

Also a custom post could do this all day long if there was a setting to pick C shortest distance solution, while allowing A to go wherever it needs to.

[crazy^millman]

What I have found is to make a line normal to my surface or at my desired angle. I then will pick that line for my Axis limits and then Conical limits and the control for the cone Axis is that line to keep things doing that limited. Another thing to look at is the tool axis control depending on what option you pick you will get different abilities to control singularity that some don't give you. Change from something like surface to Fixed angle to Axis you will see Tool Axis crosses Tilt Axis when this is checked then it will also help control singularity issues that are causing what your seeing with the C Axis spinning. Ton's of power in this toolpath and the last thing is not limiting your toolpath to part geometry. to many limit the true power of this toolpath thinking they must pick 2 chains that are on the surface or even like the shape of the part. What will happen is they get to much pinching motion with the toolpath. Visualize the name it says morph between 2 curves so as long as those 2 curves will cut a nice shape on the surfaces you are trying to machine they can be any 2 curves that are outside of the shape close to the cutting area and will give you motion you want to cut. That is where our creative side needs to kick in and we need to get away from our mathematical side. Another toolpath to consider is parallel to curve. Here again one curve can drive this toolpath in very creative ways to cut even the most difficult part. Also don't forget to look at collision avoidance tons of power in that as well.

I just remembered maintain tilt that will also help depending on the shape and part, but like above it depends on the Tool Axis control process you pick for controlling the toolpath.

[crazy^millman]

4 minutes ago, huskermcdoogle said:

Also a custom post could do this all day long if there was a setting to pick C shortest distance solution, while allowing A to go wherever it needs to.

Not really needed in this case if limits and some other things are applied correctly. Being able to prewind a head if a Head Head with limited travel it might help some, but this is a Table Table not as much as it would on a head head machine is my thinking.

[david]

Thank you to every one. Here is my file. The machine control and definition and post are customized for this machine and my hasp by In House Solutions.

Morph.mcam

[crazy^millman]

28 minutes ago, david said:

Thank you to every one. Here is my file. The machine control and definition and post are customized for this machine and my hasp by In House Solutions.

Morph.mcam

From china is not the best choice here. I can see the toolpath is having singularity issues. Look here this tells me needs to be approached differently.

 

Edited August 1, 2018 by C^Millman
From Chain not From Point

[gcode]

I changed Tool Axis Control to "Angle from Spindle, main direction"

Tilt Angle = 0  and Z axis

In Collision Control I used  Tilt Tool (Automatic)

The results look pretty good in my Okuma MU H1000 machine sim

[JohnW@Kauffman]

10 minutes ago, gcode said:

I changed Tool Axis Control to "Angle from Spindle, main direction"

Tilt Angle = 0  and Z axis

In Collision Control I used  Tilt Tool (Automatic)

The results look pretty good in my Okuma MU H1000 machine sim

This works on my Onsrud 5 axis router as well, except that I had to change the Tilt Angle to 7.

[crazy^millman]

Run your results using the Generic 5 Axis machine and see if the table spins. I was going to suggest breaking this up into 2 toolpaths. Approach half of it titling the tool 7 Degree from the one edge and then tilt the tool 7 degrees from the other edge. Another way would be from outside to center. I was also thinking of using point at the middle of the chain that way you get more tilt through the complete cut than cut what we are seeing doing it through the whole chain. Tool has enough clearance and the shape helps where collision avoidance to tilt the tool is not rally needed.

Edited August 1, 2018 by C^Millman
Was wrong about tool clearance

[david]

Thanks everyone. The next time the machine is open I will try what gcode suggested. When I posted the file, the "C" axis is moving quite a bit. Is this normal?