Centroid Community CNC Support Forum

Ok, so "precision mill" stopped the issue with the 4th axis hitting position error.

Even in a 360-degree rapid move at 100% override?

Yes, I noticed that the "load" meter did not make it up as high and it even did a little decel before stopping.

a majority of the mill table stuttering has been removed as well.

I also figured out why the pockets were coming out oddly shaped and why the control did not detect any issue: Have you ever done something like repair a lawn mower and could not figure out why it won't start? Then, you figure out that you just forgot to add fuel? Well, I did not tighten down the T slot bolts for the 4th axis enough and the end mill was pushing the 4th axis a few thousandths every time it would do a pass... I thought I checked that... So, now the pockets turn out as expected!

https://www.youtube.com/watch?v=brkABxRefUE

Glad you got it worked out.

What post are you using for Fusion 360? Did you simply edit the post to enable the 4th axis?

I use a 4th axis post that I modified, however I have a fairly unique machine (older head porting setup from centroid) and the rotary is a shelf/ table and I have only done 3+1. I have a 3+2 post with an offset 5th axis that I have been working out geometry in CAD, but I have yet to test it on the machine.

John

I am using the "Fanuc inverse time and A axis" post processor. It can be found in the Fusion 360 post processor library. Another thing I need to work out is that the post processor likes to use negative A moves and currently my control is set up to only go from 0 to 360deg. I just tried to run a "flow" tool path on a cam lobe shape and the mill fell flat on its face. I wonder if there is a setting somewhere in the control where I can select negative degree outputs for the A axis.

I mainly want the live 4th axis feature whenever some part with a draft angle needs to be milled out. It takes forever try to use the corner of the tool and step every few thou to make the draft angle. I would rather just use the side of the tool or the bottom of the tool and be done with it within minutes.

I would have gone with 5 axis, but I think one really needs at least a bed mill to make that doable so the z can have at least a foot or two of vertical travel so it can get out of the way of the rotating trunion table. One of these days i may look into a used bed mill and just cannibalize the control off of my BP mill and give it to the bed mill. Those Atrump bed mills look pretty spiffy.

There should be no reason you can't go to negative A angles.

What happens when you jog A minus, past zero?

What happens if you MDI "G0 A-90"?

What, exactly, happened when you tried to run the CAM-generated G codes?

My mistake. I must have done several rotations of the 4th axis earlier that day and not realized that I had wound it up. So, when the program started, the 4th axis just sat there spinning and I did not realize that it was trying to go back though all those encoder counts. This morning I went out there and It ran the part just fine, complete with 25 rotations of the 4th axis. Once it was done, it did turn 25 times in the other direction to get back to its zero .

https://www.youtube.com/watch?v=gAaJGp1d8VE

Ok, so I got around to installing the 2:1 pulley setup and that did not help whatsoever on the position error. At this point I think I can say that it is not a load issue.

I think I will need to go into the PID parameters and tune the 4th axis correctly. How would I go about running a mock program that allows the 4th axis servo to rotate one way and then come back the other way to show the current state of the tune?