Hi,
I am experiencing problems with my X Axis losing steps on pockets being milled on 30 degree placement on the Rotary A. A few weeks ago Dano was losing steps in his Scull Geometry and his solutions was to reduce his motor tuning in the Config area in Mach 3 Mill.
I have been experimenting and testing different combos and it went from about .008 change per part to about .004 when viewed on the A as a reference.
I have doubled checked my G-Code and it was posted correctly numerically.
I started out with my X And Y with these parameters in the motor tuning when I was getting the .008 run out per part.
Steps @ 20000 Velocity @ 15 Accel @ .095 Steps Pulse @ 4 Dir Pulse @ 0
After reviewing Dano's thread I changed to:
Steps @ 20000 Velocity @ 10 Acecel @ .04 Steps Pulse @ 4 Dir Pulse @ 0
After changing to this configuration it got better but was still very bad when the last part at 330 was compared to 0 after the routine.
I then changed to:
Steps @ 20000 Velocity @ 5 Acecel @ .02 Steps Pulse @ 4 Dir Pulse @ 0
Now I have a .004 run out on the X which adds up to the Part at 0 degrees and the part at 330 being .048 off on the X total.
All of my parts and pocket seem to be perfect and the compensation is working just fine so I am guessing my machine is running OK but for some reason I am losing steps in perfect increments as the group of pocket is presenting in a slight spiraling effect. Each pocket is being milled a two depth elevations with a .004 clean up pass on the side walls. There is no problem with the X or Y as these geometries are prefect as the walls are not stepping off like I saw on Dan's problem. My problem seems to be getting back to the Zero reference start point before the rotary A axis turns 30 degrees.
I have checked all my love joy connections and marked them and they are rock solid.
At this time I am starting to question the 20000 steps and the 4 and 0 pulse numbers but since there are numerous permutations I am afraid to mud the waters of diagnosis.
The most frustrating part of this dilemma I am experiencing is that every now and then during my testing the out come comes out with the correct geometry? banghead The cue I am building has 150 inlays and I don't want to commit to the pockets on that cue until I see it repeat. Gremlins at work to say the least.
Any ideas or suggestions would be greatly appreciated.
Thanks,
Rick G
Here is a pic on a test piece showing the problem:
Here is a piece of another test showing the 0 and 330 area and it is ok. I can't get it to test and be repeatable.
I am experiencing problems with my X Axis losing steps on pockets being milled on 30 degree placement on the Rotary A. A few weeks ago Dano was losing steps in his Scull Geometry and his solutions was to reduce his motor tuning in the Config area in Mach 3 Mill.
I have been experimenting and testing different combos and it went from about .008 change per part to about .004 when viewed on the A as a reference.
I have doubled checked my G-Code and it was posted correctly numerically.
I started out with my X And Y with these parameters in the motor tuning when I was getting the .008 run out per part.
Steps @ 20000 Velocity @ 15 Accel @ .095 Steps Pulse @ 4 Dir Pulse @ 0
After reviewing Dano's thread I changed to:
Steps @ 20000 Velocity @ 10 Acecel @ .04 Steps Pulse @ 4 Dir Pulse @ 0
After changing to this configuration it got better but was still very bad when the last part at 330 was compared to 0 after the routine.
I then changed to:
Steps @ 20000 Velocity @ 5 Acecel @ .02 Steps Pulse @ 4 Dir Pulse @ 0
Now I have a .004 run out on the X which adds up to the Part at 0 degrees and the part at 330 being .048 off on the X total.
All of my parts and pocket seem to be perfect and the compensation is working just fine so I am guessing my machine is running OK but for some reason I am losing steps in perfect increments as the group of pocket is presenting in a slight spiraling effect. Each pocket is being milled a two depth elevations with a .004 clean up pass on the side walls. There is no problem with the X or Y as these geometries are prefect as the walls are not stepping off like I saw on Dan's problem. My problem seems to be getting back to the Zero reference start point before the rotary A axis turns 30 degrees.
I have checked all my love joy connections and marked them and they are rock solid.
At this time I am starting to question the 20000 steps and the 4 and 0 pulse numbers but since there are numerous permutations I am afraid to mud the waters of diagnosis.
The most frustrating part of this dilemma I am experiencing is that every now and then during my testing the out come comes out with the correct geometry? banghead The cue I am building has 150 inlays and I don't want to commit to the pockets on that cue until I see it repeat. Gremlins at work to say the least.
Any ideas or suggestions would be greatly appreciated.
Thanks,
Rick G
Here is a pic on a test piece showing the problem:
Here is a piece of another test showing the 0 and 330 area and it is ok. I can't get it to test and be repeatable.
Last edited:
