LinkBack URL
About LinkBacks
Very carefully measure / estimate whether the flats are perpendicular to each other and whether they are parallel o the X Y axes of machine frame
?
Flat spots on helical moves (G2/3) Well I've got my mini-CNC moving around well in linear motions. However for some reason a helical coordinated motion creates 'flat spots'. (a 1" circle has 1/4" straight lines on 4 'sides')
It 'looks like' the accel/decel is too slow - but setting config values has no effect on output.
I was thinking:
1. Configuration / HAL settings
2. Backlash (seems little effect)
3. ??????????
Thanks in advance for the HELP!
Very carefully measure / estimate whether the flats are perpendicular to each other and whether they are parallel o the X Y axes of machine frame
?
Flat spots appear to be parallel to each other. Perpendicularity to the axis is difficult to judge - looks non-90º, but best I can measure says it is.
I've attached a scan of a circle with center mark for judgment.
Thanks for the help ... can you share your if ___ then ___ logic?
Well.. the flats do look like they are parallel to the oposite side.
And they sortof look parallel to the axes of motion but possibly not quite...
I would check backlash.. acceleration... and most probably the setup time settings in Your driver configuration. That is direction set up times.
This presumes that You are runing with stepper motors.
I'm guessing that you went around that circle clockwise? If so, the flats are backlash. Look, for instance, at the 12 o'clock flat; as the table moves around to the peak Y position, the circle is fine. Then after passing 12o'clock, the Y stepper reverses and starts taking up the backlash in the leadscrew. It's not until the X has gone quite a way off of the centerline that the Y slop is gone and the table actually starts following the curve again, only to have the same thing happen in X at 9 o'clock, and so on.
Also check for any loose connections in couplings or soft and too compliant couplings between motors and leadscrews.
Solved! Perfect gentlemen (very good anyway)
I found a loose 'connection' in both axes. After 'tightening', reducing the backlash had a great effect on eliminating the flats.
I appreciate all the input!
(result attached)
Excellent, I'm glad to hear you've gotten it worked out!
Posting Permissions
