During the creation of my tapping arms, I came across a certain issue that I had not previously experienced due to the high speed of feed (50 IPM) and occasional sudden shifts in direction to follow the profile. I had garnered knowledge of a similar issue from an individual who had irregular motor tuning on a knee compared to his X and Y axis. This was believed by some to be glitch in Mach.
In any event, as I discovered early in my machining of the axis arms, it isn't a good idea to have different accelerations on two axes. I wouldn't have noticed it had it not been for easing back the accel on the X axis because it was faulting too much when I run with two 6" vises on the table. I lowered the X axis accel to match the Z and forgot about it. This latest job making a tapping arm reminded me in short order that something needed tending to.
It seems odd to me that Mach3 doesn't account for this, and it ought to be classified as a bug. Mach's trajectory planner should be able to properly choreograph a coordinated move across n-axes with a different acceleration and top velocity on each. If nothing else, the lazy algorithm would simply limit all the coordinated axes to the least acceleration and velocity of any axis involved at the time.
In any event, if you think you're doing yourself a favor by finding each axis's maximum performance envelope, you might in fact be doing the opposite until Mach learns to deal with it better.
BTW, the symptom will be that the tool follows the coordinated move's path in a very sloppy way. Going around the pivot point on the tapping arm swing block it was painfully obvious to the naked eye something was wrong-no calipers or micrometers needed.
Be the first to know about updates at CNC Cookbook
Join our newsletter to get updates on what's next at CNC Cookbook.