Square Corner Problem with CNC
From the original questioner:
All ballscrews have been checked with zero backlash. Mechanically the machine is very tight. I donít know if my Fanuc 15M has the look ahead feature. How can I check that? That sounds like exactly what we need to have.
From contributor H:
As Contributor R stated the look-ahead function will help, usually when you are running a program. You will see a line highlighted and then the cursor will usually be two lines ahead. The line highlighted is the one currently being run. The cursor is indicating where the controller is reading. This is usually turned on by the machine manufacturer, as is solves and prevents a lot of issues.
The exact stop would be your best bet to get a true square corner if all of your ball screws are set and there is no backlash/play. It does mean a longer cycle, but the only other option is to program a slightly rounded corner. You do get faster production without the exact stop enabled but the controller is trying to slow down the machine when it approaches a corner. Then when it reaches that position it immediately switches and speeds up to reach the next destination. All CNC's will do this. They do calculate how to slow down and then continue. The only way to truly solve this is to use exact stop.
Some software for programming the machine will also trick the machine. They will program the machine to go beyond the corner, and then make a small arc outside of the panel and then come in tangent to the point they just left. This way the part is square in the corner and the machine doesn't slow down.
From contributor B:
I have heard of using arc in the corners before. It has always interested me. I was not sure how to program them. I have a Rover 30. How small of an arc would you do? I assume a 270 degree arc at the corners?
From contributor H:
You are correct in using a 270 degree arc. You will need to play with the size of the arc, but you should be able to use one as small as 0.050".
From contributor K:
We had the same issue with a new Shoda a few years back. It came down to the fact that the machine tables were too heavy for the parameters settings in the fanuc control. It was never fully solved and we tried adjusting the acceleration/deceleration parameters with not much success. We just learned to live with the issue.
Would you like to add information to this article?
Interested in writing or submitting an article?
Have a question about this article?
Have you reviewed the related Knowledge Base areas below?