On Jul 7 2013 9:24 PM, Jon Elson wrote: > Chris Morley wrote: >> >> For instance if your machine could move maximally at 50 inches a >> minute, why would >> you allow the TP to ask it to move 52 while G33.1? Same premise as >> jerk limiting. >> If your machine really can run 52 then why not set the limits to 52? >> > Well, this is the problem. If the spindle is connected to the Z axis > by > a tap, and > the program calls for a spindle speed such that the Z axis exceeds > its > limits of > speed or acceleration, it will break the tap or damage the workpiece. > Spindle > speed can be fairly easily calculated so the Z feeds are within the > machine's > capability. But, if a VFD is reversing the motor by forward/reverse > relays, > it may produce a strong acceleration that exceeds the Z axis' > acceleration > limit at that particular thread pitch. I was worried about this and > used a filter to slow down the rate of the deceleration and reversal. > But, > in a case without an analog spindle speed command to the spindle > drive, > this won't be possible. (Not that I'm saying such a setup is a good > idea!) > So, it is possible to write G-code with a G33.1 where the spindle > speed > or the spindle acceleration exceeds the machine's limits. LinuxCNC > has no way to know this is going to happen until it is in the middle > of > the G33.1 cycle! (Actually, it could use commanded spindle speed > times the thread pitch to compute required feed and report an error > if this is greater than the velocity limit on Z.) But, it certainly > has > no idea how quickly the spindle will reverse at the bottom of the > thread. I don't know what to do about that, except test first with > no tool and Halscope.
I had not thought about VFD reversing by forward/reverse relays... Do we have numbers to run through for comparative z feed rates? EBo -- ------------------------------------------------------------------------------ This SF.net email is sponsored by Windows: Build for Windows Store. http://p.sf.net/sfu/windows-dev2dev _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
