On Friday 20 March 2020 00:24:07 John Dammeyer wrote: > > > But it's here where I become a bit confused. A 60 tooth gear at > > > 1200 RPM produces 7200 edges per second or 1 about every 139 > > > microseconds. Say we're rigid tapping 8 threads which means a > > > total of 57600 pulses. If the hardware counter is 16 bit signed > > > that's larger than the 32767 before the counter rolls over. If > > > it's a 32 bit counter it's not an issue. > > > > I doubt if that would be a problem at 1200 because it would > > following error out from Z not being able to move that fast. > > Probably before the tap ever touched metal going in. Another problem > > would also rear its ugly heard is that the overshoot at the bottom > > of the hole would likely exceed the length of the tap at 1200. I got > > concerned about blind holes, and despite the belts yelping like > > michelin tires on hot blacktop, the overshoot at the bottom of the > > hole was over 3 turns at 300 rpms when swinging an 8" 40 lb chuck > > with a workpiece in it. Trade my 1hp motor for a 10 and a bigger vfd > > might speed up the turnaround but my place isn't wired for that kind > > of power. > > > > So I have modified most of my rigid tap routines to tell me how many > > turns and how far the overshoot is at that taps pitch and show me in > > the gui. So I can cut clear to the bottom of the blind hole w/o > > breaking the tap. > > Hi Gene > I can understand that one has to factor accel and decel into the > picture. The spindle won't stop suddenly and won't start instantly > either. So the Z axis tracks the decelerating spindle and that needs > to be factored into the depth of the hole of course. > > The same is true for threading up to a shoulder. > > I have a spring loaded tap holder so even if there's a little bit of > error my system will ultimately not break the tap. Plus in the near > future my spindle will be step/dir run with an AC servo so > theoretically the tracking of Z verses rotation will be as good as the > system can provide. I have often considered driving the spindle with big steppers, but I keep running into the howto's. I have hal code, which hasn't been migrated to this lathe yet, in both TLM and the G0704, that when the reversal comes out of motion, blocks it, but ramps the output of a limit3 towards zero at a good rate until a one shot gets a chance to time out. The one is kept retriggered by phase-A as it slows, and when its running slow enough, ISTR 50 millisecs between phase-A's the one shot then allows the reverse thru and switches the limit3 input speed back on, so it ramps back up to speed and life returns to normal. This gives Z time to proper follow the spindel. Both of those are PMDC motors, G0704 can do a full reversal at 1500 rpms in around 400 millisecs, but TLM is obviously slower because if I run the reverse any faster I'll strip the teeth off the spindle belt, so the limit3 is set slower. > But it doesn't answer my question on how it works with a spindle using > 0-10V plus DIR for the spindle while using STEP/DIR for the Z axis. > How is the Z coupled to the spindle when the control over > deceleration, as you explained about yours, is somewhat loose.
Its following the spindles encoder for the Z but it has a locking time to accel the z before it fully locks them together. Once its locked, you can kill spindle power and turn it by hand, either direction and z will follow. So its not "loose", but profiled during the turn around to stay at a max rate that stays within the practical accell limits of that machines ballistics. With the pico pwm-servo's running the PMDC's at about the equ to 2 horse motors during surges, useing a 17 amp limit, you can turn things around so fast parts breakage failures are endemic so TLM is slower on purpose, cogged belts and pulleys cost too many sheckels. Those are both 90 volt, 9.7 amp FLA motors although thats the OEM motor on the G0704, don't seem to mind a source voltage in the 125 range. Both will peak the supply above 150 volts in the slowdown phase because they recoup the motors energy by dumping it back into the analogue supplies, which overvoltages their capacitors but only for a very short time because they suck up that power while re-accelerating in the other direction a few milliseconds later. As long as the caps don't have time to heat, they don't seem to mind. The Sheldon, being run by a vfd and the pi hasn't seen enough use yet to import that code. The vfd, being run by the spinx1 is of course a bit slower, particularly when I have the 8" chuck mounted. I put powerlink belts in it so they don't complain like the worn v-belts did. Its first job was putting a new SS 6.5 Creedmoor barrel in old AT&T. The Ackley-Improved, the 4th one of those installed since I built it in the 60's was rusting out, again. And here in WV, these hills are right in your face so I've no need for 640 yard ranges common in SD or Nebraska. And the 6.5 is a lot easier on me. :) Take care & stay safe John. > John > > > > > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/gene> _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
