On Sunday 18 October 2020 17:12:16 andy pugh wrote: > In addition to working with LinuxCNC I also help look after a 1916 > fire engine. This fire engine has a 9 litre 4-cylinder engine with a > large flywheel. A consequence of this is that on some hills that it > could easily climb in third gear you can't get out of first gear if > forced to come to a halt and re-start. > The reason for this is the crash gearbox. To engage second gear you > need to de-clutch and then engage the gear once the engine revs have > fallen to match output-shaft revs. And up a hill the vehicle slows > faster than the engine, so that by the time you could engage second, > you need to stop and go back to first. > > My lathe has a 2-speed gearbox, operated by electromagnetic clutches. > It chooses the gear when the spindle is turned on based on commanded > speed. > This can be a bit sub-optimal when turning a large diameter down to a > small diameter, as it ends up with too low a surface speed if stuck in > low gear from the beginning. > > I have long had the idea of doing on-the-fly gearchanges: Wait for a > G0, de-assert spindle-at-speed, disengage the current gear, > synchronise the motor to the current spindle speed in the new gear, > then engage the new gear. > > Today I did this, and I have the same issue as the fire engine. When > switching from high to low gear it works very smoothly. The spiindle > is disengaged and starts to slow, the motor accelerates, they meet in > the middle, engage the gear and then ramp back up to the new speed. > > It works less well going from low to high gear, the spindle slows > faster than the motor, so the spindle has stopped before synch > happens. Just like the fire engine. > > It's still an improvement, but not what I was hoping for. > > I could probably make it work better by programming an active > slow-down to the motor with a faster ramp. But to do that I need to > drag the lathe away from the wall, as the VFD front panel is only > accessible from the back of the lathe. (unless there is a way to > program the VFD parameters via the modbus link)
I've assumed that the higher priced vfd's can be programmed over the com bus, but with the front panel of the cheapest clone I could find at the time I did the Sheldon, I was able to program, not for gear changing, but for the rigid tapping reversal, a shaped stop, switch directions, and the same limit3 ramps it back to the set speed. In the other direction. How fast? At 100 rpms, with a nearly 40 lb 8" chuck mounted, overshoot is 1/4 turn of the spindle. Back to speed in the other direction is about the same. Needless to say, belt slippage gets in the way at the higher speeds, with 400 revs needing almost 4 turns to come to a stop. Even so, I'd guess the complete turnaround is finished in about 1 second at 400 revs. The slippage is in the dual belts that bring the spindle power to the spindle, my pulleys have a visible belly worn into them. And made out of pure unobtainium at 75 yo. The old, badly worn v-belts screamed like Michelin tires on hot blacktop but these powerlinked belts are slipping without the screech. Spinning in bronze half shell bearings, I'm cautious at applying more tension because of the heat buildup in the bearings. 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 [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
