On Sunday 28 January 2018 08:25:49 John Alexander Stewart wrote: > Thanks David and Gene; > > Thanks for the responses. > > 1) The stiffness of mounting is one that I had not really thought much > about but makes sense.
> 2) a higher tooth count gear is something I'll look at this AM. Have > lots of old gears kicking about - and, the idea of adding a bit of > steel gooped to the side is a good idea! > > 3) This lathe is belt driven, so no rigid tapping, Why not? Unless the carriage screw itself is belt driven that is. Belt drive to the spindle is ok as long as and the bull gear the encoder is sensing is keyed to the spindle, then the rigid tapping drive to the screw motor follows the spindle even if the cutting load slows it. So thats a shrug. I guess thats just one of the reasons the long carriage bolt gets replaced by a ball screw, driven by a motor. Servo or stepper is irrelevant, but steppers are easier if lots noisier. Some of that noise is because I used a 1600 oz/in off the mill for the z drive, and its noisy as all get out at some speeds. Between around 5" and 20" you can hear it all over the garage. And while it can run 3x faster on the lathe than it could on the mill due to the lack of weight to lift, I really should replace it with another copy of whats on the mill, a 940 oz/in that can lift the mills 60+ lb head at 90 ipm. This 1600 oz/in is timing belt coupled to the screw, 32 on the motor, 42 on the screw, driving a 1450mm long 1" 5 tpi Chinese screw I paid less than 200$ for. To keep the screw clean, I made 4 end couplings that clamp a couple rubberized cloth bellows nominally 1.5" in diameter, to the end mounts and the faces of the nut, so the screw is totally encased in the bellows. I even made an air passage thru the nut so the bellows wouldn't be under vacuum or air pressure enough to notice as it moves at up to 90 ipm. Thats been working nicely for about a year now. Nut has been greased twice. If I was to replace both that motor with a 940 oz/in, and the AC powered driver same as the mill has now, about a $290-$300 project, I expect it could do 200 ipm rapids. With maybe 20% of the stepper noise. Those 1600 sized nema 34's are 4 wire only and have a 22mh coil inductance, and that, unless you can muster up a 300+ volt stepper driver, (I've not seen that announcement yet) is going to severely impact its high speed torque. The 940 is available as 8 wire, which can be paralleled for quite low inductance, so the G0704 mill's z went from 27 ipm to 90 ipm by useing the smaller motor. But rigid tapping, or g76, either one, using lcnc for the precise coupling, should Just Work if the encoder is sensing the spindle correctly and you are driving the z screw with linuxcnc. Steppers just stay in time (within their speed limits), while servo's will need an encoder to tell lcnc where the screw is. > and I had adopted > Andy Pughs' HAL feedback loop for a two-speed mill which works very > well. IIRC, Andy had it working for his 6-speed horizontal mill, so > crossing fingers, I can get it going for this 6-speed lathe. Andy's insight also showed me how to scale the encoders a/b while still giving me an index from the spindle, and thats working well. Andy's ability to understand, and recommend, is well appreciated at this old farts site. Considering the pi's less than stellar performance I had a whole lot of stuff I wanted to do on this Sheldon, but I built it to run in a 100hz thread, where it seems to work just fine. The apron panel now sports two of the $20 100ppi encoders, with variable speed per click programming that goes down to .00010" per click, so I can still drive it by hand if needed. Best of both worlds. Speed of jog is adjusted by holding a pushbutton and rotating the dial but the machine doesn't move, release the button and the machine moves with the dial until the dial hasn't moved for 5 seconds, then disables. That I need to extend as its too fast a timeout, but when I fix that, I'll probably move that variable to the ini file. Right now its hard coded in the hal file. And its all displayed in the pyvcp panel. > Funny how a project will sit for a year or two, then get exciting > again! Yes, amazing sometimes. Keeps us out of the bars too. ;-) > John. -- 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) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
