hello andy, very nice explaination, i was thiking about using spindle-synchronised motion a few months ago, but didn't know if it was possible,
Do you think that if i use a cam software with the emc2 post, i could shape the lobe using the spindle as a spindle in synchronised motion? That would be very nice because the machine has a good reduction, so the torque is not a problem, and the maximum speed now is about 200 rpm. (This is an old machine to make lobes using a master, and it uses standard HSS tools to make the lobes). Do you think that if i adapt an encoder to the spindle also a vfd, and i make the X axis accurately and fast enough to follow the moves of the spindle i can make the lobe?.. that would be excellent because i don't have to modify the entire machine. Thanks again for your help. Regards. 2010/5/5 Andy Pugh <a...@andypugh.fsnet.co.uk> > On 5 May 2010 15:47, John Kasunich <jmkasun...@fastmail.fm> wrote: > > > Just because a particular motor works well driving the spindle > > in lathe mode does NOT mean it will be even close to good enough > > for direct drive positioning. > > A good point, I have only been discussing what is technically possible in > EMC. > > I am fairly sure that various lathes do exactly what is being > proposed, but how much such spindle drives cost is something I could > not even speculate on. > This machine does it, it also looks quite expensive :-) > http://www.youtube.com/watch?v=JZcFMjLmjZc&feature=related > > <thinking aloud> > > I would guess that there is a risk of the flute-frequency setting off > oscillations in the controller. > > For the proposed application (rough-milling cams) the rotary stiffness > requirements might not be all that exacting. > > I suppose in principle, with a fast-enough X axis and enough low-speed > spindle torque you can machine cams with a conventional turning > operation and spindle-synchronised motion. > > I think this might be one of those applications where a small-budget > proof-of concept might be a good idea, though the problem here is that > it might be something where you need the expensive stuff (high torque > servo, high-count encoder) to make it work. > > > Good PID tuning might be able to reduce the steady state error, > > but when an individual flute of a spinning end mill applies a > > brief force to the axis, the axis will move. Only after it moves > > will there be a position error that the PID can use to drive the > > axis back to the proper position. > > I wonder if a Resolver (and high-resolution interface electronics) > might be a better bet than an encoder? > > > http://www.analog.com/en/other-products/militaryaerospace/ad2s80a/products/product.html > > Is a 16-bit converter and claims +/- 2 arcmin resolution. on 50mm > diameter material I think that is 0.015mm (0.0005") tangential > resolution. Say 4 counts to produce a PID response and it is probably > good enough for end-mill roughing a cam with live tooling, but not for > some other operations you might think of. (I am imagining the cutter > axis at right angles to the spindle axis and on the same plane, the > stiffness requirements with the axes parallel strike me as more of a > challenge) > > -- > atp > > > ------------------------------------------------------------------------------ > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
