How well does a VFD maintain speed with encoder feedback

[strantor 0]

Hi, I'm considering trying to build myself a lathe. I already have a lot of the materials I'll need to build it. One thing I don't have, is a box full of all the specific gears to make up all the combinations of gear ratios between the spindle and the lead screw that you would find on a regular lathe. I do however have (2) 1hp 3ph induction motors and 2 appropriately sized VFDs (one yaskawa one emerson unidrive sp). I was thinking that if I used 2 seperate motors, one for the spindle and one for the lead screw, I would actually be more versatile than a traditional lathe, as I could have any ratio of turns that I want. However, I'm a little concerned about the precision involved. With mechanical gears there is no possibility of ratio of turns changing with a change in load, but with 2 seperate induction motors this would be possible. I think the problem might be mitigated by the use of encoder feedback on the 2 motors for precision speed control, but I'm not sure if that is good enough. I've installed a few induction motors & VFDs before and everything I've done has been in the wire & cable industry - i.e. motors that run machines that twist cables. The absolute speed control of the VFD + induction motor with encoder feedback has always been more than good enough for what I needed, which was pretty high tolerance in comparison to making threads @ 32tpi. That's what has me worried; I've never really put low tolerance speed control with induction motors to the test. I've got a stong suspicion that I can pull it off by gearing everything way down (I do have some gears) to increase the position resolution. It would be slow, but hopefully it would be accurate. Anybody know how well this might work? how well can the VFD maintain an exact speed, with changing load?

[Shiner 2]

Yaskawa has free tech support that may be able to give you good numbers. This may be one of those projects where you end up tweaking some of those 300 parameters that never seem to be used..... 1-800-YASKAWA If you want some real world application feedback on whether or not this is going to be reliable, one of our engineers has been doing drive commissioning for about 30 years. I'm sure he has come across something with similar requirements. 1-716-656-9900 ask for Jim Touhy (pronounced two-E)

[gleblanc 1]

A 1 hp motor is WAY bigger than what you want for the lead screw. I think the hard part of this project is going to be fabricating a lathe bed, and figuring out how to get the headstock aligned. Next will be figuring out a cross slide and/or compound. As far as replacing the traditional change gears with something electronic, it can be done. John from Automation Artisans has made a great product in the ELS. It can take a single PPR encoder on the spindle and some stepper motors and drive both the lead screw and the cross slide. Unless you're into re-inventing the wheel, I'd take a look at that. http://www.autoartisans.com/ELS/