Greetings Ned,
I don't know yet if using encoders and a servo motor is the best way
to go for matching spindle speeds. I have had success doing a bench
test using one servo motor as the master and a second servo motor
driven by a Gecko 320 as the slave. But the slave is only spinning
itself and not the mass of a spindle and chuck. You mentioned using a
VFD to match spindle speeds. In the lathe the existing spindle is
driven by a DC motor connected to a Randtronics drive. Would both
spindles need to be driven by AC motors connected to VFDs in order for
speed matching with a VFD to work? And can you suggest any models or
brands of VFDs to look at?
Thanks,
Eric

On Thu, 11 Sep 2008 00:29:03 GMT, wrote:

Greetings Ned,
I don't know yet if using encoders and a servo motor is the best way
to go for matching spindle speeds. I have had success doing a bench
test using one servo motor as the master and a second servo motor
driven by a Gecko 320 as the slave. But the slave is only spinning
itself and not the mass of a spindle and chuck. You mentioned using a
VFD to match spindle speeds. In the lathe the existing spindle is
driven by a DC motor connected to a Randtronics drive. Would both
spindles need to be driven by AC motors connected to VFDs in order for
speed matching with a VFD to work? And can you suggest any models or
brands of VFDs to look at?

From your other posts it sounds like you're making good progress,
Eric. Have you run into something about the servo that worries you?

For VFDs, I'd look first at Yaskawa. To get the sort of speed
regulation you're looking for you'll need a vector drive with
feedback. The F7 and G7 drives, marked "CLV" (closed loop vector)
under "Control Type" on this page, accept accessory cards that take
encoder feedback. They'll follow step and direction inputs, so should
be adaptable to your current control scheme. Unfortunately you have to
buy the high-end drives to get the vector option.

http://www.yaskawa.com/site/products.../ACDrives.html

The older Yaskawa GPD drives were also nice units and available with a
vector plug-in, though I don't think they had step and direction
inputs. You'll also see them rebadged by Magnetek. You may be able to
locate one on ebay, though the speed command may be a problem if they
don't have the step and direction inputs.

You wouldn't need to have vector drives on both spindles, but that
could be a slick setup. Though I've never done it, the Yaskawas can be
networked, and I believe can be set up as slave and master.

I expect Mitsubishi has something as well, but I'm not as familiar
with their current models. It also appears you can get a vector drive
from Automation Direct that accepts step and direction, undoubtedly
cheaper than Mitsu or Yaskawa.

--
Ned Simmons

On Wed, 10 Sep 2008 23:59:06 -0400, Ned Simmons
wrote:

On Thu, 11 Sep 2008 00:29:03 GMT, wrote:

Greetings Ned,
I don't know yet if using encoders and a servo motor is the best way
to go for matching spindle speeds. I have had success doing a bench
test using one servo motor as the master and a second servo motor
driven by a Gecko 320 as the slave. But the slave is only spinning
itself and not the mass of a spindle and chuck. You mentioned using a
VFD to match spindle speeds. In the lathe the existing spindle is
driven by a DC motor connected to a Randtronics drive. Would both
spindles need to be driven by AC motors connected to VFDs in order for
speed matching with a VFD to work? And can you suggest any models or
brands of VFDs to look at?

From your other posts it sounds like you're making good progress,
Eric. Have you run into something about the servo that worries you?

For VFDs, I'd look first at Yaskawa. To get the sort of speed
regulation you're looking for you'll need a vector drive with
feedback. The F7 and G7 drives, marked "CLV" (closed loop vector)
under "Control Type" on this page, accept accessory cards that take
encoder feedback. They'll follow step and direction inputs, so should
be adaptable to your current control scheme. Unfortunately you have to
buy the high-end drives to get the vector option.

http://www.yaskawa.com/site/products.../ACDrives.html

The older Yaskawa GPD drives were also nice units and available with a
vector plug-in, though I don't think they had step and direction
inputs. You'll also see them rebadged by Magnetek. You may be able to
locate one on ebay, though the speed command may be a problem if they
don't have the step and direction inputs.

You wouldn't need to have vector drives on both spindles, but that
could be a slick setup. Though I've never done it, the Yaskawas can be
networked, and I believe can be set up as slave and master.

I expect Mitsubishi has something as well, but I'm not as familiar
with their current models. It also appears you can get a vector drive
from Automation Direct that accepts step and direction, undoubtedly
cheaper than Mitsu or Yaskawa.

--
Ned Simmons
Thanks Ned. Even though things are working well now it will eventually
involve two motors on the second spindle. Since the servo will be belt
driving the spindle, and there will be the load of the spindle and
chuck, I am a little concerned. The next test will be to load the
slave and see if it keeps up. I still need to make a strobe to watch
setup.
Eric

On Thu, 11 Sep 2008 15:29:24 GMT, wrote:


Thanks Ned. Even though things are working well now it will eventually
involve two motors on the second spindle. Since the servo will be belt
driving the spindle, and there will be the load of the spindle and
chuck, I am a little concerned. The next test will be to load the
slave and see if it keeps up. I still need to make a strobe to watch
setup.

Oh yeah, I'd forgotten about the 2-motor setup.

If you can get your hands on a large brushless servo motor cheaply,
it's possible to use it as a synchronous motor and drive it open open
loop with a conventional volts/Hz VFD. It'll lock to the drive's
output frequency without the slip inherent in a typical induction
motor, so commutation and shaft speed feedback are unnecessary.
Probably more unconventional than what you've got in mind, but I'll
explain more if you're interested.

--
Ned Simmons