[Mike Braddock]

I have an old CNC mill with a VFD and AC drive. It has a control knob so the operator sets the spindle speeds before running the program. My question is, can this be controlled in closed-loop fashion if I add an encoder to the motor shaft or spindle, or will that only work with certain types of AC motors (i.e. syncronous vs induction)? I would be using something like EMC or a Delta Tau PMAC for the motion control. Thanks for any help you can provide!

Mike

[Al_The_Man]

Generally if you want close spindle control of an induction motor using a VFD, the VFD encoder option is ordered and a pulse generator or encoder is fitted to the motor shaft.
If you want to close the loop to the controller for synchronous tapping, then usually an encoder is added to the final spindle shaft, and this is used preferably with the above, although you may be able to get results with the final shaft encoder and use a good flux vector VFD.
To control the VFD by Sxxxx value & M3/4/5, you need an analogue output from your controller plus a fwd/rev controller output to the respective inputs on the VFD.
Some VFD's will accept ±10vdc with just a Start input.
Al.

[Mike Braddock]

Thanks Al. For the immediate project, I don't want the type of syncronization neccesary for rigid tapping, but rather just some cutting force disturbance rejection. In other words, if my spindle is running at 2500 rpm in air and begins cutting, I just want it to be able to maintain a fairly constant velocity despite the cutting forces (let's say +/- 50 rpm). Currently my machine doesn't do this. I think the answer to my question is that adding a velocity feedback device (tach or hall effect sensor) will allow me to close the loop and provide *some* disturbance rejection, but how well it work will depend on the quality of my AC motor/VFD combo.

Asking a second question, what would I have to replace my current inductive AC motor and old VFD with to get the type of syncronization required for rigid tapping? Would a new syncronous AC motor and high quality VFD get the job done, or would I be better off using a DC motor and amplifier? The motor is about 3 hp. I'm interested to hear which you think would work better and why. Thanks again Al!

Mike

[Al_The_Man]

First I would check on the availability of the PG or encoder option card for the VFD, you would need to add the PG or enc. on the rear shaft of the motor.
For rigid tapping and synchronous control, many of the Spindle motors used in some of the major systems use a Vector rated motor, with PG feedback and the spindle shaft encoder to allow the Z axis to sync to a relatively stable spindle, together with a high quality VFD type controller.
Manuf. like Mitsubishi supply after market spindle controls and motors like this to replace the older expensive DC spindle controllers.
But these also are not cheap.
Al.
.

[Mike Braddock]

Thanks again Al. So for regular velocity control (not for rigid tapping) it's usually necessary to close the position loop on the spindle VFD? Does the amplifer take the derivative of this signal and use that for velocity feedback? Would closing the velocity loop in a controller and outputting a variable analog voltage to the spindle VFD controller not work (well) to control spindle speed? In other words, would it work but just not give very good velocity error?

Mike

[Al_The_Man]

It is more velocity control that a position loop.
The VFD has DSP processors that operate with induction type motor and encoder or PG to maintain the velocity.
I have never tried it by closing the velocity loop with a Induction motor/VFD back to a controller, I suspect it would be very difficult to tune without a tight inner loop from motor to VFD.
Al.

[Mike Braddock]

That makes sense. Thanks for your explanation Al.