I have one of these rotary voice coil actuators (RA 60):
http://www.beikimco.com/actuators_rotary.php

Think of it like a motor, but limited to +-15 degrees of rotation.

I want to position it to a particular encoder count. I can get velocity from the encoder.

What would be the best control method for this type of actuator? I have various types of amplifiers at my disposal, and various other stuff to do signal processing with.

Edit: Another important detail would be that I am using it pretty much unloaded.

 

These are similar to the ones found in Hard drives, and in order to be able to "control" the position of the armature, you will have to use an h-bridge to drive the coil, basically the same way they position it in a hard drive to read a specific area on the hard drive platters....

I am working a driver circuit for these types of "voice coils", after a few failed trials on "positioning" it using PWM and other sources of control, all I managed to do was to get the "coil" to "sing" at different frequencies.

The last driver circuit I used to test the "positioning" of these armatures, I used an LMD18201T H-Bridge, and I managed to "position" the armature right where I wanted it by constantly providing a PWM input to the H-Bridge's PWM pin, and used the direction input to control the position instead, once the armature was in the correct "position" I just pulsed the direction bit to keep the armature in that location.

 

ST L298 in conjunction with a uC. Just be a bit concerned about your amp load. Motor can pull upwards of 5.6A, driver is only capable of 2.5A Continuous. Might be better to go with the driver BMorse listed above as it seems capable of 3A (not much research). Might need to incorporate MOSFET's on the driver outputs to achieve higher amps? IDK

 

It won't need much current if unloaded.

 

When I posted the thread, I was thinking more in terms of control systems theory.
PID, velocity feedback, back EMF, current control.
How can I get the motor to lock in on an encoder count without dithering?

I have a few different H bridge style servo amps that have a number of different modes, plus I have an I/O board that I can use to process signals and implement PID. I just haven't had a lot of luck so far.

 

For best resolution I would go with an optical encoder if possible, but with this type of motor that may be difficult. I don't think back EMF would be plausible if you are looking for positional accuracy.

 

What is it driving? All you have said so far is "pretty much unloaded".

If we knew what it was driving and how it needed to perform in use we could suggest good driving techniques.