Transfer Function of actuator

[rollsroyce21]

Hi!!

I am trying to design a position control system based on an electromagnetic actuator. a PID controller seems like a reasonable choice for me to use but I am having problems trying to derive a transfer function for the actuator before I begin designing the controller
any help would be appreciated
thanks!

 

[sreid]

We need a little more information. Is the actuator, say, a voice coil driven by a current loop amplifier?

 

[rollsroyce21]

Yes, sorry about that
Its a SMAC LAL 30-015 moving coil actuator and it is driven by a current loop amplifier.
I am currently trying to measure its performance in terms of robustness. So I am running a MATLAB simulation first of the system and then comparing the results to the actual working model

The amp is a 24v , 5A dc supply
The actuator has a 15mm stroke and peak force of 14.5N

 

[PNachtwey]

So what do you think the basic form of the actuator is? I would think it would be something like

K*omega^2/(s*(s^2+2*zeta*omega*s+omega^2)

The second order part would represents the mass in the coil and omega should has a very high frequency. If there isn't much friction then the damping factor, zeta, will be quite low, perhaps 0.1 or smaller. Let use know what you think.

 

[sreid]

The actuator can be modled as a single pole low pass filter electrically. You need to know the inductance of the coil to calculate the bandwidth [plus the loop gain of the amplifier]. Or measure it, lock down the actuator, drive the amplifier with a low amplitude square wave and scope the current rise time.

Mechanically, the actuator follows Newton's First Law of Motion, F = M x a

If you want a "Killer" amplifier to drive the actuator, look at a LA-400 amplifier from Varedan Technologies

http://www.varedan.com