# position control of a pneumatic actuator using on/off solenoid valves

O

#### Ozgur Atesoglu

Hi,

We are to control the position of a mass using two on/off solenid valves. Since the actuation is on/off type, continuous and proportional position control may be modulated by changing the widths of the on/off pulses.

Is there a well known strategy for that type of control? Especially for on/off solenoid valves?

The way that we are pursuing will be composed of following steps;

1. Design the conventional (PI,PID, ...) controller for the position loop as the plant is being actuated proportionally (in simulation)

2. Apply the PWM scheme on the designed controller to actuate the valves (talking on software application of PWM.. say Matlab..)

3. Make necessary tunings on the controller and PWM parameters

Is it possible to control the position with these steps? Or is it necessary that the PWM control has to be designed especially for velocity loop and then covered with a simple proportional control on position?

Can you offer better ways or existing control schemes and necessary reading material?

Thanks a lot.

Ozgur

E

#### Emery Premeaux

Ozgur,

If you have good quality air lines, cylenders and valves, you should be able to "set it and forget it" By this I mean: once airpressure is adjusted on both sides of the cylender plate, so long as there are no leaks, the position of the mass will be maintained almost indefinately (not entirely the case, but fine for most uses).

this is exactly the sort of system used in garages for car lifts. They however use fluid instead of air. The reason being temperture changes cause the mass to move up and down when using air.

As for maintaining the position of the mass by way of PWM, you must consider that pneumatics are a lot more dynamic than electronic methods. Fluids pressureize varing rates. In addition, you must contend with not just the theoretical switching speed but the reality switching speed of a solenoid operated shuttle valve. Dont take my word for it though, Im not an engineer.. :/

If position of the mass is critical, here is (I think) a better way to do it:
Use the cylender as a piston, and position the mass with a servo. I can verify this on a mass that weights about 100 pounds. Set the cylender to hold the mass up at the full range of the cylender using a regulator. Then use a servo with a screw attached to push the mass down. It works very well, as the servo only has to work in one direction. Temperature changes are not much of a problem, and the servo responds hundreds of times faster than the solenoid valve does.

Note that air pressure is never releived from the cylender. To lift up for instance, the servo simply releases its holding current and re-applies it at the appropriate location. To move it down, the servo works against the air pressure under the cylender.

Hope this makes sense, and helps
Emery