Delta Computer Systems
Some machines coordinate both electrohydraulic and electromechanical controls to gain the best benefits of both. One such system is a ceramic-wafer press manufactured by Gasbarre Products Inc., DuBois, Pa. (www.gasbarre.com). Two motion axes close and compress a mold cavity, and one axis fills the cavity with ceramic powder and ejects finished parts.
Because the upper axis provides all the compression force, hydraulics makes the most sense. The hydraulic axis uses both position and force control in sequence to move the upper piston into contact with the ceramic powder (using position control) and then to compress the powder to the desired density (using force control).
The second motion axis slides the cylinder up as the compression piston moves down. Because this axis is not directly applying compression force, position control alone is sufficient to ensure proper motion. A conventional electric servomotor driving a ball-screw operates this axis.
The third axis requires high-speed back-andforth linear motion to sweep powder into the cavity and then eject the finished part, making a hydraulic actuator the preferred power source.
An electronic motion controller coordinates motion and ensures repeatability. To satisfy Gasbarre's key requirements, the motion controller would:
Gasbarre selected an RMC100 motion controller from Delta Computer Systems to coordinate all motion control. The RMC100 connects directly to the system's MDT position transducers and a load cell the force sensor generating an analog signal output. Concept Systems Inc., Albany, Oreg., a motion-control systems integrator, assisted with design, development, and testing.
Fluid-power systems often take a back seat to traditional electromechanical motion control. But designers who understand both technologies can take advantage of the best features of each, and in the process build higher-quality machines with lower life-cycle costs.
Particularly for applications that require precise control of large forces and smooth motion, fluid power can deliver significant benefits. However, that means selecting the right hydraulic-system elements and tuning the motion controller for optimal performance.
ELECTRIC VERSUS HYDRAULIC
Conventional electric motors are well suited to applications where rotational motion predominates. They are generally easy to control and can be the least expensive power source in small systems that have few axes or light loads. Linear electric motors have an advantage in positioning applications that require quick direction changes, although they can be more expensive than conventional motors. Hydraulic motors and actuators do virtually everything that electric motors can, but with several advantages.
For example, hydraulic actuators lift and hold heavy loads without brakes, move heavy objects at slow speeds, apply torque without gearing, and need less space and produce less heat than electric motors. Electric motors must be sized for the maximum load; pumps are sized for the average load. And hydraulic actuators are comparatively small, even for applications that involve heavy loads.
Hydraulic systems hold the greatest advantage when there is intermittent motion because an accumulator stores energy while the actuator is stationary. On the other hand, electric motors make sense in continuous-motion applications, such as conveyors.
Electric motors typically mount close to or directly on the motion axis. A hydraulic pump may be located remotely with only the accumulator and pressure-control valves near the actuators. This can make fluid power an ideal motive force for robotic applications with many axes. A pump mounted in the base minimizes weight on the arms and possibly reduces noise. And sharing a pump between several actuators can result in a lower cost per axis than equivalent elec-motor systems.
Fluid power has the additional advantage that pressure can be held constant without significant additional energy. By comparison, applying constant torque could overheat an electric motor. In material-transfer applications prone to mishandling or misfeeds, fluid power's morecompressible power-transport medium may be more forgiving than electromechanical power if the system jams.
Designers developing electrohydraulic systems for the first time must deal with some new issues. The most common use of hydraulic power is linear motion and the most important factor in planning linear-motion systems is sizing the actuators.
Clearly, cylinders need to be long enough for the required stroke. Specifying the cylinder diameter is where designers sometimes make mistakes. This is crucial because system natural frequency is roughly proportional to cylinder diameter. Natural frequency determines the system's maximum controlled acceleration rate. Therefore, to accelerate twice as quickly means doubling both the system natural frequency and cylinder diameter.
A common error is to use smalldiameter cylinders that move quickly under relatively light loads. Unfortunately, at higher loads a small piston provides little surface area for hydraulic pressure to act on. Consequently, the system may not attain the desired speed in the required distance. Applying a large hydraulic force can also cause a small-diameter cylinder to act as a spring. The system oscillates when acceleration stops as the hydraulic spring attempts to return to an uncompressed state.
Because fluid is compressible, system stiffness increases with piston diameter. Thus, systems with larger-diameter cylinders compress less, accelerate and decelerate quicker, and offer more-precise control. As a rule of thumb, hydraulic-cylinder diameter must double to halve the acceleration rate.
After determining cylinder diameter, size the pump to provide sufficient fluid flow for the required speed and acceleration. An oversized pump wastes energy. Fortunately, sizing is relatively simple: The volume of oil flow must match the cylinder's change in internal volume over time.
Note that higher acceleration rates place significant demands on pump flow. For instance, doubling the accelerationrequires double the cylinder diameter or four times the surface area. With four times the area and twice the speed, oil flow must be eight times higher.
An accumulator in a fluid-power system serves two purposes. First, it is a buffer, in essence time-averaging power requirements from the pump. Second, it keeps system pressure relatively constant despite the effects of motion-control inputs. This avoids the need to continually change the motion controller's input-response relationships to maintain precise control. A good rule of thumb is to size the accumulator so that pressure changes no more than 10% during the operating cycle. Further, to minimize system pressure losses, locate the accumulator close to the valve rather than the pump.
High-performance fluid-power systems typically use servo or proportional valves. With servovalves, a linear increase in current through the valve coil directly moves the spool, causing a linear-increase in oil flow through the valve. Proportional valves, on the other hand, have position feedback on the spool and a valve amplifier to linearize flow. Proportional valves are generally less expensive and more tolerant of contaminants than servovalves, but the benefits often come at the expense of performance. Valves often have an overlap or "dead band" in the center where the flow is blocked. This causes a nonlinear response for which the motion controller must compensate. Zero-overlap valves are often necessary for optimum performance.
For maximum responsiveness to control inputs, size valves to provide required flow plus 10 to 20%. However, significantly oversizing the valve will result in coarse control as it uses only a small part of the control range.
Mount the valves as close to the cylinder as possible and use tubing instead of hose. This reduces the volume of trapped oil and lessens compressibility. Also, mount the valve on top of the cylinder-so that any air in the system will automatically carry back to the fluid reservoir.
To monitor pressure, place sensors on either end of the cylinder bottom where there is less oil motion and no trapped air. A common mistake is to mount pressure sensors in the manifold, where the venturi effects of moving oil can decrease pressure readings. Turbulence in the oil flow may reduce venturi effects but, in any event, manifold pressure may not equal cylinder pressure.
Linear transducers such as linear magnetostrictive displacement transducers (MDTs) measure absolute position and do not require homing. MDTs also have pressure and temperature specifications that allow them to be inserted directly into hydraulic cylinders.
The motion controller is critical to delivering the full benefits of electrohydraulics. Look for electronic motion controllers that:
- Provide a direct interface to position and pressure transducers, avoiding extra interface costs and performance delays due to signal propagation.
- Can control proportional valves and generate precise hydraulic motion.
- Smoothly transition between position and pressure-control modes to
- avoid motion discontinuities that can impact performance and output quality.
- Can coordinate several axes simultaneously by "gearing" one axis motion to another, ensuring precise, repeatable motion.
- Support the execution of high-level commands, such as higher-order function interpolations, to smooth motion without requiring complex and time consuming low-level programming.
- Have graphical tuning tools to speed design optimization.
- Provide direct interface to factory or fieldbus networks if the machine is part of a larger plant environment.
Many of these characteristics are also useful for electromechanical control. In fact, some complex systems coordinate control of both electromechanical and hydraulic motion axes. (See Hybrid control.)
Tuning fluid-power systems is similar to tuning electromechanical systems. Electric servos have two main operating modes, velocity and torque. In velocity mode, speed is proportional to the control output from the motion controller to the drive amplifier. In torque mode, the servo's torque or acceleration is roughly proportional to the control output to the amplifier.
Hydraulic systems only operate in velocity mode, as oil flow is ideally proportional to control output from the motion controller. Velocity mode is more intuitive than torque mode and is easier to set up, by running the system with open-loop controls. In torque mode, a system requires closed-loop control because a constant open-loop voltage causes the servomotor to accelerate. A zero-control output does not stop the servo, it just allows the servo to coast to a stop.
Tuning the proportional, integral, and differential terms (P, I, and D) is similar for a velocity-mode or torquemode controller. However, the differential term is much more important for controlling an electric motor in torque mode. The differential term, which relates to the rate of change in the error between the system's actual and target positions, provides speed stability.
In contrast, electric-servo velocitymode systems are easier to set up and usually do not require a differentiator because the motor's drive amplifier provides this function. The downside is that the amplifier and motion controller must be properly tuned, increasing the effort required to ensure proper system operation. It is often easier to fix the drive-amplifier gain to a constant value and let the controller manage the motion profile solely in relation to its internal PID values. Then all gains are in one place. Because fluid-power systems always operate in velocity mode, they share the advantages of simpler tuning with velocity-mode electric motor controls.
In addition to P, I, and D, many motion controllers also provide feed-forward parameters. Feed-forward control algorithms let the system anticipate and proactively drive the motion rather than react to transducer stimulus. In electrohydraulic systems (and electromechanical systems using velocity-mode controls), velocity feed forward is the most important term in a correctly designed algorithm. It provides a component to the control output proportional to velocity. Acceleration feed forwards give the control signal a boost while accelerating and braking, but have no effect at constant velocity.
After setting the feed forwards, the designer will typically tweak the P, I, and D gains to get the desired control. Tuning the PID adjusts the amount of output needed to make the system move. The controller generates this output using five terms, generated by acceleration and velocity feed forwards, and proportional, integral, and differential gains. The goal is to make the feed forwards do most of the work. This way, the PID contribution to the control output is small and minimizes the error between target and actual position.
Another difference between fluid-power actuators and electric servos is that the latter typically require only one set of gains. Fluid-power motion controllers require two sets of gains for linear cylinder applications. This is because surface area differs on either side of the cylinder piston, due to the cylinder rod. The maximum force and therefore system gain is greater when the piston is extending than when retracting. A typical electric servocontroller will have a hard time controlling a hydraulic system because it usually has only one set of gains and can be tuned to work properly in only one direction. Hydraulic motion controllers should have two sets of gains, one for extending and one for retracting. Having two sets of gains is also handy in vertical applications where the load changes greatly depending on whether the system moves up or down.
Delta Computer Systems, (360) 254-8688,