Machinedesign 1154 0910lsseries 0 0

Servopneumatic systems

Sept. 1, 2010
Precision is one of the chief advantages of servopneumatic control systems. José Riofrio of Enfield Technologies, Trumbull, Conn., explains some of the latest developments and where they may be a fit in your next design.

MSD: What are the advantages of servopneumatic control over other alternatives?

JR • Enfield: Aside from the familiar pros and cons between hydraulics and pneumatics in general (such as force capabilities, cleanliness, and cost) there are very few, if any, additional comparisons to be made between servohydraulics and servopneumatics, and there is almost no head-to-head competition between the two. Whether to use one or the other typically becomes clear in the early stages of a design process.

However, comparing servopneumatics with servoeletrics is somewhat tricky and there are some grey areas. The biggest advantage of servopneumatics is its high energy and power densities, which can result in significant cost savings (both in terms of money and payload) in applications that simultaneously require high speeds and high forces. Generally speaking, the cost of a servopneumatic positioning system holds constant across the power spectrum: A small increase in valve and cylinder bore sizes results in significant power gains (related to the square of the bore), with very little added cost for the end user. In contrast, low-power stepper motors driving ballscrews may be affordable, but get increasingly bigger, heavier, and much more expensive as power demands increase. Additionally, unlike steppers or dc servomotors, pneumatic actuators can store energy, which allows them to hold a static force with virtually no energy consumption.

MSD: What are the roles of the special valves and controllers used in servopneumatic control systems?

JR • Enfield: Servopneumatic control systems have four main applications — position, force, pressure, and flow. The explanation here focuses on position control, though the other three are almost identical and differ mostly in the type of sensor employed.

Typical servopneumatic (and most servohydraulic) position control systems consist of a double-acting cylinder, full-scale linear position sensor, high-speed proportional 5/3 (5-port, 3-position) valve, and a suitable closed-loop controller. The controller receives a target setpoint command signal and a sensor feedback signal, and based on its algorithm, communicates a plan of action to the valve. From an end user standpoint, however, the controller and the valve can be treated as a “black box” component: The user simply inputs a command signal (target position, velocity, or trajectory) and the cylinder moves accordingly. Typically, this command signal comes from a PLC and is in the form of voltage (0 to 10 Vdc) or current (4 to 20 mA).

The proportional valve (or in some cases, servo valve) undoubtedly plays the central role in the overall system. Even though it has three discrete “positions,” it can infinitely vary its effective orifice size, essentially controlling the “extend” and “retract” cylinder speeds. The valve's high-speed feature means that it can change the size of the orifice almost instantly. By quickly and accurately changing valve positions (cylinder direction) and orifice sizes (cylinder speed), any cylinder position can be obtained and maintained. This is where the controller comes into play.

The controller needed for a servopneumatic system is comprised of two main stages: The control stage (i.e., PID) compares the command and feedback signals, while the drive stage takes the algorithmic output of the control stage (i.e., “control effort”) and turns it into a suitable amplified drive signal for the valve. The controller algorithm, regardless of its nature, always works to minimize the difference between the command and the feedback by generating a cylinder motion in the direction of the setpoint command. Note: For adequate closed-loop position control, the controller needs to be the fastest device in the system.

MSD: Any advice for engineers setting up a servopneumatic linear positioning system for the first time?

JR • Enfield: The most important aspect for correctly setting up a servopneumatic system is sizing. The cylinder bore size must be carefully selected for the required dynamics, and the valve must be sized according to the flow requirements of the system. The two most common mistakes typically made when setting up a system are undersizing the cylinder bore relative to the system dynamics, and oversizing the valve relative to the cylinder size and dynamics, either of which usually results in an uncontrollable system. Here is a good step-by-step example of how to size a position control system using servopneumatics:

  1. Determine the maximum controlled acceleration required.

  2. Based on total moving mass, determine the force required to provide the acceleration (F = m•a).

  3. Select a cylinder bore size that can provide at least twice the force required (ideally, much higher) based on the line pressure and bore area. The higher this force margin, the more controllable the system becomes, and as such it may pay off to approach a margin of 10 if the space permits it. It's useful to keep in mind that the force increases exponentially with respect to the cylinder bore: To double the available maximum force, one needs to increase the cylinder bore by only 41%. Similarly, by doubling the bore size, the available maximum force increases by a factor of four.

  4. Based on the maximum target linear velocity, determine the maximum airflow that will be needed in SLPM (standard liters per minute - SI units) or SCFM (standard cubic feet per minute - Imperial units). Most valves in the U.S. and European markets list their flow specs in at least one of these two units.

  5. Select a proportional valve that can handle the required flow rate calculated in step four. In general, it is better for controllability to undersize rather than oversize.

Page 2 of 2

MSD: Any new developments regarding servopneumatic control systems?

JR • Enfield: As microprocessors continue to become cheaper, faster, and more powerful, low-cost digital controllers will become more common. This will allow for much more capable controllers in smaller packages, which can be easily customized without PCB revisions. There also seem to be promising developments in the field of micro and nanotechnologies, which could make servopneumatics (and servohydraulics) suitable for microfluidic applications. Some companies already sell miniature valves that use piezoresistive elements to control the orifice apertures.

Servopneumatics enable superior swivelling

When motion requires gentle braking, shock absorbers are often preferred. Another option gaining use is servopneumatic components — because they combine the advantages of pneumatic and electric drives.

Compared with standard pneumatics, servopneumatics are up to 30% faster and use up to 30% less air. Servopneumatic drives are also freely positionable, making them up to 50% more economical than electrical solutions, according to engineers at Festo Corp., Hauppauge, N.Y.

The company recently introduced the DSMI-B servopneumatic swivel module that rapidly swivels without vibrating. Shock absorbers are unnecessary, reducing purchasing and maintenance costs.

The DSMI-B is suitable for a wide range of material-handling tasks when combined with the electrical terminal CPX (a modular automation platform for valves and electrical control functions) and its CMPX (end-position controller) or CMAX (positioning) modules.

The sturdy DSMI-B semi-rotary vane drive has integrated measuring for use as a soft-stop axis and for free positioning. Designed for mass moments of inertia of up to 6,000 kg/cm2, it swivels up to 270°. In one Festo ball-throwing demonstration, engineers set up the DSMI-B's arm to catapult a ball using a short stroke cylinder. While the ball is in the air, the arm swivels 180° in just 0.5 sec, brakes without vibrating, and catches the ball.

An electronic end-position controller, CMPX, gently brakes the swivel drive in its end positions, even at high speeds. This soft stopping reduces cycle times by about 30% while reducing noise and eliminating vibration. Loads to 300 kg can be dynamically moved; what's more, the system can be programmed to move to two additional, user-defined intermediate positions without a fixed stop.

Pairing the swivel module DSMI-B with servopneumatic positioning module CMAX allows free positioning, which is appropriate where loads exceed 10 kg and accuracy to within a few tenths of a millimeter is sufficient. Using the CMAX, drives can be positioned precisely to within 0.2 mm and force controlled to within 5%.

For more information, visit or call (800) 993-3786.

Precise positioning requires accurate feedback

Positioning — and its corresponding accuracy — ultimately depends on the validity of the position information provided by sensor feedback. Because servopneumatic linear positioners typically do not contain any rotating elements, rotary encoders cannot be used. Instead, position feedback must come from a linear position sensor, a vital component available in several formats:

  • Incremental linear encoder (magnetic or optical)
  • Absolute linear encoder (magnetic or optical)
  • Linear potentiometer (resistive potentiometric)
  • Linear position transducer (magnetostrictive)

Although incremental and absolute linear encoders can provide exceptional accuracy and performance, they typically have higher initial cost than other alternatives. Their quadrature or serial digital electrical interfaces can also be costly on the controller side, and the operation of optical types can be adversely affected by wet or dirty conditions. Both of these factors can negatively impact the performance-to-cost ratio of these devices.

Linear potentiometers or pots often fit the bill in terms of price as well as simplicity of electrical interface (analog voltage). However, due to mechanical contact wear, their life is limited in high-cycle applications. As the internal contact surface wears, electrical noise increases and accuracy degrades over time. Linear pots often lack the environmental ratings necessary to resist typical industrial conditions such as exposure to heavy dust or liquids, which can greatly accelerate wear and signal degradation. Another note of caution: Linear pots depend on an operating rod that must be mechanically coupled to a positioning system, and sometimes it's inconvenient or impossible to implement such a mechanical connection.

Magnetostrictive linear position transducers solve many of these issues with non-contact operation. Here's how they work: A magnet, acting as a position marker, is affixed to a linear-moving element of the servopneumatic system. The linear position transducer is mounted close to the position magnet, which “floats” above the transducer housing without touching it. The output is directly proportional to the magnet's location over the transducer's entire length. Importantly, magnetostrictive linear position transducers provide absolute position measurement: The transducer always reports actual position without the need to re-home the system, even on power-up. Several electrical interfaces are available, but the most common and least costly to implement on the controller side is analog (voltage or current).

This month's handy tips courtesy of Balluff Inc., Florence, Ky. For more information, visit or call (859) 727-2200.

Voice your opinion!

To join the conversation, and become an exclusive member of Machine Design, create an account today!