A motor turns, and through a series of belts, gears, couplings and ball screws, rotary motion changes to linear. For decades, designers have been configuring such mechanisms, essentially starting from scratch on each new project.
Now, thanks to advances in electromagnetic technology, there's an alternative. A new type of linear motor - tubular linear motors - promise to simplify design because they are nearly identical in form, fit, and function to ball screw and bull nut systems. These self-contained actuators are so similar in fact, that machines often need little or no modification to accommodate them.
Tubular linear motors include an armature and a rotor. The armature, or thruster, consists of a single conductive wire cylindrically wound and encapsulated. The stator is a cylindrical assembly of sintered NdFeB permanent magnets arrayed in an onaxis North-South stack contained in an encasing tube. Without iron core elements, there is no cogging so motion is smooth.
The moving thruster does not actually ride on the stator. A relatively large air gap of about 1 mm separates the components. Such a gap is beneficial because it lowers alignment tolerances during installation. Supporting the thruster is an independent bearing system.
As with forcer platten motors, the single stator can independently control multiple thrusters.
Like almost all linear motors, tubular designs operate in servo mode. Analog or digital Hall-effect switches embedded within the armature render sinusoidal or multi-step trapezoidal commutation. Alternately, combining a magnetic or optical incremental linear encoder with a Hall-effect switch will deliver commutation, as well as position and velocity loop closure, and ensure correct phasing on power up.
Tubular motors are brushless, so they need little maintenance and have low EMI. Present versions offer continuous force exceeding 70 lbs, peak output of approximately 300 lbs, and maximum speed approaching 400 ips.
It's all in the symmetry
Much of their power output is due to their radial symmetry. All the magnetic flux intersecting the slider coils generates thrust. The arrangement of the thruster windings and flux pattern ensure that the current and magnetic fields are perpendicular. The result is maximum force.
The symmetry also balances the magnetic fields, reducing any attractive forces between the slider and stator. Lower forces ease installation and reduce loading requirements on support bearings. Tubular motors display attractive forces of several pounds, while conventional forcerplatten motors have forces of several hundred pounds.
Like all devices, there are limitations. For example, the size of these motors can be a drawback for some applications. Support for the stator assembly is only possible at the extreme ends. Therefore, sagging limits the stroke to approximately 80 in. with a 1.5-in. diameter stator. The motor thruster also has a profile height greater than that of other linear motors.
On the job
Regardless of their design type, linear motors are subject to similar application considerations. The relatively large gap between the thruster and stator makes tubular motors less prone to fouling by ferrous chips. Thus, they can operate in applications conventional linear motors can't. Even so, it's a good idea to use protective shields or some means of actively capturing and removing residues.
Horizontal motion applications are best for two reasons. First, when de-energized, linear motors lack holding force without the addition of a shaft brake. Secondly, these motors need some type of lifting force to go against gravity. They typically don't have the benefit of forcemultiplying mechanical ratios.
Early applications for tubular motors have focused on their positioning accuracy and speed. Semiconductor wafer fabrication, for example, relies on the sub-micron positioning available with these motors. In packaging and printing functions, it's the place-to-place speed that's important.
Here's an example of an application that required both functions. Researchers at the Max Plank Institute in Berlin wanted to automate biomedical sample spotting and gridding operations. A tubular motor was chosen to drive the multi-axis robot manipulators. The manipulators pick up samples and spot them onto 22 by 22 cm membranes. The robot grids up to 230,400 samples per membrane at a rate of 250,000 samples per hour.
But this is only the beginning for these devices. As developers refine the design and add more sizes, they will become an increasingly attractive alternative to the more established linear motor architectures.
Continue on page 2
Like its rotating counterparts, a linear motor consists of an armature and a stator that generate magnetic fields to produce force and displacement. The physical principle underlying all linear motor types is shown in the illustration and expressed by the equation:
F = i L B sin Ø where i is the current flowing in a conductor of length L located in a magnetic field B. Sin Ø is the angle between the current and magnetic field vectors, and F is the force vector exerted on the conductor. The vector's cross product follows the familiar right hand rule. Modulating the current controls the magnitude and direction of the resultant force.
Most linear motors use permanent magnets made of various rare earth materials such as SmCo. The cost can be high, depending on the size of the stator. But while expensive, rare earth materials are efficient compared to conventional ferrite components. Sintered SmCo magnets reduce copper winding thermal losses by 78% over a motor of identical geometry and continuous force capacity using sintered ferrite magnets. When equipped with NdFeB magnets, losses shrink by 86% compared to a ferrite equipped motor.
The predominant architectures for linear motors are the force-platten and U-channel designs. The forcer-platten motor has permanent magnets in the stator oriented at right angles to the thrust axis. The magnets are slightly skewed in the vertical plane to reduce thrust ripple.
The coil typically contains an iron core to increase the electro-magnetic flux density and the force output. However, this composition often exhibits strong detent forces that lead to "coggy" movement. Engineers must consider the attractive forces between the stator and slider in installation and load-bearing computations. Core iron also causes eddy current losses that are proportional to motor velocity.
The amount of force these motors can produce is somewhat less than that of tubular designs. The coil segments are not perpendicular to the magnetic field, making the vector angle less than one. Thus, the segments parallel to the motion axis don't contribute to thrust output, all of which results in less than maximum force. But to obtain more output force from these designs would result in higher heat losses.
In addition, consistent force output depends on maintaining a maximum air gap of 0.5 mm between the forcer and platten. Otherwise the control system must compensate for any variation.
Forcer-platten motors also require active cooling to counteract electrical and thermal inefficiencies.
The U-channel motor armature, on the other hand, consists of a planar winding epoxy bonded to a plastic "blade" that projects between a double row of magnets. This design offers zero detent force with the resulting smoothness. In addition, there are no attractive forces between armature and stator.
The armature blade has low mechanical stiffness, which gives rise to resonance and tuning issues. And engineers must address heat transfer and dissipation for best operation.
The motor also has magnetic flux inefficiencies similar to those of forcerplatten motors. As with that design, coil sections that lie parallel to the thrust axis do not contribute to the motor output, and those not perpendicular to the magnetic field provide only a fraction of full capacity.
Matt Johnson is product manager, Industrial Devices Corp., in Petaluma, Calif.