The days of matching catalog motors with available power supplies and controllers may be over. The old way of handling motion control was to choose off-the-shelf devices with ratings usually exceeding actual application requirements. This was especially true for controllers available in three or four models each designed to accommodate a range of motor sizes. The result was often an overrated (in current, voltage, or both) and expensive controller driving a larger-than-needed motor.
Overspecified systems of this nature may be all right in small quantities. But in volume OEM production where price is an important consideration, a better matching between components becomes critical.
In addition, the performance demands made on today's motion-control systems frequently eliminate off-the-shelf products from contention. These demands concern high acceleration and deceleration rates, tight speed accuracies, and fine adjustment increments that are difficult to reach using collections of standard products.
Another approach is increasingly being used to meet such requirements. Rather than make do with standard parts, it is often better to design the motor and controller together as an optimized system. This technique involves constructing a motor from a few building block components and matching it with a power supply and controller that together provide the needed performance at a minimum cost.
Successful examples of this approach can be found in the retrofitting of electronic controls to mechanical production or manufacturing equipment in machine tools; filament-winding, packaging and labeling machinery; avionics controls and actuators; industrial sewing machines; and optical recording equipment.
A number of problems can make standard motion-control products suboptimum. One is that off-the-shelf motors generally come with a set performance and fixed dimensions. Shaft diameters and lengths, motor diameters, and supply voltage requirements are generally invariable.
It is often difficult to get motor manufacturers to make even slight changes in these specifications. For example, ordering a standard motor with a special rear shaft extension may result in a sharp jump in both price and lead time, assuming the vendor is willing to make the modification at all.
Some motor suppliers address the need for such special products through field modification centers located at a distributor site. These centers may offer some limited variations, usually only for brush-type motors. It is also interesting to note that the modifications these centers make do not address the overall system or the controllers. In many cases, other factors can complicate the design process, even when an off-the-shelf motor seems to provide the right performance. For example, space limitations may eliminate standard motors from consideration. Numerous applications today call for a motor built on and around the driven shaft, particularly when the system must be torsionally stiff (no belts), or have zero lost motion (no gears), or where shaft runout requirements are in millionths of an inch.
These systems are often typified by a need for high-speed accuracy (&\#177;0.001%) and are generally regulated with phase-locked servosystems where an optical tachometer feedback signal is compared to a reference frequency. The comparison generates a system error proportional to both shaft velocity and position. Typical 0-dB bandwidth is on the order of 100 to 300 rad/sec. To obtain acceptable system stability in such uses, it is necessary to minimize effects such as spring rate, or at least to place the response outside the system bandwidth.
The alternative approach of designing motors and controllers together as a system is made practical by modern motors that are basically constructed of a few common components. This allows manufacturers to make motors in a kit form that can be assembled around the drive spindle during manufacture.
The process of custom tuning all motor parameters for a given application is not as difficult as it may sound. While this philosophy suggests a generation of specials without end, there is actually considerable fallout of common parts and subassemblies when the motor, controller, and feedback devices are considered as a system.
One reason is that the brushless motors often used in high-performance motion-control applications have quite a simple construction. Motor components can be thought of as building blocks consisting of laminations and magnets. This allows motion-control suppliers to inventory a large array of basic building blocks used in developing motors for special requirements. All other components such as copper wire, bearings, motor housing parts, shaft and rotors, shaft encoders and readers are either generally available commodities or can be modified with short lead times.
In terms of controller performance, optimization generally considers factors such as the actual peak and rms torque requirements or whether a dc supply voltage is available. The need for an enclosed or open frame and special interface specifications can also be built into custom specifications.
For example, consider simple velocity control over a 500 to 5,000-rpm speed range. Many standard controllers require either a resolver or encoder to provide the feedback signal for velocity. But such feedback components might be an overspecification for this speed range. Instead, Hall-effect commutation sensors on the motor itself might provide velocity sensing that would work as well.
The motor-controller interface often can be optimized in other ways. One way is to design the BEMF constant of the motor and the dc bus voltage of the controller to yield a maximum speed that is less than 10 to 15% above the highest operating speed. This ensures that certain failure modes do not produce high runaway speeds.
Other optimization is possible in direct-drive applications where speeds are low (0 to 500 rpm) and torques high. Here, custom motor/controller combinations can be configured to get more power from smaller motors, use less energy for a given task, obtain faster acceleration and deceleration rates to improve throughput, reach higher or more precise operating speeds, meet precise positioning criteria, improve heat dissipation for cooling, and reduce maintenance or lengthen operating life.