Motion control helps save fish

Mark A. Gottschalk, Western Technical Editor

Seal Beach, CA--The endangered Chinook salmon, Delta smelt, and the Sacramento split-tail have a new friend--ConeScreen. Looking much like a misplaced alien spaceship from the X-Files, ConeScreen is a positive barrier designed to prevent fish from being drawn into the water-intake pumps placed in wetland areas. "Without a screen, the fish pass through the pumps and come out as fishmeal," says Rick Halstead, president of Empire Magnetics (Rohnert Park, CA), which distributes the systems. "They've found that's a little hard on them."

ConeScreens are more than just a simple passive filter. With a government requirement that the screens stay clear of debris, motorized wiper brushes sweep the face of a ConeScreen at regular intervals. The product was conceived by Intake Screens, Inc. (Clovis, CA), but Empire Magnetics developed the motion-control system that powers the brushes.

The environmental conditions under which the system must operate posed significant challenges. By nature, the screen is immersed in brackish water up to twelve feet deep. It must resist the likes of nesting birds, clogging algae, mud, logs, and significant corrosive effects. "We had an initial model where some creature found the motor cables to be delicious and gnawed through them," says Halstead. In addition, though early specifications called for 110V ac current, the final energy source is a 24V dc battery bank recharged by solar cells. Though significantly less powerful, the battery system offers greater logistical flexibility.

Engineers first selected a step motor to drive the brushes, but faced with a 24V dc power supply, they turned to a three-phase brushless dc servo with Hall-effect commutation. "The back EMF from the steppers was so high that we couldn't get any decent performance from the 24V setup," says Halstead. The servo motor makes for a simpler design too. "By going with the brushless motor we use straight 24V dc, no inverters, no conversion as was needed with the steppers," he says. "Lots of pieces were no longer needed."

The motor is mated to a Sumitomo cycloidal gearbox with a 1,225:1 ratio. With a motor torque of 84 oz-in at 1,800 rpm, the unit produces 5,562 in-lbs of torque at 5A. Rotation speed of the brushes is about 1.6 rpm.

Corrosion effects proved so great that, on one early model, corrosion ate through nickel plating and 1/4 inch of cast iron in less than 90 days. Lesson learned, engineers changed all exposed metal to stainless steel. The gearbox output shaft was turned down and a stainless-steel sleeve put over it. Stainless braid covers all cables. And a pressurized, oil-filled housing surrounds the drive assembly. It includes a compensating bladder which accommodates oil-volume changes due to temperature. Engineers fill most of the housing with rocks to reduce the amount of oil needed as well as the size of the compensating bladder.

ConeScreens are offered in 6-ft thru 14-ft diameters, and each installation costs about $50,000. It is currently the only design that meets California Department of Fish & Game criteria. "Their large surface area and self cleaning keep the water flow-rate down," says Halstead. "Little minnow-size fish can just swim away from it."

Servo positions 1,000-ton building, improves telescope tracking

Amado, AZ--How do you control the rotation of a 1,000-ton building that houses one of the world's most powerful telescopes? "Pretty carefully," says Fred Whipple Observatory's senior staff engineer Karl Harrar. Harrar, charged with upgrading the original, failing SCR drive atop Mount Hopkins, turned to Westwood, MA-based Copley Control Corp.

"After a wide search, Copley was the only company that produced a DC amplifier that could drive shunt motors with a PWM signal of more than 30 kHz," explains Harrar. "We didn't want any chance of hearing this amp. A lot of companies couldn't produce enough power either. We needed at least plus or minus 55A continuous at plus or minus 240 volts."

The observatory's design houses the entire telescope within a rotating building to minimize building size and cut construction costs. The penalty for these benefits is that the building requires a motion system to rotate the building in synch with the telescope.

Although rotating on independent supports, the building is slaved electronically to the telescope. An LVDT sensor provides the required feedback to achieve a 0.001 degree positioning resolution.

Four 3-ft diameter wheels that run on a steel, 57-ft diameter track support the telescope's co-rotating building. Two of the wheels, powered by 15 hp, 240V dc shunt motors, provide about 500,000 lb-ft of torque to each drive wheel. "Copley's wideband DC amplifiers not only operate at high switching frequency, they benefit from carrier cancellation modulation," says Copley engineer Jim Woodward.

Copley's Model 264P contributes to Harrar's success for four reasons. It doubles the ripple frequency, produces zero ripple at zero output power, provides zero-crossing linearity, and delivers high current in short circuits while drawing near-zero power from the DC supply.

"When tracking stars, building motion is very smooth and linear. I've seen a great reduction in tracking errors since we changed to the wider bandwidth PWM Copley amplifiers," says Harrar.

Simplifying control cuts development time

Richmond, CA--In the fast-paced semiconductor-manufacturing industry, getting to market on time is vital for success. Berkeley Process Control's MachineWorks helped SpeedFam--a Chandler, AZ-based supplier of high-throughput precision surface-processing systems--improve the process yield, tool reliability, and productivity of a new chemical/mechanical planarization (CMP) machine known as Auriga. Major achievements include:

SpeedFam's CMP machine, Auriga, is a 22-axis semiconductor-wafer polisher. An automated load station maneuvers dry wafers onto a rotating, 55-station, servo-controlled index table. The polish station uses a gantry mechanism to transfer unplanarized wafers onto the primary polishing table. Finally, an unload station transfers the wafers into output cassettes.

Three networked MachineWorks controllers command the load, polish, and unload stations with distributed control and I/O. MachineWork's distributed architecture not only reduces the number of physical interconnects, its on-board integration environment lets engineers code and debug individual stations in parallel, then bring them on-line together in multitasking operational mode.

The Auriga team simply transferred existing machine control sequences into MachineWorks, reducing 150,000 lines of C++ code from a similar project to 10,000 steps. Less code means shorter development time and higher reliability. The controller comes with virtually all control hardware and software built-in, including real-time command of all axis motion, I/O, operator interface, and networking.

In addition to MachineWorks, Berkeley supplied all servo amplifiers, vector drives, PLC-like I/O, and touchscreen operator interfaces. Cooperation between SpeedFam and Berkeley will continue on the next-generation CMP tool.

TECHNOLOGY FACE OFF

Miniature motor mania

Two titans of tiny electric motors face off in a discussion about product, industry, and application trends

Mark A. Gottschalk, Western Technical Editor

Earlier this year RMB Miniature Bearings introduced the first production electric motors just millimeters in diameter. Meanwhile, the engineers from Micro Mo say they have designed an even smaller motor, just 1.9-mm in diameter. To get a take on the present and future of such miniature motors, Design News tracked down RMB's product manager for the Smoovy^TM line of miniature motors, Albert Birkicht; and Micro Mo's VP Advanced Research and Planning, Stephen O'Neil.

DN: What's driving the need for miniature motors?

Birkicht: Actually, I think it's more a question of what's driving the need for miniaturization in general. Miniature motors are just one element in the drive for miniaturization in all kinds of products. If you want to make something smaller then you have to look at all the elements of the product and make them smaller. One very important part of many of these products is certainly motors. What's driving the general trend is difficult to say, but I guess it is less energy consumption. In space flight--one application we are doing--every gram of weight is very important. In medical applications, all the minimally invasive methods are pushing size down to try to do less harm to the patient. O'Neil: The real need is meeting designers' demands for very small devices, and providing sufficient levels of power to do real work. Many prototypes that I've seen over the years, for example electrostatic motors, require huge amounts of power to operate, are very expensive, and cannot be consistently reproduced. Even in the area of larger than 2-mm diameter, the Japanese produced a 2.8-mm step motor earlier this decade, but it wasn't commercially viable. Japanese companies have developed 4-mm pager motors, but the increase in cost over their 6-mm counterparts translates into minimal sales. In more sophisticated applications areas, actuators are less price sensitive, but in order to be commercially successful, they have to be affordable in the context of the end product. Then they will be "needed."

DN: What distinguishes your motor's design from the competition?

Birkicht: Good question, because I don't know very well the motor design from the competition. But I can tell you what our design parameters were when we designed it. When we started we had what we call in German a "green field". We had no motor technology in house at the time and we could look across all kinds of motor technologies to come up with what we had in mind--which was a miniature motor that was manufacturable. Our number one design parameter was manufacturability. When we took a closer look, it made sense to make it as simple as possible and not do any intricate design features. What we ended up with is a motor principle that is ancient, almost. The specialty we brought in was how to miniaturize it, and to miniaturize it in a way that was manufacturable.

O'Neil: Micro Mo is part of the Faulhaber Group, which has been in small motor development for almost 50 years. All of the high-power micromotors today are based on System Faulhaber(R) technology, which first introduced the concept of an ironless-core motor to the world. Our group has continued to refine the technology, optimize the manufacturing technology needed to produce it, and go on to develop new materials and assembly techniques to bring reliable product to market. Expertise in proprietary lubricants, magnet systems, and machining techniques really differentiate the technology. Our 1.9-mm products are the logical outcome of our Group's emphasis on producing motors with the highest power in the smallest size possible. The System Faulhaber coil we use affords minimal current draw, no cogging, high dynamic performance, and light weight.

DN: Why develop a miniature brushless DC motor versus a brush or perhaps step motor?

Birkicht: Versus brush motor, I have a standard list of advantages and disadvantages of brush versus brushless that I could read to you. I think the most important one is that brushless motors of any size are just much more versatile. You can do everything you can do with a brush motor with a brushless motor. There is also the increasing trend towards electronics. So why would you need a mechanical commutation system when you already have electronics in the application? Brushless commutation is really quite simple, and in most cases it does not add even a single cent to the cost of the application. Now comparing brushless motors and step motors, most people don't see that they are basically the same technology. The difference is more from the applications they go into.

O'Neil: Brush motors are easy to make perhaps down to the 3-mm or 4-mm range. However, the brush and commutator systems encounter real technical problems at less than 2-mm. Brush systems are the least durable because brushes wear out and must be lubricated, so this means that there may be reliability issues. Brushes also take up room and generate particulate matter which is undesirable in some applications. Step motors are a type of brushless motor that's driven by a pulse fed to the motor. I can't think of any theoretical reason why ultraminiature motors could not be made using this technology. But they are incapable of servo performance, can't reliably achieve high speeds, and the internal geometry is relatively complex, therefore, the construction is much more difficult at the micro level.

DN: Is the race to make smaller motors simply for bragging rights, or can these motors do useful work?

Birkicht: They certainly can do useful work. We have a lot of motors in all kinds of applications. If RMB had wanted to do something that brought bragging rights we would have done smaller bearings, not small motors--after all, we were a bearing company when we began this. We did it because we saw an interesting new field for our skills, which are miniaturization and micro-assembly.

O'Neil: Our goal was to develop a product with commercial potential. As for useful work, we think that one of the novel aspects of the Faulhaber Group's work relates to its expertise in precision gearing. Without having an interface to the outside world, no motor can do useful work on its own. And given the inherently small torques seen in this type of product, we felt that having a good gearhead was important to overcome this problem. This is why the 1.9-mm diameter LIGA-based gearheads were designed and developed to mate with the 1.9-mm micromotors.

DN: What do you see as the most important applications?

Birkicht:Optics and medical technology. Medical technology because of the trend towards minimally invasive surgery, and optics because to deflect a laser beam you don't need any torque to turn a mirror.

O'Neil: Some of the applications that we are working on now involve minimally-invasive diagnostic and surgical procedures, analytical instruments, data storage equipment, laser and optical system movement, and the like. These are products that will be commercialized in the near term.

DN: How small can you go? What are the limiting factors?

Birkicht: The limiting factor is torque. How small can you go? With conventional brushless DC motors I don't think you can go much smaller than the motors that Micro Mo is just now doing and still have useful torque. Below that, you are into MEMS applications. You'd get very, very little torque, but there might be some applications where you combine millions of them to do something useful. What is the limit? I suppose you could build motors with dimensions in the micrometers and still turn. Can they do useful work? That is open to discussion.

O'Neil: We are quite comfortable now at the 1.9-mm diameter level. We have done smaller designs--down to 1 mm, which is probably the threshold for this particular type of technology--but the gains you achieve in reducing motor diameter by doing this are more than offset by the power losses and inefficiencies you see. The next generation of ultraminiature motors will probably use more non-traditional materials and technologies, such as those seen in the semiconductor manufacturing arena. We have done considerable research in that area, but at this time products made using these techniques do not have much torque and are difficult to power and interface with the outside world.

Products to watch

Motion in print: Guideway catalog

Schneeberger Inc. offers a 100-page catalog for its MR series Monorail and MMS guideway with rollers. In addition to basic dimensional specs, this catalog offers design and installation guidelines, mounting instructions, and information on the handling of rail and carriages.
Schneeberger, Inc., 11 DeAngelo Dr., Bedford, MA 01730.

Increase table speeds, reduce settle

Developed for high-speed, short-stroke applications such as wire-bonding, the NXT table uses a precision dual-guideway bearing and a linear-motor drive to improve dynamic performance. Reduced-preload stresses result in a 50% lighter moving table compared to conventional designs. Using a 2-axis, grid-measuring system instead of two separate linear encoders completely eliminates orthogonal-positioning errors. Measuring 55-mm high and 150-mm square, the table's maximum X and Y stroke is 67 mm. Powerful enough to move displacement loads of 1.8 kg in the X and 1 kg in the Y direction, the NXT maintains an overall-displacement force of 1N.
Schneeberger, Inc., 11 DeAngelo Dr., Bedford, MA 01730.

Cut size and weight, maintain precision

These mini-ac servomotors use tiny encoders to provide feedback for sinusoidal commutation and achieve 2,048 ppr resolution. The SGMM-A1C31 delivers 10W of power from a 25- x 25- x 54-mm motor. Model SGMM-A2C31 is 10 mm longer and provides twice the power. Both are available with gear and/or brake options, and matched SGDA amplifiers provide 4.50 oz-in and 9.02 oz-in of torque respectively. The optional planet-cogwheel gearbox comes in 5:1, 16:1, and 25:1 gear ratios. A self-lubricating design provides low backlash; 10 arc minutes standard. Brake holding torque is 150 to 200% of rated-motor torque. Maximum speed is 5,000 rpm, and a molded-resin stator improves thermal efficiency.
Yaskowa Electric America, Inc., 2942 MacArthur Blvd., Northbrook, IL 55438; FAX (847) 291-3457.

Improve variable-speed performance

Designed to replace dc motors in variable-speed applications, new ac-induction motors combine ac-motor reliability with the precise speed control of an industry-standard three-phase ac inverter drive. Available in 31, 34, 39, and 44 frame sizes, the motors operate with drive voltages up to 230V ac. The design's maximum speed of 10,000 rpm, can increases flow rates and system throughput in pump drive and fan applications. Power output capability up to 3/4 hp can increase load carrying capacity, and special electrical insulation protects against inductive voltage spikes associated with inverter spikes.
Eastern Air Devices, Inc., 1 Progress Dr., Dover, NH 03820; FAX (603) 742-9080.

One small step for mankind

The PMU Series five-phase hybrid step motor is said to be the world's smallest, and it's matched with the world's smallest 115V ac driver. The PMU33AH motor (0.72 degree full step) produces a constant 3.3 oz-inches of torque over a speed range of 100 pps to 40,000 pps (4,800 rpm). It measures just 1.10-inches square by 1.22-inches long, occupying just 1.10³. It's matched with a driver measuring 1.38- x 4.34- x 5.32-in³. The 115V, 60 Hz ac input is converted into 4-phase or 5-phase Pentagon excitation scheme for full-step and half-step operation. It incorporates a special low- vibration circuit for quiet operation.
Oriental Motor USA, 2580 W. 237th St., Torrance, CA 90505; FAX (310) 325-1076.

Rodless actuators handle high loads

Newly designed rodless linear actuator handles higher load, higher power, and higher speeds. Called the R Series, they feature ready-to-mount motor and actuator systems in travel lengths up to 108 inches, speeds up to 120 inches/sec, and capable of handling thrust loads of up to 1,200 lbs. Three models are available, R2, R3, and R4. The smallest, R2, has a 2- x 2-inch crossection, whereas the largest, the R4, has a 4- x 5-inch crossection. All can be had with choice of acme screw, ball screw, or high-speed steel-reinforced belt drive.
Industrial Devices Corp., 64 Digital Drive, Novato, CA 94949; FAX (415) 883-2094.

Controller improves process consistency

When Joe Jackson, electrical controls designer at Smith Fiberglass Products Inc. (Little Rock, AR) wanted to improve process consistency on a strip-laying machine, he replaced the existing drive with an Aximaster 9000D servo controller from Cleveland Motion Control (CMC). The new controller maintains the exact ratio of movement between a carriage and a rotating mandrel to lay down a strip of fiber material at a specific angle. The controller interfaces the machine's PLC with an OPTO-22 I/O interface. CMC spec'd a new master motor (with encoder output) for the Aximaster to follow. "CMC helped us most during start-up, and in initial equipment specification," says Jackson.

Programmable limit switch cuts costs, enhances performance

Replacing a stand-alone programmable limit switch (PLS) with new PLS capability from Control Technology Corp. (Hopkinton, MA) could save you a bundle. In fact, Amplas (Green Bay, WI) saved about $1,200 when they used Control Tech.'s new PLS capability on their machines. Amplas invested the savings back into the product with an upgrade to touchscreen operator interfaces. The PLS function is available on Control Tech.'s MultiPro and 2000 Series Automation Controllers, and allows state synchronization of digital outputs based upon a servo axis position.