« iepriekšējāTurpināt »
THE SCOOP ON ULTRASONIC MOTORS
I was invited to Japan by Prof. Kenji Uchino of Sophia University (and more
Group at the Massachusetts Institute of Technology (MIT) Artificial
piezoelectric ultrasonic micromotors. This is a joint project with the
what I learned of the state of the art in ultrasonic motors there.
by Anita Flynn
trip was that I was able to visit many The idea behind a piezoelectric
companies that developed these motors, ultrasonic motor is to generate a travelWhile we have been building robots met the inventors, and learned of yeting wave of bending in a metallic ring. for many years in our own group, one other newer companies that are incor- The bending wave is excited by voltages problem we face today is that while porating these types of actuators into applied to piezoelectric elements sensing and control for our robots can products. I visited Ricoh, NEC, Canon, attached to the ring, and the excitation fit in small inexpensive packages, motors and Panasonic. I also met many of the at the resonance frequency of the ring and batteries tend to be the largest, academic founders of the field at the is typically in the ultrasonic range. A most costly items. Consequently, I am conference and spent a day at Sophia traveling wave in a beam causes a pardoing my Ph.D. research on ways to University.
ticle on the surface to generate elliptic make very tiny, mass-produced robots While my research focuses on using motion. This forms the stator of the and better actuators for micromachines. ferroelectric thin films to pattern ultra- motor. A rotor or carriage pressed I am focusing on piezoelectric ultra- sonic micromotors (and minimotors) against the vibrating stator is then sonic micromotors using ferroelectric on silicon wafers, typically ultrasonic propelled along through frictional thin films of PZT (lead zirconate motors are macroscopic (1 to 3 inches coupling. titanate) in a joint project with the in diameter) and formed from bulk Of course, other types of piezoelecPennsylvania State Materials Research ceramic PZT, on the order of 200 um tric actuators have been around for Laboratory and Massachusetts Insti- thick as compared to our 0.30-um-thick years, such as multilayer stack actutute of Technology (MIT) Lincoln films, which are made at Penn State in ators and inchworm motors for preciLaboratory. That is the reason I was a sol-gel process. While ferroelectric sion X-Ystages and hydrophone transinvited to give the symposium talk at thin films have been heavily supported ducers for Navy sonar. The interesting Kanagawa Science Park on 14 December by the Defense Advanced Research thing about ultrasonic motors is that 1991.
Projects Agency (DARPA) in the United they deliver continuous macroscopic Piezoelectric ultrasonic motors are States in the past few years for non- motion at inherently high torques and fairly new but widely popular in Japan, volatile memories, our contribution has low speeds without the need for gears. while almost unheard of in the United been to recognize that we can use sol. Consequently, they can be made very States. They were invented in 1981 in gel processed films to take advantage thin and compact. They are also very Japan by Toshi Sashida and there are of their piezoelectric characteristics and quiet since there are no gears. In addinow over 90 U.S. patents (all of Japanese actually pattern motors.
tion, they have quick response time authorship). The interesting part of my
and large holding torque even when
power is removed. Efficiency has reached I was surprised to see a company move Neither Ricoh nor NEC seemed to 45% to this point.
so quickly! They didn't say specifically have a clear picture of where microWith all that as background, let me what their target application was, but motors might lead them, but it was relate what I learned of ultrasonic motors generally they were investigating the amazing to see the amount of resources in Japan in my travels. Prof. Uchino area for possible uses in printheads. applied to this realm (especially in light works in the area of piezoelectric mate
of the fact that Bell Laboratories has rials and actuators and has contact with NEC
now discontinued all activity in the area, many researchers in companies. He set
while they were the initial impetus for up meetings for meat Ricoh, NEC, and We visited NEC Central R&D much of this field). Canon. I set up the meeting at Panasonic
the meeting at Panasonic Laboratories on 10 December. NEC is as we have collaborative work with them a huge company, and over 1,600 people CANON in the "Mobot" Laboratory.
worked in that research facility. We
also had two researchers there present Our visit to Canon on 11 December RICOH
their work to us. The first was on a new was slightly different. This was not
type of ultrasonic motor they had devel- central research but rather the Lens On 9 December we visited Ricoh oped. (This work was presented at the Development Group, where they have Central Research and Development 1989 Ultrasonics Symposium in developed and productized ultrasonic (R&D) Laboratories where approxi- Montreal). It was a spark plug shaped motors as the autofocus mechanism in mately 350 people are employed. In device that combined longitudinal and their EOS series single-lens reflex (SLR) their showroom of products we saw torsional vibrations to achieve elliptic cameras. They produce 50,000 of these soon-to-be-introduced products in the motion and hence rotary output. They cameras each month and the motors areas of speech recognition boards for brought in a working prototype and are truly beautiful. Canon began work PCs, rewritable magneto-optical disks, demonstrated it for us. This motor can in 1983 and holds over one-third of the conducting polymers for electronic deliver high torque and they are work- U.S. patents on these vibration wave displays, credit card thin batteries, and ing with Olympus to investigate possi- motors. neural network chips for handwriting bilities for using it as the film-winding Ring-type ultrasonic motors fit an recognition. Then we heard talks by actuator in cameras. They have worked autofocus application nicely because two researchers in the areas of micro- on this project for 4 years. They also the motors are essentially hollow-machining and sol-gel PZT films for have another flat, thin type ultrasonic perfect for passing wires through withmicroactuators. The first talk described motor for paper pushing in a very thin out a slip ring, or even for allowing a parallelogram silicon electrostatic printer.
light to be transmitted to the film from microactuator that had been a collabo- The second talk at NEC described the lens. Since no gears are needed, rative project between Ricoh, Prof. their work in silicon micromechanics. I the motor fits in the periphery of the Fujita at the University of Tokyo, and was surprised at how many different lens barrel cylinder, and its quick visiting scientist Dr. Ken Gabriel, of types of sensors and actuators they had response time and quiet sound make AT&T Bell Laboratories and now of fabricated. They started a few years for an elegant camera. Also, it is the Naval Research Laboratory (NRL), ago, and their first project was a silicon interesting to note that one current I believe. The second talk was by a pressure sensor with piezo resistors problem with ultrasonic motors is that young Japanese woman, a chemist, who and temperature compensation cir- they tend to wear out because of was making thin film PZT using sol-gel cuitry. They had also made silicon accel- friction and material fatigue. Conseprocessing. I was surprised that it was erometers targeted for the automobile quently, an application such as fan motor so similar to our work (their goal was industry. More interestingly, they had might not be amenable, but autofocusing also actuators, as opposed to memo- made linear arrays of ultrasonic prox- pictures is just fine. ries, which is what most effort is focused imity sensors, a single-crystal, wafer- Three engineers from the Canon on), but then the head of Ricoh R&D, bonded, variable capacitance micro- group presented their work to us, gave Dr. Mario Onoe, had visited the Mobot motor, a linear comb-type micromotor; us the specification sheets on three Laboratory last year and also we had and an electrostatic wobble motor that varieties of motors, and handed out delivered papers on our work both at was fairly thick, about 30 um. Basically, copies of papers they had written. One the Ultrasonics Symposium in Hawaii they have duplicated just about every talk explained the general fundamenlast December and the MEMS (Micro type of device that has been discussed tals of the motor, one talk discussed Electro Mechanical Systems) confer- in recent MEMS conferences.
how they overcame an early problem ence in Nara, Japan, last January. Still,
with squeal noise, and the third talk their patents.
nicely described finite element analysis firm called Piezotech set up to sell the One company they have licensed to, of the rotor-stator slip-stick interac- technology
which manufactures and sells the motors, tion. It was all quite impressive and Dr. Kumada also had the informa- is Fukoko-a company that Toyota works they answered many questions I had tion on the new Toshiba electromag- closely with as a supplier of rubber for formed after taking apart one of their netic (EM) motor and the 10-mm Seiko windshield wipers. motors back home and reading through ultrasonic motor:
MATSUSHITA After the presentations, they walked Toshiba EM motor: us around their laboratories and showed
I made a trip down to Osaka on us equipment they used in developing Dimensions 3 mm diameter x 5 mm thick 16 December and visited Central R&D their motors, such as optical micro- No-load speed .. 200,000 rpm
at Matsushita Electric (Panasonic) and scopes for measuring micron-scale Stall torque
met with the ultrasonic motor group. deflections on the surface of the vibrat
At each company I visited, I gave my ing stators. Canon has 25 engineers Seiko ultrasonic motor:
talk and explained our goal of building working on ultrasonic motors. They
cheap, mass-producible small robots also gave me some catalogs for the Dimensions 10 mm diameter x 5 mm thick using ultrasonic motors. The group at smallest electromagnetic motors com- No-load speed .. 6,000 rpm
Matsushita had been working on these mercially available today--Namiki
motors for 8 years and showed me four motors, which are 7 mm in diameter.
or five working models: disk type, ring They mentioned that they had heard While at the conference, I talked to type, and linear models. They were that Seiko-Densi had made a 10-mm some young engineers at various car extremely helpful and gave me insights ultrasonic motor, but they did not know companies and found out that Toyota from their experience to help me in any more details. I asked them about now has a ring-type ultrasonic motor in designing better motors with higher the Toshiba announcement of an elec- the headrest of the new Crown Majesta. output. They also offered to collabotromagnetic motor, supposedly 1 mm The new top-of-the-line cars have over rate in the future and to answer any in diameter, which had earned a two- 60 actuators in the seats, mirrors, win- questions I might write to them about paragraph article in the New York Times dows, antennas, etc. Nissan is also
later. on 5 November, but they hadn't heard working on another, different type of Matsushita does not have any of about it. The Times article gave few ultrasonic motor. Essentially, the auto- their motors in products at the moment details.
motive companies are interested in these because they typically sell low-cost
motors for their luxury cars because consumer electronics and the ceramic SOLID STATE ACTUATOR they are compact, very thin, and quiet. materials now are too expensive. For SYMPOSIUM
On 13 December I gave my talk and instance, one motor that they had built
showed videotapes of small ultrasonic and which I saw spinning had the folThe first day of the Solid State stators patterned on silicon wafers that lowing specifications: Actuator Symposium was 12 December. spun glass lenses 1.5 mm in diameter. One speaker was Akio Kumada, who Afterwards, a young engineer from 10-mm Matsushita ultrasonic motor: holds several patents on some very Olympus that I was sitting next to said unique ultrasonic motors. He gave a that he also was working on an ultra- No-load speed talk on his latest, called a Revolving sonic micromotor. Prof. Tomikawa, from
25 g f-cm Center of Gravity Resonator, which Yamagata University, who advises 20 Drive frequency
70 kHz vibrates radially, much like a hula hoop. students in this area, said he was affected Drive voltage
60 V His vision is to build a new generation by the talk and wanted to try to build a Drive current
PP of milliwatt motors that run from similar micromotor. complementary metal oxide semicon- One person noticeably absent from Interestingly, the ceramic for this 10-mm ductor (CMOS) chips and button bat- the conference was Sashida, the origi- motor would cost $1 (in lots of 1,000) teries. He brought along an impressive nal inventor of ultrasonic motors. . and that was too large a sum for them! working prototype that fit inside a watch Apparently, many companies now I also learned much from the face and rotated via an on-stator PC commercially producing ultrasonic
ultrasonic Matsushita group about the finer points board and a 1.5-V lithium battery. His motors have licensed the technology of ultrasonic motor design, such as motor draws 15 mA. He has a small from him. He runs a small develop- criteria for the teeth, or the mechanical
ment company called Shinsei-Kogyo. amplifiers, and issues in control.
........ 900 rpm
Separately from the ultrasonic motor work, Matsushita has another research group in Tokyo that is involved in the Ministry of International Trade and Industry's (MITI) large 5-year program in microrobotics. Matsushita is teamed with Mitsubishi and Murata to build small, pipe-crawling microrobots. They will be 10 mm orso in diameter and will be formed in modules and interconnected like a train. Sensors, such as ultrasonic imagers, will be actuated so that they can turn and inspect the walls of tubes.
Finally, on 17 December, I spent the morning with Prof. Uchino at his laboratory at Sophia University. Four of his students presented their work to me on new high strain materials using antiferroelectric-ferroelectric phase switching, sputtered barium titanate thin films, fatigue analysis, and novel actuators such as “moonie” actuators, which incorporate the best features of both bimorph benders and multilayer stack actuators.
All in all, it was a productive trip and I learned many things that will be helpful in designing our next generation of piezoelectric micromotors.
Anita Flynn received the B.S. and M.S. degrees in electrical engincering from the Massachusetts Institute of Technology, Cambridge, in 1983 and 1985, respectively. Subsequent to that, she spent 5 years as a research scientist at the MIT Artificial Intelligence Laboratory in the Mobile Robotics Group working on sensing and control problems in autonomous robots. Since 1990 she has becn a Ph.D. student at the Artificial Intelligence Laboratory researching piczoclectric motors for miniature robots.
TELE-EXISTENCE WORK AT THE
UNIVERSITY OF TOKYO
Tele-existence work at S. Tachi's University of Tokyo laboratory is described.
by David K. Kahaner
• Realtime interactive computer
simulation (computer simulation, In another article (see page 29) I Tachi began our conversation by real time computer simulation, etc.) described a paper by Prof. S. Tachi on showing me a chart describing the diftele-existence. His work seemed very ferent threads of research that are now • Communication with a sensation of interesting and I decided to visit and converging to be part of what is cur- presence (telephone, teleconference, look in person.
rently called artificial (or virtual) real- etc.)
ity (AR/VR). Below I have extracted Prof. Susumu Tachi
the text and reorganized it to better fit • Tele-existence/telepresence (teleResearch Center for Advanced in the format of this report. Looking operation, telerobotics, etc.) Science and Technology
backward it is fairly obvious that these (RCAST)
projects shared many common elements, Research into some of these topics University of Tokyo
although it is doubtful if the researchers began as early as the 1960s, for exam4-6-1 Komaba
themselves thought in these terms. ple, with Ivan Sutherland's computer Meguro-ku, Tokyo 153, Japan
graphics projects. The early 1980s saw Tel: +81-3-3481-4467
Virtual console (mouse, three- rapid growth due to work by Furness, Fax: +81-3-3481-4469,-4580
dimensional (3D) mouse, virtual Kruger, Sheridan, and others. Currently E-mail: firstname.lastname@example.org.
in the United States, centers of excelac.jp
lence are at the University of • Real time interactive 3D computer Washington, MIT, and the University Prof. Tachi moved to the University graphics (computer graphics (CG), of North Carolina in the academic world; of Tokyo several years ago from the 3D CG, etc.)
the National Aeronautics and Space Mechanical Engineering Laboratory
Administration (NASA) and the Naval (MEL) in Tsukuba. MEL is run by the • Virtual products (computer-aided Ocean Systems Center (NOSC) within Agency of Industrial Science and design (CAD), 3D CAD, interac- the U.S. Government; and at several Technology (AIST), which is part of tive 3D CAD, etc.)
small companies that are marketing the Ministry of International Trade and
products. (This list is meant to be sugIndustry (MITI). About 10 years ago • Cybernetic interface (man-machine gestive, not exhaustive.) Tachi also spent a year at the human interface, etc.)
Tachi's work was also mentioned in Massachusetts Institute of Technology
my Scientific Information Bulletin arti(MIT).
• Responsive environment (3D video/ cle on virtual reality (“Virtual reality,”
holography and art, interactive video 16(4),43-45 (1991)]. That report referand art, etc.)
ences a conference held July 1991 on