AU620135B2 - Creación de Inventos y Patentes con Inteligencia Artificial

AU620135B2 – Magnetostrictive torque sensor
– Google Patents

AU620135B2 – Magnetostrictive torque sensor
– Google Patents
Magnetostrictive torque sensor

Download PDF
Info

Publication number
AU620135B2

AU620135B2
AU40355/89A
AU4035589A
AU620135B2
AU 620135 B2
AU620135 B2
AU 620135B2
AU 40355/89 A
AU40355/89 A
AU 40355/89A
AU 4035589 A
AU4035589 A
AU 4035589A
AU 620135 B2
AU620135 B2
AU 620135B2
Authority
AU
Australia
Prior art keywords
signal
torque
primary
shaft
coil
Prior art date
1988-07-21
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)

Ceased

Application number
AU40355/89A
Other versions

AU4035589A
(en

Inventor
Robert D. Klauber
Erik B. Vigmostad
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)

Sensortech LP

Original Assignee
Sensortech LP
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
1988-07-21
Filing date
1989-07-19
Publication date
1992-02-13

1989-07-19
Application filed by Sensortech LP
filed
Critical
Sensortech LP

1990-02-19
Publication of AU4035589A
publication
Critical
patent/AU4035589A/en

1992-02-13
Application granted
granted
Critical

1992-02-13
Publication of AU620135B2
publication
Critical
patent/AU620135B2/en

2009-07-19
Anticipated expiration
legal-status
Critical

Status
Ceased
legal-status
Critical
Current

Links

Espacenet

Global Dossier

Discuss

Classifications

G—PHYSICS

G01—MEASURING; TESTING

G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE

G01L3/00—Measuring torque, work, mechanical power, or mechanical efficiency, in general

G01L3/02—Rotary-transmission dynamometers

G01L3/04—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft

G01L3/10—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft involving electric or magnetic means for indicating

G01L3/101—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft involving electric or magnetic means for indicating involving magnetic or electromagnetic means

G01L3/102—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft involving electric or magnetic means for indicating involving magnetic or electromagnetic means involving magnetostrictive means

G—PHYSICS

G01—MEASURING; TESTING

G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE

G01L3/00—Measuring torque, work, mechanical power, or mechanical efficiency, in general

G01L3/02—Rotary-transmission dynamometers

G01L3/04—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft

G01L3/10—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft involving electric or magnetic means for indicating

G01L3/105—Rotary-transmission dynamometers wherein the torque-transmitting element comprises a torsionally-flexible shaft involving electric or magnetic means for indicating involving magnetic or electromagnetic means involving inductive means

Description

_i 1 1 1 OPI DATE 19/02/90 P1 AOJP DATE 29/03/90 APPLN. ID 40355 89 PCT NUMBER PCT/US89/03118 INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) (51) International Patent Classification 5 (11) International Publication Number: WO 90/01152 G01L 3/10 Al (43) International Publication Date: 8 February 1990 (08.02.90) (21) International Application Number: PCT/US89/03118 (81) Designated States: AT (European patent), AU, BE (European patent), BR, CH (European patent), DE (European (22) Internatioial! Filing Date: 19 July 1989 (19.07.89) patent), FR (European patent), GB (European patent), IT (European patent), JP, KP, KR, LU (European patent), NL (European patent), SE (European patent), SU.
Priority data: 222,838 21 July 1988 (21.07.88) US Published With international search report.
(71) Applicant: SENSORTECH, L.P. [US/US]; 1608 North Fourth Street, Fairfield, IA 52556 (US).
(72) Inventors: KLAUBER, Robert, D. 1100 University Manor Drive, .38B, Fairfield, IA 52556 VIGMOSTAD, Erik, B. 605 South Maple Street, Fairfield, IA 52556
(US).
(74) Agents: McCOY, Michael, D. et al.; Bell, Seltzer, Park Gibson, P.O. Drawer 34009, Charlotte, NC 28234 (US).
(54) Title: MAGNETOSTRICTIVE TORQUE SENSOR S68B 66B
VA~
VA.: (57) Abstract A noncontacting method for sensing torque based on the principle of magnetostriction which induces a primary magnetic flux in a shaft (20) and obtains a torque dependent secondary signal which is a function of a secondary flux is characterized by obtaining an auxiliary signal via such means as an auxiliary core/coil (48/50) and using that signal with appropriate electric circuitry to eliminate non-torque induced variations in the secondary signal. The first embodiment (Figures 5A and 5B) uses the auxiliary signal to maintain the primary magnetic flux at effectively constant amplitude. The second embodiment (Figures 6A and 6B) divides the secondary signal by the auxiliary signal. A third embodiment (Figure 7) employs a plurality of sensors (66A/ 66B) strategically located around the shaft (20) to eliminate spurious signals which are due to bending stress and shaft misalignment.
SWO 90/01152 PCT/US89/03118 MAGNETOSTRICTIVE TORQUE SENSOR BACKGROUND OF THE INVENTION Field of the Invention This invention relates to a torque sensor based on the principle of magnetostriction, and more particularly, to an improved magnetostrictive torque sensor which is simpler, more accurate, and more economical than state of the art sensors as well as more suitable to mass production and usage.
Description of Prior Art Engineers and scientists have sought a simple, reliable, accurate means for measuring torque in rotating shafts for well over a century.
Applications for such a torque measuring apparatus include diagnosis, prognosis, and load level monitoring of a vast number of different types of rotary drive mechanisms such as automotive, ship, and plane engines; motors and generators of all types; oil drilling rigs; rotating machining tools; all electric power steering; robotics; and much more.
Further, measurement of mechanical power produced by an engine (or. used by a generator) cannot be made without knowing both torque and rotational speed of the shaft. Hence there has heretofore been no ready means to determine on-line power and efficiency of rotary drive devices simply, accurately, and reliably. This has proven to be *A Y L i: i 4f: 4S -F i l ir- WO 90/01152 PC/US89/0311t8 -2problematic in many areas of modern technology, but it has been particularly troublesome in attempts to develop modern automotive engine control systems which would improve fuel efficiency and optimize engine performance.
Heretofore several methods have been developed for measuring torque in rotating shafts (see below), but none has been ideal. That is, no single presently known method offers all of the following desirable properties.
1. Contact free (no slip rings, etc.) 2. Reliable (low failure rate) 3. Accurate 4. Small and unobtrusive (requiring little shaft/engine re-work) Inexpensive 6. Applicable at high as well as low speeds 7. Instantaneous torque measurement (not merely mean torque over several revolutions) 8. Amenable to mass production (not restricted to special test apparatus) There are presently only four distinct methods for measuring torque directly in a rotating shaft. They are: 1. Twist angle of shaft measurement 2. Strain gauge sensor 3. Reaction force measurement 4. Magnetostrictive sensors The twist angle method involves measurement of the angle of twist of a shaft and correlates this, using the material and dimensional characteristics of the shaft, to torque. It entails a complicated and cumbersome mechanism with low sensitivity, calibration difficulties, and the necessity of using two different locations along the j i.
1: i ii -i 8 iB i i i~:Il i I i gi~tl s
I
WO90/01152 PCT/US89/03118 -3shaft. It invariably entails extensive engine modification, a costly endeavor.
The strain gauge approach requires bonding of strain gauges to the shaft surface and relating strain measurement to torque. It is limited to low speed, is not amenable to mass production, lacks durability, and needs some means such as slip rings and brushes to bring the signal off of the shaft.
Reaction force measurement utilizes Newton’s second law for rotational motion to relate force and motion of the engine mounts to shaft torque. The method must employ a large structure, has low sensitivity, is not feasible for production runs, and measures driveline, not engine, torque.
Magnetostrictive torque sensors take advantage of the magnetostrictive property of ferromagnetic materials whereby tension stress increases (and compressive stress decreases) a given magnetic induction field the field) carried by the material. A coil of wire of arbitrary number of turns wrapped around an iron core is placed close to the shaft and an electric current passing through the wire causes a magnetic field to be induced in the rotating shaft. In magnetostrictive sensor designs such as those described in U.S. Patents 2,912,642 and 4,589,290 a second coil of arbitrary number of turns wrapped around a second iron core is then placed close to the shaft and used to measure the change in the induction (the B field) which results from the increased surface stress.caused by the applied torque.
The magnetostrictive method has several advantages over the other three methods, including 1) Non-contact: no slip rings 2) Not restricted to low speeds I 3) Measures torque of engine directly I a -4- 4) High sensitivity i Economical 6) Simple structure: no strain gauges, no large apparata 7) Only one location anywhere on shaft: little engine rework 8) Durable and reliable: no moving parts to cause mechanical failure, resistant to high pressure and temperature of engine environment 9) Readily miniaturized: can be made unobtrusive However, magnetostrictive torque sensors have heretofore been plagued with several major problems which have prevented them from becoming the standard in the field. These are 1) Output signal varies with RPM even at constant torque.
2) Output signal varies with temperature.
3) Spurious signal variation within one mechanical cycle (one revolution of shaft) prohibits accurate instantaneous measurement of torque: only average values over several shaft revolutions are possible.
4) Correction methodologies such as those described in U.S. Patents 4,589,290 and 4,697,459 and SAE paper #870472 heretofore employed for problems 1) to 3) above have not been able to reduce inaccuracies to an acceptable level.
All such correction methodologies developed to date involve complicated and extensive electronic circuitry and/or additional sensors for i temperature and RPM.
6) Additionally, all such correction methodologies utilized to date are affected by subtle individual shaft material and property variations such as residual stress, slight inhomogeneity in shaft magnetic properties, shaft Li i WO90/01152 PCT/US89/03118 i tolerances/misalignment, and shaft bending stress. i Hence such methodologies must be tailored specifically for each individual shaft and therefore are not suitable for mass production.
7) Furthermore, such correction methodologies have to be re-calibrated repeatedly over the lifetime of the shaft, since residual stress values, tolerances, misalignments, bending stresses, and even magnetic property inhomogeneities change over time (particularly in high temperature environments such as those of automobile engines.) Recalibration for mechanisms such as automobile engines is so difficult as to render such correction methodologies impractical.
OBJECTS AND ADVANTAGES OF THIS INVENTION Accordingly, several general objects and advantages of this magnetostrictive torque sensor invention over prior art are considerable improvement in accuracy to well within acceptable levels, simplicity of design, lowered cost, suitability for mass production, and practicality for continued usage. Such general objects and advantages are achieved by the following specific objects and advantages.
1) Elimination of signal dependence on shaft rotation speed. 2) Elimination of signal dependence on I temperature. I 3) Elimination of signal variations within a single shaft revolution due to shaft magnetic l| property inhomogeneities and residual stresses. i 4) Elimination of signal variations due to bending stress, shaft misalignment, and tolerance variations.
5) Instantaneous measurement of torque i resulting from advantages 3) and 4) above.
L
2 44 WO 90/01152 PCT/US89/03118 -6- 6) Elimination of signal dependence on individual shaft properties and hence suitability for mass production.
7) Elimination of need to recalibrate sensing device during course of shaft lifetime.
8) Simple, effective signal processing circuitry and sensor orientation permitting advantages 1) through 7) above.
Further objects and advantages of the invention will become apparent from a consideration of the drawings and ensuing description of it.
SUMMARY OF THE INVENTION This magnetostrictive torque sensor invention solves essentially all of the problems associated with prior art through heretofore unrealized insight into the fundamental electromechanical principles underlying previous experimental results and by using that insight to devise a superior design solution.
In prior art a coil of wire with current flowing through it induces a primary magnetic induction field (B field) in a ferromagnetic shaft which is located close to the coil. If the shaft is not moving, a constant amplitude AC current in the wire will produce a constant amplitude magnetic flux in the shaft where the flux amplitude depends on the magnetic permeability of the shaft material. As torque is applied to the shaft, the shaft experiences torsional stress as a result of the applied torque. Through the phenomenon of magnetostriction, a secondary magnetic induction field component arises at an angle to the primary magnetic field direction and the magnitude of that second magnetic field flux is dependent on the amount of torque applied to the shaft.. The secondary flux is measured via Faraday’s law by placing a second coil close to the shaft aligned in WO 90/01152 PCT/US89/03118
I
the direction of the secondary magnetic field and then measuring the voltage across the second coil.
The voltage across the second coil is therefore a direct indicator of torque. The procedure is accurate if the primary field flux can be maintained constant.
If the shaft is turning, however, and if the shaft has local variations in permeability around its circumference (true of virtually all shafts), then the flux amplitude in the shaft is not constant and varies as the shaft turns.
A different shaft, with different permeability, will also result in different primary magnetic flux, yielding a different voltage across the secondary coil, and hence prohibiting use of the sensor on a mass production scale.
This problem is solved in the present invention by using an auxiliary coil a third coil) to measure the primary magnetic flux via Faraday’s law. Feedback of the signal from the auxiliary coil to the power supply for the primary coil is done in such a manner as to maintain the auxiliary coil voltage amplitude constant, thereby maintaining the primary flux amplitude constant regardless of changes in shaft permeability due to rotation of a single shaft or use of different shafts. Hence the torque indicating output signal J from the second voltage will be independent of shaft speed or shaft material composition. 51 A second embodiment of the invention divides the secondary coil voltage by the auxiliary coil voltage to obtain a signal which has a similar integrity. The second coil voltage varies with RPM, I| shaft material, and torque whereas the auxiliary i;| coil voltage varies with RPM and shaft material, but I not torque. Hence the resultant signal after w 1 1 1 1 1 1 WO 90/01152 PCU/US89/03i 18 division is free of spurious signal components and is an accurate indication of torque.
Embodiment three employs two or more magnetostrictive torque sensors of any of embodiments one through four by placing the sensors strategically around the shaft such that a resultant signal produced by combining the signals of the individual sensors is free of erroneous signal components resulting from bending or misalignment of the shaft.
The invention, in its various embodiments, solves each of the problems associated with prior art discussed in the «Description of Prior Art» section in a superior and wholly satisfactory manner.
BRIEF DESCRIPTION OF THE DRAWINGS Drawing~s Illustrating Prior Art Figs. 1A and lB depict the behavior of an induced magnetic field B in a shaft undergoing applied torque. Fig. 1A depicts the zero torque stae;Fig. lB the state with applied torque.
Figs. 2A and 2B show the standard prior art configuration magnetostrictive torque sensor.
Fig. 2A depicts the physical arrangement of the primary and secondary coil/cores; Fig. 2B depicts the signal processor circuit block diagram.
Drawings Illustrating Problems and Limitations of Prior Art Fig. 3 is a reproduction from SAE paper #87,0472 showing the signal variation in output signal with constant torque over one mechanical cycle which is typical of prior art configurations.
Fig. 4 is also a reproduction from SAE paper #870472 and shows the variation in output signal with, shaft. rotation speed which is found in prior art configurations.
i I WO 90/01152 PC’/US89/03118 -9- Drawings Illustrating the Present Invention Fig. 5A depicts the signal processor block diagram representing one variation of embodiment number one of the present invention. Fig. 5B shows a second variation of embodiment number one.
Fig. 6A depicts the signal processor block diagram representing one variation of embodiment number two of the present invention. Fig. 6B shows a second variation of embodiment number two.
Fig. 7 depicts embodiment number three of the present invention, comprising a multiplicity of magnetostrictive torque sensors strategically located such that the sum of the voltages produced by each sensor is a signal which is effectively free of misalignment and bending stress induced signals.
DRAWING REFERENCE NUMBERS 0o rotating shaft primary core 32 primary coil 34 secondary (pickup) core 36 secondary (pickup) coil 38 voltmeter measuring output of secondary coil oscillator 42 power amplifier 44 amp meter 46 low pass filter 48 auxiliary coil (on primary core) auxiliary coil voltmeter 52 auxiliary coil (on primary core) for embodiment number two 54 auxiliary coil voltmeter for embodiment number two 56 signal divider 66A magnetostrictive sensor A 66B magnetostrictive sensor B, on opposite side of shaft L i r:i ii i-I I L I l- :ili~L–XIIL~I-~-lil~~ I- i ~I WO 90/01152 PCT/US89/03118 1 0- I: 68A air gap between sensor A (item 66A) and shaft (item 68B air gap between sensor B (item 66B) and shaft (item 70c bending compressive stress bending tensile stress PRINCIPLES AND ANALYSIS OF MAGNETOSTRICTIVE TORQUE SENSORS Conceptual Overview Figures 1A and lB show the effect on an induced magnetic field (also called induction) B in a shaft 20 when torque is applied to the shaft Figure 1A depicts the shaft 20 with zero torque applied and the induction B directed circumferentially. The induction B can also be represented in terms of two components B x and By at angles to the circumferential direction.
Figure 1B illustrates the change in the induction B when torque is applied. The torque creates tensile stress a t in the direction of the induction component Bx, and compressive stress a, in the direction of the induction component Due to the principle of magnetostriction, induction component B x increases in magnitude to B x; and induction component By decreases in magnitude to B’.
The induction components can then be added vectorially to obtain the induction with torque applied The induction B’ makes an angle 6 with respect to the original induction B, and can then be represented in terms of two components B cire and B axial directed along the circumferential and axial directions respectively. Therefore, B’cir B’cos6 B ial B’sinS.
WO 90/01152 PCT/US89/03118 WO.90/01152 PCr/US89/03118 S-11- I The angle S is in practice quite small, hence B’crc B B B ‘aia; B B6.
The magnitude of B’lia, increases with increasing applied torque and equals zero when torque equals zero.
Figures 2A and 2B depict the standard prior art configuration for inducing the magnetic induction B and for measuring the induced axial induction Blia. Figure 2A shows the primary coil 32 carrying current, typically AC, wrapped around the primary ferromagnetic U shaped core 30 in which the core 30 is aligned circumferentially to shaft 20 and in which the ends of the core 30 are shaped and placed such that a constant width air gap exists between the core 30 and the shaft 20. The current in primary coil 32 causes a magnetic flux to be carried through primary core 30 across the air gaps and along the surface of shaft 20 in a manner similar to that depicted in Figures 1A and IB.
Secondary coil 36 is typically open circuit and is wound around secondary U shaped core 34 which is aligned in the axial direction of shaft and has ends shaped and oriented such that constant width air gaps exist between the secondary core 34 and the shaft 20. Application of torque to the shaft when current is passing through the primary coil 32 causes axial induction B’l~ i to arise and magnetic flux to pass through secondary core 34.
Voltage is then produced in secondary coil 36 via I Faraday’s law and measured by voltmeter 38. This voltage is zero for zero applied torque and I increases as torque increases; hence, the voltage is a direct measure of applied torque.
*i ji I r :r 1 :l l field flux and which is a function of non-torque induced variations in magnetic permeability in said torque transmitting element and /2 -t i| WO 90/01152 PCT/US89/031i8 -12- Limitations of Idealized Approach Figure 3 is a reproduction from SAE paper #870472 showing the spurious output signal variation measured by voltmeter 38 (see Figure 2B) within each single revolution (mechanical cycle) of shaft This phenomenon is attributed to local variations in magnetic permeability and residual stress around the shaft circumference. The shaft material is not completely homogeneous either in its magnetic properties or in degree and distribution of residual stresses. As the shaft rotates, these non-uniformities pass under the sensor and alter the amount of flux transmitted through the sensor coils.
The result is a non-torque dependent variation in the output voltage.
A second cause for spurious subcycle signal alteration is variation in air gap thickness between sensor and shaft due to i) shaft misalignment and/or ii) bending stresses in the shaft. Since the flux is also dependent on this gap thickness, if the gap thickness varies then so will the output voltage.
The standard solution to this problem has been to use a low pass filter to eliminate all signal variations with frequencies greater than that of one shaft revolution (one mechanical cycle.) Unfortunately, this introduces a time constant into the device which limits the measurement bandwidth to approximately 1/2 the mechanical frequency and precludes measurement of torque changes more rapid than those occurring over several mechanical cycles.
The resulting signal becomes an average rather than an instantaneous measure of torque and is hence limited to steady state or slow transienti operations. J Figure 4 is also a reproduction from SAE paper #870472 and illustrates a second problem i i i II 1
‘A
WO 90/01152 PCT/US89/03118 i -13encountered in prior art the variation in output i o signal with RPM. Increasing the speed of rotation of the shaft 20 while torque remains constant increases the output signal. This is undesirable since the ideal signal should reflect change in torque alone. The common solution approach heretofore employed has been to monitor RPM and introduce a shaft speed dependent correction to the output signal detected by voltmeter 38. This has not been completely successful, eliminating some but not all of the error. Further, it has complicated the device by introducing additional measurement and apparata.
A third problem in prior art is temperature dependence. As the shaft temperature varies (a common phenomenon in automobile and many other engines), so does the output signal even as all other parameters remain constant. This is no doubt due in large part to i) temperature induced variations in magnetic permeabilities of the 11 shaft/sensor materials, ii) concomitant modification in tolerances which affect the air gap dimension between sensor and shaft, and iii) changes in resistivity of the wires used in the sensor. The typical prior art solution is to employ feedback from a thermocouple to further correct the output signal. As reported in SAE paper #870472 this also has had only some but not full success. This is presumably because different temperature gradation S fields within the engine would affect the sensor/shaft differently yet could have the same .temperature value for any given single point the point at which the thermocouple might be located.) Signal instability or drift over time is yet a further problem. During testing, the SAE flux in a shaft (20) and obtains a torque dependent secondary signal which is a function of a secondary tlux is cnaractenzea oy obtaining an auxiliary signal via such means as an auxiliary core/coil (48/50) and using that signal with appropriate electric circuitry to eliminate non-torque induced variations in the secondary signal. The first embodiment (Figures 5A and 5B) uses the auxiliary signal to maintain the primary magnetic flux at effectively constant amplitude. The second embodiment (Figures 6A and 6B) divides the secondary signal by the auxiliary signal. A third embodiment (Figure 7) employs a plurality of sensors (66A/ 66B) strategically located around the shaft (20) to eliminate spurious signals which are due to bending stress and shaft misalignment.
In summary. several disadvantages of. prior art magnetostrictive torque sensors are: used Causes are: i) Local random variations in magnetic material properties ii) Sinusoidal variation in air gap due to shaft misalignment WO 90/01152 PCT/US89/03118 V -14- paper 870472 researchers foundstresses i t necessary to «null out» the zeignal drifts from day to day In summary, several disadvantages of prior art magnetostrictive torque sensors are: ac) utputSignal variation within one mechanical cycle prohibits accurate instantaneous torque d) utpAverage values withov er several cycles must be used. Causes are: Analys0 i) Local random variations in Faraday’s law of electromagnetism, i.e., d v -N d -NA dB dt dt
(I)
wmagnetic magnetic flux and A is cre rosproperties sectional area, can be rewritten using thein air gap definition for inductance L of a coil/core, i.e., due to shaft misalignment where N is numbef conding stressewindings in i is current b) Signal drifts from day to day in) Output varies with RPM d) Output varies with temperature Analysis of Prior Art Limitations HenceFaraday’s law of electromagnetism, reritten v -N d( -NA L dB dt dt dt (III) where is magnetic fluly all electric core crosuit sectional area, can be rewritten using the app25 definition for inductance L of a coilthe seconre, i.e., Li (II) t ‘where N is number of coil windings and i is current I 30 _a dL In virtually all electric circuit l ,1 applications L is constant and the second term in f the far right hand side of (III) above vanishes. In i accurately, and reliably. This has proven to be WO 90/01152 PCT/U S89/03118 the present case, however, it does not vanish, and as is shown below, the second term gives rise to the phenomenon of output signal dependence on RPM depicted in Figure 4.
Inductance L depends on magnetic permeability AFe of the shaft, and permeability varies slightly in different places around the shaft circumference. This variation can be due to either i) natural changes in the material grain structure, composition, etc. or ii) residual stresses on the shaft surface from machining, forming, etc. (which change F e via magnetostriction). Hence, as the shaft rotates, the sensor «sees» a varying inductance through which its flux must pass. The time derivative of L is therefore not zero and the second term in the far right hand side of (III) above becomes important, it makes a contribution to the output voltage. Further, since the time derivative reflects the rate of change of L, the time derivative increases in magnitude for increasing values of RPM. Hence, the RMS value of as s shown below, the second term in the right hand sidee of (III) increases with increasing rotational speed of the shaft, whereas the RS value of the first term alonepeeability does not.
It is the second term which is responsible for the heretofornce. e unexplained variation in signal output with shaft RPM. In addition, as discussed previously, the second term is also partly responsible for the irregular variation in signal within one shaft revolution shown in Figure 3.
Additionally, it is the second term which causes each individual shaft to have a different deuctaendence on RPM, since each shaft has different local variations in residual stress and magnetiche ntpermeability Hence correctionmethodologies sinceh as those heretofore employed hich attempt toof change of 2] ii i>.
necessity of using two different locations along the W090/01152 PCT/US89/03118 -16correct for output signal dependence via preset feedback signal correction are not suitable for mass production. Further, the correction needed for each individual shaft will change with time as residual stresses relax and as temperature and stress alterations modify local permeability variations.
Hence the heretofore used correction methodologies would become increasingly inaccurate over time and would need continual recalibration. This is impractical in most applications, but particularly so for automotive and other vehicle engines.
Conclusion: Local variations in permeability and residual stresses result in: a) Signal irregularities within one shaft revolution which must be «averaged out» over several cycles b) Increase in output signal with RPM.
Correction methodologies used heretofore are unsuitable for mass production and become increasingly inaccurate over time.
DETAILED DESCRIPTION OF THE INVENTION Previous art has maintained a long held tradition in electromechanics. That is, it has used an electronic feedback loop to keep input current ip to primary coil 32 at a constant RMS value. The objective in doing this has been to keep primary core 30 flux Op at constant RMS (see The underlying assumption is that L, the inductance, remains constant. In the case of magnetostrictive torque sensors, however, the inductance L does not remain constant, so flux p varies even if i, remains fixed. This results in the spurious output voltage signal discussed in previous sections. The present invention, in different embodiments, involves alternative methodologies which solve this and other problems associated with prior art.
Embodiment 1: Feedback to keep dp RMS constant
V
I3) Measures torque of engine directly i
I
WO 90/01152 PCT/US89/03118 -17- Figure 5A depicts circuitry to be used with the magnetostrictive torque sensing apparatus I of Figure 2A in lieu of the signal processor of Figure 2B. The volt meter 50 measures open circuit voltage in an auxiliary coil 48 of any number of turns which in addition to primary coil 32 is also wrapped around the primary core 30. From Faraday’s Law the voltmeter 50 can be used to measure flux op in the primary core 30. Flux cp in the primarycoil 32 can be monitored and used in a feedback to the power amplifier 42 which continuously adjusts input voltage such that cpp, rather than ip, is kept at constant R14S value. Keeping the voltage amplitude of voltmeter 50 constant (by varying keeps the amplitude of primary flux p constant. In this way local variations in inductance L around the circumference of the shaft do not cause variations in the primary flux Op. The result is a circumferential induction and an axial induction Bx.i. which are virtually free of anomalies caused by surface inhomogeneities. Hence the output signal (from voltmeter 38) is essentially independent of RPM and considerably more accurate than existing devices. The output signal from voltmeter 38 is also of relatively constant amplitude within one mechanical cycle (for constant torque within one mechanical cycle) and therefore is suitable for detecting variation in torque virtually instantaneously.
In some instances this embodiment has an even simpler modification as shown in Figure For certain values of circuit parameters the feedback of the signal from voltmeter 38 does not have to be done. Note from that time varying flux in the primary core 30 means voltage drop i across the primary coil 32. There is internal .resistance in the primary coil 32 as well, so the a i 1 l l l variations such as residual stress, slight inhomogeneity in shaft magnetic properties, shaft 7i
~I
WO 90/01152 PCT/US89/03118 -18driving voltage vn must equal the total of the inductive voltage produced by time varying flux pp of the primary coil 32 plus the voltage drop across the internal resistance R, of the primary coil circuit. That is v -V Rpp Np dp Ri (IV) dt When resistance RP in the primary coil 32 is negligible (such as might be encountered with use of superconductor materials or with few windings in the primary coil 32), the second term in the right hand side of (IV) can be assumed effectively zero.
In that case the input voltage (vn) is equal and opposite to the primary coil voltage -vo 0 L associated with the oscillating primary flux pp. Hence keeping input voltage v, (rather than input current ip) at constant amplitude keeps primary flux amplitude op constant regardless of any non-uniformity in shaft magnetic properties.
Note that in this or any other embodiment the primary core/coil 30/32 can be aligned axially instead of circumferentially and the secondary core/coil 34/36 can be aligned circumferentially instead of axially. Further, even though the device performs optimally when the secondary and primary core/coils 34/36 and 30/32 are at right angles to one another and when one of the coil/cores is axially aligned, they can in fact be at any angle with respect to one another and with respect to the shaft axis.
The auxiliary coil 48 is depicted as wrapped around primary core 30 and concentric with primary coil 32, but all that is essential to the proper working of embodiment one is that a significant portion of the flux in the primary flux path of the shaft 20 pass through auxiliary coil 48.
.i
I
L*V
W0.90/01152 PCT/US89/03118 -19- Hence auxiliary coil 48 can be wrapped inside of, wrapped outside of, wound along with, located ii elsewhere on primary core 30 than, or placed in close proximity to primary coil 32; or placed IN anywhere else so long as the signal obtained from auxiliary coil 48 or a function of that signal can be used to control the primary flux in the shaft The description of embodiment one contains many specificities, but embodiment one is not limited in scope by these specificities. Embodiment one is primarily a method to keep primary flux at effectively constant amplitude, thereby eliminating spurious signal components due to material inhomogeneities, temperature, and shaft speed. This can be accomplished utilizing an auxiliary coil as recited herein to determine primary flux or by any other suitable method or means such as, bit not limited to, Hall effect sensing. Further, the determination of secondary flux can be accomplished with a secondary coil as described herein or by any other suitable method or means including, but not limited to, Hall effect sensing. In addition, primary flux can be controlled by passive control methods including, but not limited to, methods which incorporate additional circuit components such as capacitors, resistors, etc. in the primary circuit which serve to maintain primary flux at effectively constant amplitude. Embodiment one, therefore, encompasses any method which maintains primary flux at effectively constant amplitude and monitors torque via determination of secondary flux.
Embodiment 2: Divide Output Signal by Input Voltage Figures 6A and 6B depict different variations of embodiment number two of the present invention. In embodiment number two the standard prior art approach is used for the primary circuit in which primary current i. amplitude is kept seconaary r±ux is measurea via raraaay’s law by placing a second coil close to the shaft aligned in SWO 90/01152 PCT/US89/03118 constant. For the constant primary current ip amplitude local variation in magnetic permeability F, will be reflected in both the Bcir and the Bi al fields. Output voltage from secondary coil 36 depends on the time derivative of Bxai (see In Figure 6A an auxiliary coil 52 around the primary core produces a voltage signal (measured by voltmeter 54) dependent on the time derivative of the primary Bp field on Both signals incorporate the local inhomogeneities, but only the output of the secondary includes changes due to torque induced stresses. Dividing the instantaneous secondary voltage produced by secondary coil 36 by the instantaneous auxiliary coil voltage measured by voltmeter 54 by means of signal divider 56 results in a signal measured by voltmeter 38 which is essentially free of inhomogeneity induced variation.
The signal measured by voltmeter 38 nevertheless still depends directly on the applied stress in the shaft and is a good measure of instantaneous torque.
In practical application it may at times prove necessary to add a component to the circuit between voltmeter 54 and signal divider 56 which would convert instantaneous zero signal values (measured by voltmeter 54) to small finite values in order to preclude division by zero in signal divider 56.
As with embodiment negligible resistance in primary coil 32 would result in the simplification of embodiment #2 shown in Figure 6B.
With negligible primary coil 32 resistance R, I voltage in the primary coil 32 itself can be used directly to divide into the voltage from the secondary coil 36, thereby eliminating the need for auxiliary coil 52 on the primary core.
As with embodiment one, the primary core/coil 30/32 can be aligned axially instead of 1 1 \i.
1 1 1 1 1 WO,90/01152
I
PCT/US89/03118 -21-
I
circumferentially and the secondary core/coil 34/36 can be aligned circumferentially instead of axially.
Further, even though the device performs optimally when the secondary and primary core/coils 34/36 and 30/32 are at right angles to one another and when one of the coil/cores is axially aligned, they can in fact be at any angle with respect to one another and with respect to the shaft axis.
The auxiliary coil 52 is depicted as wrapped around primary core 30 and concentric with primary coil 32, but all that is essential to the proper working of embodiment two is that a significant portion of the flux in the primary flux path of the shaft 20 pass through auxiliary coil 52.
Hence auxiliary coil 52 can be wrapped inside of, wrapped outside of, wound along with, located elsewhere on primary core 30 than, or placed in close proximity to primary coil 32; or located anywhere else so long as the signal obtained from auxiliary coil 52 or a function of that signal can be used as a meaningful input to signal divider 56.
Although Figures 6A and 6B and the above discussion relates to a division of the signal produced by secondary coil 36 by either the signal produced by auxiliary coil 52 or primary coil 32, embodiment two equally relates to a division of the signal produced by auxiliary coil 52 or primary coil 32 by the signal produced by the secondary coil 36.
Further, the signals divided can be instantaneous, RMS, amplitude or any other indicator of signal strength.
The description of embodiment two contains many specificities, but embodiment two is not limited in scope by these specificities. Embodiment two is primarily a method to eliminate spurious signal components due to material inhomogeneities, temperature, and shaft speed by dividing two signals i: i: prior art- conriguralons.
WO 90/01152 PCT/US89/03118 -22- each of which contain similar such spurious signal i components, but only one of which contains a torque dependent component. These signals can be obtained as recited herein using one or more coils to measure flux levels but the method is not restricted to use of such coils. Other methods for determining flux such as, but not limited to, Hall effect sensing could be used as well. Embodiment two, therefore, encompasses any method which accomplishes division of two signals such as those described herein whether or not coils are used to obtain those signals.
Concomitant Solutions to Variation in RPM, Temnerature, Drift Problems Spurious m.aterial non-uniformity induced frequency components of the output voltage signal from secondary coil 35 are directly dependent on shaft speed. Via embodiment #1 or the spurious frequency components are eliminated from the output signal measured by voltmeter 38 and dependence of the output signal of voltmeter 38 on shaft speed is minimized and reduced to an inconsequential level.
Temperature dependence in both embodiments is minimized as well. In embodiment #1 variations due to temperature dependent permeability and mechanical tolerances will automatically be compensated for by keeping the primary flux

PCT/US89/03118 j

8. The method for sensing torque of claim 6 wherein said step of applying is further characterized by: applying the primary signal to at least one primary coil (32) of any number of turns and of such low electrical resistance that the primary signal is a sufficiently accurate approximation of the auxiliary signal; and wherein said step of dividing is further characterized by: using the primary signal as the auxiliary signal and dividing (56) one of said auxiliary signal and said secondary signal into the other (Figure 6B).

9. The method for sensing torque of claim 1 wherein the improvement is further II characterized by the step of: modifying the torque transmitting element (20) by any method which may minimize random anisotropy in magnetic permeability such as, but not limited to, vibrational shaking, such that the resultant signal is affected minimally by magnetic anisotropy. The method for sensing torque of claim 1 wherein the improvement is further characterized by the step of: using at least two resultant signals i respectively associated with different locations (66A, 66B) around the torque A| transmitting element (20) wherein, if the angular spacings between said locations are effectively equal, the at least two resultant signals are added to obtain an enhanced k1 signal; and if the angular spacings between ~4 temperature, and shaft speed by dividing two signals I L I»i :i i i WO 90/01152 L9L PCT/US89/03118 said locations are not effectively equal, the at least two resultant signals are phase shifted and then added to obtain the enhanced signal, whereby said enhanced signal is used to determine torque, and said enhanced signal is minimally affected by bending stress and misalignment of the torque transmitting element (20) (Figure 7). i ;1 n F ~-AM

AU40355/89A
1988-07-21
1989-07-19
Magnetostrictive torque sensor

Ceased

AU620135B2
(en)

Applications Claiming Priority (2)

Application Number
Priority Date
Filing Date
Title

US07/222,838

US4939937A
(en)

1988-07-21
1988-07-21
Magnetostrictive torque sensor

US222838

1988-07-21

Publications (2)

Publication Number
Publication Date

AU4035589A

AU4035589A
(en)

1990-02-19

AU620135B2
true

AU620135B2
(en)

1992-02-13

Family
ID=22833915
Family Applications (1)

Application Number
Title
Priority Date
Filing Date

AU40355/89A
Ceased

AU620135B2
(en)

1988-07-21
1989-07-19
Magnetostrictive torque sensor

Country Status (7)

Country
Link

US
(1)

US4939937A
(en)

EP
(2)

EP0430977A1
(en)

JP
(1)

JPH04500118A
(en)

CN
(1)

CN1039483A
(en)

AU
(1)

AU620135B2
(en)

BR
(1)

BR8907571A
(en)

WO
(1)

WO1990001152A1
(en)

Families Citing this family (50)

* Cited by examiner, † Cited by third party

Publication number
Priority date
Publication date
Assignee
Title

US5062062A
(en)

1988-12-30
1991-10-29
Kabushiki Kaisha Toyota Chuo Kenkyusho
Torque detecting apparatus

US5287735A
(en)

1990-12-10
1994-02-22
Sensortech L.P.
Engine misfire or roughness detection method and apparatus

US5495774A
(en)

1993-06-10
1996-03-05
Sensortech L.P.
Magnetostrictive torque sensor air gap compensator

US5269178A
(en)

1990-12-10
1993-12-14
Sensortech, L.P.
Engine misfire, knock of roughness detection method and apparatus

US5902934A
(en)

1990-12-10
1999-05-11
Sensortech, L.P.
Phase magnitude signal detector

US5591925A
(en)

1991-07-29
1997-01-07
Garshelis; Ivan J.
Circularly magnetized non-contact power sensor and method for measuring torque and power using same

US5351555A
(en)

1991-07-29
1994-10-04
Magnetoelastic Devices, Inc.
Circularly magnetized non-contact torque sensor and method for measuring torque using same

US5520059A
(en)

1991-07-29
1996-05-28
Magnetoelastic Devices, Inc.
Circularly magnetized non-contact torque sensor and method for measuring torque using same

JP2783118B2
(en)

1992-11-06
1998-08-06
三菱電機株式会社

Torque detector

US5412999A
(en)

1993-02-26
1995-05-09
Sensorteck L.P.
Position sensing with magnetostrictive stress sensor

US5493220A
(en)

1993-03-05
1996-02-20
Northeastern University
Magneto-optic Kerr effect stress sensing system

US5631559A
(en)

1993-03-05
1997-05-20
Northeastern University
Method and apparatus for performing magnetic field measurements using magneto-optic kerr effect sensors

EP0651239A3
(en)

1993-10-29
1996-06-12
Omron Tateisi Electronics Co
Magnetostrictive torque sensor, magnetostrictive torque measuring apparatus, and condition – monitoring apparatus for a cutting tool using the same.

US6779409B1
(en)

1997-01-27
2004-08-24
Southwest Research Institute
Measurement of torsional dynamics of rotating shafts using magnetostrictive sensors

US7131339B2
(en)

1997-01-27
2006-11-07
Southwest Research Institute
Measurement of torsional dynamics of rotating shafts using magnetostrictive sensors

US6047605A
(en)

1997-10-21
2000-04-11
Magna-Lastic Devices, Inc.
Collarless circularly magnetized torque transducer having two phase shaft and method for measuring torque using same

US6318189B1
(en)

1998-11-18
2001-11-20
Robert D. Donaldson
Digital torque-responsive pneumatic tool

US6220105B1
(en)

1999-04-16
2001-04-24
Magna-Lastic Devices, Inc.
Magnetoelastic disc-shaped load cell having spiral spokes

US6456059B1
(en)

1999-12-13
2002-09-24
Rockwell Automation Technologies, Inc.
Non-homogeneous material magnetic flux sensor and method

US6774618B2
(en)

1999-12-13
2004-08-10
Rockwell Automation Technologies, Inc.
Magnetic flux sensor and method

GB0028385D0
(en)

2000-11-21
2001-01-03
Inertia Switch Ltd
Torque sensing apparatus and method

US6912922B2
(en)

2000-11-21
2005-07-05
First Inertia Switch Limited
Torque sensing apparatus and method

GB2383417B
(en)

2001-12-20
2005-10-05
Weston Aerospace
Sensing method and apparatus

US7925392B2
(en)

2002-04-23
2011-04-12
Lord Corporation
Aircraft vehicular propulsion system monitoring device and method

US20050114053A1
(en)

2003-10-14
2005-05-26
Southward Steve C.
Magnetostrictive wavelet method for measuring pulse propagation time

DE10236051B4
(en)

2002-08-06
2007-11-29
Eads Deutschland Gmbh

Stress / strain gauge sensor and stress / strain measurement method

US20080173102A1
(en)

2007-01-24
2008-07-24
Nehl Thomas W
Villari torque sensor excitation and pickup arrangement for magnetrostrictive shafts

JP2008134261A
(en)

2008-02-12
2008-06-12
Honda Motor Co Ltd
Rotary torque detection mechanism and electric power steering device

EP2397829B1
(en)

2010-06-21
2016-04-27
PolyResearch AG
Dynamic signal torque sensor

DE102011080282B4
(en)

2011-08-02
2016-02-11
Siemens Aktiengesellschaft

Method and measuring device for examining a magnetic workpiece

CN102401708A
(en)

2011-11-24
2012-04-04
北京科技大学
Torque sensor based on magnetostriction effect and torque measuring method

US9212958B2
(en)

2012-12-28
2015-12-15
General Electric Company
Non-contact magnetostrictive sensing systems and methods

WO2015001097A1
(en)

2013-07-05
2015-01-08
Torque And More (Tam) Gmbh
Solid borne sound wave phase delay comparison

CN103743508B
(en)

2013-12-27
2015-12-02
中国科学院苏州生物医学工程技术研究所
Torsional moment test macro and torque sensor device thereof

CN104483054B
(en)

2014-11-28
2017-04-19
西安交通大学
Flexoelectric passive torque sensor

US9618408B2
(en)

2015-02-26
2017-04-11
General Electric Company
System and method for torque transducer and temperature sensor

DE102015206152B3
(en)

2015-04-07
2016-07-07
Schaeffler Technologies AG & Co. KG

1 – 12An arrangement and method for non-contact measurement of a moment on a machine element

DE102016205784A1
(en)

2016-04-07
2017-10-12
Robert Bosch Gmbh

Torque detecting device and vehicle

US10113921B2
(en)

2016-05-20
2018-10-30
General Electric Company
Systems and methods for determining mechanical stress of machinery

US10302687B2
(en)

2016-06-14
2019-05-28
General Electric Company
Filtration thresholding

US10184846B2
(en)

2016-08-31
2019-01-22
General Electric Company
Systems and methods for compensating for air gap sensitivity in torque sensors

JP6740908B2
(en)

2017-01-11
2020-08-19
日立金属株式会社

Method for manufacturing shaft for magnetostrictive torque sensor

DE102017111055B3
(en)

2017-05-22
2018-09-06
Helmholtz-Zentrum Dresden – Rossendorf E.V.

Apparatus, arrangement and method for characterizing the torsion, rotation and / or positioning of a shaft

US10345167B2
(en)

2017-07-12
2019-07-09
General Electric Company
Temperature compensated torque sensor

US11287337B2
(en)

2019-07-16
2022-03-29
Bently Nevada, Llc
Reference signal compensation for magnetostrictive sensor

WO2021152482A1
(en)

2020-01-27
2021-08-05
Mts Sensor Technologie Gmbh & Co. Kg
Magnetostrictive position measurement using frequency domain analysis

DE102021120525A1
(en)

2021-08-06
2023-02-09
Schaeffler Technologies AG & Co. KG

sensor arrangement

DE102021120522B4
(en)

2021-08-06
2023-06-01
Schaeffler Technologies AG & Co. KG

sensor arrangement

DE102021120663B4
(en)

2021-08-09
2023-03-30
Schaeffler Technologies AG & Co. KG

Torque and bending moment measuring arrangement

DE102021126597A1
(en)

2021-10-14
2023-04-20
Schaeffler Technologies AG & Co. KG

Torque and bending moment measuring arrangement

Family Cites Families (24)

* Cited by examiner, † Cited by third party

Publication number
Priority date
Publication date
Assignee
Title

US2365073A
(en)

1942-03-06
1944-12-12
H D Mallison
Means for indicating horsepower and horsepower factors

US2553833A
(en)

1947-07-11
1951-05-22
Morgan Rifenbergh C
Stress measuring apparatus

US2912642A
(en)

1953-04-18
1959-11-10
Asea Ab
Method and device for indicating and measuring mechanical stresses within ferro-magnetic material

US3011340A
(en)

1957-06-26
1961-12-05
Asea Ab
Means for measuring a torsional stress in a shaft of magnetostrictive material

US3340729A
(en)

1965-01-27
1967-09-12
Avco Corp
Electromagnetic torquemeter

US3681982A
(en)

1968-09-08
1972-08-08
Yaskawa Denki Seisakusho Kk
Transducer system for detecting and measuring mechanical loads or stresses

JPS5214985B2
(en)

1972-04-03
1977-04-26

FR2194311A5
(en)

1972-07-25
1974-02-22
Ctre Etu Rech Machine

US4135391A
(en)

1977-11-22
1979-01-23
Asea Aktiebolag
Torque transducer

DE3031997C2
(en)

1980-08-25
1986-01-16
Licentia Patent-Verwaltungs-Gmbh, 6000 Frankfurt

Arrangement for non-contact measurement of static and dynamic torques

DE3206008C1
(en)

1982-02-19
1983-08-25
Licentia Patent-Verwaltungs-Gmbh, 6000 Frankfurt

Proceed for the non-contact measurement of static and dynamic torques

DE3206503C1
(en)

1982-02-24
1983-08-25
Licentia Patent-Verwaltungs-Gmbh, 6000 Frankfurt
Method for compensating for the electrical interference voltage which occurs during torque measurement in accordance with the eddy-current principle on shafts, and is generated by the heterogeneous nature of the permeability of the surface of the shaft

US4479390A
(en)

1983-01-28
1984-10-30
Borg-Warner Corporation
Transducer for measuring torque and/or speed of rotating shaft

US4506554A
(en)

1983-06-07
1985-03-26
Asea Aktiebolag
Magnetoelastic torque transducer

DE3437379A1
(en)

1983-10-12
1985-04-25
Bently Nevada Corp., Minden, Nev.
Device for measuring the rotary or bending force exerted on a shaft

JPS6088335A
(en)

1983-10-19
1985-05-18
Nissan Motor Co Ltd
Torque detector

JPS60140133A
(en)

1983-12-28
1985-07-25
Toyota Central Res & Dev Lab Inc
Torque sensor

US4566338A
(en)

1984-07-20
1986-01-28
Trw Inc.
Noncontact torque sensor

JPH0672825B2
(en)

1984-08-30
1994-09-14
株式会社豊田中央研究所

Torque measuring device

JPS61162726A
(en)

1985-01-11
1986-07-23
Toyota Central Res & Dev Lab Inc
Stress detector

DE3511442A1
(en)

1985-03-29
1986-10-02
Internationale Fluggeräte und Motoren GmbH, 6940 Weinheim
Method and device for contactless measurement of static and dynamic torques

US4697459A
(en)

1985-09-04
1987-10-06
Kabushiki Kaisha Toyota Chuo Kenkyusho
Torque measuring apparatus

JPS6275328A
(en)

1985-09-30
1987-04-07
Toshiba Corp
Torque sensor

JPH06183927A
(en)

1992-12-17
1994-07-05
Ichimaru Pharcos Co Ltd
Cosmetic containing hydrogenated milk fat

1988

1988-07-21
US
US07/222,838
patent/US4939937A/en
not_active
Expired – Fee Related

1989

1989-07-19
EP
EP89908932A
patent/EP0430977A1/en
active
Pending

1989-07-19
AU
AU40355/89A
patent/AU620135B2/en
not_active
Ceased

1989-07-19
WO
PCT/US1989/003118
patent/WO1990001152A1/en
not_active
Application Discontinuation

1989-07-19
EP
EP89402052A
patent/EP0352187A1/en
not_active
Ceased

1989-07-19
JP
JP1508423A
patent/JPH04500118A/en
active
Pending

1989-07-19
BR
BR898907571A
patent/BR8907571A/en
not_active
Application Discontinuation

1989-07-20
CN
CN89104920.7A
patent/CN1039483A/en
active
Pending

Also Published As

Publication number
Publication date

WO1990001152A1
(en)

1990-02-08

EP0352187A1
(en)

1990-01-24

JPH04500118A
(en)

1992-01-09

CN1039483A
(en)

1990-02-07

US4939937A
(en)

1990-07-10

AU4035589A
(en)

1990-02-19

BR8907571A
(en)

1991-06-18

EP0430977A1
(en)

1991-06-12

Información de contacto