JP2009131104A - Transfer robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transfer robot with direct drive motors available in an atmosphere without air, which monitors bearings effectively and quickly returns to its operational state when an abnormality occurs. <P>SOLUTION: The abnormality of the bearings 19 and 19' of the direct drive motors D1 and D2 making up the transfer robot and that of the link support bearing can be detected by comparing a difference between a torque command value at the earliest stage (first elapsed time) at which the transfer robot starts operation and a torque command value with an elapse of a predetermined operation time (second elapsed time after the first elapsed time), which difference results when a table T is caused to come closer to or go away from the drive motors D1 and D2 (by rotating the drive motors D1 and D2 reverse to each other). <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a transfer robot using a direct drive motor used in an atmosphere outside the atmosphere such as a vacuum.

  For example, in a semiconductor manufacturing apparatus or the like, a workpiece is processed in an ultra-high vacuum atmosphere in a vacuum chamber in order to eliminate impurities as much as possible. As an actuator used in that case, for example, in a drive motor of a workpiece positioning device, a lubricant containing a volatile component such as general grease cannot be used for a drive shaft bearing. Lubrication is enhanced by plating soft metals such as gold and silver on the inner and outer rings of the bearing. In addition, there is a fact that a stable material with excellent heat resistance and low emission gas is selected for the coil insulating material of the drive motor, the wiring coating material, and the adhesive of the laminated magnetic pole.

  In particular, in recent years, the degree of integration of semiconductors has increased, and at the same time, higher density has been promoted by reducing the pattern width of the IC. In order to manufacture a wafer that can cope with this miniaturization, a high degree of uniformity in wafer quality is required. In order to meet the demand, it is important to further reduce the impurity gas concentration in the low-pressure gas processing chamber of the wafer. Further, in order to perform microfabrication as required, an extremely high precision positioning device is required. From the above viewpoint, the following problems are pointed out when the conventional actuator is examined.

  That is, in the case of a drive motor used in a vacuum chamber equipped with an ultra-vacuum atmosphere, even if a stable material with excellent heat resistance and low emission gas is selected for the coil insulation material or wiring coating of the drive motor, As long as it is an organic insulating material, it is microscopically porous and has numerous holes on its surface. Once this is exposed to the atmosphere, gas, water molecules, etc. are taken in and occluded in the holes on the surface. A long time is required for degassing to remove these occluded impure molecules by vacuum evacuation, and a reduction in production efficiency is unavoidable. Furthermore, since heat cannot be released due to air convection in a vacuum, when the coil temperature rises locally, the resistance of that portion increases, heat generation is accelerated, and the coil insulation film is burned out. Easy to invite. On the other hand, while using an inorganic material for the coil insulating material, it is conceivable to reduce adsorbed impure molecules by using a stainless steel sheath wire for the wiring. However, in that case, not only the cost becomes very high, but also the ratio of conductors such as copper in the coil winding space decreases, resulting in an increase in electrical resistance, resulting in a decrease in motor capacity. There is.

In order to solve such a problem, Patent Document 1 discloses a direct drive motor in which a stator is disposed inside a partition wall, a rotor is disposed outside the partition wall, and the rotor is driven to rotate to drive a frog leg arm. . According to the direct drive motor of Patent Document 1, coil insulation materials and wiring coatings attached to the stator are disposed inside the partition maintained at atmospheric pressure, and therefore, the storage impureness when they are disposed in the vacuum chamber. The problem of molecular discharge and the problem of heat generation can be avoided.
JP 2006-109656 A

  By the way, vacuum bearing oil is used for bearings that support the rotor in the direct drive motor for the above-described applications and joint portions of a transfer robot that transfers a semiconductor wafer or the like using the direct drive motor. However, since the lubricant used in such an ultra-high vacuum atmosphere has a shorter lubrication life compared to general grease, spike-like torque fluctuations and abnormal increase in average torque caused by poor lubrication of bearings As a result, vibration may occur in the arm or table of the transfer robot, and the semiconductor wafer as the transfer object may be damaged. In order to prevent process failures and accidental line stoppage associated with semiconductor wafer transfer robots, the bearings arranged to support the rotation of the drive source and the bearings arranged at the joints of the arm are monitored for signs of abnormalities. In such a case, it is important to immediately stop the apparatus and perform maintenance such as replacement. However, conventionally, there has been no effective technique for monitoring a bearing used in a semiconductor wafer transfer robot.

  The present invention has been made in view of the above problems, and its object is to perform effective monitoring of bearings in a transfer robot equipped with a direct drive motor used in an atmosphere outside the atmosphere, The object is to provide a transfer robot that can quickly restore operation if there is an abnormality.

The transfer robot of the present invention is a transfer robot that drives a table via a link mechanism by power from two direct drive motors.
Each direct drive motor
A housing;
A partition wall extending from the housing and isolating the atmosphere side from the atmosphere outside;
A rotor disposed outside the atmosphere with respect to the partition;
A rotor support bearing disposed outside the atmosphere with respect to the partition wall and rotatably supporting the rotor;
A stator disposed on the atmosphere side with respect to the partition wall and driving the rotor;
An angle detector for detecting a rotation angle of the outer rotor;
A motor control circuit for energizing the stator based on a torque command value or a current command value calculated according to the rotation angle of the outer rotor detected by the angle detector;
By comparing the torque command value or the current command value calculated at a predetermined time by the motor control circuit, the friction torque of the link support bearing disposed at the joint portion of the rotor support bearing and the link mechanism is detected. It is characterized by that.

  The transfer robot according to the present invention has a feature that a direct drive motor suitable for use in an atmosphere outside the atmosphere is provided as a drive source, and there is no rotation introducer such as a magnetic coupling and a speed reducer. Therefore, it can be said that only the rotor support bearing and the link support bearing cause the mechanical loss when the direct drive motor rotates.

Factors that change the torque command value or current command value when the direct drive motor configured in this way is rotationally driven are:
(1) Iron loss such as switching loss and eddy current loss of partition walls,
(2) Cogging torque inherent to direct drive motors,
(3) Kinetic energy for accelerating / decelerating the moment of inertia of the rotor and link mechanism,
(4) Friction torque of rotor support bearing and link support bearing,
It is limited to. Of these factors, the friction torque of the rotor support bearing and link support bearing is constant and quantitative, and can be easily quantified. Therefore, by removing factors other than friction torque from the torque command value or current command value calculated in the motor control circuit, abnormalities in the friction torque of the rotor support bearing and link support bearing can be detected. Thus, abnormality of the rotor support bearing and the link support bearing can be detected, and maintenance such as replenishment of lubricating oil and replacement of parts can be performed by stopping the transfer robot and the like as necessary.

The angle detector is
An outer magnetic coupling rotor disposed outside the atmosphere with respect to the partition wall and rotating together with the rotor;
An inner magnetic coupling rotor disposed on the atmosphere side with respect to the partition wall and rotating in synchronization with the outer magnetic coupling rotor by a magnetic coupling action;
An inner bearing disposed on the atmosphere side with respect to the partition wall and rotatably supporting the inner magnetic coupling rotor;
A resolver device arranged on the atmosphere side with respect to the partition wall,
The resolver rotor of the resolver device is preferably configured to be rotatably supported by an inner bearing together with the inner magnetic coupling rotor.

  According to the present invention, the angle of the rotor disposed outside the atmosphere with respect to the partition wall is transmitted to the resolver device disposed on the atmosphere side with respect to the partition wall by a magnetic coupling action. The inner bearing only supports the inner magnetic coupling rotor and the resolver rotor in rotation, and the load capacity is small. Therefore, the friction torque is smaller than that of the rotor support bearing and link support bearing, and the friction torque of the rotor described above is reduced. Does not affect detection.

  In the direct drive motor configured as described above, the resolver rotor of the resolver device is rotatably supported by the inner bearing and is not mechanically connected to the rotor support bearing. Therefore, even if an abnormal torque change is detected in the rotor support bearing and the replacement work is performed, the accuracy of the angle detector is not affected by the rotor mounting error, etc. This eliminates the need for measurements to be performed, so the transport robot and the like can be restored quickly. Due to these effects, by applying the direct drive motor of the present invention to a transfer robot used in an atmosphere outside the atmosphere, it is possible to effectively monitor the bearing and provide a transfer robot that can be quickly restored to operation. can do.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view of a frog-leg-arm type transport robot using a motor system including a direct drive motor according to the present embodiment. In FIG. 1, two direct drive motors D1 and D2 are connected in series. A first arm A1 is connected to the rotor of the lower direct drive motor D1, and a first link L1 is pivotally connected to the tip of the first arm A1. On the other hand, the second arm A2 is connected to the rotor of the upper direct drive motor D2, and the second link L2 is pivotally connected to the tip of the second arm A2. The links L1 and L2 are pivotally connected to a table T on which the wafer W is placed. Although not shown, link support bearings are provided in the joint portions of the arm A1 and the link L1, the joint portion of the arm A2 and the link L2, the link L1 and the joint portion of the table T, and the joint portion of the link L2 and the table T, respectively. Is arranged. In addition, as a link support bearing, what was described in Unexamined-Japanese-Patent No. 2006-329420, for example can be used.

  As is clear from FIG. 1, if the rotors of the direct drive motors D1 and D2 rotate in the same direction, the table T also rotates in the same direction. If the rotor rotates in the opposite direction, the table T The motors D1 and D2 are approached or separated from each other. Therefore, if the direct drive motors D1 and D2 are rotated at an arbitrary angle, the wafer W can be transferred to an arbitrary two-dimensional position within a range where the table T can reach.

  Thus, for example, in a wafer transfer arm disposed in a vacuum chamber in a semiconductor manufacturing apparatus, for example, an apparatus having a plurality of arms such as a scalar type or a frog-leg type shown in the figure, a plurality of rotary motors are required. . In a vacuum environment, it is necessary to reduce the surface area of contact with the outside as much as possible, and at the same time to reduce the number of mounting holes for a motor or the like as much as possible in order to effectively use the space. In addition, in order to transport the wafer W horizontally and with minimal vibration, it is necessary to firmly hold the moment acting on the tip of the arm with the rotor support. Therefore, a plurality of direct drive motors D1 and D2 are connected coaxially at the housing part, and the connecting part is tightly joined with a seal (tightly joined by welding, O-ring, metal gasket, etc.), and the motor rotor is disposed. It is also necessary to separate the open space and the housing external space.

  Further, in order to convey the wafer W horizontally and with less vibration, it is necessary to firmly hold the moment acting on the tips of the arms A1 and A2 by the rotor support portion. Furthermore, when driving multiple axes in a vacuum environment, if the current rotation position of the arm is not recognized when the power is turned on, the arm A1, A2, etc. are hit against the wall of the vacuum chamber or the shutter of the vacuum chamber. There is a possibility. A motor system in which direct drive motors that can meet such demands are connected coaxially will be described.

In this embodiment, a surface magnet type 32-pole 36-slot outer rotor brushless type direct drive motor is used. The slot combination of 32 poles and 36 slots has a configuration four times that of the slot combination of 8 poles and 9 slots, which is generally known to have a small cogging force but a large magnetic attraction force in the radial direction and a large vibration during rotation. It is. Since the magnetic attractive force in the radial direction is canceled by increasing ²ⁿ times (n is an integer), vibration during rotation without increasing the roundness and coaxiality of the stator and rotor and the rigidity of the mechanical parts Can be made small and cogging is inherently small, so that a very smooth rotation can be obtained. On the other hand, by using such a very multipolar motor, the electrical angle cycle is greater than the mechanical angle cycle, so that the positioning controllability is good. Therefore, as in the present invention, it is suitable for a direct drive motor that drives a robot apparatus without using a speed reducer. Further, since the thickness and salient pole width of the stator connecting portion and the yoke thickness of the rotor can be reduced without reducing the total magnetic flux amount, a direct drive motor having a thin and large diameter and narrow width as in the present invention is used. Is preferred.

  FIG. 2 is a view of the configuration of FIG. 1 taken along the line II-II and viewed in the direction of the arrow. The internal structure of the two-axis motor system will be described in detail with reference to FIG. First, the direct drive motor D1 will be described. A hollow cylindrical main body 12 that is coaxially joined to the central opening 10a of the disk 10 installed on the surface plate G and fixed to each other by bolts 11 has a cup-shaped partition wall 13 attached to the upper end thereof. The center of the main body 12 can be used for passing wiring to the stator. The main body 12 and the disk 10 constitute a housing.

  The partition wall 13 is made of stainless steel, which is a non-magnetic material, and has a thick bottom portion 13a fitted to the main body 12 and a thin wall portion extending from the periphery thereof so as to penetrate the direct drive motors D1 and D2 in the axial direction. It consists of a cylindrical part 13b. Therefore, the partition wall 13 is commonly used for the direct drive motors D1 and D2. The lower end of the cylindrical portion 13b is joined to the holder 15 so as to be sealed by TIG welding, and the holder 15 is fixed to the disk 10 with bolts 16. Here, by setting the welded portion of the cylindrical portion 13b to substantially the same thickness, heat is prevented from escaping only to the component on one side, and the fitting portion can be welded uniformly. The contact surface between the holder 15 and the disc 10 is provided with a groove processing for fitting the seal member. After the seal member OR is fitted into the groove, the holder 15 and the disc 10 are fastened by the bolts 16. The part is separated from the atmosphere side. The partition wall 13 is made of SUS316 made of austenitic stainless steel, which has high corrosion resistance and is less magnetic, and the holder 15 is also made of SUS316 because of its weldability with the partition wall 13.

  Further, the partition wall 13 and the holder 15 are airtightly joined, and the holder 15 and the disk 10 and the disk 10 and the surface plate G are hermetically sealed by O-rings OR, respectively. Therefore, the internal space surrounded by the disk 10 and the partition wall 13 is airtight from the outside. In addition, the partition 13 does not necessarily need to be a nonmagnetic material. Further, instead of using the O-ring OR, the members may be hermetically sealed by electron beam welding or laser beam welding.

  A bearing holder 17 is integrally formed on the outer peripheral upper surface of the disc 10. An outer ring of a four-point contact ball bearing 19 used in a vacuum is fitted on the bearing holder 17 and fixed with bolts 20. On the other hand, the inner ring of the bearing 19 is fixed to a double cylindrical cylindrical member 23 into which the first outer rotor member 21 is fitted, and is fixed by a bolt 22 that fastens the first outer rotor member 21 together. That is, the 1st outer side rotor member 21 is rotatably supported with respect to the partition 13 by the cylindrical member 23 which supports arm A1 (FIG. 1). The first outer rotor member 21 and the cylindrical member 23 constitute an outer rotor.

  The disc 10 (including the bearing holder 17) is made of austenitic stainless steel having high corrosion resistance. The disc 10 also serves as a fitting and fixing device with the surface plate G that is a chamber, and a sealing device on its lower surface. , A groove 10b for fitting the O-ring OR is provided.

  The bearing 19 is a four-point contact ball bearing that can load radial, axial, and moment loads with a single bearing. By using this type of bearing, only one bearing of the direct drive motor D1 is required, so that the two-axis coaxial motor system of the present invention can be thinned. The bearing 19 is made of martensitic stainless steel, which has high corrosion resistance for both the inner and outer rings and can be hardened by quenching. The rolling elements are ceramic balls, and the lubricant is vacuum grease that does not solidify even in vacuum.

  The bearing 19 may be a metal lubrication that is plated with a soft metal such as gold or silver on the inner ring and the outer ring and does not emit outgas even in a vacuum, and is a four-point contact ball bearing. The first outer rotor member 21 from the arm A1 can receive a moment in a tilting direction, but is not limited to a four-point contact type, and a cross roller, a cross ball, and a cross taper bearing can also be used. Alternatively, fluorine film treatment (DFO) may be performed to improve lubricity.

  The cylindrical member 23 has a surface for fitting and fixing the inner ring of the bearing 19. The four-point contact ball bearing 19 is a very thin bearing, and rotational accuracy and friction torque are greatly affected by differences in accuracy of members to be assembled and linear expansion coefficients. Therefore, in the case of the present embodiment, the inner ring of the bearing 19 that is a rotating ring is an interference fit or an intermediate fit on a cylindrical member 23 that is easy to obtain processing accuracy and whose linear expansion coefficient is substantially the same as the bearing ring material of the bearing. The outer ring of the bearing 19, which is a fixed ring, is configured to have a clearance fit with a bearing holder made of austenitic stainless steel or an aluminum holder 17, thereby preventing a decrease in rotational accuracy of the bearing 19 and an increase in friction torque due to a temperature rise. It has become.

  The first outer rotor member 21 includes a permanent magnet 21a, an annular yoke 21b made of a magnetic material for forming a magnetic path, and a wedge made of a non-magnetic material for mechanically fastening the permanent magnet 21a and the yoke 21b. (Not shown). The permanent magnet 21a is a segment type in which each of the N pole and S pole magnets is made of a magnetic metal and is divided into 16 poles in a 32 pole configuration, and each of the permanent magnets 21a has a sector shape. The arc centers of the inner diameter and the outer diameter are the same, but the tangential intersection of the circumferential end face is closer to the permanent magnet 21a, so that the wedge is tightened with a screw from the outer diameter side of the yoke 21b, thereby fixing the permanent magnet 21a to the yoke 21b. It is concluded to. By setting it as such a structure, a permanent magnet can be fastened without using the fixing member which generate | occur | produces outgas, such as an adhesive agent. The permanent magnet 21a is a neodymium (Nd—Fe—B) based magnet having a high energy product, and is coated with nickel in order to improve corrosion resistance. The yoke 21b is made of low carbon steel having high magnetism and is plated with nickel in order to improve rust prevention and corrosion resistance and prevent wear during bearing replacement after processing and molding.

  A first stator 29 is arranged on the inner side in the radial direction of the partition wall 13 so as to face the inner peripheral surface of the first outer rotor member 21. The first stator 29 is attached to a cylindrically deformed lower portion of the flange portion 12a extending in the radial direction at the center of the main body 12, and is formed of a laminated material of electromagnetic steel plates. The motor coil is concentratedly wound after the bobbin is fitted. The outer diameter of the first stator 29 is approximately the same as or smaller than the inner diameter of the partition wall 13.

  A first inner rotor 30 is disposed adjacent to and parallel to the first stator 29. The first inner rotor 30 is rotatably supported by ball bearings 33 with respect to a resolver holder 32 that is bolted to the outer peripheral surface of the main body 12. A permanent magnet 30a is attached to the outer peripheral surface of the first inner rotor 30 via a back yoke 30b. The permanent magnet 30a is composed of 32 poles and 16 magnets of N poles and S poles alternately made of magnetic metal like the permanent magnet 21a of the first outer rotor member 21. Accordingly, the first inner rotor 30 is rotated along with the first outer rotor member 21 driven by the first stator 29.

  The bearing (inner bearing) 33 that rotatably supports the first inner rotor 30 is a four-point contact ball bearing that can apply radial, axial, and moment loads with a single bearing. By using this type of bearing, since only one bearing is required, the direct drive motor D1 can be thinned. Since the inside of the partition wall 13 is an atmospheric environment, a bearing using grease lubrication based on general bearing steel and mineral oil can be applied.

  Since the inside of the partition wall 13 is an atmospheric environment, the permanent magnet 30a is bonded and fixed to the back yoke 30b. The permanent magnet 30a is a neodymium (Nd—Fe—B) based magnet having a high energy product, and has a nickel coating to prevent demagnetization due to rust. The yoke 30b is made of low carbon steel having high magnetism, and is chromate plated for rust prevention after processing and forming.

  Resolver rotors 34 a and 34 b are assembled as detectors for measuring the rotation angle on the inner periphery of the first inner rotor 30, and resolver stators 35 and 36 are disposed on the outer periphery of the resolver holder 32 so as to face it. Although attached, in this embodiment, the high-resolution incremental resolver stator 35 and the absolute resolver stator 36 capable of detecting which position of the rotor is located in one rotation are arranged in two layers. Therefore, even when the power is turned on, the rotation angle of the absolute resolver rotor 34b is known, no return to origin is required, and the electrical phase angle of the magnet with respect to the coil is known. Therefore, the rotation used for driving current control of the direct drive motor D1 Angle detection is possible without using a pole detection sensor.

  The resolver holder 32 and the first inner rotor 30 are made of carbon steel, which is a magnetic material, so that electromagnetic noise from the motor field and the motor coil is not transmitted to the resolver stators 35, 36 that are angle detectors. Chromate plating is applied after molding to prevent rust.

  In the high-resolution variable reluctance resolver (resolver device) used in the present embodiment, the incremental resolver rotor 34a has a plurality of slot teeth having a constant pitch, and the incremental resolver stator 35 has an outer peripheral surface on its outer peripheral surface. In addition, teeth that are out of phase with respect to the incremental resolver rotor 34a are provided at each magnetic pole in parallel with the rotation axis, and a coil is wound around each magnetic pole. When the incremental resolver rotor 34a rotates integrally with the first inner rotor 30 supported by the bearing 33, the reluctance with the magnetic pole of the incremental resolver stator 35 changes, and the fundamental wave of the reluctance change with one revolution of the incremental resolver rotor 34a. The reluctance change is detected so that the component has n cycles, and is digitized by the resolver control circuit shown in FIG. 3 and used as a position signal to rotate the incremental resolver rotor 34a, that is, the first inner rotor 30. An angle (or rotational speed) is detected. The resolver rotors 34a and 34b and the resolver stators 35 and 36 constitute a detector.

  According to the present embodiment, the first inner rotor (inner magnetic coupling rotor) 30 rotates at the same speed with respect to the first outer rotor member (outer magnetic coupling rotor) 21 by the magnetic coupling action. Since it is rotated, the rotation angle of the first outer rotor member 21 can be detected through the partition wall 13. Further, in the present embodiment, the resolver alone has the bearing 33 without using the motor forming parts and the housing. Therefore, the eccentricity adjustment by the resolver alone and the position adjustment of the resolver coil are performed before being incorporated in the housing. Therefore, there is no need to provide adjustment holes and notches in the housing and both flanges.

  Next, although the direct drive motor D2 is demonstrated, the main body 12 comprises a housing here. The cylindrical member 23 of the direct drive motor D1 described above extends upward to a position where it overlaps with the direct drive motor D2, and the inner peripheral surface of the four-point contact ball bearing 19 ′ used in vacuum An outer ring is fitted in a fitting manner and is fixed by a bolt 20 '. On the other hand, the inner ring of the bearing 19 ′ is fixed by a bolt 22 ′ that fits around the circumferential surface of the double cylindrical ring-shaped member 23 ′ and fastens the second outer rotor member 21 ′ together. That is, the second outer rotor member 21 ′ is rotatably supported with respect to the partition wall 13 by the ring-shaped member 23 ′ that supports the arm A <b> 2 (FIG. 1). The second outer rotor member 21 'and the ring-shaped member 23' constitute an outer rotor.

  The bearing 19 ′ is a four-point contact ball bearing that can load radial, axial, and moment loads with a single bearing. By using this type of bearing, only one bearing of the direct drive motor D2 is required, so that the biaxial coaxial motor of the present invention can be thinned. The inner and outer rings are made of martensitic stainless steel, which has high corrosion resistance and can be hardened by quenching. The rolling elements are ceramic balls, and the lubricant is vacuum grease that does not solidify even in vacuum.

  The bearing 19 'may be made of a metal lubrication in which a soft metal such as gold or silver is plated on the inner ring and the outer ring to prevent outgassing even in a vacuum, and is a four-point contact ball bearing. The first outer rotor member 21 'from the arm A1 can receive a moment in the tilting direction, but is not limited to a four-point contact type, and a cross roller, a cross ball, and a cross taper bearing can also be used. May be used, or fluorine film treatment (DFO) may be performed to improve lubricity.

  The ring-shaped member 23 ′ has a surface for fitting and fixing the inner ring of the bearing 19 ′. The four-point contact ball bearing 19 'is a very thin bearing, and rotational accuracy and friction torque are greatly affected by differences in accuracy of members to be assembled and linear expansion coefficients. Therefore, in the case of the present embodiment, the inner ring of the bearing 19 ′ has an interference fit or an intermediate fit on a yoke 21b that is easy to obtain machining accuracy and whose linear expansion coefficient is substantially the same as the bearing ring material of the bearing. The outer ring is configured to have a clearance fit with a bearing holder made of austenitic stainless steel or an aluminum boss, thereby preventing a decrease in rotational accuracy of the bearing 19 ′ and an increase in friction torque due to a temperature rise.

  The second outer rotor member 21 ′ is composed of a permanent magnet 21a ′, an annular yoke 21b ′ made of a magnetic material for forming a magnetic path, and a non-magnetic for fastening the permanent magnet 21a ′ and the yoke 21b ′ mechanically. It is comprised by the wedge (not shown) which consists of a magnetic body. The permanent magnet 21a 'is a segment type in which each of the N poles and the S poles is made of a magnetic metal and is divided into 16 poles, and each of the permanent magnets 21a' has a sector shape. The arc centers of the inner diameter and the outer diameter are the same, but the tangent intersection of the circumferential end surface is closer to the permanent magnet 21a ′, so that the wedge is tightened with a screw from the outer diameter side of the yoke 21b ′, thereby permanent magnet 21a ′. Is fastened to the yoke 21b '. By setting it as such a structure, a permanent magnet can be fastened without using the fixing member which generate | occur | produces outgas, such as an adhesive agent. The permanent magnet 21a 'is a neodymium (Nd-Fe-B) magnet having a high energy product, and is coated with nickel in order to improve corrosion resistance. The yoke 21b 'is made of low carbon steel having high magnetism and is plated with nickel in order to improve rust prevention and corrosion resistance and prevent wear during bearing replacement after processing and molding.

  A second stator 29 ′ is disposed on the inner side in the radial direction of the partition wall 13 so as to face the inner peripheral surface of the second outer rotor member (outer magnetic coupling rotor) 21 ′. The second stator 29 ′ is attached to the upper part of the flange 12 a that extends in the radial direction in the center of the main body 12, is formed of a laminated material of electromagnetic steel plates, and each salient pole is insulated. After the bobbin is fitted, the motor coil is concentratedly wound. The outer diameter of the second stator 29 ′ is approximately the same as or smaller than the inner diameter of the partition wall 13.

  A second inner rotor (inner magnetic coupling rotor) 30 ′ is disposed adjacent to and parallel to the second stator 29 ′. The second inner rotor 30 ′ is rotatably supported by a ball bearing 33 ′ with respect to a resolver holder 32 ′ that is bolted to the outer peripheral surface of the main body 12. A permanent magnet 30a 'is attached to the outer peripheral surface of the second inner rotor 30' via a back yoke 30b '. The permanent magnet 30a 'has a configuration of 32 poles as in the case of the permanent magnet 21a' of the second outer rotor member 21 ', and 16 magnets of N and S poles are alternately made of magnetic metal. Accordingly, the second inner rotor 30 'is rotationally driven by the second stator 29' in synchronization with the second outer rotor member 21 '.

  A bearing (inner bearing) 33 ′ that rotatably supports the first inner rotor 30 ′ is a four-point contact ball bearing that can apply radial, axial, and moment loads with a single bearing. By using this type of bearing, since only one bearing is required, the direct drive motor D2 can be thinned. Since the inside of the partition wall 13 is an atmospheric environment, a bearing using grease lubrication based on general bearing steel and mineral oil can be applied.

  Since the inside of the partition wall 13 is an atmospheric environment, the permanent magnet 30a 'is fixedly bonded to the back yoke 30b'. The permanent magnet 30a 'is a neodymium (Nd-Fe-B) based magnet having a high energy product, and is coated with nickel to prevent demagnetization due to rust. The yoke 30b 'is made of a low-carbon steel having high magnetism, and is chromated for rust prevention after work forming.

  Resolver rotors 34 a ′ and 34 b ′ are assembled as detectors for measuring the rotation angle on the inner periphery of the second inner rotor 30 ′, and the resolver stator 32 ′ is disposed on the outer periphery of the resolver holder 32 ′ so as to face it. Although 35 'and 36' are attached, in this embodiment, a high resolution incremental resolver stator 35 'and two layers of an absolute resolver stator 36' capable of detecting which position of the rotor is in one rotation. Is arranged. Therefore, even when the power is turned on, the rotational angle of the absolute resolver rotor 34b 'is known, no return to origin is required, and the electrical phase angle of the magnet with respect to the coil is known. Therefore, the relative rotational angle of the direct drive motor D2 is This is possible without using a detection sensor.

  The resolver holder 32 ′ and the second inner rotor 30 ′ are made of carbon steel, which is a magnetic body, so that electromagnetic noise from the motor field and the motor coil is not transmitted to the resolver stators 35 ′, 36 ′, which are angle detectors. The material is chromate plated for rust prevention after processing and molding.

  According to the present embodiment, the second inner rotor 30 ′ rotates at the same speed by the magnetic coupling action with respect to the second outer rotor member 21 ′. The rotation angle can be detected through the partition wall 13. Further, in the present embodiment, the resolver alone has the bearing 33 ′ without using the motor forming parts and the housing. Therefore, before being incorporated in the housing, the eccentricity adjustment and the position of the resolver coil by the resolver alone are provided. Since adjustments such as adjustment can be performed, there is no need to provide adjustment holes or notches in the housing and both flanges.

  In the high-resolution variable reluctance resolver (resolver device) used in the present embodiment, the incremental resolver rotor 34a 'has a plurality of slot teeth having a constant pitch, and the outer peripheral surface of the incremental resolver stator 35'. Are provided with teeth shifted in phase with respect to the incremental resolver rotor 34a ′ at each magnetic pole in parallel with the rotation axis, and a coil is wound around each magnetic pole. When the incremental resolver rotor 34a ′ rotates integrally with the second inner rotor 30 ′ supported by the bearing 33 ′, the reluctance with the magnetic pole of the incremental resolver stator 35 ′ changes, and the incremental resolver rotor 34a ′ rotates once. The reluctance change is detected so that the fundamental wave component of the reluctance change becomes n periods, digitized by the resolver control circuit shown in FIG. 3 as an example, and used as a position signal, so that the incremental resolver rotor 34a ′, that is, the first 2 The rotation angle (or rotation speed) of the inner rotor 30 ′ is detected. The resolver rotors 34a 'and 34b and the resolver stators 35' and 36 'constitute a detector.

  According to the present embodiment, the outer rotor (second outer rotor 21 ′ and ring-shaped member 23 ′) of the direct drive motor D2 is replaced with the outer rotor (first outer rotor 21 and cylindrical member of the other direct drive motor D1. 23), the bearing 19 'is supported by the bearing 19' so that if the outer rotor of the direct drive motor D2 is removed, the bearing 19 'supporting the outer rotor can be exposed, and then the outer rotor of the direct drive motor D1. Since the bearing 19 supporting the outer rotor can be exposed and can be easily inspected and removed, the maintainability is also improved. Furthermore, since it is only necessary to remove the outer rotor outside the partition wall 13, it is not necessary to remove the partition wall structure, and a leak check or the like is not required at the time of reassembly, thereby improving assemblability.

  The first stator 29 and the second stator 29 'are arranged vertically with the flange portion 12a as the center, and the resolver is arranged on the inner side in the radial direction. The main body 12 has a hollow structure, and the flange portion 12a has at least one radial through hole 12d communicating with the center, through which the motor wiring is drawn out to the center of the main body 12. It has become. On the other hand, both ends of the main body 12 are provided with at least one notch 12e and 12e, respectively, through which the resolver wiring is drawn out to the center of the main body 12. With this structure, it is possible to arrange the resolver of the direct motor D1, the stator 29, the stator 29 ′ of the direct motor D2, and the resolver in this order from the housing side. The angle of the stator and resolver can be adjusted. Therefore, if a facility for rotationally driving the reference outer rotor is prepared separately, the angle of the resolver relative to the stator can be adjusted with high accuracy by setting the main body 12 incorporating the stator and resolver in the facility. It is possible to prevent the angle positioning accuracy from being lowered due to the deviation of the commutation, and to improve the compatibility of the drive control circuit with the biaxial coaxial motor of the present invention.

  FIG. 4 is a block diagram showing a drive circuit for the direct drive motors D1 and D2. When a motor rotation command is input from an external computer, the motor control circuit DMC1 for the direct drive motor D1 and the motor control circuit DMC2 for the direct drive motor D2 respectively drive signals from the CPU to the three-layer amplifier (AMP). (Torque command value or current command value) is output, and a drive current is supplied from the three-layer amplifier (AMP) to the direct drive motors D1 and D2. As a result, the outer rotor members 21 and 21 'of the direct drive motors D1 and D1 rotate independently to move the arms A1 and A2 (FIG. 1). When the outer rotor members 21, 21 ′ rotate, resolver signals are output from the resolver stators 35, 36, 35 ′, 36 ′ whose rotation angles have been detected as described above, and are output to the resolver digital converter (RDC). The CPU input after the digital conversion in step 1 determines whether or not the outer rotor members 21 and 21 'have reached the command position, and stops the drive signal to the three-layer amplifier (AMP) if the command position is reached. Thus, the rotation of the outer rotor members 21 and 21 ′ is stopped. This enables servo control of the outer rotor members 21 and 21 '.

  When driving multiple axes in a vacuum environment, there is a possibility that the arm A1 or the like may hit the wall of the vacuum chamber or the shutter of the vacuum chamber if the current rotational positions of the arms A1 and A2 are not recognized when the power is turned on. However, in the present embodiment, the variable reluctance comprising the absolute resolver stators 36 and 36 ′ for detecting the absolute position of one rotation of the rotating shaft and the incremental resolver stators 35 and 35 ′ for detecting the rotational position with finer resolution. Since the mold resolver is employed, the rotational position control of the outer rotor members 21, 21 ′, that is, the arms A1 and A2, can be performed with high accuracy.

  Here, the resolver is adopted for detecting the rotation of the inner rotor 30. However, since the detector can be arranged on the atmosphere side inside the partition wall 13, a servo motor generally used for high-precision positioning is driven with high precision and smoothness. For example, an optical encoder that is employed as a position detecting means, a magnetic encoder using a magnetoresistive element, or the like can be used.

  Hereinafter, a method for detecting an abnormality in the bearings 19, 19 'or link support bearings used in the direct drive motors D1, D2 for driving the transfer robot will be described. FIG. 5 shows that in the direct drive motors D1 and D2 of the present embodiment, the rotor support bearings 19 and 19 ′ and the link support bearing are in a normal state, that is, at the earliest time when the operation as a transfer robot for semiconductor wafers is started. The torque command value (a) calculated inside the motor control circuit DMC1 and the inside of the motor control circuit DMC2 when the operation (FIG. 1) is performed to separate or approach (reverse rotation between D1 and D2) The torque command value (b) calculated by On the other hand, in FIG. 6, in the direct drive motors D1 and D2, an abnormality occurs in one or a plurality of locations of the bearings 19 and 19 ′, which are the rotor support bearings, and the link support bearings. The torque command value (a) calculated inside the motor control circuit DMC1 when the same operation as that in FIG. 5 is performed in the state where the torque fluctuation or the average torque is abnormally increased, and the inside of the motor control circuit DMC2 The torque command value (b) calculated by

  5 and 6, the absolute value of the torque command value of the motor control circuit DMC2 is large because the 19 ′ supporting the second outer rotor member (rotor) 21 ′ of the direct drive motor D2 is the direct drive motor. This is because it is loaded on the first outer rotor member (rotor) 21 of D1 and rotates at twice the rotational speed by the differential operation.

The torque command values in FIGS. 5 and 6 are the same for both motor control circuits DMC1 and DMC2.
(1) Iron loss such as switching loss and eddy current loss of partition walls,
(2) Cogging torque inherent to the direct drive motors D1 and D2,
(3) Kinetic energy for accelerating and decelerating the moments of inertia of the rotors 21, 21 'and the link mechanisms (A1, A2, L1, L2), etc.
(4) Friction torque of bearings 19, 19 ′ and link support bearings,
Is included.

  Among these factors, iron loss such as switching loss and bulkhead eddy current loss depends only on the rotational speed, and therefore does not change over time. As for the cogging torque, the amplitude does not change over time, and the phase does not change unless the initial angle of the operation for separating the table T is changed. Further, the moment of inertia of the rotors 21 and 21 'does not appear in the torque command when the rotor rotational speed is constant. The moment of inertia of the link mechanism (A1, A2, L1, L2) changes according to the position of the table T, but does not change with time.

  Therefore, abnormalities in the bearings 19, 19 ′ and link support bearings used in the direct drive motors D1, D2 constituting the transfer robot caused the table T to be separated or approached (reverse rotation between D1 and D2). The torque command value at the earliest time (first elapsed time) at which the operation of the transfer robot has started, and the torque command value after the predetermined time has elapsed (second elapsed time after the first elapsed time) Can be detected by comparing the difference (the difference between the solid line and the dotted line in FIG. 6). By comparing FIGS. 5 and 6, it is understood that an abnormality has occurred in any of the bearings in both the direct drive motors D1 and D2.

  FIG. 7 is a cross-sectional view similar to FIG. 2 according to the second embodiment. About this Embodiment, a different site | part is demonstrated with respect to embodiment of FIGS. 2-4, and it abbreviate | omits description by attaching | subjecting the same code | symbol about the site | part which has the same function.

  In the present embodiment, the annular portion 113a is airtightly bolted to the stepped portion 110a of the upper disc portion 110 attached to the upper surface of the cylindrical main body 112 via an O-ring OR. The lower portion of the annular portion 113a is a flange portion 113c that extends thinly outward in the radial direction. The upper end of the thin cylindrical portion 113b is TIG welded to the bent outer edge. The thickness of the attachment portion of the annular portion 113a is thicker than the thickness of the flange portion 113c and the thin cylindrical portion 113b. The lower end of the thin cylindrical portion 113b is TIG welded to the holder 15 as in the above-described embodiment. The annular portion 113a, the flange portion 113c, the thin cylindrical portion 113b, and the holder 15 constitute a partition wall 113. The disk portion 110, the main body 112, and the disk 10 constitute a housing.

  In the present embodiment, the upper surface of the upper disc portion 110 is closed by the lid member 101, and the bearing holder 107 attached to the outer periphery thereof supports the bearing 19 '. Therefore, the cylindrical member 123 of the direct drive motor D1 does not extend to the direct drive motor D2 side. The upper disk part 110, the lid member 101, and the bearing holder 107 are made of austenitic stainless steel having high corrosion resistance.

  The outer diameter portion 110a of the mounting seat surface of the bearing holder 107 in the upper disc portion 110 is located radially inward from the thin cylindrical portion 113b. Therefore, if the bearing holder 107 is removed from the upper disc portion 110, 2 The two outer rotor members 21, 21 ′ can be removed upward without disassembling the upper disk part 110. Therefore, it is not necessary to disassemble the airtight structure during maintenance or the like, and the work can be facilitated. That is, the maximum outer diameter portion of the housing (main body 12 and upper disk portion 110) supporting the partition wall structure is the outer rotor (outer rotor members 21 and 21 ′ and ring-shaped members 23 and 23 of the direct drive motors D1 and D2. Since it is smaller than the inner diameter of '), if the bearing holder 107 is removed from the housing, the outer rotors of the direct drive motors D1 and D2 can be removed from the partition wall 13, thereby facilitating inspection and removal. Will also improve. Furthermore, since only the bearing holder 107 needs to be removed, there is no need to remove the partition wall structure, and a leak check or the like is not required at the time of reassembly, thereby improving assemblability.

  Furthermore, in the present embodiment, the thickness of the flange portion 113c is thinner than the thickness of the annular portion 113a of the partition wall 113, so that the partition wall 113 has a dimensional accuracy, mechanical accuracy, and temperature change. Even when axial stretching stress is generated, the thin flange portion 113c is deformed, so that the axial stress and bending stress of the partition wall 113 can be relieved, thereby preventing a seal failure or breakage. . Moreover, since it is not necessary to process the annular portion 113a and the upper disk portion 110 and the main body 12 to which the annular portion 113a is attached with high accuracy, a lower cost direct drive motor can be provided.

In the present embodiment, a frog-leg transfer robot composed of direct drive motors D1 and D2 can be driven as in the above-described embodiment, but the fixing method of the rotor support bearings in direct drive motors D1 and D2 is different. ,
(1) the rotor support bearing 19 of the direct drive motor D1,
(2) the rotor support bearing 19 ′ of the direct drive motor D2,
(3) a link support bearing of the arm A1 and the link L1 to which the direct drive motor D1 is coupled;
(4) a link support bearing for the arm A2 and the link L2 to which the direct drive motor D2 is coupled;
It is possible to identify friction torque detection and abnormality.

  Hereinafter, means for separately detecting only the friction torque of each rotor support bearing will be described. FIG. 8 shows an operation of turning the table T (FIG. 1) in the configuration of the present embodiment when the rotor support bearings 19 and 19 ′ are in a normal state, that is, at the earliest time when the operation of the semiconductor wafer or the like is started. Torque command value (a) calculated inside the motor control circuit DMC1 and torque command value calculated inside the motor control circuit DMC2 when D1 and D2 rotate at the same speed in the same direction) (B) is shown. On the other hand, in FIG. 9, only in the direct drive motor D2, an abnormality occurs in the bearing 19 ′ that is the rotor support bearing, and as a result, spike-like torque fluctuations and an abnormal increase in average torque occur. 8 shows the torque command value (a) calculated inside the motor control circuit DMC1 and the torque command value (b) calculated inside the motor control circuit DMC2 when the same operation as in FIG. 8 is performed. Is.

  In such a turning operation of the table T, the arms A1 and A2 and the links L1 and L2 are not relatively displaced, so the link support bearing does not rotate, and in addition, the rotor support bearings 19 and 19 ′ are the same as those in the above-described embodiment. Therefore, only the friction torque of each of the rotor support bearings 19 and 19 ′ can be detected. Here, as shown in FIG. 9B, a torque spike is seen in the torque command value of the direct drive motor D2, and the friction torque is increased compared to the initial value, so that some abnormality has occurred in the bearing 19 ′. I understand that.

  Next, a means for separating and detecting only the friction torque of the link support bearing will be explained. FIG. 10 shows a table T (FIG. 1) in the configuration of the present embodiment, in which the rotor support bearings 19 and 19 'and the link support bearing are in a normal state, that is, at the very beginning of operation as a transfer robot for semiconductor wafers or the like. Torque command value (a) calculated in the motor control circuit DMC1 and torque command value calculated in the motor control circuit DMC2 when the separating operation (reverse rotation between D1 and D2) is performed (B) is shown. On the other hand, in FIG. 11, in the direct drive motors D1 and D2, an abnormality occurs in one or a plurality of locations of the bearing 19 ′, which is the rotor support bearing, and the link support bearing, resulting in spike-like torque fluctuations. In a state where the average torque is abnormally increased, the torque command value (a) calculated inside the motor control circuit DMC1 when the same operation as that in FIG. 10 is performed, and calculated inside the motor control circuit DMC2. The torque command value (b) to be performed is shown.

  In such a separation or approaching operation of the table T, since the link support bearing rotates together with the rotor support bearings 19 and 19 ′, the friction torque of the rotor support bearings 19 and 19 ′ and the friction torque of the link support bearing are mixed. The torque command value in the corrected state is shown. However, as shown in FIGS. 8 and 9, by removing the friction torque of only the rotor support bearings 19 and 19 ′ obtained by the turning operation, the link support bearings loaded on the respective direct drive motors D1 and D2. Only the friction torque can be detected. Here, as shown in FIG. 11A, a torque spike is seen in the torque command value of the direct drive motor D1, but it is determined from FIG. 9A that the bearing 19 is normal. It can be seen that some abnormality occurred only in the support bearing. Further, comparing FIG. 9B and FIG. 11B, the torque spike of the torque command value of the direct drive motor D2 is increased. It can be seen that an abnormality occurred.

  In the above embodiment, the example using the surface magnet type 32-pole 36-slot outer rotor brushless motor has been described. However, the present invention is not limited to this type of motor, and any brushless motor can be applied. Other magnetic pole types, for example, a permanent magnet embedded type, other slot combinations, or an inner rotor type may be used.

  Further, as a countermeasure against interference of each axis, a configuration may be adopted in which the number of rotor poles and the number of slots of adjacent axes in the axial direction differ. For example, in the case of 2-axis coaxial, the first axis is 32 poles and 36 slots, the second axis is 24 poles and 27 slots, and in the case of 4-axis coaxial, the first axis and the third axis are 32 poles and 36 slots, and the second axis If the fourth axis has a configuration of 24 poles and 27 slots, it is possible to prevent mutual interference such as generation of thrust in the rotating direction on the rotor and the magnetic coupling device due to the magnetic field of each axis.

  The rotor permanent magnet is a neodymium (Nd-Fe-B) -based magnet and has been described using a nickel coating as an example of a coating for enhancing corrosion resistance. However, the material is limited to this material and surface treatment. However, it should be changed as appropriate depending on the environment in which it is used. For example, a samarium-cobalt (Sm · Co) -based magnet that is difficult to demagnetize at high temperatures should be used depending on the temperature conditions during baking. If used in ultra-vacuum, a titanium nitride coating with a high outgas barrier should be applied.

  In addition, the yoke has been described using an example in which nickel is plated with low carbon steel as a material. However, the yoke is not limited to this material and surface treatment, and may be appropriately changed depending on the environment used. In particular, regarding surface treatment, if it is used in ultra-vacuum, it should be subjected to Kanigen plating, clean s plating, titanium nitride coating, etc. with few pinholes.

  The method of fastening the permanent magnet to the yoke has been described using an example in which a non-magnetic wedge is tightened with a screw from the outer diameter side of the yoke, but it is appropriately changed depending on the environment in which it is used. May be bonded or other fastening methods.

  In addition, the bearings 19 and 19 ′ have been described using an example of a grease grease four-point contact ball bearing for vacuum. However, the bearings 19 and 19 ′ are not limited to this type, material, and lubrication method. It can be changed as appropriate according to the rotational speed, etc., and may be a cross roller bearing. In the case of a 4-axis coaxial motor, it may be structured to be supported by another bearing in order to further increase mechanical rigidity. When multi-point contact bearings cannot be used, such as when rotating at high speeds, bearings that support the rotor of each axis and other bearings may be structured to apply preload as deep groove ball bearings or angular bearings, When used in, a metal lubrication that does not emit gas, such as plating of a soft metal such as gold or silver on the raceway, may be used.

  In addition, the inner rotor functioning as a magnetic coupling has been described as using a permanent magnet and a back yoke. However, the material and shape of the permanent magnet and the back yoke are not limited thereto. For example, depending on the mass of the resolver and the friction torque of the bearing, the number of poles may not be the same as that of the rotor, and the width may not be the same. A salient pole that does not use a permanent magnet may be used.

  Further, although an example in which a resolver is used as an angle detector has been described, it can be appropriately changed depending on manufacturing cost and resolution, and for example, an optical rotary encoder may be used.

  Moreover, although the example which used the grease lubrication 4-point contact ball bearing was demonstrated as bearings 33 and 33 'which rotatably support the rotation side of an angle detector, it is not limited to this form and a lubrication method. If the multi-point contact bearing cannot be used, such as high-speed rotation or reduction of friction torque, the angular bearing and deep groove ball bearing must be It is good also as a structure which arranges two for every axis | shaft and applies preload.

  In addition to the other partition walls, the structural parts and the material, shape, and manufacturing method of the partition parts and the partition walls are appropriately changed depending on the manufacturing cost, the environment used, the load conditions, the configuration, and the like.

  The present invention has been described above with reference to the embodiments. However, the present invention should not be construed as being limited to the above-described embodiments, and can be modified or improved as appropriate. For example, the motor system according to the present embodiment can be used not only in a vacuum atmosphere but also in an atmosphere outside the atmosphere. For example, in the case of a semiconductor manufacturing process, a reactive gas for etching may be introduced into the vacuum chamber after evacuation, but in the motor system of the present embodiment, the inside and the outside are shielded by the partition wall. Therefore, there is no possibility that the motor coil, the insulating material, and the like are etched.

1 is a perspective view of a frog-leg-arm type transfer robot using a motor system including a direct drive motor according to the present embodiment. It is the figure which cut | disconnected the structure of FIG. 1 by the II-II line | wire and looked at the arrow direction. It is a figure which shows the example of a resolver control circuit. It is a figure which shows the example of a motor control circuit. In the direct drive motors D1 and D2 of the present embodiment, the rotor support bearings 19 and 19 'and the link support bearing are in a normal state, that is, at the earliest time when the operation as a transfer robot for semiconductor wafers is started, the table T (FIG. 1) Torque command value (a) calculated inside the motor control circuit DMC1 and the torque command calculated inside the motor control circuit DMC2 when performing the operation of separating them (reverse rotation between D1 and D2) It is the graph which showed value (b) with time. In the direct drive motors D1 and D2, an abnormality occurs in one or more of the bearings 19 and 19 ′, which are the rotor support bearings, and the link support bearing, resulting in spike-like torque fluctuations and an abnormal increase in average torque. When the same operation as that in FIG. 5 is performed, the torque command value (a) calculated inside the motor control circuit DMC1 and the torque command value calculated inside the motor control circuit DMC2 ( It is the graph which showed b) with time. It is sectional drawing which shows 2nd Embodiment. In the direct drive motors D1 and D2 in FIG. 7, the rotor support bearings 19 and 19 ′ are in a normal state, that is, the operation of turning the table T (FIG. 1) at the very beginning of operation as a transfer robot for semiconductor wafers (D1). And D2 are rotated at the same speed in the same direction), the torque command value (a) calculated inside the motor control circuit DMC1 and the torque command value calculated inside the motor control circuit DMC2 ( It is the graph which showed b) with time. In the direct drive motors D1 and D2 in FIG. 7, an abnormality has occurred in one or a plurality of locations of the rotor support bearings 19 and 19 ′, and as a result, spike-like torque fluctuations and abnormal increase in average torque have occurred. 8, the torque command value (a) calculated inside the motor control circuit DMC1 and the torque command value (b) calculated inside the motor control circuit DMC2 when performing the same operation as in FIG. It is the graph shown in. In the direct drive motors D1 and D2 of FIG. 7, the rotor support bearings 19 and 19 ′ and the link support bearing are in a normal state, that is, at the very beginning of operation as a transfer robot for semiconductor wafers, the table T (FIG. 1) is separated. Torque command value (a) calculated in the motor control circuit DMC1 and torque command value (inside the motor control circuit DMC2) when the operation (reverse rotation between D1 and D2) is performed. It is the graph which showed b) with time. In the direct drive motors D1 and D2 in FIG. 7, an abnormality occurs in one or a plurality of portions of the rotor support bearings 19 and 19 'and the link support bearing, and as a result, spike-like torque fluctuations and an abnormal increase in average torque occur. In the generated state, the torque command value (a) calculated inside the motor control circuit DMC1 and the torque command value (b) calculated inside the motor control circuit DMC2 when performing the same operation as in FIG. ).

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Disc 11 Bolt 12 Main body 12a Flange part 12b Outer periphery 12d Hole 12e Notch 13, 113 Partition 15 Holder 16 Bolt 17, 17 'Bearing holder 18, 18' Bolt 19, 19 'Four-point contact ball bearing 20, 20' Bolt 21, 21 'Outer rotor member 21a, 21a' Permanent magnet 21b, 21b 'Yoke 22, 22' Bolt 23 'Ring member 23 Cylindrical member 24, 24' Bolt 25, 25 'Magnetic shield plate 29, 29' Stator 30 30 ′ inner rotor 30a, 30a ′ permanent magnet 30b, 30b ′ back yoke 32, 32 ′ resolver holder 33, 33 ′ bearing 34, 34 ′ detection rotor 34a, 34a ′ incremental resolver rotor 34b, 34b ′ absolute resolver rotor 35, 35 'Incremental resolver stator 3 6, 36 'Absolute resolver stator 41 Magnetic shield plate 101 Lid member 107 Bearing holder 110 Upper disk portion 113a Bulkhead holder 113b Thin cylinder 125 Magnetic shield plates A1, A2 Arm D1, D2 Direct drive motor DMC1 Motor control circuit DMC2 Motor control circuit G Surface plate L1, L2 Link OR O-ring T Table

Claims (2)

In the transfer robot that drives the table via the link mechanism by the power from the two direct drive motors,
Each direct drive motor
A housing;
A partition wall extending from the housing and isolating the atmosphere side from the atmosphere outside;
A rotor disposed outside the atmosphere with respect to the partition;
A rotor support bearing disposed outside the atmosphere with respect to the partition wall and rotatably supporting the rotor;
A stator disposed on the atmosphere side with respect to the partition wall and driving the rotor;
An angle detector for detecting a rotation angle of the outer rotor;
A motor control circuit for energizing the stator based on a torque command value or a current command value calculated according to the rotation angle of the outer rotor detected by the angle detector;
By comparing the torque command value or the current command value calculated at a predetermined time by the motor control circuit, the friction torque of the link support bearing disposed at the joint portion of the rotor support bearing and the link mechanism is detected. A transport robot characterized by that. The angle detector is
An outer magnetic coupling rotor disposed outside the atmosphere with respect to the partition wall and rotating together with the rotor;
An inner magnetic coupling rotor disposed on the atmosphere side with respect to the partition wall and rotating in synchronization with the outer magnetic coupling rotor by a magnetic coupling action;
An inner bearing disposed on the atmosphere side with respect to the partition wall and rotatably supporting the inner magnetic coupling rotor;
A resolver device arranged on the atmosphere side with respect to the partition wall,
The transport robot according to claim 1, wherein the resolver rotor of the resolver device is configured to be rotatably supported by an inner bearing together with the inner magnetic coupling rotor.

Images