JP2020079954A - Substrate processing apparatus - Google Patents

Abstract

To minimize an Abbe error in measurement, in encoder measurement of a position in a circumferential direction of a cylindrical member.SOLUTION: A substrate processing apparatus is provided that comprises: a rotary drum, a first processing part which processes a sheet substrate at a first specific position within a range in a circumferential direction at which the sheet substrate is closely supported by the rotary drum; a scale disc which is provided on the end face side in a direction of a central line of the rotary drum so as to rotate around the central line together with the rotary drum, and has a scale annularly engraved at a predetermined pitch in the circumferential direction of the rotation of the rotary drum formed on a side face vertical to the central line, for encoder measuring a change in a position in the circumferential direction of the sheet substrate; and a first encoder head which is in the same direction as the first specific position related to the circumferential direction of the rotation of the rotary drum, is arranged so as to face a side face of the scale disc and reads out the scale, wherein a radius of an outer peripheral surface of the rotary drum and a distance from the central line of the read-out position of the scale by the first encoder head are the same.SELECTED DRAWING: Figure 23

Description

  The present invention relates to a substrate processing apparatus.

  Among exposure apparatuses used in a photolithography process, there is known an exposure apparatus that exposes a substrate by using a cylindrical or cylindrical mask as disclosed in the following Patent Document (for example, Patent Document 1).

  In order to satisfactorily project and expose the image of the pattern of the mask onto the substrate not only when using the plate-shaped mask but also when exposing the substrate using the cylindrical or cylindrical mask, Patent Document 1 discloses By forming a mark (scale, alignment mark, etc.) for position information with a predetermined positional relationship with respect to the pattern in a predetermined area of the pattern forming surface of the cylindrical mask, and detecting the scale with an encoder system, A configuration for acquiring position information of a pattern in the circumferential direction (or the rotation axis direction) of the pattern forming surface is described.

  Further, in order to continuously expose a flexible long sheet substrate by using a cylindrical mask, the long sheet substrate is wound around and supported by a feed roller, and the sheet substrate wound around the feed roller is cylindrical. There is also proposed an exposure apparatus which enables device production (exposure processing) with high mass productivity by bringing a mask close to each other and rotating a feed roller and a cylindrical mask (for example, refer to Patent Document 2).

JP-A-7-153672 JP-A-8-213305

  In the processing apparatus that processes the object to be processed (sheet substrate) on the curved surface of the cylindrical member such as the feed roller as described above, it is required to accurately detect the position of the cylindrical member in the circumferential direction. Similar to Patent Document 1, even when a mark (scale) for position information acquisition is engraved on the outer peripheral surface of the cylindrical member for encoder measurement, if the measurement Abbe condition largely deviates, manufacturing error of scale scale Or, due to expansion and contraction due to temperature, an error may occur in the result of encoder measurement, and the accuracy of detecting the position of the cylindrical member in the circumferential direction may decrease.

  It is an object of an aspect of the present invention to reduce an Abbe error during measurement when measuring the position of a cylindrical member in the circumferential direction by an encoder.

  According to the first aspect of the present invention, there is provided a substrate processing apparatus that conveys a long sheet substrate having flexibility in a long direction and performs a predetermined process on the sheet substrate, the substrate processing apparatus being fixed from a center line. A rotating drum that closely supports a part of the sheet substrate in the longitudinal direction with an outer peripheral surface that is curved in a cylindrical shape with a radius, and rotates around the center line to convey the sheet substrate in the longitudinal direction, A first processing unit that processes the sheet substrate at a first specific position within a circumferential range in which the sheet substrate is closely supported by the rotating drum, and encoder position change in the circumferential direction of the sheet substrate is measured. In order to do so, it is provided on the end face side of the rotating drum in the direction of the center line so as to rotate around the center line together with the rotating drum, and annularly at a predetermined pitch along the circumferential direction of rotation of the rotating drum. A scale disk having engraved scales formed on a side surface perpendicular to the center line; and a side surface of the scale disk that has the same orientation as the first specific position in the circumferential direction of rotation of the rotary drum. A first encoder head that is arranged so as to face each other and that reads the scale; and a radius of the outer peripheral surface of the rotary drum, and a read position of the scale by the first encoder head from the center line. Provided is a substrate processing apparatus in which the distance is set to be the same.

FIG. 1 is a schematic diagram showing an overall configuration of a substrate processing apparatus (exposure apparatus) according to the embodiment. FIG. 2 is a schematic diagram showing the arrangement of the illumination area and the projection area in FIG. FIG. 3 is a schematic diagram showing a configuration of a projection optical system applied to the substrate processing apparatus (exposure apparatus) of FIG. FIG. 4 is a perspective view of a second drum member (rotary drum) applied to the substrate processing apparatus (exposure apparatus) of FIG. FIG. 5 is a perspective view for explaining the relationship between the detection probe and the reading device applied to the substrate processing apparatus (exposure apparatus) of FIG. FIG. 6 is an explanatory diagram of the positions of the encoder scale disk and the reading device according to the embodiment as seen in a plane orthogonal to the rotation center line. FIG. 7 is an enlarged view schematically showing the scale scale. FIG. 8 is a schematic diagram showing the positional relationship between the scale and the encoder head. FIG. 9 is a flowchart showing a procedure for correcting the scale pitch error of the scale. FIG. 10 is a diagram showing the relationship between the scale disk having scales on the outer peripheral surface and the encoder head. FIG. 11 is a diagram showing an example of the correction map. FIG. 12 is a conceptual diagram showing the timing when these read values are acquired from the pair of encoder heads. FIG. 13 is a conceptual diagram showing the timing when these read values are acquired from the pair of encoder heads. FIG. 14 is a flowchart showing the procedure for correcting the scale error. FIG. 15 is an explanatory diagram for explaining a roundness adjusting mechanism that adjusts the roundness of the encoder scale disk. FIG. 16 is an explanatory diagram for explaining a roundness adjusting mechanism that adjusts the roundness of the encoder scale disk. FIG. 17 is a schematic diagram showing a first modification of the substrate processing apparatus (exposure apparatus). FIG. 18 is an explanatory diagram for explaining the position of the reading device when the encoder scale disk according to the first modified example of the substrate processing apparatus (exposure apparatus) is viewed in the rotation center line direction. FIG. 19 is a schematic diagram showing the overall configuration of a second modified example of the substrate processing apparatus (exposure apparatus). FIG. 20 is a schematic diagram showing the overall configuration of a third modification of the substrate processing apparatus (exposure apparatus). FIG. 21 is a schematic diagram showing the overall configuration of a fourth modification of the substrate processing apparatus (exposure apparatus). FIG. 22 is a signal waveform diagram for simply explaining the actual reading operation of the scale by the encoder head shown in each of FIGS. 4 to 6, 10 and 16 to 21. FIG. 23 is a diagram showing the configuration in the case of forming a scale on the side end surface of the scale disk, as seen from the direction in which the rotation center line extends, as in the case of FIG. 24 is a cross-sectional view taken along the line A-A′ of the configuration of FIG. 23 taken along a plane including the installation azimuth line and the rotation center line. FIG. 25 is a diagram showing the arrangement of the scale disk and the encoder head in the XZ plane, as in FIG. 6 described above. FIG. 26 is a flowchart showing the procedure of a device manufacturing method for manufacturing a device using the substrate processing apparatus (exposure apparatus) according to the embodiment.

  Modes (embodiments) for carrying out the present invention will be described in detail with reference to the drawings. The present invention is not limited to the embodiments described below. The following embodiments will be described as an exposure apparatus used in a so-called roll-to-roll system in which various types of processing for manufacturing one type of device are continuously performed on a substrate. Further, in the following, an XYZ rectangular coordinate system is set, and the positional relationship of each part will be described with reference to the XYZ rectangular coordinate system. As an example, the predetermined direction in the horizontal plane is the X-axis direction, the direction orthogonal to the X-axis direction in the horizontal plane is the Y-axis direction, and the direction orthogonal to each of the X-axis direction and the Y-axis direction (that is, the vertical direction) is the Z-axis direction. And

  FIG. 1 is a schematic diagram showing an overall configuration of a substrate processing apparatus (exposure apparatus) according to the embodiment. FIG. 2 is a schematic diagram showing the arrangement of the illumination area and the projection area in FIG. FIG. 3 is a schematic diagram showing a configuration of a projection optical system applied to the substrate processing apparatus (exposure apparatus) of FIG. As shown in FIG. 1, the substrate processing apparatus 11 includes an exposure apparatus (processing section) EX and a sheet substrate transport device (hereinafter, appropriately referred to as a transport device) 9. The exposure apparatus EX is supplied with the substrate P (sheet, film, etc.) by the transportation apparatus 9. For example, a substrate P of a flexible long sheet pulled out from a supply roll (not shown) is sequentially processed by a substrate processing device (exposure device) 11 through a substrate processing device for a pre-process, There is a device manufacturing system in which, after being sent to a substrate processing apparatus for a post process by a transfer device 9, it is wound up on a recovery roll. In this way, the substrate processing apparatus 11 can be used as a part of a device manufacturing system (flexible display manufacturing line).

  The exposure apparatus EX as the substrate processing apparatus 11 is a so-called scanning exposure apparatus, and is a pattern formed on the cylindrical mask DM while driving the rotation of the cylindrical mask DM and the feeding of the flexible substrate P in synchronization with each other. Image is projected onto the substrate P via the projection optical system PL (PL1 to PL6) having a projection magnification of 1× (1). In the exposure apparatus EX shown in FIG. 1, the Y axis of the XYZ orthogonal coordinate system is set parallel to the rotation center line AX1 of the first drum member 21 that forms the cylindrical mask DM. Similarly, the rotation center line AX2 of the second drum member 22 as a cylindrical member that indicates a part of the substrate P in the longitudinal direction in a cylindrical shape is set parallel to the Y axis of the XYZ orthogonal coordinate system.

  As shown in FIG. 1, the exposure apparatus EX includes a mask holding device 12, an illumination mechanism IU, a projection optical system PL, and a control device 14. The exposure apparatus EX moves (rotates) the cylindrical mask DM held by the mask holding apparatus 12 and transfers the substrate P by the transfer apparatus 9. A part (illumination region IR) of the cylindrical mask DM held by the mask holding device 12 together with the illumination mechanism IU is illuminated by the illumination luminous flux EL1 with uniform brightness. The projection optical system PL projects the image of the pattern in the illumination region IR on the cylindrical mask DM onto a part of the substrate P (projection region PA) being transported by the transport device 9. With the movement of the cylindrical mask DM, the portion on the cylindrical mask DM arranged in the illumination region IR changes. Further, as the substrate P moves, the part on the substrate P arranged in the projection area PA changes. By doing so, the image of the predetermined pattern (mask pattern) formed on the surface of the cylindrical mask DM is projected onto the cylindrical surface of the substrate P via the projection optical system PL (PL1 to PL6). .. The control device 14 controls each unit of the exposure apparatus EX and causes each unit to execute processing. Further, in the present embodiment, the control device 14 controls the transport device 9.

  The control device 14 may be a part or all of a higher-level control device that collectively controls a plurality of substrate processing devices of the device manufacturing system described above. Further, the control device 14 may be a device that is controlled by the host control device and is different from the host control device. The control device 14 includes, for example, a computer system. The computer system includes, for example, a CPU (Central Processing Unit), various memories, an OS (Operating System), and hardware such as peripheral devices. The operation sequence, parameters, and the like of each unit of the substrate processing apparatus 11 are stored in a computer-readable recording medium in the form of a computer program, and various processes are performed by the computer system reading and executing the program.

  The computer system also includes a homepage providing environment (or display environment) when it can be connected to the Internet or an intranet system. The computer-readable recording medium includes a flexible disk, a magneto-optical disk, a ROM, a portable medium such as a CD-ROM, and a storage device such as a hard disk built in the computer system. The computer-readable recording medium holds a computer program dynamically for a short period of time, such as a communication line for transmitting the program through a network such as the Internet or a communication line such as a telephone line. In some cases, such as a volatile memory inside a computer system that is a server or a client, which holds a program for a certain period of time. Further, the computer program may be for realizing a part of the functions of the substrate processing apparatus 11, or may be for realizing the functions of the substrate processing apparatus 11 in combination with a program already recorded in the computer system. The host controller can be realized by using a computer system, like the controller 14.

  As shown in FIG. 1, in the mask holding device 12, the first drum 26 is held by the first drum member 21 holding the cylindrical mask DM, the guide roller 23 supporting the first drum member 21, and the control command from the controller 14. A drive roller 24 that drives the first drum member 21 and a first detector 25 that detects the position of the first drum member 21 are provided.

  The first drum member 21 is a cylindrical member having a curved surface that is curved at a constant radius from a rotation center line AX1 (hereinafter, also referred to as the first center axis AX1 as appropriate) that serves as a predetermined axis, and is located around the rotation center line AX1. To rotate. The first drum member 21 has a first surface P1 on which the illumination area IR of the cylindrical mask DM is arranged, and the first surface P1 has a line segment (bus line) and a first central axis parallel to this line segment. It is a cylindrical surface formed by rotating around AX1. The cylindrical surface is, for example, an outer peripheral surface of a cylinder or an outer peripheral surface of a cylinder. The first drum member 21 is made of, for example, glass, quartz, or the like, has a cylindrical shape having a constant thickness, and its outer peripheral surface (cylindrical surface) serves as the first surface P1. That is, in the present embodiment, the illumination region IR of the cylindrical mask DM is curved in a cylindrical surface shape having a constant radius r1 from the rotation center line AX1. As described above, the first drum member 21 has a curved surface (a cylindrical surface having a predetermined curvature) curved at a constant radius from the rotation center line AX1.

  The cylindrical mask DM is prepared as a transmissive flat sheet mask in which a pattern is formed on one surface of a strip-shaped ultrathin glass plate having a high flatness (for example, a thickness of 100 μm to 500 μm) by a light shielding layer such as chromium. To be done. The mask holding device 12 is used in a state in which the cylindrical mask DM made of an ultrathin glass plate is curved following the curved surface of the outer peripheral surface of the first drum member 21 and is wound (attached) on this curved surface. The cylindrical mask DM has a pattern non-formation region in which no pattern is formed, and the pattern non-formation region is attached to the first drum member 21. The cylindrical mask DM can be attached to and detached from the first drum member 21.

  The cylindrical mask DM is made of an ultra-thin glass plate, and instead of winding the cylindrical mask DM around the first drum member 21 made of a transparent cylindrical base material, the first drum member 21 is made of a transparent cylindrical base material such as quartz. However, a mask pattern made of a light shielding layer such as chrome may be directly formed on the outer peripheral surface by drawing. In this case as well, the first drum member 21 functions as a support member for the pattern of the cylindrical mask DM.

  The first detector 25 optically detects the rotational position of the first drum member 21, and is composed of, for example, a rotary encoder. The encoder may be absolute or increment format. The first detector 25 outputs information indicating the detected rotational position of the first drum member 21, for example, a two-phase signal from an encoder head, which will be described later, to the control device 14. The first drive unit 26 including an actuator such as an electric motor adjusts the torque and the rotation speed for rotating the drive roller 24 according to a control signal input from the control device 14. The control device 14 controls the rotational position of the first drum member 21 by controlling the first drive unit 26 based on the detection result of the first detector 25. Then, the control device 14 controls one or both of the rotation position and the rotation speed of the cylindrical mask DM held by the first drum member 21.

  The transport device 9 includes a drive roller DR4, a first guide member 31, a second drum member 22 forming a second surface P2 on which the projection area PA of the substrate P is arranged, a second guide member 33, drive rollers DR4, DR5, and The second detector 35 and the second drive unit 36 are provided.

  In the present embodiment, the substrate P that has been conveyed to the drive roller DR4 from the upstream side of the conveyance path of the substrate P, that is, the side opposite to the conveyance (movement) direction of the substrate P, passes through the drive roller DR4 and passes through the first guide. It is conveyed to the member 31. The substrate P that has passed through the first guide member 31 is supported by the surface of the cylindrical or cylindrical second drum member 22 having a radius r2, and is transported to the second guide member 33. The substrate P that has passed through the second guide member 33 is transported downstream of the transport path. The rotation center line AX2 of the second drum member 22 and each rotation center line of the drive rollers DR4 and DR5 are set to be parallel to the Y axis.

  For example, the first guide member 31 and the second guide member 33 move in the transport direction of the substrate P to adjust the tension in the transport direction that acts on the substrate P in the transport path. The first guide member 31 (and the drive roller DR4) and the second guide member 33 (and the drive roller DR5) are, for example, the width direction of the substrate P (the direction orthogonal to the transport direction of the substrate P, and the Y direction). ), the position in the Y direction of the substrate P wound around the outer periphery of the second drum member 22 can be adjusted. It should be noted that the carrier device 9 only needs to be able to carry the substrate P along the projection area PA of the projection optical system PL, and the configuration of the carrier device 9 can be appropriately changed.

  The second drum member 22 is a cylindrical member having a curved surface (cylindrical surface with a predetermined curvature) curved at a constant radius from a rotation center line AX2 (hereinafter also referred to as a second center axis AX2 as appropriate) serving as a predetermined axis. , A rotary drum that rotates around the second central axis AX2. The second drum member 22 forms a second surface (support surface) P2. The second surface P2 is a part of the substrate P onto which the imaging light flux from the projection optical system PL is projected, and supports a part including the projection area PA in an arc shape (cylindrical shape). In the present embodiment, the second drum member 22 is a part of the transport device 9 and also serves as a support member (substrate stage) that supports the substrate P to be exposed. That is, the second drum member 22 may be a part of the exposure apparatus EX. In this way, the second drum member 22 is rotatable about its rotation center line AX2 (second center axis AX2), and the substrate P follows the outer peripheral surface (cylindrical surface) of the second drum member 22. And the projection area PA is arranged on a part of the curved substrate P. Therefore, the substrate P turns and moves in the peripheral surface portion including the projection area PA of the cylindrical surface having the radius r2.

  In the present embodiment, the second drum member 22 rotates by the torque supplied from the second drive unit 36 including an actuator such as an electric motor. The second detector 35 is composed of, for example, a rotary encoder or the like, and optically detects the rotational position of the second drum member 22. The second detector 35 outputs information indicating the detected rotational position of the second drum member 22 (for example, a two-phase signal from encoder heads EN1, EN2, EN3, EN4, EN5 described later) to the control device 14. .. The second drive unit 36 adjusts the torque and the rotation speed for rotating the second drum member 22 according to the control signal supplied from the control device 14. The control device 14 controls the rotational position of the second drum member 22 by controlling the second drive unit 36 based on the detection result of the second detector 35, and controls the first drum member 21 (cylindrical mask DM). The second drum member 22 is synchronously moved (synchronized rotation). The details of the second detector 35 will be described later.

  The exposure apparatus EX is an exposure apparatus on the assumption that a so-called multi-lens type projection optical system PL is mounted. The projection optical system PL includes a plurality of projection modules that project a part of the image of the pattern of the cylindrical mask DM. For example, in FIG. 1, the left side of the center plane P3 including the rotation center line AX1 of the cylindrical mask DM and the second center axis AX2 of the second drum member 22 and parallel to the YZ plane (the transfer direction of the substrate P is Three projection modules (projection optical systems) PL1, PL3, PL5 are arranged on the opposite side) at regular intervals in the Y direction, and three projection modules (projection) on the right side of the center plane P3 (the side in which the substrate P is conveyed). Optical system) PL2, PL4, PL6 are arranged at regular intervals in the Y direction.

  In such a multi-lens type exposure apparatus EX, the end portions in the Y direction of the areas (projection areas PA1 to PA6) exposed by the plurality of projection modules PL1 to PL6 are superposed on each other by scanning to form a desired pattern. Project the whole picture. In such an exposure apparatus EX, even when the size of the pattern on the cylindrical mask DM in the Y direction becomes large and it becomes necessary to handle the substrate P having a large width in the Y direction, the projection module PL and the projection apparatus PL are projected. Since it is only necessary to add a module on the illumination mechanism IU side corresponding to the module PL in the Y direction, there is an advantage that the display panel size (width of the substrate P) can be easily increased.

  The exposure apparatus EX does not have to be of the multi-lens type. For example, when the dimension of the substrate P in the width direction is small to some extent, the exposure apparatus EX may project the image of the full width of the pattern on the substrate P by one projection module. Further, each of the plurality of projection modules PL1 to PL6 may project a pattern corresponding to one device. That is, the exposure apparatus EX may project patterns for a plurality of devices in parallel by a plurality of projection modules.

  The illumination mechanism IU includes a light source device 13 and an illumination optical system. The illumination optical system includes a plurality (for example, six) of illumination modules IL arranged in the Y-axis direction corresponding to each of the plurality of projection modules PL1 to PL6. The light source device 13 includes, for example, a lamp light source such as a mercury lamp, a solid-state light source such as a laser diode and a light emitting diode (LED), or a gas laser light source. Illumination light emitted from the light source device is, for example, bright lines (g line, h line, i line) emitted from a lamp light source, far ultraviolet light (DUV light) such as KrF excimer laser light (wavelength 248 nm), ArF excimer laser light. (Wavelength 193 nm) and the like. The illumination light emitted from the light source device 13 has a uniform illuminance distribution, and is distributed to the plurality of illumination modules IL via a light guide member such as an optical fiber.

  Each of the plurality of illumination modules IL includes a plurality of optical members such as lenses. In the present embodiment, the light emitted from the light source device 13 and passing through any of the plurality of illumination modules IL is referred to as an illumination luminous flux EL1. Each of the plurality of illumination modules IL includes, for example, an integrator optical system, a rod lens, a fly-eye lens, and the like, and illuminates the illumination area IR with the illumination light flux EL1 having a uniform illuminance distribution. In the present embodiment, the plurality of illumination modules IL are arranged inside the cylindrical mask DM. Each of the plurality of illumination modules IL illuminates each illumination region IR of the mask pattern formed on the outer peripheral surface of the cylindrical mask DM from the inside of the cylindrical mask DM.

  FIG. 2 is a diagram showing an arrangement of the illumination area IR and the projection area PA in this embodiment. In FIG. 2, a plan view (left side view in FIG. 2) of the illumination area IR on the cylindrical mask DM arranged on the first drum member 21 as viewed from the −Z side and a second drum member 22 are illustrated. A plan view (right side view in FIG. 2) of the projection area PA on the arranged substrate P viewed from the +Z side is illustrated. The symbol Xs in FIG. 2 indicates the rotation direction (movement direction) of the first drum member 21 (cylindrical mask DM) or the second drum member 22.

  The plurality of illumination modules IL illuminate the first to sixth illumination regions IR1 to IR6 on the cylindrical mask DM, respectively. For example, the first illumination module IL illuminates the first illumination area IR1 and the second illumination module IL illuminates the second illumination area IR2.

  The first illumination region IR1 is described as a trapezoidal region elongated in the Y direction, but in the case of a projection optical system that forms an intermediate image plane, such as the projection optical system (projection module) PL, the position of the intermediate image. A field diaphragm plate having a trapezoidal aperture can be arranged at. Therefore, the first illumination area IR1 may be a rectangular area including the trapezoidal opening. The third illumination region IR3 and the fifth illumination region IR5 are regions having the same shape as the first illumination region IR1, and are arranged at regular intervals in the Y-axis direction. The second illumination region IR2 is a trapezoidal (or rectangular) region that is symmetrical to the first illumination region IR1 with respect to the center plane P3. The fourth illumination region IR4 and the sixth illumination region IR6 are regions having the same shape as the second illumination region IR2, and are arranged at regular intervals in the Y-axis direction.

  As shown in FIG. 2, each of the first to sixth illumination regions IR1 to IR6 is a hypotenuse of the trapezoidal illumination regions adjacent to each other in the Y-axis direction when viewed along the circumferential direction of the first surface P1. Are arranged so that the triangular parts of are overlapped (overlapped). Therefore, for example, the first area A1 on the cylindrical mask DM passing through the first illumination area IR1 by the rotation of the first drum member 21 is the cylindrical mask DM passing through the second illumination area IR2 by the rotation of the first drum member 21. It partially overlaps the upper second area A2.

  In the present embodiment, the cylindrical mask DM includes a pattern forming area A3 in which a pattern is formed and a pattern non-forming area A4 in which a pattern is not formed. The pattern non-forming area A4 is arranged so as to surround the pattern forming area A3 in a frame shape, and has a characteristic of blocking the illumination luminous flux EL1. The pattern forming area A3 of the cylindrical mask DM moves in the movement direction Xs with the rotation of the first drum member 21, and each partial area of the pattern forming area A3 in the Y-axis direction is the first to sixth illumination areas. Pass any of IR1 to IR6. That is, the first to sixth illumination regions IR1 to IR6 are arranged so as to cover the entire width of the pattern forming region A3 in the Y-axis direction.

  As shown in FIG. 1, each of the plurality of projection modules PL1 to PL6 arranged in the Y-axis direction corresponds to each of the first to sixth illumination modules IL in a one-to-one correspondence, and the illumination is performed by the corresponding illumination module IL. The image of the partial pattern of the cylindrical mask DM appearing in the illuminated area IR is projected onto each projection area PA on the substrate P. For example, the first projection module PL1 corresponds to the first illumination module IL, and an image of the pattern of the cylindrical mask DM in the first illumination region IR1 (see FIG. 2) illuminated by the first illumination module IL on the substrate P. Is projected on the first projection area PA1. The third projection module PL3 and the fifth projection module PL5 correspond to the third to fifth illumination modules IL, respectively. The third projection module PL3 and the fifth projection module PL5 are arranged at positions overlapping the first projection module PL1 when viewed in the Y-axis direction.

  In addition, the second projection module PL2 corresponds to the second illumination module IL, and an image of the pattern of the cylindrical mask DM in the second illumination region IR2 (see FIG. 2) illuminated by the second illumination module IL on the substrate P. To the second projection area PA2. The second projection module PL2 is arranged in a symmetrical position with respect to the first projection module PL1 with the center plane P3 interposed therebetween when viewed from the Y-axis direction.

  The fourth projection module PL4 and the sixth projection module PL6 are arranged corresponding to the fourth and sixth illumination modules IL, respectively, and the fourth projection module PL4 and the sixth projection module PL6 are viewed from the Y-axis direction, It is arranged at a position overlapping the second projection module PL2. With such an arrangement, the odd-numbered first projection area PA1, third projection area PA3, and fifth projection area PA5 are displaced from the center plane P3 by a certain amount in the -X direction and arranged in a line in the Y-axis direction. To be done. The even-numbered second projection area PA2, fourth projection area PA4, and sixth projection area PA6 are arranged in line in the Y-axis direction with a certain amount of deviation from the center plane P3 in the +X direction.

  In the present embodiment, the light reaching each of the illumination regions IR1 to IR6 on the cylindrical mask DM from each illumination module IL of the illumination mechanism IU is an illumination light flux EL1. Further, the light that has been subjected to intensity distribution modulation according to the pattern of the cylindrical mask DM appearing in each of the illumination areas IR1 to IR6 and is incident on each of the projection modules PL1 to PL6 and reaching each of the projection areas PA1 to PA6 is an imaging light flux EL2. And Then, of the imaging light flux EL2 reaching each of the projection areas PA1 to PA6, the chief ray passing through each center point of the projection areas PA1 to PA6 is, as shown in FIG. 1, a second central axis AX2 of the second drum member 22. Seen from the above, they are arranged at positions (specific positions) at an angle θ in the circumferential direction with the center plane P3 interposed therebetween.

  In each of the first to sixth projection areas PA1 to PA6, the end portions (trapezoidal triangular portions) of the projection areas (odd-numbered and even-numbered) adjacent to each other in the direction parallel to the second central axis AX2 are the second surface. It is arranged so as to overlap in the circumferential direction of P2. Therefore, for example, the third area A5 on the substrate P passing through the first projection area PA1 by the rotation of the second drum member 22 is on the substrate P passing through the second projection area PA2 by the rotation of the second drum member 22. It partially overlaps with the fourth area A6. The first projection area PA1 and the second projection area PA2 are arranged such that the exposure amount in the region where the third region A5 and the fourth region A6 overlap is substantially the same as the exposure amount in the non-overlapping region. Shape etc. are set.

  Next, a detailed configuration of the projection optical system PL of this embodiment will be described with reference to FIG. In addition, in the present embodiment, each of the second projection module PL2 to the fifth projection module PL5 has the same configuration as the first projection module PL1. Therefore, the configuration of the first projection module PL1 will be described as a representative of the projection optical system PL, and description of each of the second projection module PL2 to the fifth projection module PL5 will be omitted.

  The first projection module PL1 shown in FIG. 3 includes a first optical system 41 for forming an image of the pattern of the cylindrical mask DM arranged in the first illumination region IR1 on the intermediate image plane P7, and the first optical system 41. The second optical system 42 for re-imaging at least a part of the intermediate image thus formed on the first projection area PA1 of the substrate P, and the first field stop 43 arranged on the intermediate image plane P7 on which the intermediate image is formed. ..

  The first projection module PL1 also includes a focus correction optical member 44, an image shift correction optical member 45, a rotation correction mechanism 46, and a magnification correction optical member 47. The focus correction optical member 44 is a focus adjustment device that finely adjusts the focus state of a pattern image (hereinafter, referred to as a projected image) of a mask formed on the substrate P. The image shift correction optical member 45 is a shift adjustment device that slightly laterally shifts the projected image in the image plane. The magnification correction optical member 47 is a shift adjustment device that slightly corrects the magnification of the projected image. The rotation correction mechanism 46 is a shift adjustment device that slightly rotates the projected image in the image plane.

  The imaging light flux EL2 from the pattern of the cylindrical mask DM is emitted from the first illumination region IR1 in the normal direction (D1), passes through the focus correction optical member 44, and is incident on the image shift correction optical member 45. The imaging light flux EL2 that has passed through the image shift correction optical member 45 is reflected by the first reflecting surface (plane mirror) p4 of the first deflecting member 50, which is an element of the first optical system 41, and passes through the first lens group 51. The light is reflected by the first concave mirror 52 arranged at the pupil position, passes through the first lens group 51 again, is reflected by the second reflecting surface (flat mirror) p5 of the first deflecting member 50, and enters the first field stop 43. .. The image-forming light flux EL2 that has passed through the first field stop 43 is reflected by the third reflecting surface (flat mirror) p8 of the second deflecting member 57 that is an element of the second optical system 42, passes through the second lens group 58, and exits to the pupil. It is reflected by the second concave mirror 59 arranged at the position, passes through the second lens group 58 again, is reflected by the fourth reflecting surface (flat mirror) p9 of the second deflecting member 57, and enters the magnification correcting optical member 47. .. The imaging light flux EL2 emitted from the magnification correction optical member 47 enters the first projection area PA1 on the substrate P, and the image of the pattern appearing in the first illumination area IR1 is magnified (×) in the first projection area PA1. Projected in 1).

  When the radius of the cylindrical mask DM shown in FIG. 1 is r1 and the radius of the cylindrical surface of the substrate P wound around the second drum member 22 is r2, and the radius r1 is equal to the radius r2, the projection modules PL1 to PL1. The chief ray of the imaging light flux EL2 on the mask side of PL6 is tilted so as to pass through the central axis AX1 of the cylindrical mask DM. The inclination angle is the same as the inclination angle θ (±θ with respect to the center plane P3) of the principal ray of the imaging light flux EL2 on the substrate P side.

  In order to give the tilt angle θ described above, the angle θ1 of the first reflecting surface p4 of the first deflecting member 50 with respect to the optical axis AX3 shown in FIG. 3 is made smaller than 45° by Δθ1, and the second deflection with respect to the optical axis AX4. The angle θ4 of the fourth reflecting surface p9 of the member 57 is made smaller than 45° by Δθ4. Δθ1 and Δθ4 are set to have a relationship of Δθ1=Δθ4=θ/2 with respect to the angle θ shown in FIG.

  FIG. 4 is a perspective view of the second drum member 22 (rotary drum) applied to the substrate processing apparatus (exposure apparatus) of FIG. FIG. 5 is a perspective view for explaining the relationship between the detection probe and the reading device applied to the substrate processing apparatus (exposure apparatus) of FIG. FIG. 6 is an explanatory diagram showing the positions of the encoder scale disk and the reading device according to the embodiment as seen in the XZ plane orthogonal to the rotation center line AX2. In FIG. 4, for convenience, only the second to fourth projection areas PA2 to PA4 are shown, and the first, fifth, and sixth projection areas PA1, PA5, and PA6 are omitted.

  The second detector 35 shown in FIG. 1 optically detects the position (more specifically, the rotational position) of the second drum member 22, and as a scale member as shown in FIGS. And a high circularity encoder scale disk (disk) SD and encoder heads EN1, EN2, EN3, EN4, EN5 as reading units.

  The encoder scale disk SD is fixed to one end of the second drum member 22 that is orthogonal to the rotation axis ST of the second drum member 22. Therefore, the encoder scale disk SD rotates integrally with the rotation axis ST around the rotation center line AX2. The encoder scale disk SD may be fixed to both ends of the second drum member 22. That is, the encoder scale disk SD may be fixed to at least one end of the second drum member 22.

  On the outer peripheral surface of the encoder scale disk SD, a plurality of scales (engraved lines) GP are provided as scales for position detection to detect the position or the amount of position change of the second drum member 22 (cylindrical member) in the circumferential direction. Has been done. In the following, the scale GP will be appropriately referred to as the scale GP. The portion of the encoder scale disk SD on which the scale GP is engraved is the scale portion. The plurality of scales GP have, for example, a grid line having a pitch of 20 μm arranged annularly along the direction in which the second drum member 22 rotates, and together with the second drum member 22, the circumference of the rotation axis ST (second central axis AX2). To rotate.

  The encoder heads EN1, EN2, EN3, EN4, EN5 are arranged around the scale GP as seen from the rotation axis ST (second rotation center line AX2) so as to face the scale GP with a constant gap. Each of the encoder heads EN1 to EN5 is a non-contact sensor that projects a measurement beam on the scale GP and photoelectrically detects the beam (diffracted light) reflected by the scale GP. Further, when viewed from the rotation axis ST (second rotation center line AX2) of the second drum member 22, the encoder heads EN1 to EN5 are arranged in different azimuths (angle positions) in the circumferential direction of the scale disk SD. There is.

  Each of the encoder heads EN1 to EN5 is a reading device having a measurement sensitivity (detection sensitivity) with respect to a change in displacement in the tangential direction (in the XZ plane) of the scale GP. As shown in FIG. 4, when the installation azimuths of the encoder heads EN1 to EN5 (the angular directions in the XZ plane around the rotation center line AX2) are represented by the installation azimuth lines Le1, Le2, Le3, Le4, Le5, As shown in FIG. 6, the encoder heads EN1 and EN2 are arranged so that the installation azimuth lines Le1 and Le2 form an angle of ±θ° with respect to the center plane P3. In the present embodiment, the angle θ is set to 15° as an example, but the present invention is not limited to this.

  The projection modules PL1 to PL6 shown in FIG. 3 are also processing units that use the substrate P as an object to be processed, irradiate the substrate P with a light flux that forms a pattern image, and perform an exposure process. From this, the exposure apparatus EX includes the first processing unit including the odd-numbered projection modules PL1, PL3, and PL5 and the second processing unit including the even-numbered projection modules PL2, PL4, and PL6, and the substrate P is mounted on the substrate P. On the other hand, the principal ray of the imaging light flux EL2 from each of the first processing section and the second processing section (for example, the principal ray passing through the center point of the projection area PA) is projected on the substrate P when viewed in the XZ plane. Vertical incidence. In this way, the position at which the chief ray is incident on the substrate P is set as the specific position at which the substrate P is processed (here, the irradiation of the image forming light beam).

  The specific position is set at a position of an angle ±θ in the circumferential direction from the center plane P3 on the substrate P supported by the outer peripheral surface of the second drum member 22, as viewed from the second central axis AX2 of the second drum member 22. ing. As shown in FIGS. 4 and 6, the installation azimuth line Le1 of the encoder head EN1 is a chief ray passing through the center points of the respective projection areas (projection fields of view) PA1, PA3, PA5 of the odd-numbered projection modules PL1, PL3, PL5. Is arranged so as to match the inclination angle θ with respect to the center plane P3. Similarly, the installation azimuth line Le2 of the encoder head EN2 has an inclination angle with respect to the center plane P3 of the principal ray passing through the center points of the respective projection areas (projection fields of view) PA2, PA4, PA6 of the even-numbered projection modules PL2, PL4, PL6. It is arranged so as to coincide with θ. Therefore, the encoder heads EN1 and EN2 read the scale on the scale GP located in the direction connecting each specific position and the second central axis AX2.

  The encoder head EN4 is arranged upstream of the encoder head EN1 in the transport direction of the substrate P, that is, before the exposure position (projection area). The encoder head EN4 is arranged on the installation azimuth line Le4. The installation azimuth line Le4 is located at a position where the installation azimuth line Le1 of the encoder head EN1 is rotated by about 90° around the axis of the rotation center line AX2 toward the upstream side in the transport direction of the substrate P. The encoder head EN5 is arranged on the installation azimuth line Le5. The installation azimuth line Le5 is located at a position where the installation azimuth line Le2 of the encoder head EN2 is rotated by about 90° around the axis of the rotation center line AX2 toward the upstream side in the transport direction of the substrate P.

  As exemplified above, the principal ray of the imaging light flux EL2 passing through the centers of the odd-numbered projection areas PA1, PA3, PA5 and the principal ray of the imaging light flux EL2 passing through the centers of the even-numbered projection areas PA2, PA4, PA6. When the inclination angle ±θ with respect to the central plane P3 with respect to the light beam is 15°, the opening angle between the installation azimuth line Le1 and the installation azimuth line Le2 in the XZ plane is 30°. Therefore, the opening angle in the XZ plane between the installation azimuth line Le4 and the installation azimuth line Le5 is also approximately 30°.

  By arranging the encoder heads EN4 and EN5 as described above, when the directions of the installation azimuth lines Le4 and Le5 on which the encoder heads EN4 and EN5 for reading the scale GP are arranged are viewed from the rotation center line AX2 in the XZ plane. In addition, the principal ray of the imaging light flux EL2 with respect to the substrate P is substantially orthogonal to the direction of incidence on a specific position on the substrate P. Therefore, even if the second drum member 22 shifts in the Z direction due to a slight backlash (about 2 μm to 3 μm) of the bearing that supports the rotation axis ST, this shift causes the shift in the projection regions PA1 to PA6. The position error in the obtained direction along the imaging light flux EL2 can be measured with high accuracy by the encoder heads EN1 and EN2.

  The encoder head EN3 is arranged on the installation azimuth line Le3. The installation azimuth line Le3 is such that the installation azimuth line Le2 of the encoder head EN2 rotates about 120° around the axis of rotation center line AX2, and the installation azimuth line Le4 of the encoder head EN4 rotates around the axis of rotation center line AX2. The line Le2 is set at a position rotated by approximately 120° in the opposite direction to the rotation direction. That is, when viewed in the XZ plane, the three installation azimuth lines Le2, Le3, Le4 extending from the rotation center line AX2 are set at intervals of approximately 120°.

  The encoder scale disk SD that is a scale member has, for example, a metal, glass, or ceramics having a low coefficient of thermal expansion as a base material. The encoder scale disk SD is made to have a diameter as large as possible (for example, a diameter of 20 cm or more) in order to improve measurement resolution. Although the diameter of the encoder scale disk SD is smaller than the diameter of the second drum member 22 in FIG. 4, the diameter of the outer peripheral surface of the second drum member 22 around which the substrate P is wound, By making the diameters of the scale GP of the encoder scale disk SD uniform (substantially equal), the so-called measurement Abbe error can be further reduced.

  The minimum pitch of the scale GP engraved in the circumferential direction of the encoder scale disc SD is limited by the performance of a scale marking line device or the like that processes the encoder scale disc SD. Therefore, if the diameter of the encoder scale disk SD is increased, the angle measurement resolution corresponding to the minimum pitch can be increased accordingly. When the directions of the installation azimuth lines Le1 and Le2 in which the encoder heads EN1 and EN2 for reading the scale GP are arranged are viewed from the rotation center line AX2, the principal ray of the imaging light flux EL2 is incident on the substrate P with respect to the substrate P. When the second drum member 22 is shifted in the X direction due to a slight backlash (about 2 μm to 3 μm) of the bearing that supports the rotating shaft ST, the projection is performed by this shift. Positional errors in the feed direction (Xs) of the substrate P that can occur in the areas PA1 to PA6 can be measured with high accuracy by the encoder heads EN1 and EN2.

  As shown in FIG. 5, an image of a part of the mask pattern projected by the projection optical system PL shown in FIG. 1 and the substrate P are relatively arranged on a part of the substrate P supported by the curved surface of the second drum member 22. For alignment (alignment), a plurality of alignment microscopes AMG1 and AMG2 for detecting alignment marks and the like formed in advance on the substrate P are provided. The alignment microscopes AMG1 and AMG2 are pattern detection devices arranged around the second drum member 22. The alignment microscopes AMG1 and AMG2 are detection probes for detecting a specific pattern discretely or continuously formed on the substrate P. The detection region by the detection probe is arranged upstream of the specific position in the transport direction of the substrate P.

  As shown in FIG. 5, the alignment microscopes AMG1 and AMG2 have a plurality of (for example, four) detection probes arranged in a line in the Y-axis direction (width direction of the substrate P). The alignment microscopes AMG1 and AMG2 are detection probes at both ends of the second drum member 22 in the Y-axis direction, and can constantly observe or detect the alignment marks formed near both ends in the width direction of the substrate P. Then, the alignment microscopes AMG1 and AMG2 are detection probes other than both side ends of the second drum member 22 in the Y-axis direction (width direction of the substrate P), and are formed on the substrate P along the longitudinal direction, for example. It is possible to observe or detect the alignment mark formed in the blank portion or the like between the pattern formation regions of the display panel.

  As shown in FIG. 5, lines that pass through the centers (detection centers) of the respective observation regions on the substrate P by the alignment microscopes AMG1 and AMG2 and are orthogonal to the second central axis AX2 are observation azimuth lines AM1 and AM2. In this case, the observation azimuth lines AM1 of the four alignment microscopes AMG1 are arranged parallel to the Y-axis direction, and similarly, the observation azimuth lines AM2 of the four alignment microscopes AMG2 are arranged parallel to the Y-axis direction.

  As shown in FIGS. 5 and 6, when viewed in the XZ plane, the installation azimuth line Le4 of the encoder head EN4 is set to the same azimuth as the observation azimuth lines AM1 of the four alignment microscopes AMG1. The installation azimuth line Le5 of the encoder head EN5 is set to the same azimuth as the observation azimuth lines AM2 of the four alignment microscopes AMG2.

  In this way, the detection probes of the alignment microscopes AMG1 and AMG2 are arranged around the second drum member 22 when viewed from the second central axis AX2. The detection probes of the alignment microscopes AMG1 and AMG2 have a direction (installation azimuth lines Le4 and Le5) that connects the position where the encoder heads EN4 and EN5 are arranged and the second central axis AX2 with the second central axis AX2 and the alignment microscope. It is arranged so as to coincide with the direction connecting the detection centers of AMG1 and AMG2. It should be noted that the encoder heads EN4 and EN5 corresponding to the respective observation areas (detection centers) of the alignment microscopes AMG1 and AMG2 and the rotation center lines on which the encoders EN1 and EN2 corresponding to the respective projection areas PA1 to PA6 of the projection modules PL1 to PL6 are arranged. The position in the AX2 surrounding direction is set between the sheet entrance area IA shown in FIG. 6 where the substrate P starts to come into contact with the second drum member 22 and the sheet removal area OA where the substrate P is removed from the second drum member 22. It

  The alignment microscopes AMG1 and AMG2 are arranged in front of the exposure position (projection area PA). The alignment microscopes AMG1 and AMG2, for example, send an image of an alignment mark (formed in an area within a square of several tens of μm to several hundreds of μm) formed in the vicinity of an end portion of the substrate P in the Y direction at a predetermined speed. In this state, an image pickup device such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor) or the like is used for high-speed image detection (sampling). At the moment when the sampling is performed, the control device 14 stores (latches) the rotational angle position of the encoder scale disk SD that is sequentially measured by the encoder head EN4, so that the mark position on the substrate P and the second drum member are stored. The correspondence with the rotational angle position of 22 is obtained.

  If the mark detected by the alignment microscope AMG1 is also detected by the subsequent alignment microscope AMG2, the expansion and contraction of the substrate P and a slight slip on the second drum member 22 can be measured. The angular position Φa1 measured by the encoder head EN4 when the alignment microscope AMG1 samples the mark and the angular position Φa2 measured by the encoder head EN5 when the alignment microscope AMG2 samples the same mark are stored.

  An up/down counter (counter) that is connected to each of the two encoder heads EN4 and EN5 (and EN1, EN2, and EN3) and outputs a measurement value corresponding to the angular position is, for example, the outer circumference of the scale disk SD. It is assumed that the origin mark (not shown) engraved on the surface is simultaneously reset to zero at the moment when the specific encoder head (any one of EN1 to EN5) detects it or at an arbitrary time. The difference value between the angular positions Φa1 and Φa2 obtained in this way is compared with the opening angle Φ0 of the installation azimuth lines Le4 and Le5 of the two alignment microscopes AMG1 and AMG2 that have been precisely calibrated in advance. When there is an error between the difference value (Φa1−Φa2) and the opening angle Φ0, the substrate P is slightly above the second drum member 22 between the sheet entry area IA and the sheet removal area OA. There is a possibility that it is slipping on or stretched in the feed direction (circumferential direction).

  Generally, the positional error at the time of patterning is determined according to the fineness of the device pattern formed on the substrate P and the overlay accuracy. For example, a linear pattern with a width of 10 μm is accurately overlaid on the underlying pattern layer. For the matching exposure, only a fractional error or less, that is, a positional error of about ±2 μm in terms of the dimension on the substrate P is allowed. In order to realize such highly accurate measurement, the measurement direction of the mark image by each of the alignment microscopes AMG1 and AMG2 (the tangential direction of the outer periphery of the second drum member 22 in the XZ plane) and the encoder heads EN4 and EN5. It is necessary to align the measurement direction (the outer circumferential tangent direction of the scale GP in the XZ plane) within the allowable angular error.

  As described above, the encoder heads EN4 and EN5 are arranged so as to coincide with the measurement direction (the tangential direction of the circumferential surface of the second drum member 22) of the alignment mark on the substrate P by the alignment microscopes AMG1 and AMG2. .. Therefore, when the position of the substrate P (mark) is detected by the alignment microscopes AMG1 and AMG2 (at the time of image sampling), the second drum member 22 (encoder scale disk SD) is orthogonal to the installation azimuth line Le4 or Le5 in the XZ plane. Even in the case of shifting in the circumferential direction (tangential direction), it is possible to perform highly accurate position measurement in consideration of the shift of the second drum member 22.

  Further, if the scale pitch of the scale GP is always constant, it is assumed that there is no speed unevenness in the rotation, and the change interval of each reading value of the encoder heads EN1, EN2, EN3, EN4, and EN5 (the up/down pulse generation time to the counter ) Is constant. However, the deformation of the encoder scale disk SD when mounting the encoder scale disk SD on the second drum member 22, the position (tilt) error when mounting the encoder heads EN1, EN2, EN3, EN4, and EN5, and the manufacture of the encoder scale disk SD Due to influences such as time accuracy, eccentricity at the time of mounting, and the like, the scale GP may have an inherent error (pitch error of the scale itself, pitch unevenness due to eccentricity, deformation, and the like). Further, the scale GP may also have an inherent error due to a constantly changing element such as expansion and contraction of the encoder scale disk SD due to a temperature change during operation of the substrate processing apparatus 11. In the present embodiment, the measurement error associated with the peculiar error of the scale GP that occurs due to the above-mentioned cause is obtained. Then, based on the obtained measurement error, a correction map (correction amount data) for correcting the peculiar error amount of the scale GP is created, and each read value (actual measurement value) of the plurality of encoder heads EN1 to EN5 is calculated. The correction is performed based on the correction map, and the measurement error amount due to the peculiar error of the scale GP is canceled or reduced.

  In the present embodiment, the correction map of the measurement error due to the pitch error and the pitch unevenness of the scale of the scale GP is arranged on the outer circumference of the scale disk SD coaxially attached to the second drum member 22 as the cylindrical member. It is executed based on the actual measured values of the plurality of encoder heads. Here, the correction map is created by the encoder head EN4 as the first reading unit, the encoder head EN5 as the second reading unit, and the control device 14 as the correction unit and the map creation unit. In the present embodiment, for convenience, the encoder head EN4 is used as the first reading unit and the encoder head EN5 is used as the second reading unit. However, the first reading unit and the second reading unit have at least two mounting angle intervals that are known in advance. Any encoder head may be used.

  The control device 14 as the correction unit, the difference value (m4-m5) between the read value by the encoder head EN4 (assumed to be the count value m4 by the counter) and the read value by the encoder head EN5 (assumed as the count value m5 by the counter), Alternatively, a difference value (K45− Based on m4−m5), the pitch error of the graduation that has occurred over one round of the graduation GP is obtained, for example, for each predetermined angular position from the origin position of the graduation GP. Then, the control device 14 stores the data of the scale pitch error for one round of the scale GP as a correction map, and based on the correction map, the read value of the encoder head EN4, the read value of the encoder head EN5, or another encoder. Each read value of the heads EN1 to EN3 is corrected.

  FIG. 7 is an enlarged view schematically showing the scale of the scale GP. FIG. 8 is a schematic diagram showing the positional relationship between the scale GP and the encoder heads EN4 and EN5. As shown in FIG. 7, the scale of the scale GP is composed of, for example, a convex portion GPt having a rising portion GPa and a falling portion GPb, and a concave portion GPU between adjacent convex portions GPt. In the present embodiment, one convex portion GPt and one concave portion GPU are assumed to be one unit of the scale GP, that is, one pitch of the scale. To simplify the explanation, each of the encoder heads EN1 to EN5 outputs the up pulse U when reading the rising part GPa of the scale and outputs the down pulse D when reading the falling part GPb. ..

  The scale of the scale GP is such that the distance SS1 from the rising portion GPa to the rising portion GPa of the adjacent scale GP or the distance SS2 from the falling portion GPb to the falling portion GPb of the adjacent scale GP is the pitch (interval between the scales). )become. Assuming that the scale pitch determined when the encoder scale disk SD is designed is SS, if the scale GP is manufactured accurately, the distance SS1 or SS2 will match the scale pitch SS at any part on the scale GP. ..

  Therefore, when the scale GP moves in the direction indicated by the arrow R in FIG. 7, when viewed from the pulses output by the encoder head EN, two up-pulses U and one down-pulse U are output, or When two down pulses D and one up pulse U are output, the outer peripheral surface (scale GP) of the encoder scale disk SD (see FIG. 6) has moved (rotated) by the scale pitch SS. Become. If the types of pulses output by the encoder head EN are not distinguished, the outer peripheral portion of the encoder scale disk SD has moved by the scale pitch SS each time three pulses are detected.

  In actual encoder measurement, a two-phase signal (sin wave, cos wave) is generated from the encoder head, and interpolation signal processing is performed based on the two-phase signal to further subdivide the scale pitch SS. .. Therefore, the up pulse U and the down pulse D that are actually counted by the digital counter are generated at every position obtained by equally dividing the scale pitch SS into a few fractions to a few tenths.

  FIG. 8 is a schematic view in which a plurality of scales GP provided on the outer peripheral portion of the scale disk SD are arranged linearly. In FIG. 8, it is assumed that the scale GP provided on the outer peripheral portion of the encoder scale disk SD moves in the direction indicated by the arrow R. An encoder head EN4 as a first reading unit and an encoder head EN5 as a second reading unit are arranged in this order in the moving direction of the scale GP. When viewed from the scale GP, the two encoder heads EN4 and EN5 relatively move in a direction opposite to the moving direction of the scale GP.

  As shown in FIG. 6, the pair of encoder heads EN4 and EN5 includes a line connecting the encoder head EN4 and the second central axis AX2 (installation azimuth line Le4) and a line connecting the encoder head EN5 and the second central axis AX2. The central angle (encoder mounting angle) formed with the (installation azimuth line Le5) is θs. Further, as shown in FIG. 8, the linear distance (inter-head distance) in the circumferential direction of the scale GP at the position where the pair of encoder heads EN4 and EN5 read the scale GP on the surface of the encoder scale disk SD is XS.

  When the pair of encoder heads EN4 and EN5 are attached to the frame or the like of the substrate processing apparatus 11, the encoder attachment angle θs and the head-to-head distance XS are constant. As described above, due to the deformation of the encoder scale disk SD, the accuracy in manufacturing the encoder scale disk SD, the eccentricity at the time of mounting, the expansion and contraction of the encoder scale disk SD caused by the temperature change, etc., the scale pitch SS becomes smaller than that of the encoder scale disk SD. It is not always constant in the circumferential direction. For example, in a very schematic manner, as shown in FIG. 8, in the areas a, c, and d on the scale GP, the rising portion GPa and the rising portion GPa stand between one encoder mounting angle θs and one head distance XS. There are three scales GP with the descending portion GPb as one set. However, there are 2.5 scales GP in the region b and 6 scales in the region e.

  In the example shown in FIG. 8, when the scale pitch SS is a design value, for example, a prescribed number (three in this example) of scale GP are present between one encoder mounting angle θs and one head-to-head distance XS. It shall be. Actually, the number of scales GP existing between one encoder mounting angle θs and one head-to-head distance XS is increased or decreased from the above-mentioned specified number due to the above-mentioned error factor of the scale GP. In the region a in FIG. 8, the graduation pitch is SSa, but in the region b, the number of graduations GP is smaller than the specified number, so that the graduation pitch SSb in the region b is larger than the graduation pitch SSa in the region a. Further, since the number of graduations GP in the region e is larger than the specified number, the graduation pitch SSe in the region e is smaller than the graduation pitch SSa.

  For example, it is assumed that the designed scale pitch SS is 100 and the designed head-to-head distance XS is 300. In the areas a, c, and d of FIG. 8, the actual head-to-head distance X calculated based on each value (counting by the counter) obtained by reading the scale GP by the encoder heads EN4 and EN5 is 300. This matches the designed head-to-head distance XS. On the other hand, in the area b in FIG. 8, the actual head-to-head distance X calculated based on each value read by the encoder heads EN4 and EN5 (count value by the counter) is 250, and in the area e, the encoder head EN4 , The actual head-to-head distance X based on the read value of EN5 (count value by the counter) is 600.

  As described above, the difference between the designed head-to-head distance XS and the actual head-to-head distance X is due to the scale pitch error of the scale GP. After fixing the pair of encoder heads EN4 and EN5 to the device, if the device is installed in an environment of constant temperature, the head-to-head distance XS does not change. Therefore, for example, with reference to the head-to-head distance XS after fixing the encoder heads EN4 and EN5, a map of the scale pitch error of the scale GP obtained from the read values of the encoder heads EN4 and EN5 (for each 360° angular position for one rotation). Error amount or error correction amount) is created. After creating the map, the error amount or the correction amount corresponding to the angular position is mapped based on the reading value (counter value of the counter) of the scale GP by the encoder heads EN4 and EN5 (or other heads EN1 to EN3). If it is called from and is sequentially corrected, the movement distance error of the scale GP in the circumferential direction can be corrected in real time.

When correcting the actual scale pitch error and the moving distance error of the scale GP, for example, the scale pitch SS when there is no error in the scale GP, the designed head-to-head distance XS, and the actual head-to-head distance X are used, and the scale GP is further used. The actual graduation pitch (actual graduation pitch) SSr in the case where an error has occurred is calculated from, for example, equation (1). By using the actual scale pitch SSr, it is possible to correct the error of the scale GP and consequently the error of the moving distance of the scale GP.
SSr=SS×XS/X... (1)

Further, the error of the scale GP is corrected based on the number of the scale GP (measurement scale line) NS existing between the head-to-head distance XS read by the pair of encoder heads EN4 and EN5, and the moving distance error of the scale GP is corrected. You may correct. The number NS of graduation lines existing between the head-to-head distance XS can also be obtained by the count value by the counter of the up pulse U and the down pulse D obtained from the pair of encoder heads EN4 and EN5. When the measurement graduation line NS is used, the actual graduation pitch SSr can be obtained by the equation (2) including the head-to-head distance XS.
SSr=XS/NS...(2)

  The error of the graduation pitch and the error of the moving distance of the graduation GP are determined by the control device 14 as a correction unit included in the position detecting device of the cylindrical member by a calculation using, for example, the formula (1) or the formula (2). It is applied as a correction amount for the read values of the encoder heads EN4 and EN5. Therefore, the cylindrical member position detection device and the substrate processing device 11 including the control device 14 generate the error map or the correction map even if an error occurs in the scale pitch SS due to deformation of the encoder scale disk SD provided with the scale GP. Since the error can be corrected almost in real time by using the encoder scale disk SD and the second drum member 22, accurate position measurement (position measurement in the circumferential direction) can be realized. Next, the correction of the scale pitch error and the movement distance error will be described.

  FIG. 9 is a flowchart showing a procedure for correcting the scale pitch error of the scale. FIG. 10 is a diagram showing the relationship between the scale disk SD having scales on the outer peripheral surface and the encoder heads EN4 and EN5. FIG. 11 is a diagram showing an example of the correction map. When correcting the scale pitch error of the scale GP, as shown in FIG. 10, the head-to-head distance XS of the pair of encoder heads EN4 and EN5 is measured in advance and stored in the storage unit of the control device 14. The head-to-head distance XS is obtained at a position where the pair of encoder heads EN4 and EN5 read the scale GP on the outer peripheral surface of the encoder scale disk SD. The position where the pair of encoder heads EN4 and EN5 reads the scale GP can be a position where the encoder scale disk SD is a perfect circle and the encoder scale disk SD is not eccentric with respect to the rotation center line AX2. .. In this case, as shown in FIG. 10, on the curved surface Pd (scale surface of the scale GP) curved at the radius of the design value of the encoder scale disk SD (distance from the center to the outer peripheral surface) ra, as shown in FIG. The head-to-head distance XS of the encoder heads EN4 and EN5 is measured. In the example shown in FIG. 10, the scale GP of the encoder scale disk SD rotates (turns) in the direction indicated by the arrow R, that is, from the encoder head EN4 to the encoder head EN5.

  As shown in FIG. 9, in step S101, when the process of the substrate processing apparatus 11 shown in FIG. 1 is not started (step S101, No), the correction of the scale GP or the like is not executed. In step S101, when the processing of the substrate processing apparatus 11 is started and the scale disk SD is stably rotating (step S101, Yes), in step S102, the control device 14 sets the encoder heads EN4 and EN5 at a predetermined timing. These readings (counter count values) are obtained from. The acquisition at a predetermined timing means, for example, that the controller 14 reads each of the encoder heads EN4 and EN5 every time the scale GP of the encoder scale disk SD rotates about the rotation center line AX2 by a predetermined angle α (degrees). Acquiring a value, that is, latching and storing the count value of each counter. Hereinafter, the angle α may be appropriately referred to as a rotation angle α. When the encoder scale disk SD is rotating at a constant angular velocity (constant circumferential velocity), the control device 14 may acquire each reading value of both encoder heads EN4 and EN5 at every predetermined time t. In this example, α (degrees) is preferably a divisor of 360, but the angle α is not limited to this. The angle α is determined according to the tendency of the scale pitch error (the fluctuation range of the error or the rate of change in the circumferential direction).

  When the encoder scale disk SD is rotating at a constant angular velocity, the control device 14 acquires these read values from both encoder heads EN4 and EN5 at each time t. When the control device 14 acquires these read values from both encoder heads EN4 and EN5 for each predetermined angle α, for example, a rotation angle detecting means of the encoder scale disk SD is prepared. Then, the control device 14 acquires these read values from both encoder heads EN4 and EN5 at each timing when the rotation angle detection means detects that the encoder scale disk SD has rotated by the angle α. Further, either one of both encoder heads EN4 and EN5 may be used as the rotation angle detecting means. For example, when the encoder head EN4 is used as the rotation angle detecting means, the control device 14 detects from the encoder heads EN4 and EN5 each time the encoder head EN4 detects that the encoder scale disk SD has rotated by the angle α. Get these readings. The fact that the encoder scale disk SD has rotated by the angle α can be detected, for example, by the encoder head EN4 detecting the number of scales GP corresponding to the rotation angle α and outputting the number of pulses corresponding to this.

  Next, proceeding to step S103, the control device 14 obtains the actual scale pitch SSr as the correction value of the scale GP based on the read value obtained at step S102. For example, when obtaining the actual scale pitch SSr using the above equation (2), the control device 14 obtains the measurement scale number NS from the read values of the encoder heads EN4 and EN5. The measurement scale number NS is the difference (NSa-NSb) between the read value NSa of the number of scale GP by the encoder head EN4 and the read value NSb of the number of scale GP by the encoder head EN5. Then, the control device 14 reads the head-to-head distance XS stored in its own storage unit and obtains the actual scale pitch SSr from the equation (2). The actual scale pitch SSr becomes a correction value for the interval of the scale GP in the range of the angle α (degrees). When the actual scale pitch SSr is obtained using the above-mentioned formula (2), the inter-measuring encoder distance X is obtained from the product of the number of measurement scales NS and the scale pitch SS when the scale GP has no error, and the distance between the heads is calculated. The number NS of measurement scales may be obtained from the distance XS, the scale pitch SS, and the distance X between the measurement encoders.

  The measurement scale number NS can be obtained as follows, for example. When the encoder head EN4 reads the reference position (scale reference position) GPb of the plurality of scales GP included in the encoder scale disk SD, the control device 14 resets the encoder head EN4 (0 according to the Z phase of the encoder head EN4. After the point reset), the number of scale GP detected by the encoder head EN4 is counted. Next, when the encoder head EN5 reads the scale reference position GPb, the control device 14 resets the encoder head EN5 (zero point reset by the Z phase of the encoder head EN5). Then, the control device 14 acquires the number of scales GP of the encoder head EN4 when the encoder head EN5 is reset to 0, and the number of scales GP when the encoder head EN5 reads the scale reference position GPb, that is, 0. Find the difference between and. This difference is the measurement scale number NS. After this, the control device 14 continues counting the scale GP by the encoder heads EN4 and EN5, and at the same time, the count value (scale) of the read values of both encoder heads EN4 and EN5 at a predetermined angle α or at a predetermined time t. The GP count value) is acquired, and the difference between them is obtained, and this is set as the measurement scale number NS at a predetermined angle α or a predetermined time t. In this example, when the encoder scale disk SD makes one revolution, the scale reference position GPb returns to the original position, but at this time, the counters connected to the encoder heads EN4 and EN5 do not need to be reset, or reset. You may.

  When the actual scale pitch SSr (corresponding to the correction value of the scale GP) at a certain angle α is obtained in step S103, the control device 14 associates it with the angle α in the correction map TBc shown in FIG. 11 in step S104. Describe. For example, No. of the correction map TBc. In 2, the angle 2×α and the corresponding actual scale pitch SSr2 (=XS/NS2) are described. The correction map TBc is stored in the storage unit of the control device 14. The control device 14 obtains the actual scale pitch SSr over the entire circumference of the plurality of scales GP (the entire circumference of the encoder scale disk SD). During the processing of the substrate processing apparatus 11, the control device 14 determines a correction value corresponding to the angle of the encoder scale disk SD detected by at least one of the encoder head EN4 and the encoder head EN5, that is, the actual scale pitch SSr from the correction map TBc. Read out and correct the error of the scale GP. As shown in FIG. 6, when the substrate processing apparatus 11 has three or more encoder heads EN1, EN2, EN3, EN4, EN5, the control device 14 uses the correction map TBc for other than the encoder heads EN4, EN5. You may correct the error of the scale GP. Next, the process proceeds to step S105, and when the process of the substrate processing apparatus 11 is not completed (step S105, No), the control device 14 continues steps S102 to S104, and the process of the substrate processing apparatus 11 is performed. When finished (Yes at Step S105), the control device 14 finishes the correction of the scale GP and the like.

  In the above example, during the processing of the substrate processing apparatus 11, that is, when the processing unit is performing the predetermined processing (for example, the exposure processing) on the substrate, the control device 14 causes the intervals between the plurality of scales GP to be apparent. Above, correct so that it becomes constant. By doing so, the error of the scale GP generated during the operation of the substrate processing apparatus 11, that is, the error of the reading scale interval can be corrected in real time, so that the processing accuracy of the substrate processing apparatus 11 is improved.

  When the disk SD makes one round, that is, when the scale reference position GPb at the position of the encoder head EN4 returns to the position of the encoder head EN4, the actual scale pitch SSr is obtained as the correction value of the scale GP for the entire circumference. . The controller 14 may similarly similarly correct the scale GP each time the disk SD makes one revolution, or after the disk SD makes one revolution, finishes the correction of the scale GP at a predetermined timing (for example, a predetermined time). The correction of the scale GP may be restarted after a lapse of time or when there is a predetermined temperature change. In the former case, the correction map TBc is updated at any time, so that the encoder scale disk SD and the scale GP can be promptly dealt with even when they are deformed or changed in size. In the latter case, the update frequency of the correction map TBc can be suppressed, and accordingly, the hardware resources of the control device 14 can be effectively used.

  When the control device 14 acquires these read values from the encoder heads EN4 and EN5, the shorter the acquisition timing or the smaller the interval between the encoder heads EN4 and EN5, the higher the correction accuracy of the scale GP. The distance between the encoder heads EN4 and EN5 is restricted to some extent in consideration of the size of the encoder heads EN4 and EN5 and the arrangement of other components. Therefore, there is an advantage that the versatility is improved by shortening the timing when acquiring these read values from the encoder heads EN4 and EN5.

  12 and 13 are conceptual diagrams showing the timings when these read values are acquired from the pair of encoder heads. In the above-described example, every time the plurality of scales GP of the encoder scale disk SD rotate about the rotation center line AX2 by a predetermined rotation angle α (degrees), the control device 14 reads these read values from the encoder heads EN4 and EN5. Got When the rotation angle α (degrees) at this time is set to a divisor of 360, the encoder heads EN4 and EN5 read the same position every circumference while the encoder scale disk SD is rotating a plurality of times (see FIG. 12). ). In this case, in order to improve the correction accuracy of the scale GP, it is necessary to reduce the predetermined rotation angle α, but it is not possible to reduce the rotation angle α inevitably due to restrictions of the device.

  In this embodiment, since the encoder scale disk SD having a plurality of scales GP is a rotating body (continuous body), it is possible to continuously measure using the encoder heads EN4 and EN5 without necessarily ensuring the periodicity for each rotation. Is. Therefore, for example, by setting the rotation angle α (degrees) to a number that is not a divisor of 360 degrees, the periodicity of the reading positions of the encoder heads EN4 and EN5 is destroyed when the encoder scale disk SD rotates a plurality of times. You can In particular, by setting α (degree) to a number that is not a divisor of 360 degrees and is a prime number, it is possible to more effectively break the periodicity described above. As a result, even if the predetermined rotation angle α is large, the amount of deviation that occurs each time a plurality of graduations GP (encoder scale disk SD) overlaps the lap becomes small, and as a result, the graduation GP by the encoder heads EN4 and EN5 becomes smaller. The measurement interval can be shortened (see FIG. 13).

  For example, it is preferable that the angle α be a value that does not become a divisor of 360 degrees between about 10 degrees and 35 degrees, but the value of 360/α is about 1 to 4 digits after the decimal point, preferably 1 digit after the decimal point. It may be a value that is divisible by 4 digits. As an example, when the angle α is a prime number such as 11 degrees, 17 degrees, 19 degrees, and 23 degrees, 360/α cannot be divided even by four digits after the decimal point. On the other hand, when the angle α is set to any of 12.5 degrees, 16 degrees, 25 degrees, and 28.8 degrees, 360/α is divisible by one digit after the decimal point. When the angle α is 19.2 degrees and 32.0 degrees, 360/α is divisible by two digits after the decimal point. When the angle α is 12.8 degrees, 360/α is divisible by 3 digits after the decimal point. Further, when the angle α is set to 25.6 degrees, 360/α is divisible by 4 digits after the decimal point. It should be noted that when the rotation angle α is between 10° and 35°, angles that are divisors (360/α is an integer) are 10°, 12°, 14.4°, 15°, 18°, 20°, 22. 5 degrees, 24 degrees, and 30 degrees. Further, the rotation angle α may be in the range of 1 degree to 10 degrees, but when avoiding a divisor such that 360/α is an integer, the rotation angle α is 7 degrees or 9 degrees. Depending on the required resolution of the error correction, the difference between the read values by the encoder heads EN4 and EN5 is calculated for each rotation angle α of less than 1 degree, for example, every 0.5 degrees, and a pitch error map is created. You may.

  FIG. 14 is a flowchart showing the procedure for correcting the scale error. In the example shown in FIG. 14, when the plurality of scales GP of the encoder scale disk SD rotate about the rotation center line AX2 by a predetermined angle α (degrees), a pair of encoder heads EN4 and EN5 read them. The processing procedure when the rotation angle α is a prime number that is not a divisor of 360 is shown. In this example, the rotation angle α can be, for example, 7 degrees, 11 degrees, or the like.

  Steps S201 to S204 are the same as steps S101 to S104 in the above example in which α (degrees) is a divisor of 360, so description thereof will be omitted. In step S205, the control device 14 repeats steps S202 to S205 when the encoder scale disk SD has not rotated to the specified number of rotations since the start of obtaining the correction value (step S205, No). In step S205, the control device 14 proceeds to step S206 when the encoder scale disk SD has rotated to the specified rotation speed since the start of obtaining the correction value (step S205, Yes). Since step S206 is the same as step S105 in the above example in which α (degree) is a divisor of 360, description thereof will be omitted. The prescribed number of revolutions in step S205 may be two revolutions or more, but as the prescribed number of revolutions increases, the effect of improving the correction accuracy of the scale GP decreases. For this reason, it is preferable to set the prescribed number of revolutions in an appropriate range of two or more revolutions.

  Next, the arrangement of the encoder heads EN4 and EN5 will be described. As shown in FIG. 6, the encoder head EN4 as the first reading unit and the encoder head EN5 as the second reading unit include a second drum member as a cylindrical member rather than an exposure apparatus EX (see FIG. 1) as a processing unit. 22 is preferably arranged on the side opposite to the rotation direction. More specifically, it is preferable that the substrate P supported by the second drum member 22 is arranged on the side opposite to the rotation direction of the second drum member 22 with respect to the portion exposed by the exposure apparatus EX. That is, the scale GP is read at two positions before the portion where the substrate P supported by the second drum member 22 is exposed by the exposure apparatus EX, and the correction value is obtained. In the example shown in FIG. 6, the rotation direction of the second drum member 22 is the direction from the encoder head EN4 to the encoder head EN5. By arranging the encoder heads EN4 and EN5 in this way, the position information in the circumferential direction of the second drum member 22 after the scale GP is corrected is used as feedback for control of processing (exposure processing in this example). Therefore, the accuracy of processing is improved.

  In the present embodiment, both the encoder heads EN4 and EN5 are arranged on the side opposite to the rotation direction of the second drum member 22 with respect to the portion where the substrate P is exposed by the exposure apparatus EX, but one of them is arranged. You may arrange|position in the part to be exposed. For example, the encoder head EN5 may be the first reading unit, the encoder head EN1 may be the second reading unit, and the correction value of the scale GP may be obtained based on the difference between the reading values of the two.

  Further, in the present embodiment, since the alignment microscopes AMG1 and AMG2 are arranged at the positions corresponding to the encoder heads EN4 and EN5, by measuring the change on the surface of the substrate P with the alignment microscopes AMG1 and AMG2, the processing positions It is also possible to predict the change of the substrate P at the time and correct it during the processing. Further, in addition to the encoder heads EN4 and EN5, at least one of the encoder heads EN1 and EN2 arranged at a position different from these, for example, the processing position, is used to shake the rotation center line AX2 (orthogonal to the rotation center line AX2. It is also possible to measure the circularity (shape distortion), the eccentricity of the second drum member 22, or the like, and correct the process based on the measured value.

  When measuring the shake of the rotation center line AX2, the eccentricity of the second drum member 22, or the like, as a third reading unit arranged on the side opposite to the encoder heads EN4 and EN5 with respect to the processing position (position of exposure processing) It is preferable to use the encoder head EN3 (see FIG. 6) with the encoder heads EN4 and EN5. By doing so, the measurement results such as the shake of the rotation center line AX2 can be compared before and after the processing position, and the intermediate value can be used as the correction value for the shake of the rotation center line AX2. In addition, by measuring the shake of the rotation center line AX2 using the encoder heads EN4 and EN5 and the encoder head EN3 that are arranged in front of and behind the processing position, and correcting based on the measured value, the rotation center line AX2 is corrected. The accuracy in correcting the shake and the like of the AX2 is improved. When the encoder head EN3 is used as the third reading unit, a straight line connecting the encoder head EN5 and the rotation center line AX2 (installation azimuth line Le5) and a straight line connecting the encoder head EN3 and the rotation center line AX2 (installation azimuth line Le3). The angle formed by is not limited to 210 degrees shown in FIG. 6, and the encoder head EN3 may be on the processing position side.

  When the encoder head EN4 as the first reading unit and the encoder head EN5 as the second reading unit, and the encoder head EN3 as the third reading unit are used, the encoder head EN4 and the encoder head EN5 in the circumferential direction of the encoder scale disk SD are used. The distance between and is preferably smaller than the distance between the encoder head EN5 and the encoder head EN3. By reducing the distance between the encoder head EN4 and the encoder head EN5, it is possible to improve the correction accuracy of the scale GP. Further, by widening the distance between the encoder head EN5 and the encoder head EN3, it is possible to improve the sensitivity when detecting the eccentricity of the second drum member 22 and the like.

  Next, the interval at which the encoder head EN4 as the first reading unit and the encoder head EN5 as the second reading unit are arranged will be described. The encoder head EN4 and the encoder head EN5 are formed by a line connecting the encoder head EN4 and the rotation center line AX2 (installation azimuth line Le4) and a line connecting the encoder head EN5 and the rotation center line AX2 (installation azimuth line Le5). It is preferable that the encoder mounting angle θs, which is the central angle, is arranged to be an angle other than 90 degrees, 180 degrees, and 270 degrees. By doing so, the two encoder heads EN4 and EN5 can also detect the shake of the rotation center line AX2, the eccentricity of the second drum member 22, and the like. Further, the encoder mounting angle θs is preferably within 45 degrees, and is preferably an angle other than 120 degrees and 240 degrees. By doing so, the two encoder heads EN4 and EN5 detect the shake of the rotation center line AX2, the eccentricity of the second drum member 22, and the like, and the correction accuracy of the scale GP is also improved. Next, a case where the substrate processing apparatus 11 has a mechanism for adjusting the roundness of the encoder scale disk SD will be described.

  15 and 16 are explanatory views for explaining the roundness adjusting mechanism for adjusting the roundness of the encoder scale disk. Although the diameter of the encoder scale disk SD is illustrated to be smaller than the diameter of the second drum member 22 in FIGS. 4 and 5 described above, the outer circumference of the second drum member 22 around which the substrate P is wound. It is preferable that the diameter of the surface and the diameter of the scale GP of the encoder scale disk SD are made uniform (substantially coincident with each other). By doing so, the so-called measurement Abbe error can be further reduced. In this case, the exposure apparatus EX preferably includes a roundness adjusting mechanism Cs for adjusting the roundness of the encoder scale disk SD as shown in FIG.

  The encoder scale disk SD that is a scale member is an annular member. The encoder scale disk SD having the scale GP on the outer peripheral surface is fixed to at least one end of the second drum member 22 orthogonal to the second central axis AX2 of the second drum member 22. The encoder scale disk SD includes a groove Sc provided in the encoder scale disk SD along the circumferential direction of the second central axis AX2, and a second drum having the same radius as the groove Sc and along the circumferential direction of the second central axis AX2. It faces the groove Dc provided in the member 22. The encoder scale disk SD has a bearing member SB such as a rolling element (for example, a ball) interposed between the groove Sc and the groove Dc.

  The roundness adjusting mechanism Cs is provided on the inner peripheral side of the encoder scale disk SD and includes an adjusting member 60 and a pressing member PP. Then, the roundness adjusting mechanism Cs is a pressing mechanism that can vary the pressing force in the direction from the second central axis AX2 to the scale GP, which is a direction parallel to the installation azimuth line Le4, with the rotation center line AX2 as the center. A plurality of (for example, eight locations) are provided at a predetermined pitch in the circumferential direction. The adjusting member 60 has a male screw portion 61 which is inserted into the female screw portion FP3 of the encoder scale disk SD and the female screw portion FP4 of the second drum member 22 through the pressing member PP, and a head portion 62 which contacts the pressing member PP. Have. The pressing member PP is an annular fixing plate having a smaller radius than the encoder scale disk SD along the circumferential direction at the end of the encoder scale disk SD. The encoder scale disk SD is attached to at least one end of the second drum member 22 by a plurality of fastening members, that is, an adjusting member 60 including a male screw portion 61 and a head portion 62, in the circumferential direction of the second drum member 22. Fixed.

  The installation azimuth line Le4 is extended to the inner circumference side of the encoder scale disk SD, and the inclined surface FP2 is formed on the inner circumference side of the encoder scale disk SD and in a cross section parallel to the second central axis AX2 and including the second central axis AX2. Are formed. The inclined surface FP2 is an inclined surface such that the thickness in the direction parallel to the second central axis AX2 becomes smaller as it approaches the second central axis AX2. The pressing member PP is formed with an inclined surface FP1 that becomes thicker in the direction parallel to the second central axis AX2 as it approaches the second central axis AX2. The pressing member PP is fixed to the encoder scale disk SD by the adjusting member 60 so that the inclined surface FP2 and the inclined surface FP1 face each other.

  The roundness adjusting mechanism Cs screws the male screw portion 61 of the adjusting member 60 into the female screw portion FP3 of the encoder scale disk SD, whereby the pressing force of the inclined surface FP1 of the pressing member PP is transmitted to the inclined surface FP2 and the encoder scale A small amount of elastic deformation is performed from the inner side of the disk SD toward the outer peripheral side. On the other hand, by rotating the male screw portion 61 to the opposite side, the pressing force suppressed by the inclined surface FP1 of the pressing member PP is transmitted to the inclined surface FP2, and a slight amount from the outer peripheral side of the encoder scale disk SD toward the inner side. Elastically deforms.

  The roundness adjusting mechanism Cs adjusts the diameter of the scale GP in the circumferential direction by a small amount by operating the male screw portion 61 in the adjusting member 60 provided with a plurality of circumferentially centered on the rotation center line AX2 at a predetermined pitch. can do. Further, since the roundness adjusting mechanism Cs can slightly deform the scale GP on the above-described installation azimuth lines Le1 to Le5, the circumferential diameter of the scale GP can be adjusted with high accuracy. Therefore, by operating the adjusting member 60 at an appropriate position in accordance with the circularity of the encoder scale disk SD, the circularity of the scale GP of the encoder scale disk SD can be increased, and a slight eccentricity error with respect to the rotation center line AX2 can be obtained. It is possible to improve the position detection accuracy in the rotational direction with respect to the second drum member 22 by reducing The adjustment amount adjusted by the roundness adjusting mechanism Cs varies depending on the diameter of the encoder scale disk SD or the radial position of the adjusting member 60, but is about several μm at maximum.

  As shown in FIG. 16, the encoder scale disk SD is fixed to the second drum member 22 by eight adjusting members 60. In this case, the encoder head EN4 as the first reading unit and the encoder head EN5 as the second reading unit have an encoder mounting angle θs that is a central angle between the encoder head EN4, the second central axis AX2, and the encoder head EN5. It is preferable that the adjustment member 60 and the second central axis AX2 adjacent to each other are arranged so as to be smaller than the central angle β.

  Since the encoder scale disk SD is fixed to the second drum member 22 by the adjusting member 60, deformation may occur near the adjusting member 60. As described above, by setting θs<β, the encoder heads EN4 and EN5 can reliably detect the error of the scale GP due to the deformation between the adjacent adjusting members 60. As a result, the correction accuracy of the scale GP is improved. Next, a modified example of the substrate processing apparatus (exposure apparatus) will be described.

(First Modification of Substrate Processing Apparatus (Exposure Apparatus))
FIG. 17 is a schematic diagram showing a first modification of the substrate processing apparatus (exposure apparatus). FIG. 18 is an explanatory diagram for explaining the position of the reading device when the encoder scale disk according to the first modified example of the substrate processing apparatus (exposure apparatus) is viewed in the rotation center line direction. In the above-described embodiment, the case where the position of the second drum member 22 supporting the substrate P in the circumferential direction is detected is taken as an example. The present invention is not limited to this, and when detecting the position in the circumferential direction of the first drum member 21 that holds the cylindrical mask DM as in the exposure apparatus EX1 of this modification, a pair of encoder heads are used to detect the position. It is possible to correct the error of the scale GP (scale) for detecting the position of the one drum member 21 in the circumferential direction.

  The encoder scale disk SD included in the exposure apparatus EX1 is fixed to both ends of the second drum member 22 orthogonal to the rotation axis AX2 of the second drum member 22. The scale GP is provided on the outer peripheral surfaces of both encoder scale disks SD. Therefore, the scale GP is arranged at both ends of the second drum member 22. The encoder heads EN1 to EN5 for reading the scales GP are arranged on both end sides of the second drum member 22, respectively.

  The first detector 25 (see FIG. 1) described in the above embodiment optically detects the rotational position of the first drum member 21, and has a high roundness encoder scale disk (scale member). It includes SD and encoder heads EH1, EH2, EH3, EH4 and EH5 which are reading devices. The encoder scale disk SD is fixed to at least one end portion (both end portions in FIG. 16) of the first drum member 21 orthogonal to the rotation axis of the first drum member 21. Therefore, the encoder scale disk SD rotates integrally with the rotation axis ST around the rotation center line AX1. A scale GPM is engraved on the outer peripheral surface of the encoder scale disk SD. The encoder heads EH1, EH2, EH3, EH4, EH5 are arranged around the scale GP as viewed from the rotation axis STM. The encoder heads EH1, EH2, EH3, EH4, and EH5 are arranged so as to face the scale GPM, and the scale GPM can be read without contact. The encoder heads EH1, EH2, EH3, EH4, EH5 are arranged at different positions in the circumferential direction of the first drum member 21. The first drum member 21 rotates from the encoder head EH4 toward the encoder head EH5.

  The encoder heads EH1, EH2, EH3, EH4, and EH5 are reading devices that have measurement sensitivity (detection sensitivity) with respect to displacement changes in the tangential direction (in the XZ plane) of the scale GPM. As shown in FIG. 17, when the installation azimuths of the encoder heads EH1 and EH2 (angle directions in the XZ plane about the rotation center line AX1) are represented by installation azimuth lines Le11 and Le12, the installation azimuth lines Le11 and Le12 are shown. , The encoder heads EH1 and EH2 are arranged so that the angle is ±θ° with respect to the center plane P3. The installation azimuth lines Le11 and Le12 coincide with the angular direction in the XZ plane about the rotation center line AX1 of the illumination light flux EL1 shown in FIG. Here, the illumination mechanism IU, which is the processing unit, performs a process of transmitting the illumination light beam EL1 to a predetermined pattern (mask pattern) on the cylindrical mask DM that is the object to be processed. Thereby, the projection optical system PL can project the image of the pattern in the illumination region IR on the cylindrical mask DM onto a part of the substrate P (projection region PA) being transported by the transport device.

  The encoder head EH4 rotates the installation direction line Le11 of the encoder head EH1 toward the rear side in the rotation direction with respect to the center plane P3 of the first drum member 21 by about 90° around the axis of the rotation center line AX1. It is set on the line Le14. Further, the encoder head EH5 is installed by rotating the installation azimuth line Le12 of the encoder head EH2 toward the rear side in the rotation direction with respect to the center plane P3 of the first drum member 21 by about 90° around the axis of the rotation center line AX1. It is set on the azimuth line Le15. Here, when approximately 90° is 90°±γ, the range of γ is the same as in the above-described embodiment.

  The encoder head EH3 rotates the installation azimuth line Le12 of the encoder head EH2 about 120° around the axis of rotation center line AX1 and rotates the encoder head EH4 about 120° around the axis of rotation center line AX1. It is set on the line Le13. The encoder heads EH1, EH2, EH3, EH4, and EH5 arranged around the first drum member 21 in this embodiment are arranged so that the encoder heads EN1 arranged around the second drum member 22 in the above-described embodiment. , EN2, EN3, EN4, EN5 are in a mirror image inverted relationship.

  As described above, the exposure apparatus EX1 is the second drum member 22 that is a cylindrical member, the scale GP, the projection modules PL1 to PL6 that are the processing units of the exposure apparatus EX1, and the first reading device that reads the scale GP. The encoder heads EN4 and EN5 are provided, and the encoder heads EN1 and EN2 that are second reading devices that read the scale GP are provided.

  The first drum member 21 has a curved surface that is curved with a constant radius from a first central axis AX1 that is a predetermined axis, and rotates around the first central axis AX1. The scale GPM is annularly arranged along the circumferential direction of the first drum member 21, and rotates together with the first drum member 21 around the first central axis AX1. The illumination mechanism IU, which is the processing unit of the exposure apparatus EX1, is disposed inside the first drum member 21 when viewed from the second central axis AX2, and the substrate P on the curved surface at a specific position in the circumferential direction of the first drum member 21. A process of transmitting the two illumination light fluxes EL1 to the (object to be processed) is performed. The encoder heads EH4 and EH5 are arranged around the scale GPM when viewed from the first central axis AX1, and the specific position described above is approximately 90 degrees around the first central axis AX1 about the first central axis AX1. It is placed in a rotated position and the scale GPM is read. The encoder heads EH1 and EH2 read the scale GPM at the specific position described above. Then, the exposure apparatus EX1 corrects the error (pitch error) of the scale GPM provided on the outer peripheral portion of the encoder scale disk SD attached to the first drum member 21, based on the reading values of the encoder heads EH4 and EH5. Therefore, the exposure apparatus EX1 can accurately measure the position of the first drum member 21 in the circumferential direction and perform processing on the object to be processed on the curved surface of the second drum member 22, that is, the substrate P. As described above, in the present embodiment, when the error is corrected based on the reading values of the encoder heads EH4 and EH5, specifically, the control device 14 of the exposure apparatus EX, the reading value of the encoder head EH4 and the encoder head EH5. The pitch error is corrected based on the difference from the read value of. However, when another encoder head is provided around the scale GPM of the first drum member 21 at an installation angle close to one of the encoder heads EH4 and EH5, a direct difference between the read values of both encoder heads EH4 and EH5 is provided. Not limited to the calculation, the pitch error can be obtained by calculation based on the read values of the three encoder heads EH4 and EH5 and the other encoder heads. The measurement of the pitch error by the three encoder heads will be described later in detail.

(Second Modification of Substrate Processing Device (Exposure Device))
FIG. 19 is a schematic diagram showing the overall configuration of a second modified example of the substrate processing apparatus (exposure apparatus). In the exposure apparatus EX2, a light source device (not shown) emits the illumination light flux EL1 with which the cylindrical mask DM is illuminated. When the illumination light flux EL1 emitted from the light source of the light source device is guided to the illumination module IL and a plurality of illumination optical systems are provided, the illumination light flux EL1 from the light source is divided into a plurality of illumination light fluxes EL1. Lead to module IL.

  The illumination light flux EL1 emitted from the light source device enters the polarization beam splitters SP1 and SP2. In the polarization beam splitters SP1 and SP2, in order to suppress energy loss due to the separation of the illumination light beam EL1, it is preferable that the incident illumination light beam EL1 is totally reflected. Here, the polarization beam splitters SP1 and SP2 reflect the light flux of S-polarized linearly polarized light and transmit the light flux of P-polarized linearly polarized light. Therefore, the light source device emits the illumination light flux EL1 that is the linearly polarized (S-polarized) light flux that enters the polarization beam splitters SP1 and SP2 to the first drum member 21. As a result, the light source device emits the illumination light flux EL1 having the same wavelength and phase.

  The polarization beam splitters SP1 and SP2 reflect the illumination light beam EL1 from the light source while transmitting the projection light beam EL2 reflected by the cylindrical mask DM. In other words, the illumination light beam EL1 from the illumination optical module ILM enters the polarization beam splitters SP1 and SP2 as reflected light beams, and the projection light beam EL2 from the cylindrical mask DM enters the polarization beam splitters SP1 and SP2 as transmitted light beams. ..

  In this way, the illumination module IL, which is the processing unit, performs a process of reflecting the illumination light flux EL1 into a predetermined pattern (mask pattern) on the cylindrical mask DM that is the object to be processed. Thereby, the projection optical system PL can project the image of the pattern in the illumination region IR on the cylindrical mask DM onto a part (projection region) of the substrate P being transported by the transport device.

  When a predetermined pattern (mask pattern) that reflects the illumination light beam EL1 is provided on the curved surface of such a cylindrical mask DM, a scale GPm can be provided on the curved surface together with this mask pattern. When this scale GPm is formed simultaneously with the mask pattern, the scale GPm is formed with the same accuracy as the mask pattern. Therefore, the curved surface detection probes GS1 and GS2 that detect the scale GPm can sample the mark image of the scale GPm at high speed and with high accuracy. At the moment when this sampling is performed, the correspondence relationship between the rotation angle position of the first drum member 21 and the scale GPm is obtained, and the rotation angle position of the first drum member 21 that is sequentially measured can be stored.

  The encoder heads EH4 and EH5 on the cylindrical mask DM side read the scale GPm. Then, the exposure apparatus EX2 corrects the error of the scale GPm provided on the surface of the cylindrical mask DM based on the difference between the read values of the encoder heads EH4 and EH5. Therefore, the exposure apparatus EX2 can accurately measure the position of the cylindrical mask DM in the circumferential direction, and can process the substrate P on the curved surface of the second drum member 22.

  The encoder heads EN4 and EN5 on the second drum member 22 side read the scale GPd of the encoder scale disk SD attached to the second drum member 22. Then, the exposure apparatus EX2 corrects the error of the scale GPd provided on the surface of the encoder scale disk SD based on the difference between the read values of the encoder heads EN4 and EN5. For this reason, the exposure apparatus EX2 can accurately measure the position of the second drum member 22 in the circumferential direction and process the substrate P on the curved surface of the second drum member 22.

(Third Modification of Substrate Processing Apparatus (Exposure Apparatus))
FIG. 20 is a schematic diagram showing the overall configuration of a third modification of the substrate processing apparatus (exposure apparatus). The exposure apparatus EX3 includes polygon scanning units PO1 and PO2 that enter an exposure beam from a light source device (not shown). To scan. The exposure apparatus EX3 included in the substrate processing apparatus can irradiate the substrate P at a specific position with the exposure light and draw a predetermined pattern without the cylindrical mask DM.

  The encoder heads EN4 and EN5 on the second drum member 22 side of the exposure apparatus EX3 read the scale GPd of the encoder scale disk SD attached to the second drum member 22. Then, the exposure apparatus EX2 corrects the error of the scale GPd provided on the surface of the encoder scale disk SD based on the difference between the read values of the encoder heads EN4 and EN5. For this reason, the exposure apparatus EX2 can accurately measure the position of the second drum member 22 in the circumferential direction and process the substrate P on the curved surface of the second drum member 22.

  As shown in FIG. 20 (and FIG. 19), the encoder heads EN4 and EN5 on the second drum member 22 side are arranged corresponding to the alignment microscopes AMG1 and AMG2 arranged in two rows in the circumferential direction. Was done. However, among the alignment microscopes AMG1 and AMG2, only the alignment microscope AMG2 (and the corresponding encoder head EN5) may be arranged. Even in such a case, it is preferable to provide the encoder head EN4. When only the alignment microscope AMG2 (and the corresponding encoder head EN5) is arranged and the encoder head EN4 cannot be installed in the vicinity thereof at the rotation angle α, the encoder heads EN1 and EN2 arranged corresponding to the exposure position are used to scale. An error map such as a pitch error or eccentricity of the scale GPd of the disk SD may be created.

  Further, at least one error map such as pitch error and eccentricity of the scale GPd obtained by using the two encoder heads EN4 and EN5, and pitch error and eccentricity of the scale GPd obtained by using the two encoder heads EN1 and EN2. By comparing with at least one error map of, it is verified whether there is a large difference between both error maps, and if there is a difference more than the allowable value, the error map can be recreated. The accuracy and reliability of the error map can be improved.

(Fourth Modification of Substrate Processing Device (Exposure Device))
FIG. 21 is a schematic diagram showing the overall configuration of a fourth modification of the substrate processing apparatus (exposure apparatus). The exposure apparatus EX4 is a substrate processing apparatus that performs so-called proximity exposure on the substrate P. In the exposure apparatus EX4, the gap between the cylindrical mask DM and the second drum member 22 is set to be small, and the illumination mechanism IU directly irradiates the substrate P with the illumination light beam EL1 for non-contact exposure. In the present embodiment, the second drum member 22 rotates by the torque supplied from the second drive unit 36 including an actuator such as an electric motor. The drive roller MGG connected by, for example, a magnetic gear drives the first drum member 21 so as to rotate in the opposite direction to the rotation direction of the second drive unit 36. The second drive unit 36 rotates the second drum member 22, rotates the drive roller MGG and the first drum member 21, and synchronizes the first drum member 21 (cylindrical mask DM) and the second drum member 22. Move (synchronize).

  The exposure apparatus EX4 also includes an encoder head EN6 that detects the position PX6 of the scale GP at a specific position where the principal ray of the imaging light flux EL2 is incident on the substrate P with respect to the substrate P. Here, since the diameter of the outer peripheral surface of the outer peripheral surface of the second drum member 22 around which the substrate P is wound is aligned with the diameter of the scale GP of the encoder scale disk SD, the position PX6 is located from the second central axis AX2. It matches the specific position described above. The encoder head EN7 is set on the installation azimuth line Le7 obtained by rotating the installation azimuth line Le6 of the encoder head EN6 toward the rear side in the feeding direction of the substrate P by about 90° around the axis of the rotation center line AX2. ..

  The exposure apparatus EX4 uses, for example, the encoder head EN3 as a first reading unit and the encoder head EN7 as a second reading unit. The encoder heads EN3 and EN7 read the scale GPd of the encoder scale disk SD attached to the second drum member 22. The control device 14 corrects the error of the scale GPd provided on the surface of the encoder scale disk SD based on the difference between the read values of the encoder heads EN3 and EN7. For this reason, the exposure apparatus EX4 can accurately measure the position of the second drum member 22 in the circumferential direction and process the substrate P on the curved surface of the second drum member 22. The encoder heads EN3 and EN7 are encoders formed by a line connecting the encoder head EN3 and the second central axis AX2 (installation azimuth line Le3) and a line connecting the encoder head EN7 and the second central axis AX2 (installation azimuth line Le7). The mounting angle θs may be less than 90 degrees, preferably 45 degrees or less.

  The above-described embodiment and the first to fourth modifications of the substrate processing apparatus (exposure apparatus) exemplify the exposure apparatus as the substrate processing apparatus. The substrate processing apparatus is not limited to the exposure apparatus, and the processing unit may be an apparatus that prints a pattern on the substrate P that is the object to be processed by an ink dropping device such as an inkjet device. Further, the processing unit may be an inspection device.

  By the way, in the description of the reading operation of the scale GP by the encoder heads EN4 and EN5 using FIG. 7 and FIG. The down pulse D is output when the falling part GPb is read, and the interval between two adjacent rising parts GPa or the interval between adjacent falling parts GPb is defined as the pitch SS of the scale GP. However, in an actual encoder measurement system, for example, as disclosed in Japanese Patent Laid-Open No. 9-196702, a two-phase signal (a sine wave having a phase difference of 90 degrees) output from a signal generator (encoder head) is used. Signal and cosine wave signal) to generate up-pulse U and down-pulse D at intervals obtained by subdividing the actual size pitch SS of the graduation GP into a few fractions to a few tenths by an interpolation circuit or a comparator. Is configured as follows.

  FIG. 22 briefly describes an actual reading operation of the scale GP (GPd, GPM) by the encoder heads EN1 to EN7 and EH1 to EH5 shown in each of FIGS. 4 to 6, 10 and 16 to 21. 3 is a signal waveform diagram for FIG. As shown in FIG. 22, each of the encoder heads EN1 to EN7 and EH1 to EH5 outputs two measurement signals (here, represented by rectangular waves) EcA and EcB having a phase difference of 90 degrees. One cycle of the measurement signals EcA and EcB corresponds to 1/n of the pitch SS of the scale GP. Although n (integer) varies depending on the optical reading form in the encoder head, it is set to any value of a multiple series such as 1, 2, 4, 8,... In a normal encoder measurement system, while the scale disk SD rotates in the forward direction and the scale GP always moves in one direction with respect to the encoder head, interpolation up based on the measurement signals EcA and EcB is performed. The pulse signal EcU continues to be generated. When the scale disk SD reverses, the down pulse signal interpolated based on the measurement signals EcA and EcB continues to be generated from that point.

  In FIG. 22, a processing circuit that sends out an up pulse signal EcU (or a down pulse signal EcD) that generates a pulse at an interval obtained by dividing one cycle of the measurement signals EcA and EcB into eight is used. The up/down counter as a counter sequentially counts up the number of pulses when the up pulse signal EcU is input, and sequentially counts down the number of pulses when the down pulse signal EcD is input. Here, for example, assuming that one cycle of the measurement signals EcA and EcB corresponds to ⅛ of the actual size pitch SS of the scale GP, the up/down counter has the scale surface of the scale disk SD (scale GP) in order. While moving by one pitch SS in the direction, 64 pulses of the up pulse signal EcU are counted. Therefore, when the pitch SS of the scale GP is 20 μm, the increment of the count value for one pitch by the up/down counter is 64, and the measurement resolution (movement amount per pulse of the signal EcU) as the encoder measurement system is 0. .3125 μm (20 μm/64). As described above, the measurement resolution of the encoder measurement system is finely obtained by interpolating the actual size of the pitch SS of the scale GP to a fraction of several to several tenths. It is obtained with accuracy according to the degree of the interpolation.

  When the circumferential distance (diameter x π) of the scale surface of the scale disk SD is divided by a finite number of scales (number of grids), the actual size of the pitch SS of the actual scale GP is a fraction with respect to 20 μm. It may be accompanied. On the other hand, the pitch SS is set so that the measurement resolution has a good value (for example, 0.25 μm), and the circumferential distance of the scale surface of the scale disk SD is divisible by the pitch SS within a predetermined accuracy. Alternatively, the diameter of the scale surface may be set.

  By the way, on the scale surface of the scale disk SD or the like, an origin mark which is an origin of one rotation of the scale disk SD is engraved together with the scale GP, and each of the encoder heads EN1 to EN7 detects the origin mark. At that moment, the origin signal (pulse) EcZ is output. The up/down counter resets the count value up to that point to zero in response to the origin signal EcZ, and then continues counting the number of pulses of the up pulse signal EcU (or the down pulse signal EcD) again. Therefore, each of the up/down counters provided corresponding to each of the encoder heads EN1 to EN7 has a pulse of the up pulse signal EcU (or the down pulse signal EcD) with the moment when the origin signal EcZ is received as a reference (zero point). Adding (or subtracting) numbers.

  From the above, for example, when the pitch error of the scale GP is obtained from the difference between the measured values of the encoder head EN4 and the encoder head EN5, the scale disk SD (second drum member 22) is set in FIGS. As described above, every time it rotates by the angle α, the difference between the count value of the up/down counter corresponding to the encoder head EN4 and the count value of the up/down counter corresponding to the encoder head EN5 is simply calculated digitally. Good. As for the angle α, the count value of the up/down counter corresponding to either one of the encoder head EN4 and the encoder head EN5 (or another one of the encoder heads may be increased) is increased by a constant value corresponding to the angle α. It can be detected by determining whether (or decreased).

(Modification 1)
In the above-described embodiment and modification, the scale GP constituting the encoder measurement system is engraved on the cylindrical outer peripheral surface of at least one of the scale disk SD as the rotating body and the second drum member 22. However, the scale GP may be formed at a predetermined pitch along the circumferential direction on the side end surface of at least one of the scale disk SD and the second drum member 22 that is perpendicular to the rotation center line AX2. FIG. 23 is a view of the configuration in the case where the scale GP is formed on the side end surface of the scale disk SD as described above, as seen from the direction in which the rotation center line AX2 extends (Y-axis direction), as in the case of FIG. 24 is a cross-sectional view taken along the line AA′ of the configuration of FIG. 23 taken along a plane including the installation azimuth line Le4 and the rotation center line AX2.

  In FIG. 23, the ring-shaped scale disk SD is attached to eight side end surfaces of the second drum member 22 with adjusting members (screws) 60. The attachment angle β of the adjusting member (screw) 60 is 45° here. A scale GP having a constant pitch SS and an origin mark Zs are formed on the side surface parallel to the XZ plane of the scale disk SD along the circumference of a radius ra from the rotation center line AX2. As shown in FIG. 24, the encoder heads EN4 and EN5 are arranged in the Y-axis direction so as to face the scale GP with a constant gap. As shown in FIG. 23, the reading position RP4 of the encoder head EN4 is set on the radius ra and on the installation azimuth line Le4. The reading position RP5 of the encoder head EN5 is set on the radius ra and on the installation azimuth line Le5. The radius ra is the radius of the outer peripheral surface 22s of the second drum member 22, which closely supports the substrate P, as shown in FIG. Therefore, the maximum diameter of the ring-shaped scale disk SD is set to be slightly larger than the radius ra. By disposing the scale disk SD and the encoder heads EN4 and EN5 in this way, the Abbe error during measurement can be minimized. Other encoder heads (EN1 to EN3, EN6 to EN7) for the ring-shaped scale disk SD in FIGS. 23 and 24 are also arranged so as to satisfy the Abbe condition for measurement, similarly to the heads EN4 and EN5.

(Modification 2)
In the above-described embodiment and modified example, in order to measure the pitch error of the scale GP, the difference value between the measured values of two encoder heads (for example, encoder heads EN4 and EN5) arranged close to each other is calculated as a scale disk. It is assumed that the SD (the second drum member 22) is stored each time it rotates by the angle α (α<θs), and a map regarding the pitch error for the entire circumference of the scale disk SD is created. In that case, in order to improve the accuracy of the map, the angle θs formed by the reading positions (corresponding to RP4 and RP5 in FIG. 23) on the scale surface of each of the two encoder heads (for example, encoder heads EN4 and EN5) is set. It is preferable to make it as small as possible. However, the angle θs may not be sufficiently small depending on the outer shapes and dimensions of the encoder heads EN4 and EN5 or the angles between the installation azimuth lines Le4 and Le5 determined by the arrangement of the alignment microscopes AMG1 and AMG2.

  Therefore, in the second modification, for example, three or more encoder heads EN4 and EN5 used for the pitch error measurement in the above-described embodiment and an encoder head EN1 (or EN2) arranged in the vicinity thereof are added. The measurement value by each of the encoder heads is used to further refine the pitch error map. FIG. 25 is a view of the arrangement of the scale disk SD (here, ring-shaped) and the encoder heads EN1, EN2, EN4, EN5 as seen in the XZ plane, as in FIG. 6 above. A scale GP and an origin mark Zs are formed along the outer peripheral surface of the SD. Further, it is assumed that the scale disk SD is fixed to the side end surface of the second drum member 22 by adjusting members (screws) 60 at 16 places in the circumferential direction. Therefore, the attachment angle β of the adjusting member (screw) 60 is 22.5°.

  As shown in FIG. 25, the encoder head EN1 whose reading position is set on the installation bearing line Le1 corresponding to the odd-numbered exposure position, and the reading position on the installation bearing line Le2 corresponding to the even-numbered exposure position. The encoder head EN2 to be set is arranged at an angle ±θ with respect to the center plane P3 in the XZ plane. The angle θs formed by the installation azimuth lines Le4 and Le5 passing through the reading positions of the encoder heads EN4 and EN5 has a relationship of θs>β. Further, the angle formed by the installation azimuth line Le1 passing through the reading position of the encoder head EN1 and the installation azimuth line Le4 passing through the reading position of the encoder head EN4 is θq. Further, the count values of up/down counters provided corresponding to the encoder heads EN1, EN2, EN4, and EN5 are Cm1, Cm2, Cm4, and Cm5, respectively.

  When the scale disk SD (second drum member 22) shown in FIG. 25 rotates clockwise in the XZ plane, the origin marks Zs formed on the scale surface of the scale disk SD are encoder heads EN4, EN5, EN1. , EN2 in that order, across each reading position. Therefore, at the moment when the origin mark Zs crosses the reading position of the encoder head EN4, the count value Cm4 of the corresponding up-down counter is reset to zero, and the origin mark Zs corresponds to the moment when it crosses the reading position of the encoder head EN5. At the moment when the count value Cm5 of the up/down counter is reset to zero and the origin mark Zs crosses the reading position of the encoder head EN1, the count value Cm1 of the corresponding up/down counter is reset to zero and the origin mark Zs is set to the encoder head EN2. At the moment of crossing the reading position, the count value Cm2 of the corresponding up/down counter is reset to zero. When the scale disk SD rotates clockwise, each count value Cm1, Cm2, Cm4, Cm5 after all four up/down counters are reset to zero always has a relationship of Cm2<Cm1<Cm5<Cm4. There is.

Therefore, when the pitch error is obtained by using the three encoder heads EN1 (count value Cm1), EN4 (count value Cm4), and EN5 (count value Cm5) to create an error map (correction map), a scale disk is used. Every time the SD (second drum member 22) rotates by a constant angle α (α<β<θs), the measurement value ΔMs related to the pitch error for each unit angle α is obtained by the following equation (1). This measured value ΔMs corresponds to the number NS of the graduations GP shown in FIG. 11 above, but is actually the count value of the up-pulse (or down-pulse) EcU pulse shown in FIG. ..
ΔMs=(Cm4+Cm1)/2−Cm5... Formula (1)

  In the formula (1), the calculated value of (Cm4+Cm1)/2 is set to an angular position which is an intermediate point between the reading position RP4 of the encoder head EN4 and the reading position RP1 of the encoder head EN1, as shown in FIG. The count value expected to be obtained when the reading position RPi of the encoder head is set on the virtual installation azimuth line Lei is shown. Therefore, the measured value ΔMs obtained by the equation (1) or (2) described later is calculated by the count value Cmi (calculated value) at the reading position RPi by the virtual encoder head and the reading position RP5 of the encoder head EN5. It becomes the difference with the count value Cm5 of. By obtaining the measured value ΔMs for 360 degrees for each unit angle α, the pitch error map of the scale (scale GP) such as the scale disk SD or the pitch error correction map can be created.

  By the way, in the case of the arrangement shown in FIG. 25, in the period in which the origin mark Zs passes between the reading position RP4 of the encoder head EN4 and the reading position RP1 of the encoder head EN1, the count value Cm4 and the measured value Cm5 are obtained. Is after zero reset, the continuity of the three count values Cm1, Cm4, and Cm5 may not be guaranteed. During that period, the magnitude relationship among the count values Cm1, Cm4, and Cm5 (absolute value) is Cm4<Cm1<Cm5 or Cm5<Cm4<Cm1. Therefore, the maximum count value (fixed value) counted by the up/down counter from the time of zero reset to the time of the next zero reset is set as Cmf, and each angle α corresponds to each encoder head EN1, EN4, EN5. When reading the count values Cm1, Cm4, Cm5, when the origin mark Zs is between the read position RP4 of the encoder head EN4 and the read position RP5 of the encoder head EN5, the count value Cm4 of the up/down counter is the maximum count value. A new count value Cm4′ to which Cmf is added may be used instead of the count value Cm4 in the equation (1). Similarly, when the origin mark Zs is between the reading position RP5 of the encoder head EN5 and the reading position RP1 of the encoder head EN1, a new maximum count value Cmf is added to each of the count values Cm4 and Cm5 of the up/down counter. Different count values Cm4′ and Cm5′ may be used instead of the count values Cm4 and Cm5 in the equation (1).

  In the case of the modified example 2, the first reading unit is configured to include two encoder heads EN1 and EN4 (or one encoder head EN5), and the second reading unit is one encoder head EN5 (or two encoder heads EN5). EN1 and EN4) are included. In the above configuration, when the angle θs and the angle θq are set to have an appropriate relationship, the angle formed by the virtual reading position RPi and the reading position RP5 becomes the attachment angle β of the adjusting member (screw) 60 (22. 5°), and the pitch error (pitch unevenness) due to slight deformation of the scale surface of the scale disk SD that remains after the adjustment of the roundness, eccentricity, etc. by the adjustment member 60 is performed with a span of the angle β or less. Can be measured in detail.

In the method of setting the virtual reading position RPi on the scale surface as described above, for example, it is set at the midpoint between the reading positions RP4 and RP1 of the two encoder heads EN4 and EN1 shown in FIG. The calculated count value obtained by the virtual first reading position RPi and the virtual second reading position RPi set at the midpoint between the reading positions RP5 and RP2 of the two encoder heads EN5 and EN2. The pitch error may be obtained from the difference from the calculated count value. The measured value ΔMs for each unit angle α in that case is calculated by the following equation (2).
ΔMs=(Cm4+Cm1)/2−(Cm5+Cm2)/2... Formula (2)

(Device manufacturing method)
FIG. 26 is a flowchart showing the procedure of a device manufacturing method for manufacturing a device using the substrate processing apparatus (exposure apparatus) according to the embodiment. In this device manufacturing method, first, the function/performance of a display panel is formed by a self-luminous element such as an organic EL, and necessary circuit patterns and wiring patterns are designed by CAD or the like (step S201). Next, the cylindrical mask DM for the required layers is manufactured based on the pattern for each layer designed by CAD or the like (step S202). Further, a supply roll FR1 around which a flexible substrate P (resin film, metal foil film, plastic, etc.) serving as a base material of the display panel is wound is prepared (step S203). The roll-shaped substrate P prepared in step S203 has its surface modified as necessary, a pre-formed underlayer (for example, fine irregularities by imprint method), and a light-sensitive substrate. It may be a laminate of a functional film and a transparent film (insulating material) in advance.

  Next, a backplane layer composed of electrodes, wirings, insulating films, TFTs (thin film semiconductors), etc. that form the display panel device is formed on the substrate P, and organic EL or the like is formed so as to be laminated on the backplane. A light emitting layer (display pixel portion) is formed by the self-luminous element (step S204). This step S204 also includes a conventional photolithography process of exposing the photoresist layer using the exposure apparatuses EX, EX2, EX3, and EX4 described in each of the above embodiments, but the photosensitivity instead of the photoresist is used. A substrate P coated with a silane coupling material is subjected to pattern exposure to form a pattern having hydrophilicity/hydrophobicity on the surface, and a photosensitive catalyst layer is subjected to pattern exposure to form a pattern of a metal film (wiring, electrodes) by electroless plating. Etc.) and a printing step of drawing a pattern with a conductive ink containing silver nanoparticles, and the like.

  Next, the substrate P is diced for each display panel device continuously manufactured on the long substrate P by the roll method, and a protective film (environmental barrier layer) or a color filter is provided on the surface of each display panel device. A device or the like is assembled by laminating sheets or the like (step S205). Next, an inspection process is performed to determine whether the display panel device functions normally or satisfies the desired performance and characteristics (step S206). A display panel (flexible display) can be manufactured as described above.

  In the above embodiment, the light-transmitting reticle in which a predetermined light-shielding pattern (or phase pattern/dimming pattern) is formed on the light-transmitting substrate is used, but instead of this reticle, for example, US Pat. No. 6,778,257 is used. As described in the specification, a variable shaped reticle (also called an electronic reticle, active reticle, or image generator) that forms a transmission pattern, a reflection pattern, or a light emission pattern based on electronic data of a pattern to be exposed. May be used. Further, instead of the variable-shaped reticle including the non-emissive image display element, a pattern forming apparatus including the self-emissive image display element may be provided.

  Further, the exposure apparatus of the above embodiment is manufactured by assembling various subsystems including the constituent elements recited in the claims of the present application so as to maintain predetermined mechanical accuracy, electrical accuracy, and optical accuracy. To be done. In order to ensure these various accuracies, before and after assembling the exposure apparatus, adjustment for achieving optical accuracy for various optical systems, adjustment for achieving mechanical accuracy for various mechanical systems, and various electrical Adjustments are made to the system to achieve electrical accuracy. The process of assembling various subsystems into an exposure apparatus includes mechanical connection between various subsystems, wiring connection of electric circuits, piping connection of atmospheric pressure circuits, and the like. It goes without saying that there is an individual assembly process for each subsystem before the assembly process for the exposure apparatus from these various subsystems. When the process of assembling the various subsystems into the exposure apparatus is completed, comprehensive adjustment is performed to ensure various accuracies of the exposure apparatus as a whole. It is desirable that the exposure apparatus be manufactured in a clean room where the temperature and cleanliness are controlled.

  Further, the constituent elements of the above-described embodiments can be appropriately combined. In addition, some components may not be used. Further, the constituent elements can be replaced or changed without departing from the scope of the present invention. Further, as far as legally permitted, the descriptions of all the publications and US patents relating to the exposure apparatus and the like cited in the above-mentioned embodiments are incorporated as part of the description of the present specification. As described above, all other embodiments and operational techniques made by those skilled in the art based on the above embodiments are included in the scope of the present invention.

11 Substrate Processing Device 14 Control Device 21 First Drum Member 22 Second Drum Member 25 First Detector 35 Second Detector 60 Adjusting Member 61 Male Screw Part 62 Head Part AX1 Rotation Center Line (First Central Axis)
AX2 rotation center line (second center axis)
Cs Roundness adjusting mechanism DM Cylindrical mask EN, EN1, EN2, EN3, EN4, EN5, EH1, EH2, EH3, EH4, EH5 Encoder head EX, EX1, EX2, EX3, EX4 Exposure device GP, GPd, GPM, GPm Scale NS Measurement scale number P Substrate SD Encoder Scale disk TBc Correction map

Claims (7)

A substrate processing apparatus which conveys a long sheet substrate having flexibility in a long direction and performs a predetermined process on the sheet substrate,
A rotation for closely supporting a part of the sheet substrate in the longitudinal direction with an outer peripheral surface curved in a cylindrical shape with a constant radius from the center line, and rotating around the center line to convey the sheet substrate in the longitudinal direction. A drum,
A first processing unit that performs processing on the sheet substrate at a first specific position within a circumferential range in which the sheet substrate is closely supported by the rotating drum;
In order to perform encoder measurement of the position change in the circumferential direction of the sheet substrate, it is provided on the end surface side of the rotary drum in the direction of the center line so as to rotate around the center line together with the rotary drum. A scale disk, in which a graduated ring is engraved in an annular shape at a predetermined pitch along the circumferential direction of rotation and is formed on a side surface perpendicular to the center line,
A first encoder head, which has the same azimuth as the first specific position with respect to the circumferential direction of rotation of the rotary drum, is arranged to face the side surface of the scale disk, and reads the scale;
The radius of the outer peripheral surface of the rotating drum and the distance from the center line of the reading position of the scale by the first encoder head are set to be the same.
Substrate processing equipment. The substrate processing apparatus according to claim 1, wherein
The scale disk is fixed to the end surface side of the rotating drum by a fastening member arranged at each of a plurality of circumferential positions at intervals of a predetermined mounting angle β around the center line,
Substrate processing equipment. The substrate processing apparatus according to claim 1, wherein
The sheet substrate has a specific pattern formed discretely or continuously along the longitudinal direction,
Within a circumferential range in which the sheet substrate is closely supported by the rotary drum, at a second specific position set on the upstream side of the first specific position in the transport direction of the sheet substrate, Further comprising a detection probe for detecting the specific pattern formed on the sheet substrate,
Substrate processing equipment. The substrate processing apparatus according to claim 3, wherein
A second encoder head that is arranged in the same direction as the second specific position with respect to the circumferential direction of rotation of the rotating drum and is arranged so as to face the scale of the scale disk, and that reads the scale.
Substrate processing equipment. The substrate processing apparatus according to claim 4, wherein
With respect to the circumferential direction of rotation of the rotary drum, the azimuth is rotated about the center line from the second specific position by an angle θs, and the azimuth is arranged so as to face the scale of the scale disk. Further comprising a third encoder head for reading,
The angle θs and the mounting angle β are set in a relation of θs<β,
Substrate processing equipment. The substrate processing apparatus according to claim 5, wherein
Further comprising a controller that measures a pitch error of the scale based on a change in a difference between the read value of the scale by the second encoder head and the read value of the scale by the third encoder head.
Substrate processing equipment. The substrate processing apparatus according to any one of claims 1 to 6,
The radius from the center line of the scale disk is made larger than the radius of the outer peripheral surface of the rotating drum,
Substrate processing equipment.

Images