Equipping a component carrier using offset information between structural features formed on opposite sides of a reference component

Abstract

Method for equipping a component carrier (180) with electronic components (190, 490c), the method comprising optically capturing a first image of a first side of a first component on which a first structural feature (296a, 496a) of the first component (190, 490c) is recognizable; optically capturing a second image of a second side of the first component (190, 490c) on which a second structural feature (294, 296b, 496b) of the first component (190, 490c) is recognizable, the first side and the second side being opposite one another; Determining a spatial offset (c1) between the first structural feature (296a, 496a) and the second structural feature (294, 296b, 496b); optically capturing an image of a side of a second component (190, 490c); Determining a spatial position of the second component (190, 490c) based on the captured image of the side of the second component (190, 490c); and Mounting the second component (190, 490c) on the component carrier (180) based on the determined spatial position of the second component (190, 490c) and the determined spatial offset (c1) between the first structural feature (296a, 496a) and the second structural feature (294, 296b, 496b) of the first component.

Description

Technical area

[0001] The present invention relates to a method and a device for equipping a component carrier with electronic components, which are optically measured on two opposite sides before being mounted on the component carrier. Background of the invention

[0002] In the production of packaged electronic components, unpackaged (semiconductor) chips, so-called "bare dies", are mounted on a component carrier or carrier. In the so-called "embedded wafer level package" (eWLP) process, one or more chips per package are placed with the active side facing up on an adhesive film attached to the carrier. A large number of placed chips are then encapsulated with a plastic mass, which later forms the housing. The entire encapsulated product is then baked under high pressure and subsequently detached from the carrier or adhesive film. In subsequent process steps, the chips are then contacted, electrically connected if necessary, and solder balls are applied to serve as electrical connections. Finally, the entire further processed encapsulated product is sawn or otherwise separated into individual components.

[0003] In concrete terms, eWLP is a package design for integrated circuits in which the electrical connection contacts are created on a wafer artificially created from chips and encapsulant. All necessary processing steps for forming a package are also performed on the artificial wafer. Compared to traditional packaging technologies, which use so-called "wire bonding," this allows the production of extremely small and flat packages with excellent electrical and thermal properties at particularly low production costs. This technology can be used to manufacture components, for example, as ball grid arrays (BGAs).

[0004] The handling of unpackaged chips in an eWLP process is typically carried out using a (modified) placement machine compared to the familiar surface mount technology. Such a placement machine features a placement head with which the chips are mounted or placed at predefined placement positions on the respective carrier. The requirements for positioning accuracy are particularly high.

[0005] For high-precision placement, for example for eWLP processes, it is often necessary to mount the (unpackaged) components to be mounted on a component carrier based on the structures on their top side. However, a conventional component camera on a placement machine can only measure structures on the underside of the component. An offset or spatial displacement between structures on the top side and structures on the underside can be determined by an additional measurement of the top side with an (additional) camera. Such additional measurement of the component slows down the placement process. This may be of little significance for placement orders where only a few components need to be mounted on a component carrier with such high precision. However, when manufacturing packaged electronic components using "eWLP", a component carrier is loaded with a large number of components, for example 100.000, so that even the slightest “slowdown” in the assembly process leads to a significant loss of efficiency.

[0006] DE 10 2014 101 901 A1 discloses a method for mounting an electronic component on a component carrier, which comprises (a) optically capturing a first image of a first side of the component, on which a first structural feature of the component can be recognized, by means of a first camera, (b) optically capturing a second image of a second side of the component, on which a second structural feature of the component is recognizable, by means of a second camera, wherein the first side and the second side are opposite one another and wherein the second structural feature is configured to be connected at a predetermined position on the component carrier, (c) orienting the electronic component such that a center of the first structural feature is aligned with a desired position relative to the component carrier, wherein the second structural feature may be offset from the predetermined position accordingly, and (d) mounting the electronic component onto the component carrier with the center of the first structural feature aligned relative to the component carrier, the second structural feature offset from the pre-determined position.

[0007] DE 10 2015 112 518 B3 discloses a placement machine for populating a carrier with unpackaged chips. This placement machine comprises (a) a chassis; (b) a feed device for providing a wafer having a plurality of chips, wherein the feed device is attached to the chassis; (c) a lifting device attached to the chassis; (d) a carrier receiving device for receiving the carrier to be populated, wherein the carrier receiving device is attached to the lifting device and is movable relative to the chassis along a z-direction by means of the lifting device; and (e) a placement head for retrieving chips from the provided wafer and for placing the retrieved chips at predefined placement positions on the received carrier.The carrier receiving device comprises (d1) a one-piece support element, which represents at least an upper part of the carrier receiving device, wherein the carrier to be loaded can be placed on the one-piece support element; and (d2) a transport device for transporting the carrier on an upper side of the one-piece support element along a transport direction. The transport device is spatially integrated into the one-piece support element at the upper side of the one-piece support element.

[0008] JP 2016-92225 A discloses a transfer device 101 for visual inspection of electronic components. The transfer device comprises (a) a base member that rotates about the rotation axis, (b) a plurality of component holding devices configured as suction cups, which are arranged along a circumferential direction of the base member and move during rotation, (c) drive means for rotating each component holding device about the rotation axis, (d) a guide mechanism for guiding each component holding device about the rotation axis, (e) a component supply device, (f) a visual inspection device, and (g) a component dispensing device.Since the component holding device 142 can be present simultaneously in the operating areas of the component supply device 110, the visual inspection device 200 and the component ejection device 170, the operations of component supply, visual inspection and component ejection can be carried out simultaneously.

[0009] The invention is based on the object of specifying a method and a device for a highly precise and yet time-efficient assembly of a component carrier with components, wherein information about structural features present on different sides of the components is taken into account. Summary of the invention

[0010] This object is achieved by the subject matter of the independent patent claims. Advantageous embodiments of the present invention are described in the dependent claims.

[0011] According to a first aspect of the invention, a method for populating a component carrier with electronic components is described. The described method comprises (a) optically capturing a first image of a first side of a first component, on which a first structural feature of the first component can be seen, in particular by means of a first camera; (b) optically capturing a second image of a second side of the first component, on which a second structural feature of the first component can be seen, in particular by means of a second camera; (c) determining a spatial offset between the first structural feature and the second structural feature; (d) optically capturing an image of a side of a second component, in particular by means of the first camera or the second camera;(e) determining a spatial position of the second component based on the captured image of the side of the second component; and (f) mounting the second component on the component carrier based on the determined spatial position of the second component and the determined spatial offset between the first structural feature and the second structural feature of the first component.

[0012] The described method is based on the finding that high-precision assembly of a plurality or series of components can be achieved efficiently and quickly by using a first component as a reference component, and determining the spatial offset between an optically detectable structure on the top side of the reference component and an optically detectable structure on the bottom side of the reference component only for this component. This determination of the spatial offset should be carried out with particularly high accuracy, whereby the time factor can play a rather subordinate role for this determination.The second component is then optically captured on only one of its sides using a single camera. During subsequent assembly of the second component, not only the image information from the image of the second component captured from only one side is taken into account, but also the spatial offset determined for the first reference component. The described method assumes that, at least for a certain number of components (preferably from a batch or, in the case of semiconductor chips, from a wafer), a possible spatial offset between the top and bottom sides is at least approximately equal.

[0013] In this document, the term "electronic component" or "component" refers to all components that can be mounted on a component carrier. The term "component" can encompass packaged components and, in particular, unpackaged components or chips. These include two- or multi-pin SMT components or other highly integrated flat, round, or differently shaped components such as ball grid arrays, bare dies, flip chips, or individual parts such as semiconductor chips from a semiconductor wafer, which are further processed into finished components, particularly after structuring and dicing the wafer.

[0014] The term "component carrier" in this document can be understood to mean any type of component-capable media, in particular substrates or printed circuit boards. A component-capable medium, in particular a printed circuit board, can be rigid or flexible. It can also have at least one first rigid region as well as at least one flexible region. A component-capable medium can also be a temporary carrier to which unpackaged chips are attached for the purpose of manufacturing components, for example, using a so-called "embedded wafer level package" (eWLP) process. Such a temporary carrier can be an adhesive film that is stretched onto a frame structure in a known manner. The adhesive film can be a heat-removable film (so-called thermo-release foil), so that the adhered chips can be easily detached from the (previously) adhesive film using thermal energy.

[0015] In this document, the term "assembly" or "mounting" refers to any type of attachment of an electronic component to the component carrier. The attachment can be permanent, for example, when assembling a printed circuit board with subsequent melting and solidification of solder material. The attachment can also be temporary, for example, when assembling a clamped carrier foil to produce an artificial wafer for the production of packaged semiconductor components.

[0016] In preferred embodiments of the invention, the first component used as a reference and the second component are of the same type. This allows it to be assumed with particularly high accuracy that the spatial offset determined for the first component is equal to a (non-determined) spatial offset between corresponding structures of the second component.

[0017] The specific offset relates in particular to an offset in a plane oriented parallel to the two sides of the first component. In the case of a cuboid-shaped component, these are preferably the two large main sides, with electrical connection contacts of the component being formed on at least one of the two main sides.

[0018] Depending on the specific application, the structural features may be, for example, electrical connection contacts of the component in question, edges of a housing or packaging of the component in question, edges of the (semiconductor) chip of an unpackaged component, a (small) chip of a (LED) component having a transparent packaging, etc.

[0019] The described spatial offset, which is also referred to as an offset in this document, can be a real offset and / or a virtual offset. A real offset can arise, for example, in semiconductor chips due to inaccuracies in various lithography steps during chip production at the wafer level. A virtual offset can occur, for example, if the optical axes of the two cameras are not perfectly aligned (parallel) to each other. Although no spatial offset need exist in reality, a virtual offset must still be taken into account for high-precision assembly through suitable position compensation during the assembly of the second component.

[0020] According to one embodiment of the invention, the method further comprises (a) determining a spatial position of the first component based on (i) the acquired first image of the first side of the first component and / or (ii) the acquired second image of the second side of the first component; and (b) mounting the first component on the component carrier based on the determined spatial position of the first component and the determined spatial offset between the first structural feature and the second structural feature of the first component.

[0021] In this embodiment, the first (reference) component used to determine the spatial offset is a real component, which later becomes part of the resulting component assembly content of the assembled component carrier. This eliminates the need for separate reference components, nor does the component used as a reference need to be removed from the assembly process after the offset has been determined. This has the advantage that the described method can be easily repeated from time to time during a longer assembly of the component carrier with a large number of components, and the value of the spatial offset can be continually updated.This makes it possible to at least partially eliminate inaccuracies in placement that arise from the fact that the spatial offset changes from component to component within a series of components (for example, due to production conditions that are not perfectly constant over time). Undesirable thermal drifts of components of a placement machine with or in which the described method is carried out can also be at least partially eliminated.

[0022] According to a further embodiment of the invention, the spatial position of the second component relates to an optically detectable structure (on the relevant side of the second component).

[0023] The optically detectable structure can be any functional feature of the second component that is characteristic and / or necessary for the operation of the second component. Depending on the nature of this functional feature, it can either (a) be located on a first side and captured by said first camera or (b) be located on an opposite second side of the second component and accordingly captured by said second camera.

[0024] According to a further embodiment of the invention, the optically detectable structure comprises electrical connection contacts of the second component.

[0025] Mounting the second component based on the position of the electrical connection contacts (and not based on other optically detectable structures that are not or only of minor importance for the function of the second component) has the advantage that the second component can always be reliably electrically contacted on a component carrier, such as a printed circuit board or a semiconductor substrate. This applies even if the electrical connection surfaces involved (of the connection contacts and / or the connection pads on the component carrier) are very small and / or located very close to one another. This aspect is becoming increasingly important with the increasing miniaturization of electronic assemblies.

[0026] According to a further embodiment of the invention, the second component is a light-emitting semiconductor component and the optically detectable structure comprises a light-emitting surface of the semiconductor component.

[0027] Mounting the light-emitting semiconductor component based on the exact position of the light-emitting surface can be particularly advantageous in optoelectronic applications, because it allows the light source (and not so much the electrical connection contacts) of the light-emitting semiconductor component to be positioned precisely. This allows optical beam paths to be "laid" or formed with high precision on a component carrier. In other words, mounting the light-emitting semiconductor component with respect to the light-emitting surface is always advantageous when the "optical" accuracy requirements for the position of a light source are higher than the "electrical" accuracy requirements for correct electrical contact.

[0028] The light-emitting semiconductor component can be a light-emitting diode (LED) or a laser diode, for example a surface emitter laser (VCSEL).

[0029] According to a further embodiment of the invention, the first structural feature comprises edges of the first side of the first component. Alternatively, the second structural feature comprises edges of the second side of the first component. Furthermore, the captured image of the side of the second component shows edges of the second component. This has the advantage that determining the spatial position of the second component is based on clearly recognizable structures and can thus be performed with high accuracy and reliability.

[0030] A further significant advantage of this embodiment in some applications, particularly when populating semiconductor chips that are taken directly from a wafer, arises when the components are not perfectly cuboid-shaped, but have slanted side surfaces or side surfaces inclined towards a base surface. Such slanted side surfaces can arise, for example, from a suboptimal mechanical or laser sawing process for separating the semiconductor chips. By determining the spatial offset described in this document, in this embodiment between the edges of one side and the structural feature on the other side, such an inclination of at least one side surface can be taken into account. Of course, in this embodiment too, it is assumed that the inclination of the components "involved" in the described method is at least approximately the same.

[0031] According to a further embodiment of the invention, the method further comprises, for each of at least one further component: (a) optically capturing a further image of a side of the further component; (b) determining a further spatial position based on the captured further image of the further component; and (c) mounting the further component on the component carrier based on the determined further spatial position of the further component and the determined spatial offset between the first structural feature and the second structural feature of the first component.

[0032] The first component can therefore be used as a reference component not only for the assembly or mounting of the second component but also for any number of subsequent components, which are (to be) mounted or placed on the component carrier with a high spatial accuracy based on the determined spatial structural information from the first side of the first component and the second side of the first component.

[0033] Depending on the specific application and the required accuracy, the number of additional components processed may be less than 5000, preferably less than 2000 and more preferably less than 500.

[0034] The number of additional components that can be further processed with high accuracy based on the structural information of the first component may depend, in particular, on any structural deviations of the components relative to the first (reference) component, and which structural deviations are to be expected or anticipated. Thus, the number of additional components may depend on the size of a production batch, assuming that all components in this production batch have at least approximately the same spatial structural properties.

[0035] When processing unpackaged electronic components (chips) that are taken directly from a wafer and placed on the component carrier, it can be assumed, for example, that all of the chips on this wafer are so spatially and structurally similar that the first chip (from this wafer) can be used as a suitable reference chip for all of the other chips on this wafer. This is the case, for example, if all of the chips on this wafer were singulated using the same sawing device, which ensures that the individual chips are at least approximately the same size and / or shape. For this type of further processing of a wafer, the accuracy can be further increased if, for example, only those chips that are arranged in the same row on the wafer with respect to the first chip (reference chip) are processed as the "other components".This means that for each row of chips on the wafer, a first component is used as a reference component or as a reference chip.

[0036] It should be noted here that in this type of semiconductor chip processing, the component carrier is typically a temporarily adhesive film that is ready for assembly and is mounted on or attached to a frame. The entirety of all chips mounted on this film is often referred to as an artificial wafer.

[0037] According to a further embodiment of the invention, the method further comprises temporarily placing at least the first component on an optically transparent storage element, wherein the first component is detected both from above and from below through the optically transparent storage element. This has the advantage that at least the first component used as the reference component can be optically detected from both sides without it having to be moved. Any handling of the first component after the detection of the first image and before the detection of the second image, which handling would always result in a certain inaccuracy in the determination of the spatial offset, is therefore not necessary. The detection of the first image and the detection of the second image can therefore take place at least approximately simultaneously, which is advantageous, among other things, for rapid implementation of the described method.

[0038] The optically transparent storage element can be implemented, for example, using a glass plate. A suitable surface treatment of the glass plate, for example, by applying an anti-reflective coating, can help ensure that the component in question can be detected with high optical accuracy even through the glass plate. The at least one reference marking can be attached or formed on the storage element, for example, by means of a suitable milling or etching of at least one surface of the glass plate. A reference marking can also be implemented using suitable reference components that are placed on the optically transparent storage element.

[0039] According to a further embodiment of the invention, the optically transparent storage element has at least one reference marking. Furthermore, the method comprises (a) detecting a first reference position of the reference marking in a first camera image from a first camera that has recorded the first image of the first side of the first component; (b) detecting a second reference position of the reference marking in a second camera image from a second camera that has recorded the second image of the second side of the first component; and (c) determining the relative spatial position and / or the relative orientation between the first camera and the second camera based on the spatial positions of the two reference positions in the two camera images.

[0040] Precise knowledge of the relative positioning and orientation of the two cameras can be used to take into account and appropriately compensate for errors resulting from spatially imperfect positioning and alignment of the two cameras during image analysis. This can further increase placement accuracy.

[0041] The described determination of the relative positioning or alignment can be performed, in particular, after an initial setup of the two cameras (in a placement machine). Furthermore, the described determination can also be performed repeatedly during a (larger) placement job. This allows unwanted temporal drifts in the relative positioning or alignment of the two cameras to be detected and appropriately compensated.

[0042] According to a further embodiment of the invention, the optically transparent storage element has at least one reference marking. The method further comprises (a) cyclically capturing the reference marking in a plurality of camera images; (b) determining, for each and in each camera image, a position of the reference marking; and (c) determining a temporal drift of the position of the reference marking in the plurality of camera images. The mounting of the second component and / or the at least one further component is further performed based on the determined drift.

[0043] Through the described cyclical or repeated detection of the reference mark, temporal drifts can be detected, which influence all spatial measurements, in particular the determination of the spatial offset between the first structural feature and the second structural feature and the determination of the spatial position of the second component and / or the additional components. With (quantitative) knowledge of such drift behavior, such temporal changes can be at least partially compensated for by appropriately modifying the mounting position of the second and / or the additional components. This allows the placement accuracy to be maintained at a high level even over a longer period of time.

[0044] Drifts, which undesirably affect spatial measurements, occur, for example, in a camera. Thermal expansion of camera components, such as a mount for an optical system or a sensor chip, affects the optical imaging within the camera, resulting in structures—here the reference marker(s)—migrating from camera image to camera image in the images captured by the camera in question.

[0045] In other words, the detection of the first component as a reference component represents an "initial measurement." After this initial measurement, the position of at least one reference mark is observed in further camera images. Based on the "wandering" of the reference mark(s), the drift is then determined, which is compensated for during the assembly of the second component(s) by appropriately repositioning a placement head that places or mounts the respective component on the component carrier.

[0046] As already mentioned above, such drifts are primarily caused by thermal expansion and / or tension in the respective camera. However, it should be noted that such drifts can also occur elsewhere in a placement machine, for example, on a gantry system that carries a placement head and / or the respective camera and is responsible for moving or positioning these components.

[0047] Please note that there is no fundamental upper limit for the number of reference marks that are cyclically recorded within the scope of the method described here, and for each of which a temporal drift is determined. In general, the higher the number of reference marks considered, the more accurately thermal expansion and / or distortions in the respective placement machine can be detected and compensated for by appropriate positioning of the placement head.

[0048] Preferably, the at least one reference marking is arranged at a location where it does not interfere with the detection of the respective component. This means that within a detection range of the respective camera, a sub-area is provided for imaging the respective component, and that within the detection range, at least one other sub-area spaced therefrom is provided for displaying the at least one reference marking. In preferred embodiments, some reference markings are located in two opposing edge regions of the detection range or in a frame-shaped detection range surrounding a central detection range provided for imaging the respective component.

[0049] According to a further embodiment of the invention, the at least one reference marking and a component are detected together. This has the advantage that no additional camera images need to be recorded to determine the temporal drift, so that the described method can be carried out quickly.

[0050] To put it simply, with each image acquisition of a component, the first component, the second component, and / or the at least one further component, at least one reference mark is also simultaneously acquired. The position of the reference mark in the various image acquisitions is then determined using a suitable image analysis, which, with suitable processor performance, can be performed in a time-neutral manner and does not slow down the execution of the described method.

[0051] According to a further exemplary embodiment of the invention, at least two selected spatial image regions are defined on a camera chip, which show the respective component and the at least one reference marking. Only image data from the selected spatial image regions are then used to determine a position of the reference marking and to determine a position of a component and / or a further position of a further component. This has the advantage that the amount of data that must be transferred from the respective camera chip to a data processing device so that suitable image analysis can be carried out there can be significantly reduced. Such a reduction in the amount of data can make an important contribution to ensuring that the speed at which the method described here can be carried out is not limited by necessary data transfers.All information obtained from image recordings about the relevant placement process, which makes an important contribution to high placement accuracy, can be carried out in a time-neutral manner with regard to other (mechanical) process steps, such as, in particular, handling of the relevant components.

[0052] According to a further embodiment of the invention, the method further comprises (a) rotating the first component by 180° about a rotation axis oriented perpendicular to the first side and / or perpendicular to the second side of the component; (b) optically capturing a third image of the first structural feature of the first component rotated by 180° using a first camera; (c) optically capturing a fourth image of the second structural feature of the first component rotated by 180° using a second camera; and (d) determining a further spatial offset between the first structural feature and the second structural feature when the first component is in the 180° rotated position. The mounting of the second component on the component carrier is further carried out based on the determined further spatial offset.

[0053] By determining the spatial offset between the two structural features twice, once when the first component is in a first angular position and a second time when the component is in a second angular position rotated by 180°, deviations from a desired orientation can be detected by at least one of the two cameras. Such detection is based on the following considerations: (A) Typically, the two cameras are oriented relative to each other such that their optical axes are parallel or, more preferably, collinear. For a cuboid-shaped component, this means that both the first side of the component and the second side of the component are captured by the respective camera at a perpendicular viewing angle. If an undesired tilt of a camera occurs, it captures the component at a slightly oblique angle. Due to parallax effects, the structural feature located on the side facing the respective camera appears at a slightly shifted position in the respective camera image. This results in a virtual spatial offset between the two structural features. (B) With a perfectly parallel orientation of the two cameras, there is no such virtual spatial offset. A determined real spatial offset between the two structural features in the second angular position is then exactly opposite to the corresponding real spatial offset in the first angular position. This means that a "constant component" in the two spatial offsets is a direct measure of the direction and extent of the camera tilt. With knowledge of the geometric dimensions of the components involved in the respective optical images, the (undesired) tilt of the respective camera can be taken into account during the mounting of the second component and / or during the mounting of further components and compensated for by appropriately positioning the placement head when placing the respective component.

[0054] It should be noted that the described (quantitative) determination of an undesired tilt of a camera axis during a lengthy component carrier assembly process can also be repeated cyclically with additional components. In this way, a temporal drift related to a variable tilt of the camera in question can be determined and progressively, i.e., proactively, taken into account or compensated for during the assembly of subsequent components.

[0055] According to a further aspect of the invention, a placement machine for populating a component carrier with electronic components is described. The described placement machine comprises (a) a chassis; (b) a receiving device attached to the chassis for receiving a component carrier to be populated; (c) a gantry system with a stationary component that is stationary mounted on the chassis and with a movable component that can be positioned relative to the stationary component; (d) a placement head that is attached to the movable component and is configured to pick up components and, after suitable positioning of the movable component, to populate the component carrier with the components, wherein each component is mounted on the component carrier at a predetermined placement position;(e) a first optical detection device for optically detecting a component from a first side; (f) a second optical detection device for optically detecting the component from a second side opposite the first side; and (g) a data processing device which is communicatively coupled to the gantry system, the placement head, and the two optical detection devices and which is configured to control or execute the method described above.

[0056] The described placement machine is also based on the finding that a first component, as a reference component, is optically measured from two opposite sides, and in the process, information about a component-specific or component-specific spatial offset between at least one optically detectable structure on a top side of the reference component and at least one optically detectable structure on a bottom side of the reference component can be determined. This component-specific offset information can then be taken into account during the assembly of subsequent components, assuming that a similar offset exists for subsequent components. As already described above, this enables highly precise placement of a plurality or a series of components to be achieved efficiently and quickly.

[0057] The optical detection device can, for example, each be a camera. In this embodiment, the two optical detection devices are implemented by (a) a single camera that captures the component in question from one side, and (b) an illumination device arranged on the opposite side of the component. The illumination device illuminates the component in such a way that the camera captures a shadow cast by the component and can recognize at least the outer contours of the component based on the resulting shadow image.

[0058] According to one embodiment of the invention, the first optical detection device is a camera stationary with respect to the chassis. The first optical detection device is preferably a so-called component camera, which detects the components from below. The optically transparent storage element described above can be attached directly or indirectly to this first optical detection device or first camera. The component to be detected can then be picked up by the placement head and placed on the storage element, where it is then detected from below by the first camera, optionally together with reference markings formed on the optically transparent storage element.

[0059] According to a further embodiment of the invention, the second optical detection device is a camera movable relative to the chassis.

[0060] In conventional placement machines, such a (second) camera is often referred to as a circuit board camera or a component carrier camera. It detects markings on a component carrier to be assembled and uses this information to determine the precise positioning of the component carrier within the placement area of ​​the placement machine. The second optical detection device or second camera can be attached to the moving component of the gantry system. Direct or indirect attachment to the placement head is also possible. Alternatively, a separate positioning system can be provided for the second camera.

[0061] It should be noted that embodiments of the invention have been described with reference to different subject matter. In particular, some embodiments of the invention are described with device claims, and other embodiments of the invention are described with method claims. However, it will immediately become clear to those skilled in the art upon reading this application that, unless explicitly stated otherwise, in addition to a combination of features belonging to one type of subject matter, any combination of features belonging to different types of subject matter is also possible.

[0062] Further advantages and features of the present invention will become apparent from the following exemplary description of currently preferred embodiments. The individual figures in the drawings of this document are to be considered merely schematic and not to scale. Short description of the drawing Fig. 1 shows a placement machine according to an embodiment of the invention, which has two cameras and a data processing device in order to determine a spatial offset between structural features present on opposite sides of a component. Fig. Figure 2 illustrates a semiconductor device with inclined side surfaces. Fig. Figures 3a to 3c illustrate different possibilities for optically detecting a component on two opposite sides. Fig. 4a illustrates a determination of the relative positions between a first camera and a second camera. Fig. 4b illustrates a simultaneous measurement of a bottom side of a component with a first camera and a determination of a temporal drift for an optical distortion of the first camera using reference markings formed on a glass plate with an opening. Fig. 4c illustrates a simultaneous measurement of the bottom and top sides of a semiconductor device to determine an offset between (i) an optically recognizable first structure formed on the bottom side and (ii) an optically recognizable second structure formed on the top side. Fig. 5a and Fig. 5b shows optically transparent storage elements with reference markings attached or formed thereon. Detailed description

[0063] It should be noted that in the following detailed description, features or components of different embodiments that are identical or at least functionally equivalent to the corresponding features or components of another embodiment are provided with the same reference numerals or with reference numerals whose last two digits are identical to the reference numerals of corresponding identical or at least functionally equivalent features or components. To avoid unnecessary repetition, features or components already explained with reference to a previously described embodiment will not be explained in detail later.

[0064] Furthermore, it should be noted that the embodiments described below represent only a limited selection of possible variants of the invention. In particular, it is possible to combine the features of individual embodiments in a suitable manner, so that a multitude of different embodiments can be considered as obviously disclosed to those skilled in the art with the variants explicitly presented here.

[0065] It should also be noted that spatial terms such as "front" and "back," "top" and "bottom," "left" and "right," etc., are used to describe the relationship of one element to another element or elements as illustrated in the figures. Accordingly, the spatial terms may apply to orientations that differ from the orientations illustrated in the figures. However, it should be understood that all such spatial terms refer to the orientations illustrated in the drawings for convenience of description and are not necessarily limiting, since the particular device, component, etc. illustrated, when in use, may assume orientations that differ from those illustrated in the drawings.

[0066] Fig. 1 shows a schematic representation of a placement machine 100 according to an embodiment of the invention. In the application described here, the placement machine 100 is used to remove components 190 formed as unpackaged semiconductor chips directly from a (sawn or singulated) wafer 185 and to place or mount them on a component carrier 180, here a carrier film clamped by a frame in a manner not shown. The component content mounted on the carrier film, which represents a so-called artificial wafer, can then be used to manufacture packaged electronic components, for example, using the eWLP process described above.

[0067] The placement machine 100 has a chassis 102, which represents a frame or support structure for various components of the placement machine 100. A Fig. 1, the component feed device provided with the reference numeral 120 provides the wafer 185 for the placement process. Component carriers 180 to be populated are transported by means of a transport device 104 into a placement area of ​​the placement machine 100 and there provided for the placement process. At a placement position, the respective component carrier 180 is fixed in a manner not shown, so that the transport device also represents a receiving device 104 for the component carrier 180 to be populated. However, by means of the transport device 104, the component carriers 180 to be populated are not only fed to the placement process, but are also transported away after at least partial placement of the component carriers 180, so that a next component carrier 180 can subsequently be populated.

[0068] The actual placement process is performed by a placement head 130. The placement head 130 is mounted on a movable support arm 114 so that it can be moved in a direction parallel to the transport direction (arrow x) of the component carriers 180 (double arrow x). The movable support arm 114 is attached to a stationary support arm 112, which is fixedly connected to the chassis 102 and spans the transport device 104 like a bridge. The movable support arm 114 can be moved transversely (double arrow y) to the transport direction. The stationary support arm 112 represents a stationary component of a gantry system 110 of the placement machine 100, and the movable support arm 114 represents a movable component of the gantry system 110. The personnel system 110 thus enables a two-dimensional movement or positioning of the placement head 130 in an xy plane spanned by the x-direction and the y-direction.

[0069] Before the assembly process of at least one component 190, the placement head 130 is moved by means of the gantry system 110 to the component feed device 120, where it picks up the at least one component 190. It is then moved over the component carrier 180 to be assembled, where it places the at least one component 190 on the component carrier 180.

[0070] The placement machine 100 further comprises two cameras, a first stationary camera 140 and a second movable camera 150.

[0071] The first stationary camera 140 is arranged according to the embodiment shown here between the component feeder 120 and the placement area in which the component carrier 180 to be placed is located. The stationary camera 140 is directly or indirectly connected to the chassis 102 and has a field of view which is directed upwards, ie in Fig. 1 from the plane of the drawing. The first camera 140 can measure a component 190, which is held by the placement head 130 by means of a suction gripper (not shown), from below, provided the placement head 130 is located above the first camera 140. The first camera 140 is therefore also referred to as a component camera.

[0072] The second movable camera 150 is mounted on the movable support arm 114 according to the embodiment shown here and can be moved along the x-direction illustrated by a double arrow, just like the placement head 130. The second camera 150, which can also be mounted directly on the placement head 130, has a downward-directed field of view, ie from above into the plane of the drawing of Fig. 1. The second camera 150 can thus detect markings (not shown) applied or formed on the component carrier 180, so that the second camera 150 can also be referred to as a component carrier camera.

[0073] The placement machine 100 further comprises a data processing device 160, which, schematically represented by dashed lines, is communicatively coupled to the two cameras 140 and 150 as well as to the placement head 130. The data processing device 160 ensures the operation of the placement head 130 by appropriately controlling drive motors or actuators. Furthermore, according to the exemplary embodiment illustrated here, the data processing device 160 also performs image processing of the images captured by the two cameras 140 and 150.

[0074] Fig. 2 illustrates a semiconductor component 190 that has been singulated from a wafer by means of a sawing process. The inventors of the invention described in this document have recognized that in practice, such a sawing process is never perfect and results in a component 190 having a component body 292 with inclined side surfaces 293. The structures 294 formed on the upper (second) side of the component body 292 are electrical connection contacts 294. A camera capturing the component 190 from above (in Fig. 2) can detect edges 296b on the top side of the component body 292 in addition to the electrical connection contacts 294. A camera capturing the component 190 from below (also not shown) can detect the edges 296a on the underside of the component body 292.

[0075] The skew angle of the side surfaces 293 typically varies between different wafers. However, the inventors have recognized that, for one and the same wafer, the components 190 originating from that wafer generally have a substantially constant skew angle. The skew of the side surfaces 293 results in a distance c1 between an edge 296b of the component body 292 on the top side and the corresponding edge 296a of the component body 292 on the bottom side. With a skew angle of zero, c1 would naturally be zero.

[0076] Detection of the center of the component 190 is typically performed by means of (a) a camera viewing the component 190 from above and (b) another camera viewing the component from below. From above, the respective camera detects a distance a1 between the structure of the component terminals 294 and the Fig. 2 right upper edge 296b of the component body 292. The camera viewing the component 190 from below detects a different center position of the component 190. This is in Fig. 2 by the fact that a smaller distance b1 is shown compared to the distance a1, which represents the (horizontal) distance between the structure of the component connection contacts 294 and the Fig. 2 lower right edge 296a of the component body 292.

[0077] Fig. 3a to 3c illustrate various possibilities for optically detecting a component 190 on two opposite sides. For simplicity of illustration, the component 190 (with slanted side surfaces) is not shown as in Fig. 2 are not illustrated by a parallelogram but by a trapezoid.

[0078] Fig. 3a shows a currently preferred embodiment in which, for the optical detection of the component 190 on both sides, the two Fig. 1, the first stationary camera 140 and the second movable camera 150. For this purpose, the component 190 is placed on an optically transparent support element 370, which according to the exemplary embodiment shown here is an optically high-quality glass plate. The glass plate 370 is located between the two cameras 140 and 150. According to the exemplary embodiment shown here, the glass plate 370 is attached directly to a housing 341 of the camera 140. Furthermore, according to the exemplary embodiment shown here, the glass plate 370 also has reference markings 372, which are used to calibrate the relative positions or alignments of the two cameras 140 and 150 to one another. A first illumination device 342, which is assigned to the first camera 140, ensures that the component 190 (as well as the reference markings 372) is illuminated from below.The component 190 and (also the reference markings 372) are thus illuminated from below through the glass plate 370. The optical detection of the component 190 and the reference markings 372 by the first camera 140 also occurs through the glass plate 370. Furthermore, a second illumination device 152 is provided, which is assigned to the second camera 150 and which illuminates the component 190 and the reference markings 372 from above. In a corresponding manner, the second camera 150 detects the component 190 and the reference markings 372 from above, without the corresponding light rays having to penetrate the glass plate 370.

[0079] To determine the "skew" and the associated spatial offset between (a) the (edges on the) top side of the component 190 or the structure of the electrical connection contacts 294 formed thereon and (b) (the edges on the bottom side) of the component 190, the component 190 is removed from a wafer and placed on the glass plate 370 (between the reference markings 372). The component 190 is then optically scanned from both sides, and the spatial offset is determined, possibly taking into account the positions of the reference markings 372 shown in the two corresponding camera images. The evaluation of the two resulting camera images is performed by a data processing device (not shown) connected downstream of the two corresponding cameras.

[0080] Fig. 3b shows a two-sided measurement of the component 190, using only one camera 150, which captures the component 190 from above. To detect the top side of the component 190, the second illumination device 352 is activated. The detection of the component 190 thus takes place under bright-field illumination. To detect the underside of the component 190, the second illumination device 352 is switched off and the first illumination device 342, which is located below the glass plate 370, is activated. The camera 150 thus detects the component 190 based on a shadow image of the component 190 (under dark-field illumination).

[0081] Fig. Figure 3c shows another variant of a two-sided measurement of the component 190, in which a rotatable handling tool 350 is used. The top side of the component 190 is captured by a component camera 150 in the same way as in the variants of Fig. 3a and Fig. 3b is the case. To optically capture the underside of the component 190, the camera 150 is moved away from the glass plate 370 (see double arrow 351) so that access to the component 190 is possible from above. This access is achieved by the rotatable handling tool 375, which has a base body 377 and a plurality of component holding devices 379 projecting radially from the base body. According to the exemplary embodiment shown here, the component holding devices are suction grippers 379. Furthermore, according to the exemplary embodiment shown here, the handling tool 375 can be moved translationally (see double arrow 376) and rotated about an axis of rotation 375a. As a result, the component 190 can be removed from the glass plate 370 and brought into the optical capture range of a camera 140, which, optionally with the aid of a first illumination device 342, captures the component 190 from below.

[0082] The following describes embodiments of the invention that enable particularly precise determination of the spatial offset between structures on the top side of a component and structures on the bottom side of the component. This high level of accuracy is achieved in particular by calibrating the cameras used and by appropriately considering the (time-varying) positions, orientations, and / or internal geometries of the cameras, which naturally lead to certain imaging errors. These imaging errors can be at least approximately compensated for in a suitable manner (particularly in image processing) by the calibration measures described below.

[0083] Calibration is based on a measurement of the aforementioned reference markings, which are formed on an optically transparent support element for a component to be detected. These reference markings are captured by both cameras, i.e., the first stationary component camera and the second movable component carrier camera. The reference markings are located at the edge of the field of view of the stationary camera in the focal plane of the stationary camera and can also be captured by appropriately positioning the movable camera. This means that both cameras measure identical structures precisely at the location relevant for offset determination. (A) To determine the relative alignment of the two cameras, the same reference markers are captured by both cameras, and their precise spatial positions are measured in both camera images. The measured spatial positions are then stored in a memory as reference values ​​for subsequent drift compensation. (B) To determine the spatial offset between structures on the top side of the component and structures on the bottom side, the component in question is placed on the optically transparent support element, measured on both sides, and the spatial offset is determined by evaluating the two corresponding camera images. The component is then picked up by a placement head and mounted on a component carrier, taking the determined spatial offset into account. (C) Drift compensation: The reference marks are measured cyclically or repeatedly (from time to time) during the processing of a (longer) placement job. The difference between the measured positions of the reference marks and the value determined at time (A) is determined. This results in updated calibration values ​​that can be taken into account when mounting subsequent components.

[0084] A variant of this cyclic drift compensation by measuring the reference marks with both cameras is to measure the reference marks as part of the measurements of the deposited components from below with the stationary camera (see the process for determining the offset above). In this case, only the drift of the stationary (component) camera can be compensated, but no (significant) additional measurement time is required.

[0085] Some camera sensors offer the option of defining multiple image areas using so-called "multi-windowing" functionality. To minimize image data transfer time, only the relevant image areas (i.e., only the image areas of the component and the reference markings) are transferred to a downstream data processing device. One possible approach would be to measure the reference markings using both cameras at longer time intervals in two additional measurements to compensate for any drift (as described above). Between these measurements, several (almost) time-neutral measurements of the reference markings are taken with the stationary camera from below, together with measurements of stored components, in order to correct at least some of the drift (i.e., drift of the stationary camera) with a higher temporal resolution and at least approximately time-neutrally.

[0086] (D) Furthermore, errors in determining the offset caused by a non-parallelism of the optical axes of both cameras can be detected and compensated for. For this purpose, the offset between the top and bottom of a component is determined twice: once in a first angular position and a second time in a second angular position in which the component is rotated by 180°. The axis of rotation is perpendicular to the surface of the optically transparent storage element. If the component has an offset between the top and bottom, this offset is determined in opposite directions in both measurements due to the 180° rotation. However, if the optical axes of both cameras are tilted, the offset is identical in both measurements. The proportion of the tilt can thus be calculated by determining the constant component of the offset in both measurements.This can then be used to correct further measurements, since the tilt should not change significantly over time.

[0087] Fig. Figure 4a illustrates a determination of the relative positions between (a) a first stationary camera 140 with an optical axis 441 and (b) a second movable camera 150. As can be seen from Fig. As can be seen in Figure 4a, both cameras measure (simultaneously) the position of a reference marker 374. Based on the positions of the reference marker in both resulting camera images, the relative spatial position and / or orientation between the two cameras can be determined.

[0088] Fig. 4b illustrates a simultaneous measurement with a first stationary camera 140 (a) of a bottom side of a component 490b and (b) of two reference markings 374. By repeatedly measuring the positions of the reference markings 374 in a temporal sequence of camera images, each of which preferably shows a different component 490b, a temporal drift for an optical distortion of the camera 140 can be determined.

[0089] As from Fig. As can be seen in Figure 4b, according to the embodiment shown here, the reference markings 374 are attached to a reference plate 470, which has an opening through which the camera 140 captures the component 490b. According to the embodiment shown here, the component 490b is a component packaged in a housing with bent connection contacts.

[0090] Fig. 4c illustrates a simultaneous measurement of the bottom and top sides of a semiconductor device 490c to determine an offset between (i) an optically recognizable first structure 496a formed on the bottom side and (ii) an optically recognizable second structure 496b formed on the top side. This type of two-sided measurement essentially corresponds to the offset measurement described above with reference to Fig. 3a has been explained.

[0091] Fig. 5a and Fig. 5b each show an optically transparent storage element 370 with reference markings 372 attached or formed thereon. In the case of Fig. In the embodiment shown in Figure 5a, the reference markings 372 are located on the side areas of the storage element 370. The area in the middle is intended for the storage of a component and is free of reference markings. Fig.In the example shown in Fig. 5b, the number of reference markings 372 is somewhat higher and the reference markings 372 are arranged in a frame-like edge region around the center of the storage element 370.

[0092] It should be noted that the term "comprising" does not exclude other elements, and that "a" does not exclude a plurality. Elements described in connection with different embodiments may also be combined. It should also be noted that reference numerals in the claims should not be construed as limiting the scope of the claims. REFERENCE SYMBOL: 100 placement machines 102 chassis 104 Pick-up device / transport device 110 Portal system 112 stationary component / stationary support arm 114 movable component / movable support arm 120 component feeding device 130 placement head 140 first camera / stationary camera / component camera 150 second camera / movable camera / component carrier camera 160 data processing facility 180 component carriers 185 wafers 190 components / chips 292 component bodies 293 sloping side surfaces 294 connection contacts 296a Edge on first side / bottom 296b Edge on second side / top side a1 distance b1 distance c1 spatial offset 341 housing 342 first lighting device 351 Movement second camera 352 second lighting device 370 optically transparent storage element / glass plate 372 reference marks 375 rotating handling tool 375a axis of rotation 376 Movement of handling tool 377 basic bodies 379 Component holding devices / suction grippers 441 optical axis 470 Reference plate 490b component (with bent terminals) 490c component 496a first structural feature (on underside) 496b second structural feature (on top)

Claims

[1] Method for loading a component carrier (180) with electronic components (190, 490c), comprising the method optical capture of a first image of a first side of a first component, on which a first structural feature (296a, 496a) of the first component (190, 490c) is recognizable; optical capture of a second image of a second side of the first component (190, 490c) on which a second structural feature (294, 296b, 496b) of the first component (190, 490c) is recognizable, wherein the first side and the second side are opposite each other; Determining a spatial offset (c1) between the first structural feature (296a, 496a) and the second structural feature (294, 296b, 496b); optical acquisition of an image from one side of a second component (190, 490c); Determining a spatial position of the second component (190, 490c) based on the captured image of the side of the second component (190, 490c); and Mounting the second component (190, 490c) onto the component carrier (180) based on the determined spatial position of the second component (190, 490c) and the determined spatial offset (c1) between the first structural feature (296a, 496a) and the second structural feature (294, 296b, 496b) of the first component. [2] Method according to claim 1, further comprising Determining a spatial position of the first component (190, 490c) based on (i) the captured first image of the first side of the first component (190, 490c) and / or (ii) the captured second image of the second side of the first component (190, 490c); and Mounting the first component (190, 490c) onto the component carrier (180) based on the determined spatial position of the first component (190, 490c) and the determined spatial offset between the first structural feature (294, 296a, 496a) and the second structural feature (296b, 496b) of the first component (190, 490c). [3] Method according to one of the preceding claims, wherein the spatial position of the second component (190, 490c) relates to an optically detectable structure (296a, 294, 296b, 496a, 496b). [4] Method according to the preceding claim, wherein the optically detectable structure comprises electrical connection contacts (294) of the second component (190, 490c). [5] Method according to one of the two preceding claims, wherein the second component is a light-emitting semiconductor component (190, 490c) and the optically detectable structure comprises a light-emitting surface of the semiconductor device (190, 490c). [6] Method according to one of the preceding claims, wherein the first structural feature comprises edges (296a) of the first side of the first component (190, 490c) or the second structural feature comprises edges (296b) of the second side of the first component (190, 490c) and wherein the captured image of the side of the second component (190, 490c) shows edges (296b) of the second component (190, 490c). [7] Method according to one of the preceding claims, further comprising, for each of at least one further component (190, 490c): optical acquisition of another image from one side of the further component (190, 490c); Determining a further spatial position based on the captured further image of the further building element (190, 490c); and Mounting the further component (190, 490c) onto the component carrier (180) based on the determined further spatial position of the further component (190, 490c) and the determined spatial offset (c1) between the first structural feature (296a, 496a) and the second structural feature (294, 296b, 496b) of the first component (190, 490c). [8] Method according to any one of the preceding claims, further comprising temporary placement of at least the first component (190, 490c) on an optically transparent storage element (370), wherein the first component (190, 490c) is captured both from above and from below through the optically transparent storage element (370). [9] Method according to the preceding claim 8, wherein the optically transparent storage element (370) has at least one reference mark (372, 374), wherein the method further comprises Capturing a first reference position of the reference mark (372, 374) in a first camera image from a first camera (140) which has taken the first image of the first side of the first component (190, 490c); Capturing a second reference position of the reference mark (372, 374) in a second camera image from a second camera (150), which has captured the second image from the second side of the first component (190, 490c); and Determining the relative spatial position and / or the relative orientation between the first camera (140) and the second camera (150) based on the spatial positions of the two reference positions in the two camera images. [10] Method according to the preceding claim 8, wherein the optically transparent storage element (370) has at least one reference mark (372, 374), wherein the method further comprises Cyclical acquisition of the reference mark (372, 374) in a plurality of camera images; Determine, for each and in each camera image, a position of the reference mark (372, 374); and Determining a temporal drift of the position of the reference mark (372) in the majority of camera images; wherein The assembly of the second component (190, 490c) and / or, if referring back to claim 6, of the at least one further component (190, 490c) is furthermore carried out based on the determined drift. [11] Method according to the preceding claim, wherein the at least one reference marking (372, 374) and a component (190, 490c) are jointly captured. [12] Method according to the preceding claim, wherein at least two selected spatial image areas are defined on a camera chip, which show the respective component (190, 490c) and the at least one reference mark (372, 374), wherein only image data of the selected spatial image areas are used to determine a position of the reference mark (372, 374) and to determine a position of a component (190, 490c) and / or a further position of a further component (190, 490c). [13] Method according to any one of the preceding claims, further comprising Rotating the first component (190, 490c) by 180° about an axis of rotation that is oriented perpendicular to the first side and / or perpendicular to the second side of the component (190, 490c); optical acquisition of a third image of the first structural feature (296a, 496a) of the first component rotated by 180° (190, 490c) using a first camera (140); optical acquisition of a fourth image of the second structural feature (294, 296b, 496b) of the first component rotated by 180° (190, 490c) using a second camera (150); and Determining a further spatial offset between the first structural feature (296a, 496a) and the second structural feature (294, 296b, 496b) when the first structural element (190, 490c) is in the position rotated by 180°; wherein The mounting of the second component (190, 490c) onto the component carrier (180) is further based on the determined additional spatial offset (c1). [14] Placement machine (100) for placing electronic components (190, 490c) onto a component carrier (180), comprising the placement machine (100) a chassis (102); a receiving device (104) attached to the chassis (102) for receiving a component carrier (180) to be populated; a portal system (110) with a stationary component (112) which is stationary on the chassis (102) and with a movable component (114) which can be positioned relative to the stationary component (112); a placement head (130) which is attached to the moving component (114) and which is configured to pick up components (190, 490c) and, after suitable positioning of the moving component (114), to populate the component carrier (180) with the components (190, 490c), wherein each component (190, 490c) is mounted on the component carrier (180) at a predetermined placement position; a first optical detection device (140) for optically detecting a component (190, 490c) from a first side; a second optical detection device (150) for optically detecting the component (190, 490c) from a second side opposite the first side; and a data processing device (160) which is communicatively coupled to the portal system (110), the placement head (130) and the two optical detection devices (140, 150) and which is configured to control the method according to one of the preceding claims. [15] Placement machine (100) according to the preceding claim, wherein the first optical detection device is a camera (140) stationary in relation to the chassis. [16] Placement machine (100) according to one of the two preceding claims, wherein the second optical detection device is a camera (150) movable in relation to the chassis.

Images