Optics: measuring and testing – By polarized light examination – With light attenuation
Reexamination Certificate
1999-05-19
2001-09-18
Font, Frank G. (Department: 2877)
Optics: measuring and testing
By polarized light examination
With light attenuation
C250S559290, C348S126000, C382S146000
Reexamination Certificate
active
06292261
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATION
This application is related to the subject matter of the application titled Rotary Sensor System with a Single Detector, filed on the same day with the present application.
COPYRIGHT RESERVATION
A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
FIELD OF THE INVENTION
This invention relates to optical sensor systems for determining the location or orientation of an object, such as an electronic component, including such systems which report a physical condition of the object, such as the orientation of the object, the height of a particular feature on the object, the distance between features on the object, the presence of expected features on the object or the coplanarity of features on the object.
BACKGROUND OF THE INVENTION
Component Position Computation
There are many optical-based approaches to identifying various parameters related to an object, such as an electronic component. The various parameters which may be reported include orientation of the component, location of features on the component, indications of the quality of the component, such as a bent lead (lead tweeze) in the case of a Quad Flat Pack (QFP) component, or a measurement of a feature located on the bottom of the component, such as the height of a ball on a Ball Grid Array (BGA) component. The system for identifying these parameters is generally mounted in a pick and place machine which picks up the component and places the component on a printed circuit board. All of these approaches rely on a quill in the pick and place machine to pick up the component, and all generally utilize the motion control system in the pick and place machine in the measurement process. Some of the approaches use cameras to collect data representative of an image of the component and others collect data representative of an outline of the component, and all have some sort of processor that analyzes the collected data in order to identify some specific portion of the component, such as the edges of the part or the tips of the component leads. The sections below briefly describe a variety of the approaches used today, and describe their relative advantages and disadvantages.
In general, the systems described below can be characterized as being either “on-head” or “off-head.” On-head sensor systems (considered together with their host pick and place machines) sense for the orientation of the component while the component travels to the target circuit board. On-head systems are preferred, as they minimize the amount of time required to place a component on a board, thereby increasing the throughput of the pick and place machine.
Off-head sensor systems perform the data collection in a fixed location on the pick and place machine. With an off-head system, the quill must first move the component to the dedicated station and after acquiring the image, move the component to the target circuit board for placement. Off-head systems typically reduce the machine's throughput since the head makes additional stops or travels further, but are used because they are inexpensive, reduce the mass which must be carried on the head and one sensor can readily service multiple quills.
Another way to characterize various sensor systems is by the relative location of the detector and the light source with respect to the component.
FIGS. 1A-C
show the three schemes of front-lit, back lit and shadow cast sensor systems, respectively. In front-lit systems, light from a source illuminates the features of interest on the component and the light reflected off of the features is optionally directed through optics forming an image on a detector. In back-lit systems, light from a source is incident on the component, and then optionally directed through focusing optics before it reaches a detector. Because the detector does not collect light scattered off the bottom of the component in either the back-lit or the shadow-cast systems, neither type of system is capable of inspecting features on the bottom of components, such as balls on a Ball Grid Array (BGA) or pins on a pin grid array (PGA). In shadow cast systems, light from a source optionally travels through collimating optics, onto the component, optionally directed through an optical system and then casts a shadow on the detector.
One approach in use today for assessing the orientation of an electronic component on a pick and place quill data representative of a shadow of the component, sometimes exclusively using the data representative of the edges of the shadow to perform its orientation computation. One such sensor is available from CyberOptics® Corporation and is called a LaserAlign® Sensor. In this shadow cast sensor system, light is cast onto the component from one side of the component and a linear detector on the other side of the component captures data representative of the component's shadow. Appropriate electronics analyze the data to compute the x,y,&thgr; coordinates of a pre-selected feature on the component.
FIG. 2
shows this approach schematically and the method is described further in U.S. Pat. No. 5,278,634 to Skunes et al., assigned to CyberOptics Corporation. This approach takes advantage of the motion control system already installed on the pick and place machine, which has an accurate encoder for reporting the angular position of the quill. At several angular positions, the system collects data from the detector representative of the shadow of the component. As appropriate for the type of component, the angular positions do not have to be equally spaced over 360 degrees nor is the component required to spin at a constant angular velocity. For each angular position, the sensor system computes the width of the component's shadow. By assessing the shadow center position at the minimum shadow width angle the component location in one axis can be determined. By measuring two shadow positions at the two minimum widths 90° apart, the position of a rectangular component can be computed. This system does not provide any further information not already derived from the shadow edges. Hence, it cannot identify misplaced or missing balls on BGAs or flip-chips. One main advantage of the sensor, however, is that it can be placed on-head, which allows for increased throughput of the pick and place machine.
Another on-head system employs a video camera on the moving head. In this approach, the camera stares down parallel to the pickup quill, and a right angle prism (or equivalent) is moved under the camera and component during travel, shown in
FIGS. 3A-B
. In this approach, there is typically one camera for each pick-up nozzle. This system has three drawbacks. First, there is substantial cost and mass associated with the right angle prism and the mechanism to slide it reliably into place. Second the minimum time between pickup and placement is limited by the time required to raise the nozzle, slide the prism into position, acquire the image, slide the prism out of position, and then lower the nozzle. Finally, the prism must be roughly as thick as the width of the component to be measured. Hence, the z motion of the component during pickup and placement is much longer than is required for other approaches.
One off-head approach to assessing component orientation is to place one or more fixed position upward looking video cameras (e.g., 2D array detectors) in the pick and place machine at a dedicated station. For small components, or ones with relatively coarse pitch between leads, one image of the component is taken and its position determined from analyzing a single image. For larger components with fine pitch, several images may be needed to provide adequate resolution. This approach utilizes standard components (e.g., cameras, o
Case Steven K.
Cohn Brian D.
Fishbaine David
Jalkio Jeffrey A.
Konicek John P.
CyberOptics Corporation
Dicke, Billig & Czaja P.A.
Font Frank G.
Nguyen Sang H.
LandOfFree
Rotary sensor system with at least two detectors does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Rotary sensor system with at least two detectors, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Rotary sensor system with at least two detectors will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2481790