Semiconductor Chip Bonding Alignment Using Camera Offset Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for installing semiconductor chips on substrates face challenges in maintaining positional accuracy due to fluctuations in camera offset amounts caused by apparatus distortion and temperature changes, leading to increased costs and complexity with the use of expensive two-vertical-field-of-view cameras and dedicated alignment marks.
Innovation Solution
An apparatus and method that utilize a bonding head with a first camera for capturing mounting surface images, a second camera for capturing bonding tool images, and a controller to calculate and store area correction amounts for camera offsets, allowing for improved positional accuracy without the need for expensive cameras or dedicated substrates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a two-vertical-field-of-view camera is used to capture images of both the chip and substrate simultaneously, then positional accuracy of chip installation is improved, but device cost increases significantly
Solution Approach 1:
The patent divides the imaging function into two separate cameras: a first camera mounted on the bonding head to capture chip images, and a second camera mounted on the substrate table to capture substrate images. This segmentation replaces the expensive two-vertical-field-of-view camera with two simpler, more cost-effective cameras that together achieve the same positioning functionality.
Solution Approach 2:
The patent introduces a marker attached to the substrate as an intermediary reference object. The second camera captures images of this marker to determine substrate position, which then serves as a reference for calculating the relative position between the bonding head and substrate. This intermediary approach enables accurate positioning without requiring complex dual-field-of-view imaging.
2Measurement precision
If a dedicated substrate with alignment mark is used to calculate correction amount, then camera offset correction accuracy is improved, but manufacturing complexity and time increase
Solution Approach 1:
The patent makes the substrate itself serve multiple functions: it is both the target for chip mounting and the carrier of the alignment marker. The marker can be directly formed on the substrate surface through standard substrate processing techniques, eliminating the need for separate dedicated substrates or additional alignment mark attachment steps.
Solution Approach 2:
The substrate provides its own alignment reference through the marker formed on its surface, eliminating the need for external alignment marks or dedicated reference substrates. The same substrate that receives the chip also provides the reference for positioning, making the system self-sufficient and reducing overall manufacturing complexity.
3Ease of manufacture
If camera offset amount is not corrected, then manufacturing process is simpler, but positional accuracy of semiconductor chip installation deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the positions of the chip (from the first camera) and substrate marker (from the second camera) are continuously monitored. The controller calculates the relative position based on these images and automatically adjusts the bonding head position to compensate for camera offset errors, ensuring accurate chip placement without requiring manual calibration.
Solution Approach 2:
The patent dynamically changes the operational parameters (camera offset correction values) based on actual measured positions from the captured images. The controller calculates correction amounts from the relative positions detected by both cameras and applies these corrections in real-time, allowing the system to adapt to varying offset conditions while maintaining simple operation for the user.
Data Source
AI summary
An apparatus for producing a semiconductor device comprises a stage, a bonding head, a bonding tool and a first camera that are attached to the bonding head, and a controller, the apparatus moreover being such that the controller is configured to execute for each of one or more points: a process of mounting an inspection chip on a mounting surface; a process of acquiring, as an inspection image, an image of the mounting surface after the inspection chip has been mounted thereon captured by the first camera; a process of calculating, as an area correction amount C, a correction amount for a camera offset amount Ocm on the basis of the position of the inspection chip in the inspection image; and a process of associating the calculated area correction amount C and the position of a discretionary point and then storing the associated information in a storage device.