You might not know that the assembly process of a laptop keyboard involves many little-known technical details. For example, beneath each key lies a delicate “scissor-switch” mechanism—this very design is what enables the keyboard to respond smoothly to every keystroke.

So, when installing the scissor-switches, how can we ensure that each “scissor-switch” component is positioned on the keyboard with its “front side” facing upward?

To ensure that each “scissor-foot” component can be precisely pressed into the fixture in the correct orientation, STEP has designed an advanced automated assembly technology and a sophisticated positioning system. These technologies not only enhance installation efficiency but also guarantee the consistency and stability of each key.

In an automated assembly device for aluminum plates used in the scissor feet of laptop computer keyboards, the scissor feet are randomly placed face-up or face-down into a feeder tray. The feeder tray typically consists of a grid with multiple rows and 8 columns. A camera identifies whether each scissor foot in the tray is oriented face-up or face-down.

Each workstation uses one robotic arm, and each robotic arm is equipped with eight suction nozzles. The robotic arms alternately pick up materials from two material trays—one on the left and one on the right—aiming to minimize the number of picking operations required to collect a full set of OK materials (i.e., eight scissor feet all facing upward). Once the materials are picked up, they are pressed down onto the keyboard fixture.

When the OK materials in the material tray are insufficient to form a complete set, the material tray needs to be refreshed. Typically, the refresh process takes more than 8 seconds. Therefore, reducing the refresh time is crucial for improving the efficiency of laptop keyboard assembly. The selection method used by the robotic arm to pick up materials is an essential step in high-speed manufacturing. In response to this market demand, we will now introduce STEP innovative and upgraded screening solution.

The method of material retrieval by robotic manipulator

For ease of understanding, let’s randomly generate some tray data. The figure below shows the matrix-based data identified from a particular tray, where dark blue represents the front-side components and light blue represents the back-side components. The robotic arm’s picking requirement is to take out 8 front-side components from the tray.

The first material retrieval plan is as follows:

Similarly, the second material-handling scheme is as follows:

Is there a more efficient material-handling solution?

Of course, it’s possible to offset the picking process. Suppose there are only two rows of data left in the material tray matrix. In the first pick, the robotic arm picks materials 1, 2, 4, and 5 from the first row. At this point, only nozzles 3, 6, 7, and 8 remain unoccupied. The robotic arm can then shift one column to the left and move down to perform the second pick. In this way, after just two picks, the entire set of picking requirements will be fulfilled.

Pain points of traditional screening methods

(1) Frequent material retrieval, low efficiency, and significant wear.

• Frequent sampling: Typically, it takes 3 to 4 sampling attempts to complete a set of data.

• Frequent operation of the robotic arm: Due to the high number of material-handling cycles, the robotic arm must perform actions—such as moving, picking up materials, and placing them—frequently, increasing the workload and accelerating wear and tear.

• Frequent tray refreshing: Traditional methods require frequent refreshing of material trays—after each material retrieval, the materials need to be rearranged or partially cleared out, which is time-consuming and may result in empty trays or uneven material distribution, thereby affecting the efficiency of subsequent material retrieval.

(2) Taking photos and performing recognition takes a long time.

• Take a photo after each material retrieval: After completing the retrieval of materials from each group, you must use a camera to capture an image of the new material status. Each time after taking a photo, the computer needs to process the image to determine the next step in the material retrieval plan.

• Delay in photo capture and image recognition: The photo capture and image recognition processes are time-consuming and can be affected by environmental factors such as lighting, camera quality, and the position of the material, thereby reducing recognition accuracy and speed.

(3) Insufficient utilization of left and right material tray resources.

• The left and right workstations fail to complement each other effectively: Traditional methods do not achieve efficient resource complementarity between the left and right workstations, resulting in separate management of materials. This leads to an imbalance in material usage during production, increasing the frequency of retooling and prolonging downtime on the production line.

• The left and right bins cannot share data: Traditional methods fail to fully utilize the material information from both bins, making it impossible for them to complement or adjust their supplies during the picking process. As a result, one bin ends up with excessive leftovers while the other is already empty, leading to resource waste.

STEP Innovates and Upgrades Screening Methods

(1) Reduce the number of material retrieval operations.

It can calculate the optimal material-handling solution, reducing the number of material-handling operations per batch (typically from 2 to 3 times) and thereby lowering the frequency and time consumption of tray refreshing.

(2) Calculate all material retrieval data in one go.

Calculate all data at once before picking materials, eliminating the need for image recognition with each photo. The robotic arm picks materials continuously, saving time. By gaining advance insight into the material status on the tray, we can prepare in advance for refreshing operations, ensuring the continuity and stability of assembly work and guaranteeing uninterrupted operation of the production line.

(3) Support complementary left-right discs

The new method treats the material on the right tray as an extension of the left tray, thereby improving resource utilization. The complementary strategy is equivalent to increasing the capacity of the material trays, enhancing picking efficiency, reducing the number of operations, and optimizing overall utilization.

Introduction to Innovative and Upgraded Configurations

1 Customer Configuration

• 1 set of ADT-6320E-B08 EtherCAT bus motion control card

• 30 sets of Ω6 EtherCAT bus-type servo drives

• 4 sets of industrial cameras

• 2 sets of AR5520 industrial robots

2 Related parameters

• Accuracy: ±0.01 mm

3 Industry Advantages

• Able to transcend departmental boundaries and quickly identify the optimal material-handling path.

• Possesses powerful estimation capabilities, and the screening method can accurately calculate the total quantity of material groups that can be obtained in a single step.

• Supports a cross-disk complementary operation mode, further enhancing the overall utilization rate of material trays.

Email:
market@stepelectric.com

Address:
No. 1560, Siyi Road, Jiading District, Shanghai Municipality