Four-Point Supported Belt Sanding Robot for Better Industrial Surface Sanding

By Samudrapom DamReviewed by Susha Cheriyedath, M.Sc.Mar 27 2024

In an article recently published in the journal Scientific Reports, researchers proposed a 4-point supported belt sanding robot for sanding large convex surfaces.

Limitations of conventional sanding robots

The movement of sanding robots that are typically used while sanding objects with a large area is similar to drawing a line along a curved surface. However, realizing a uniform sanded area using these robots is difficult.

Moreover, these sanding robots also need a long working period. Most of the current sanding robots primarily focus on the delicate sanding of small objects. However, several sanding objects possessing large curved surfaces that are made using hard metals like iron exist in the industry.

Robot sanding of large curved metal surfaces is not feasible using conventional methods. Belt sanding can attain a high material removal rate with fine surface quality as a finishing process. Additionally, the process can be performed at a low temperature, which is beneficial for several intractable materials like titanium and aluminum alloys.

Thus, belt sanding is suitable in cases where low sanding temperatures and high material removal are required, such as cases involving complex-shaped targets. Specifically, the belt's flexibility is advantageous for complex-shaped structures, which makes a belt contact method feasible for uniformly sanding large curved metal surfaces as the method minimizes the support surface.

The proposed approach

In this study, researchers proposed a 4-point supported belt sanding robot that utilizes the sanding belt's flexibility to reach surfaces with different curvatures while sanding large convex areas. The study aimed to develop large-area sanding robots for tank lorries, storage tanks, and ships.

This proposed belt sanding robot can enable uniform and rapid sanding by overcoming the limitations of sanding robots that move in point contact with a sanding object. This adaptive belt tension robot was equipped with a 4-point supported belt mechanism for sanding variable curved surfaces/curved large-area sanding.

Additionally, a sanding normal force prediction formula was proposed to describe the contact surface's sanding performance. This equation consisted of the normal force due to the horizontal and vertical elongation of the belt and the concentrated load function owing to the belt movement.

In the 4-point supported belt sanding robot, the sanding belt was rotated clockwise by the drive wheel motor with 2000 rpm and 2 kW specifications. The robot arm comprised two contact wheels, with the contact wheel at the robot arm's end adjusted to the curvature and the tension maintained. Then, it came into contact with the target surface at the robot arm's contact point.

A spring structure was used by the robot arm to preserve the reaction force upon contact under the conditions of the user. The belt length shrank as the robot arm folds, which were compensated by a tension maintenance wheel that was fixed to the robot body. This tension maintenance wheel was fixed to maintain the tension of the belt when the moving stage moved forward. In the absence of this wheel, the deviation of the belt cannot be prevented as the belt length changes when the arm is folded.

Researchers also proposed an image analysis technique to quantitatively measure the sanding area. They performed a video image analysis to calculate the sanding area and also determined whether the area was uniformly sanded. The S/N ratio for the test conditions and image area was utilized as a performance index of the sanding robot.

Research findings

The test bench dimensions (W × D × H) were 1700 mm × 1450 mm × 900 mm. Experiments were performed on very large convex specimens with radii of 725, 1000, and 2100 mm. Researchers experimentally obtained the optimal values of the design variables and applied them to confirm the sanding area when sanding was performed using a 4-point and 2-point supported belt sanding robot. Eventually, the optimal design variables were applied.

The optimal case was the area per sand when the optimized design variables were applied. Results from the optimal case demonstrated that the measured abrasive area was 34.08%, 35.26%, and 56.33% at curvature 1, curvature 2, and curvature 3, respectively, while using a 4-point support mechanism.

A comparative analysis between the sanded area of the conventional 2-point support mechanism/general belt-sanding robot and the sanded area of the proposed 4-point support mechanism/4-point supported belt sanding robots showed that the sanding performance was improved by 43% in the 4-point support mechanism.

To summarize, the findings of this study demonstrated that the proposed adaptive belt sanding robot equipped with a 4-point support mechanism is effective for sanding large objects with convex surfaces and concave surfaces.