Gear Tooth Profile Inspection Using NACHI MZ07 + QRS Vision

Industry > AutomotiveApplication > InspectionSector > Manufacturing

#GearToothInspection#VisionInspection#2D3DVisionSystem#NACHIMZ07#RobotPickingAndSorting#PrecisionMeasurement#QualityImprovement#DefectDetection

Process Overview

This solution automates gear tooth profile inspection and sorting by combining QRS 2D/3D vision inspection with the NACHI MZ07 robot.

It detects the position of randomly stacked gears, performs shape inspection, and classifies parts into OK/NG automatically.

Since the vision system is developed by QRS, it can be tuned and customized to match on-site requirements, enabling accurate inspection across various product shapes.

Components

Robot	NACHI MZ07 (6-axis articulated robot) Payload: 7 kg Horizontal reach: 723 mm Weight: 36 kg Repeatability: ±0.02 mm IP rating: IP67
Peripherals	Electric gripper: DH Robotics AG-95 Vision (Distance type): QRS Q3D-IS Gear position recognition 3.2 MP, 2048 × 1536 pixels Format: structured light Scanning range: 176 × 132 ~ 706 × 529 mm Measurement distance: 300 / 600 / 1200 mm Detectable: color, shape, position recognition, dimensional measurement Vision (Tooth profile inspection): QRS Q2D Deep learning algorithm embedded Alignment: check product alignment shape and alignment direction Measurement: measure product shape, length, and weld bead thickness Inspection: inspect product shape, defects, color, and surface damage Work cell Robot base Inspection result display monitor (QRS) Workpiece stand

Robot

NACHI MZ07 (6-axis articulated robot)
Payload: 7 kg
Horizontal reach: 723 mm
Weight: 36 kg
Repeatability: ±0.02 mm
IP rating: IP67

Peripherals

Electric gripper: DH Robotics AG-95
Vision (Distance type): QRS Q3D-IS
Gear position recognition
3.2 MP, 2048 × 1536 pixels
Format: structured light
Scanning range: 176 × 132 ~ 706 × 529 mm
Measurement distance: 300 / 600 / 1200 mm
Detectable: color, shape, position recognition, dimensional measurement
Vision (Tooth profile inspection): QRS Q2D
Deep learning algorithm embedded
Alignment: check product alignment shape and alignment direction
Measurement: measure product shape, length, and weld bead thickness
Inspection: inspect product shape, defects, color, and surface damage
Work cell
Robot base
Inspection result display monitor (QRS)
Workpiece stand

Workflow

STEP 1.	3D vision detects gear position (3D)
STEP 2.	Robot picks the detected gear
STEP 3.	2D vision performs OK/NG tooth profile inspection (2D)
STEP 4.	Place the part into the designated output position (OK/NG).

※ The content provided by Marosol is protected under U.S. and Canadian copyright and intellectual property laws. Unauthorized reproduction, distribution, or use of this content is strictly prohibited.

Key Features

Fast and accurate sorting performance

High-speed, high-repeatability industrial robot enables stable and rapid handling operations.
Vision-based position detection is performed during robot operation, reducing cycle time.

High-precision 2D/3D vision inspection capability

Even complex-shaped parts can be detected quickly with 3D recognition and localization.
Micron-level dimensional measurement is supported through 2D inspection for precise quality checks.

Implementation Results

ROI Period	2 years
Key Metrics	Reduced manpower required for quality inspection and sorting tasks. Lowered defect-related costs and improved customer-perceived product quality.
Client Feedback	Vision inspection automation caused frequent errors depending on product and environmental conditions, and it was very difficult to respond when new products were added. However, with high-precision vision performance, we were able to improve overall quality.