Instead of being reserved for end-of-the-line engine assembly verification, machine vision can now be cost-effectively deployed at various key process points along the way.

Over the past several years, many automotive manufacturers have been motivated to increase the reliability of their products. Implementing machine vision for automated inspection helps these manufacturers achieve this reliability. Other driving forces include improving quality, streamlining production, decreasing scrap rates, and managing inventory and gathering process control data by reading part codes.

By far, one of the most important components in an automobile is the engine. It is the heart of an automobile. Therefore, it is imperative that this vital component operates correctly once it leaves the assembly line and is installed into a car.

In years past, machine vision was strictly reserved for end-of-the-line automotive engine assembly verification. However, vision systems can now be cost-effectively deployed at various key process points along the way. By distributing vision sensors at multiple points and using Ethernet communications to connect them, manufacturers can respond more quickly to manufacturing problems, achieve better process control and minimize production costs.

Why Use Vision Technology

Machine vision technology can perform a wide range of engine inspection operations, including verifying the correct placement of cylinder heads, reading 2D codes on engine blocks and fuel injectors, and measuring critical dimensions on head gaskets.

Most engine assembly processes can benefit from automated inspection. However, many vision systems are expensive. Therefore, manufacturers often only use them at the end of the assembly line.

End-of-the-line inspection systems are fine for final part checking. However, conducting an inspection after an engine has undergone numerous value-adding assembly stages is not cost-effective. Defects need to be detected at the point of occurrence. If not, a line may be shut down for several hours. Operators then have to trace processes back upstream to determine the cause of the problem.

An incorrectly assembled engine necessitates rework, which can be expensive and time-consuming. This often costs manufacturers up to thousands of dollars per engine. It may take hours to manually disassemble a faulty engine, at an average of $50 per hour for labor. Add in the costs associated with shipping a new engine to the customer and returning the defective engine, and the costs quickly increase. Some defective engines may even pass final test, be assembled into a vehicle, and reach the consumer. This results in an even more serious quality issue.

Multiplexed Vision Cameras

How can a manufacturer successfully and inexpensively integrate a vision system throughout the entire engine assembly process? One way is by multiplexing vision cameras from a single processor. This method allows manufacturers to expand an end-of-the-line vision system to other points in the assembly line.

A multiplexer is a device that allows multiple cameras to be connected to a frame grabber. Frame grabbers are PC plug-in cards that transfer captured images from a vision camera to a PC processor. Most frame grabbers support up to four cameras on their own. Multiplexers are used when an application requires more than four cameras.

The benefit with vision cameras is that the hardware costs associated with each inspection point are minimal. Vision cameras have a downside, though. When linking off of a single processor, vision processing is divided up among multiple cameras. The result is reduced vision performance at each point. Another downside is scalability, which can be defined as how well a solution to a problem will work when the size of the problem increases. With a multiplexed camera configuration, scalability is inherently poor, because only a finite number of cameras can be used.

There is also a large amount of software complexity and integration costs involved in managing and maintaining multiplexed cameras. Special programming is typically required to synchronize camera triggers and outputs. These programs can be very time-consuming and expensive to develop. Also, when a problem develops at one station, the entire system often has to be taken off-line to remedy the problem.

Vision Sensors

Another machine vision inspection approach involves sensors. Vision sensors are compact, online inspection devices that combine camera, processor, vision software tools and communications in a self-contained unit.

Because of the afford-able nature of many vision sensors and improvements in processor technology, it is now possible to have one processor per camera. This results in maximum vision performance and uptime at every point on the production line where a sensor is placed. Also, because each sensor is an independent device, the complexity associated with synchronizing multiple cameras off of one processor is eliminated.

A company that only used machine vision at final inspection to check for the presence of components can now justify adding vision sensors upstream to check the placement of fuel rails, gaskets and pistons, as they are added.

Networking Sensors—The Ethernet Approach

Because vision sensors can be distributed cost-effectively throughout the engine assembly process, manufacturers need a centralized way to manage the growing number of vision sensors operating on the assembly floor.

Many of today’s vision sensors offer built-in Ethernet networking capabilities. Ethernet networking provides access to real-time vision data, streamlined system integration, and the ability to remotely manage, monitor and control vision activity. With Ethernet connectivity, vision sensors can be linked throughout the factory floor, turning distributed vision into a fully scalable, manageable vision area network. This enables users to set up and modify vision applications from remote sites, monitor inspection activity from any location in the plant and share up-to-the-minute production data with management.

Networked vision can be implemented in two ways. First, two or more vision sensors can be linked over Ethernet to form a dedicated vision area network. In a vision area network, vision sensors can exchange data and be managed by a host—either a PC or another vision sensor.

A network of vision sensors linked over Ethernet enables direct peer-to-peer communications between each sensor over a single line, so there are no complicated cabling schemes. Because a host manages the network of sensors, vision data from all four sensors can be collected at a central point and viewed on a single video monitor. The host can also archive failed images from each vision sensor, which can be used to better determine why certain failures occurred.

The second way of implementing networked vision is to uplink a vision area network to existing plant and enterprise networks. This can provide a number of benefits. It enables users to manage vision activity from remote locations. For example, users could set up and modify vision applications, share applications with other plant sites, and remotely troubleshoot problems with technicians, all without ever leaving the office. Additionally, uplinking to plant and enterprise networks enables manufacturers to gain direct access to the data related to the quality of their products directly from the vision sensors from anywhere in the plant, the enterprise or within their global organizations. Quality engineers may want to view statistical process control data, while management may want to keep an eye on production output. All it takes is a workstation with transmission control protocol and Internet protocol capability.