Fast, accurate, automated vision inspection

Machine Builder Provides Vision System That Can Handle Three Parts Per Second With 99.97% Accuracy. Its Name Is 'Vision System'

By Brian Altman, Altman Manufacturing

1 of 2 < 1 | 2 View on one page

Virtually every manufacturer recognizes the role that quality plays in establishing lasting, trusted relationships with its customers — an essential element that helps enhance its growth and long-term stability. Ensuring that level of quality, particularly in long-run, difficult-to-automate, component manufacturing and assembly operations often could be a daunting task, for which the only solution was throwing people at it.

For a long time, these manufacturers had to rely almost entirely on the manual inspection of these components. For an equally long time, machine vision inspection systems were costly, limited in capability and had uniquely developed software that was very difficult to configure, use or modify.

Now, as amazing as the human eye is, it's unreliable in repetitive work, especially when the work involves judging slight differences in size. To avoid the mistakes that workers are bound to make from time to time, most manufacturers today will rely on automation for their quality checks whenever they can.

Dare to Do the Difficult
Established in 1942 and headquartered in Lisle, Ill., Altman Manufacturing is a family-owned manufacturer of automation, fabrication and tool-and-die-cutting machinery that specializes in developing equipment and processes to mechanize and finish hard-to-automate products, including rubber and plastic goods without a defined geometry.

About three years ago, a customer that manufactures Elastomer rubber components for use in the medical industry approached Altman Manufacturing to develop a machine that it could use to automate its process to inspect finished products.

"To illustrate their challenge, our customer brought us three different products they produced: a seal, an item similar to an O-ring and a filter/seal combination," explains Paul Altman, Jr., who leads engineering at Altman Manufacturing. "While they were made of the same silicone material, the three products were different sizes, shapes and colors, and the filter, in particular, had holes that could easily be filled with excess silicone or flash during the manufacturing process. The customer wanted a system that would inspect these different parts, and also could scale to inspect any new parts they manufactured in the future."

While defective parts can be detected during manual inspections, the fact that any defective part might make it to assembly posed challenges for the customer, who could be back billed for any product that didn't work as promised.

Also read: 10 advantages of using vision to guide robots

The time and labor required to inspect products using the customer's existing manual system were costly; inspections were completed by workers who sat at inspection tables and manually checked the small parts — some of which were less than an inch in diameter — as they moved down a conveyor. Even the most efficient operator was only 87% effective, and because the manual inspection process was time-consuming and tedious, the effectiveness of the inspections often diminished in the last hours of each shift.

In fact, a typical inspector is 92% effective during the first hour and 80% effective thereafter. This forces many companies to implement a process in which inspections are staggered, and the same part is inspected multiple times by different inspectors, increasing inspection cost.

If a defective product is found, it could be reworked in the case of flash or overmold, or it might be discarded, for example, if it had a hole.

Seeing the "Proof"
The company has worked with virtually all providers of vision inspection systems, and Altman knew almost immediately which vision system would be the most effective for this application. "We choose the vision system that is best for the application, and what makes Teledyne Dalsa ideal for inspecting Elastomer rubber products are the company's unique algorithms, such as 'contour,' which is perfect for checking the edges of a rubber piece," Altman explains.

"Teledyne Dalsa also provides the flexibility to turn certain features, such as the area calculator, on and off as needed. For general inspections and finding flash, and because of the product's easy-to-use drag-and-drop functionality, we think this is the best vision system for this application."

However, before any system is put into production, Altman evaluates the parts using a sample system for pre-production inspection that was developed with Teledyne Dalsa. Using the sample inspection system allows him to view different parts, and design appropriate lighting to ensure the camera delivers the optimum results.

1 of 2 < 1 | 2 View on one page
Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


  • Above, the article states that "Since the machine can accommodate multiple parts, the parts are hand-fed by an operator, who places them on the belt" Also above, a throughput of 3 parts per second is mentioned. This seems inconsistent. Does the throughput depend on how quickly the operators can place the parts on the conveyor? Three parts per second seems unreachable if that's the case. Also, one of the goals of automated inspection is to save labor. Manual loading seems to defeat that goal. What am I missing?


RSS feed for comments on this page | RSS feed for all comments