|
 Pioneered in Europe in the early 1970s, in-mold labeling (IML) is a process whereby a label becomes forged into the wall of a container during injection molding. Over the past few decades, the process has become common practice among European injection molders. Though IML can often be expensive to implement, it offers a number of advantages over conventional glue-on labeling. For example, because the labeling occurs during the molding process, no extra labeling steps or equipment are required. Additionally, by eliminating the separate labeling station, line throughput often can be improved.
Although U.S. companies have been relatively slow to adopt IML, use of the technology has begun to emerge in recent years. Among U.S. companies riding the IML wave is Precise Technology Inc, a contract injection molder serving the packaging, consumer/industrial and healthcare markets. The company’s Bridgeport, NJ plant is currently the largest molding facility in the states using IML, where they are using the process on containers manufactured for a customer that produces baby wipes.
Automating labeling
 “A key advantage for us to use IML is consumer product safety,” explains Jim Bruynell, maintenance manager at the Bridgeport plant. “Because the label becomes part of the container during molding, it can’t be peeled off.” Another benefit is that the entire container can be recycled, since the label and the tub are both made of polypropylene.
However, Bruynell also points out that the IML process poses some challenges with respect to correct label placement. “We use robots to place each label into the mold cavity, and the robot creates a electro-static field around the label so that it automatically sticks to the mold cavity surface, prior to injecting the molten plastic into the mold. If the field around the label is not strong enough, it may allow the label to move when we inject the plastic, which can cause the label to shift to the left or right.”
Besides improper label registration, other problems can occur, owing to simple human error. For example, labels that have been mis-loaded into the molder may be applied backwards or upside-down, and because the plant produces containers for more than seven different styles of baby wipes, occasionally the wrong label might be applied. In addition to contending with these problems, Precise needs to ensure that a blank white box printed on the label, to which a lot code is later applied at the end-customer’s facility, is in the same position every time.
To guarantee that every container leaving the Bridgeport facility has a perfect label, the plant relies on In-Sight 2000 machine vision sensors from Cognex Corporation. The sensors are used to verify label position within a ±2 mm tolerance, double-check that each label is the correct one for the container, and inspect the lid hinge to verify the lid has been properly snapped into place.
In-Sight vision sensors are compact, general-purpose vision sensors featuring a digital machine vision camera, high-speed vision processing unit, a full library of vision software tools for measuring and inspecting the containers, and a vision spreadsheet interface for application set up. The set up process involved selecting vision tools and parameters from simple drop-down menus. The spreadsheet then automatically generated tool results into worksheet cells, which were then linked together to set up the measurements.
According to Steve LeBlanc of Serview, a Bristol, PA-based systems integrator that helped Precise implement the verification system, vision sensors were an appropriate choice over PC-based vision systems for this application. “The sensors had the right vision tools for the job, and were also fairly straightforward to use for those who may not have had previous experience with machine vision,” he said. “Also, the sensors are less expensive than PC-based systems, making the overall installation more cost-effective for the company.”
PC-based vision versus vision sensors
 A machine vision system is essentially a computer equipped to see and spot defects and other problems in manufactured parts or assemblies. Vision systems look for manufacturing flaws using a combination of microprocessor technology and image analysis software to interpret video images and generate information about them. The systems can then communicate this information to other equipment, such as PLCs or robotic arms that can remove bad parts from the manufacturing line.
Hardware, software and an application development environment make up the core of any vision system. Vision hardware includes a camera that captures an image of the item to be inspected, lighting to enhance the contrast of features of interest, and optics, which accurately represent the image to the camera by minimizing distortion and loss of resolution. This hardware, works with a processor, or “vision engine” to capture, digitize, and display images for analysis to generate answers, such as whether a part is defective.
Vision software “tools” are the backbone of the vision engine. By comparing specific features of interest within the image to stored data that comprises a standard, these vision tools perform image processing and/or analysis of the captured image. There is a wide assortment of vision tools available for performing many different types of inspection operations that enable vision systems to make decisions about a part’s quality, location, size, and identity. Vision tool selection and capability depends on the type of vision system platform used.
Generally, today’s vision systems are divided into two groups: PC-based and sensor-based. The key differentiators include development environment, capability, architecture, and cost. The development environment allows users to “build” (set up, and program or configure) vision applications to meet specific needs. While PC-based systems have a programmable environment, sensor-based systems generally provide a configurable environment that is easier to use.
PC-based machine vision systems offer the most capable vision tools, and provide the fastest performance because they rely on the latest CPU architectures. Performance increases with each boost in PC processor speed, and as a result, they are generally used for more complex or mathematically intensive applications. However, applying PC-based systems requires more vision expertise and knowledge of low-level programming languages such as C++ or
VisualBasic.
In contrast, vision sensors generally require no programming, and provide more user-friendly interfaces. For the most part, vision sensors are systems that are self-contained and do not require the use of a PC, VME, PCI, or similar architecture to run vision tools. While low cost and ease of deployment remain the key attributes of sensor-based platforms, over the last several years vision sensors have become increasingly sophisticated while the cost of PC-based systems has come down simultaneously.
As this gap between PC-based and sensor-based platforms continues to narrow, new users more often than not initiate feasibility studies using a vision sensor. Though ultimately, application complexity and other variables will dictate the final hardware and software requirements, vision sensors offer a price tag that makes the investment more easily cost-justifiable. In addition, vision sensors are stand-alone systems that are easily integrated with any machine to provide single-point inspections with dedicated processing, and most vision sensors offer built-in Ethernet communications for factory-wide networkability.
Label spotting
 Once a container has been molded and the lid has been snapped into the top, the entire assembly is flipped upside down (so the tub bottom faces up) and moved down a conveyor towards the inspection point. Two vision sensors are used to provide full inspection coverage of the left and right hand portions of each container. The cameras are mounted in a light-shielding cabinet approximately 18 inches above the inspection point.
Looking down at a 45-degree angle, the cameras, triggered by photo optic sensors, capture an image of each side of a container, and the images are immediately transferred to the vision processor. There, they are analyzed using the In-Sight PatFind pattern-matching tool, which compares each image to a pre-trained reference image of what the label should look like. Once this match has occurred, various edge detection and inspection tools are used to measure the exact X, Y and theta position of the label, and to ensure that the lid has been properly assembled to the container. More than 60 containers are inspected per minute.
To optimize the accuracy and reliability of the inspections, LaBlanc used multiple lighting techniques to enhance the contrast of the features of interest. In general, these techniques include a combination of minimizing shadows, freezing the motion of moving objects, increasing the sharpness of edges, removing specular reflection (glare), and making the foreground and background distinctly different gray values. “The application involves a myriad of inspections,” says LaBlanc, “from measuring label position to assembly verification of the container. There was also a variety of part appearance variations to deal with, such as glare on the part, different colored plastics, and label graphic variations. So, we implemented a highly engineered illumination system involving back lighting, front lighting, and structured lighting techniques.” For this application, high-frequency fluorescents were used for front and back lighting, and lasers were used in structured lighting techniques to achieve the desired brightness and contrast for the application. Back lighting provides maximum contrast between the part outline and its background, and is good for measuring external part edges. Structured lighting helps to measure height or depth, and accentuates surface profile on low contrast parts. Diffuse front lighting is applied on three-dimensional parts where shadows may be a problem because it tends to minimize shadows with soft, even illumination from all directions. “What comes out of all this,” remarks LaBlanc, “are really nice, high contrast images that work very well for the vision sensor’s measurement and inspection tools.”
Each In-Sight 2000 vision sensor used in this application consists of a vision processing unit based on digital signal processing, a separate 640 x 480 x 8-bit progressive scan digital camera, onboard light control, built-in discrete input/output, a standard VGA display output, a handheld control pad, and a library of vision software tools. Consequently, no PC is required, only a dumb monitor.
If a container passes inspection, it proceeds down the conveyor to a diverter and is then placed into a stacker. Containers that fail inspection are ejected off the conveyor into a dumpster and ultimately discarded. A VGA monitor above the inspection shows a live image of the inspections as they occur, green and red “LEDs” indicating pass or fail, a readout of the total number of parts inspected per shift, as well as the number of bad parts per shift.
According to Bruynell, since the vision sensors were installed, they have been able to inspect the labels 24/7 with a high degree of accuracy and repeatability. “The accuracy is superb, and we know the labels we end up shipping are the same every time. This allows us to ensure 100% product quality for our customer, and ties in with our overall continuous improvement strategy.”
To date, the company has implemented the two-vision sensor solution on four lines, and currently has plants to tie the inspection results data into their ERP system for improved production reporting.
For more information:
Precise Technology Inc,
www.rsleads.com/311df-105
Cognex,
www.rsleads.com/311df-107
|
