Robotics Industry Insights
- This industry insights is filed under:
Error-Proofing in Today’s Competitive Manufacturing Environment
by Rick Bondy, Field Marketing Product Manager
SICK, Inc. Posted 06/22/2006
Error-proofing is an essential systematic process for improving the reliability, quality and stability of lean manufacturing methods. You can never completely eliminate error, but error-proofing helps mitigate the impact of human error involved in the manufacturing process.
The goal is to catch the mistake on the assembly/production line, before it is on the finished product…or in the hands of the consumer. If a product is defective and needs to be reworked or returned, there are labor and materials costs that will be incurred. When error-proofing processes are implemented, the speed and accuracy of systems are improved, as is the overall quality of products.
While the concept of error-proofing originated in the automotive industry, the use of non-contact sensors, especially laser diode style sensors, in error-proofing applications has grown significantly over the last decade. More and more robotics manufacturers are interested in using non-contact sensors for automated error-proofing over the traditional probe method that requires frequent calibration.
Today’s manufacturers can select from a wide range of long-range, laser-based sensors and other sensing technology to ensure that products are made correctly at several stages during the production process. These sensors not only verify part presence, or that an assembly step has been completed, but also ensure that a part has been properly manufactured – measuring for machining defects down to the micron level.
Sensors for error-proofing applications:
- Laser/Visible Red LED
- Non-safety rated Light Grids
- Optical Displacement
- Distance Measurement
Which Sensor Makes Sense?
If the wrong sensing technology is used for an error-proofing application, there is a loss of efficiency. SICK recommends you focus on three decision-making criteria when determining the best sensor for an error-proofing application.
- Performance - the sensor must perform to achieve the results expected by the customer
- Repeatability - sensor must perform consistently in the real-world application at the accepted technology level of the customer
- Cost – the price for the application solution must be within the customer’s expectations
How to Apply Different Sensing Technology
Laser Sensors/Visible Red LED
Utilizing Class I and Class II laser-based sensors to detect very small targets has gained wide acceptance in factory automation applications. In addition, OSHA and other regulatory agencies deem these lasers safe for factory automation applications.
Due to limited space in today’s machine designs, the majority of photoelectric sensors used on equipment are in diffuse mode, since there is no room to mount reflectors or sender/receivers. This requires flexibility to ensure that the sensors accurately detect the presence of the target but ignore nearby backgrounds. Laser-based sensors have an extremely small focused light spot, which helps in situations where there is tight fixturing.
For example, a powertrain engine plant needed to confirm that a sealing ring (2 mm thick) had been inserted into the piston casing (figure 1). The sensor must be mounted 50 mm minimum from the target due to carrier movement. The sensor cannot detect the bottom of the piston ring groove and must allow for ring gap. The solution was to use a laser-based sensor.
The light spot size was only 0.2 mm diameter (at 100mm sensing distance) and was smaller than the ring thickness. The set point adjusted to detect the ring but not the piston background. To eliminate a false signal if the ring gap were detected, a second sensor was used at 90 degree position location (figure 2).
These sensors play an important role in the error-proofing process. Luminescent sensors can detect the presence of a luminescent mark on a part as it travels on a conveyor (figure 3). The mark indicates quality approval – that the item has already been inspected.
For example, automotive engine coils will be irregularly located on a moving conveyor belt. In this application, the sensors must be mounted 50 mm, at a minimum, from the conveyor side and must detect the marks, regardless of the position of the coils. The solution was to use six luminescent sensors.
Three sensors were utilized per conveyor side, mounted approximately 50 mm from conveyor rail. The sensors were aligned so that all potential target positions would be detected. It was vital to ensure that only inspected heater cores were installed – avoiding potentially costly problems later.
Color sensors are ideal for identifying parts that are similar in shape but come in different colors – such as a vehicle center console (figure 4). Color sensors use red, green and blue LEDs to detect color. If the customer orders an upgraded wood grain console, a color sensor can be used to ensure that the standard color console is not used.
For this type of application, the manufacturer used the teach-in function to program the sensor to provide one output for the standard target and another output for the wood grain target. However, color sensors are also distance sensitive – so if the target is too far away, it will not function properly.
Light grids can be used for performing in-line profile detection and object recognition applications, where safety light curtains are not required or are not cost-effective. Light grids operate by projecting a series of beams from a sender unit to a receiver.
If the correct sequence of beams is blocked as an object passes through on a conveyor, the light grid will send a signal to allow the robot to pick up the part. Light grids come in a variety of detection heights and ranges and offer simple “plug and play” connections, making them easy to install and setup.
The Latest Developments in Error-Proofing
The trend is not only to confirm that the part/target is present but also to confirm that the part/target is correctly positioned to ensure the correct process. This often requires a
4-20mA analog signal from the sensor correlated to the sensing distance.
The next generation of error-proofing sensors, such as optical displacement sensors, use an analog signal or an RS485 or RS232 data interface . Error-proofing applications for optical displacement sensors can be used for the adjustment and control (X, Y, Z axis) of a robot arm.
For example, the sensor can determine the position of the robot arm so it can pick up a part, such as an engine block. The robot then places the block onto a fixture and holds it there until additional sensors measure the distance to the target to verify that the engine block has been machined properly, measuring the surface down to the micron level.
Another advantage is that some optical displacement sensors use advanced CMOS technology as compared to CCD technology, which requires careful application considerations with highly reflective machined surfaces. CMOS does not bloom on polished surfaces and provide a false signal, which could occur during robot error-proofing applications due to reflections from metal parts.
Distance measurement sensors are also being used more in error-proofing applications. These sensors measure the distance to the target using time of flight technology, whereby the sensor measures the amount of time required for the reflected light to return to the sensor.
Distance measurement sensors can be used for surface inspection to check a part for the presence of special recesses, holes, brackets and equipment, enabling quality assurance…even at long sensing distances. These sensors can also be used for end effector positioning to insert parts onto a vehicle correctly.
The purpose of error-proofing is to detect product defects as early in the manufacturing process as possible, rather than at the final inspection. There are many types of sensor technologies to choose from when designing an error-proofing system. However, the best solution depends on the application. From simple product presence verification… to product profiling…to measuring the surface of a machined part for correct hole size and placement, error-proofing has proven it is a valuable and essential tool to stay competitive. A successful error-proofing program will improve your manufacturing and operations processes – and your bottom line.
Originally published by RIA via www.robotics.org on 06/22/2006