Seeing is believing: sensing and machine vision form the nerve system of collaborative manufacturing, not just in typical QC and QA applications but in design and assembly. See below how sensors and vision systems help keep our money real and our borders safe.
Keeping Fuselage In Line
Three years ago, Lockheed Martin Aeronautical Systems (LM Aero) opened the initial stages of its new production line for the U.S. Air Force's F/A-22 Raptor fighter jet. The F/A-22 fuselage is manufactured in three pieces, and the ability to quickly and accurately align and mate the three sections was a top priority for the tooling department. LM Aero "required a TIR (Total Indicated Readout) of 0.004 in. from the designed center positions of 16 wing nodes spanning 18 ft on each side of the plane," notes James Keywood, F/A-22 tooling dept manager during the design and build phase. And the entire assembly jig had to move down the first 278 ft of the assembly line and then return to the start of the line under its own power. LM Aero engineers designed a "skate" concept and Delta Sigma Corp won the bid over the competition to automate the skates. In turn, Delta Sigma selected DVT machine vision cameras for the system.
Dr. Brett Haisty, vp of engineering at Delta Sigma, explains: "We designed a set of 'pins' and put the cameras in them so that they fit in a set of tooling holes that look into the wing lugs. This accurately simulates the wing joining the fuselage. Since the cameras are inside the pins that attach the wings, they measure the misalignment, report the errors back to the control computer, which, in turn, calculates the moves required of every axis to solve for 14 axes simultaneously."
Lockheed aligned and mated the first jet "in about 10 minutes," adds Keywood. LM Aero has adopted the solution to assemble the Raptor's mid-fuselage section, with two systems installed already and a third planned for third quarter 2006. (For more, click here.)
What's In Your Wallet?
The U.S. Dept of the Treasury's Bureau of Engraving and Printing (BEP) is responsible for producing all of the currency for the world's largest economy. From the paper on which the currency is printed to its sheer output of approximately 37 million notes each day, nearly every aspect of BEP's design and printing process is distinctive. BEP must verify that its printed materials are clear and accurate, and needs to ensure that various security measures are in place to help deter counterfeiting. To help meet these demanding requirements, the Bureau converted several of its inspection stations from human spot checkers to automated visual inspection of each note.
BEP selected ipd's Sherlock machine vision software application for the development of three machine vision systems: automated inspection systems for printing plate quality assurance, automation of finished note measurement for print registration consistency, and online measurement of a quality and security feature on newly printed bills. BEP's production engineering group developed a Plate Measurement Device (PMD) with positioning technology and machine vision to automatically measure the layout pattern of security features. The team developed the PMD's machine vision inspection system using a variety of advanced, off-the-shelf optoelectronic components. A frame grabber integrated within the host PC captures images from a 1K x 1K monochrome camera. Using a precision lens with a 0.5 in. field of view, the camera achieves a measurement tolerance of ± 0.001 in. (For more, click here.)
Counting On Those Tires
Measuring and understanding vertically loaded tire tread footprint pressure distribution is the key to any tire design process. Improper tire tread pressure distribution can result in bad road traction and high road noise, early tire replacement, and in extreme cases, tire failure. Tiremaker Kumho America improves its tire design process with Sensors Products' Tactilus, an electronic sensor system that evaluates tactile contact pressures in real-time. The setup includes a silicon substrate element with embedded matrix-based sensors, a controller, and 2D/3D imaging software.
"I used the system over a period of time to obtain several static footprint pressure distribution profiles of a loaded tire," notes Kumho America technical center manager Dr. Hamid Aboutorabi. "The profiles revealed some very uneven tire tread surface contact areas. The pressure distribution patterns we discovered helped us redesign our tires in pursuit of optimum tread surface contact and road handling properties." (For more, click here.)
Automating Colors
Collins & Aikman (C & A), a manufacturer of carpet sets for the automotive industry, recently decided to migrate the color matching process from a manual to a fully automated process in a move to assure color match quality from sample to sample. The company decided that color sensors using fiber optic technology would fail because of inadequate separation of sensor lenses, while contrast sensors would not read colors off the carpet texture. Nap or pile can lie differently from one piece of carpet to the next, changing its perceived color.
Managers at C & A's Greeneville, SC plant asked Balluff for a solution, and the Balluff team chose the BFS 26K color sensor as a basis for the final color matching process. Up to four Balluff sensors read the vinyl pads and two more read the carpet. The BFS 26K carpet sensors are programmed to recognize the carpet color regardless of nap/pile orientation.
C & A's Ted Manning comments, "Our automotive customers have absolute standards when it comes to color match on these carpet pieces. Balluff has helped us substantially increase our yields of perfect carpet sets, while simultaneously raising our productivity and lowering our scrap rate." (For more, click here.)
