3D Line Confocal Imaging of Glass and Ceramics
Line confocal sensors, and the scanners based on them, can be used in the imaging of shiny and mirror-like surfaces, transparent materials, and multi-layered structures in various metrology and inspection applications.
In the late 2000s, researchers at Technical Research Center of Finland (VTT) invented a new optical method to image 3D microtopography of surfaces at sub-micron resolution. This method, called line confocal imaging (LCI), is based on capturing a continuous line of light reflections in visible spectrum from more than 2,000 lateral surface points simultaneously.
Line confocal sensors, and the scanners based on them, can be used in the imaging of shiny and mirror-like surfaces, transparent materials, and multi-layered structures in various metrology and inspection applications on discrete parts, assemblies, webs, and other continuous products. Imaging results can be used to determine a product’s form, surface topography, waviness, roughness and texture, thickness, flatness, 3D volume, etc. Additional features of the LCI method include its tomographic functionality, which enables the capture of 3D structures under transparent material layers, as well as its capability to simultaneously acquire 2D gray-scale (intensity) images from single or multiple surfaces with a large depth of focus that covers the sensor’s entire z range, up to 5.5 mm.
Each sensor has two front lenses―one for its transmitter and another for the receiver. Figure 1 depicts an LCI sensor’s imaging principle. The sensor’s transmitter has a light source that emits white light containing all visible wavelengths. A complex optical assembly separates the light into wavelengths and focuses a horizontal line of each color at a different distance from the sensor, forming a focal plane. Depending on the vertical position of the imaged surface within the plane, corresponding wavelengths from 2,048 lateral measurement points are reflected back to the sensor’s receiver.
The receiver’s spectral camera captures wavelength and intensity information from each point and forms related height profile and intensity lines. When the surface moves in front of the sensor, a 3D point cloud and a 2D grayscale image are generated from the scanned surface line by line. The resulting data can be processed, analyzed and reported with various 3D surface analysis and image processing software packages.
Depending on the imaged material, surface type and resolution requirements, three sensor resolutions are available. The sensor with highest resolution offers vertical (z) resolution of 100 nm and z range of 1 mm. The next model up in size has z range of 2.8 mm and z resolution of 0.55 µm. The largest model has z range of 5.5 mm and z resolution of 0.98 µm.
Lateral resolution in measurement line direction (x) varies from 2.2 to 8 µm, depending on the sensor model. Y resolution depends on line spacing, which also affects the maximum surface motion speed in the application. The length of the sensors’ measurement line is from 4.5-16.4 mm. The sensors can be integrated in both off-line and real-time in-production metrology systems, as they can operate at a speed of up to approximately 5,000 lines per second. This results in a data acquisition rate of over 10 million 3D surface points per second. Data from multiple interfaces in multilayered structures and 2D intensity image data, also potentially from several vertical layers within the scanned area, further increase this number.
LCI sensors and systems work well in applications that require high-speed imaging of challenging materials at sub-micron resolution. Scanning an area that would require several minutes for traditional 3D imaging methods, such as point confocal or interferometric technologies, can be completed in a few seconds.
Transparent surfaces and products made of highly reflective and mirror-like materials are suited for LCI. The sensors’ large numerical aperture and high tolerance for surface angle variation allow for the imaging of steep mirrored slopes and glossy curved surfaces. LCI does not suffer from speckle noise; this enables surface imaging at much higher resolution than laser triangulation sensors.
The quality of raw image data produced by LCI sensors is excellent and rarely needs filtering or other manipulation of any kind. Since the sensors capture 2,048 measurement points simultaneously, vibration of the imaged surface or the sensor itself seldom causes issues as relative point height positions within the measurement profile line remain unaffected.
Applications on Glass and Ceramic Surfaces
Most LCI sensors and systems are currently used in research and development, as well as quality control for manufactured parts, assemblies, and continuous products. LCI can be used on glass, ceramics, polymers, composites, metals and other materials.
The technology is able to measure and inspect precision molded, cast, cut and machined parts; etched and sandblasted surfaces; flat glass; and glass and ceramic coatings. In both laboratory and real-time production conditions, LCI sensors can image dimensions of parts and assemblies such as optical components, mirrors, and microfluidic devices; surface roughness and texture; depth and shape of etched 3D features; flatness and waviness; edge shape and radius; and nicks, cracks, and other defects.
An approximately 3 µm-deep scratch on a glass surface is imaged in Figure 2. Figure 3 depicts tomography of a 2.5D cell phone display panel with the cover glass, touch sensor and OLED layers. The glossy polymer frame and a section of the phone’s aluminum housing are shown as well.
In addition to topographic and tomographic 3D imaging, LCI sensors can be used to measure the thickness of transparent glass layers. For in-process thickness measurement applications where 3D imaging is not required, a dedicated monochromatic dual-point IR sensor can measure the thickness, distance, and surface angle of transparent and semi-transparent glass layers and coatings at a speed of up to 10,000 measurements per second with a resolution of 50 nanometers.*
Bare LCI sensors can be integrated into custom systems for off-line and online metrology applications. A software development kit allows integrators to control sensor functions and read 3D profile and 2D intensity data as the specific application may require.
One application that can benefit from high-speed non-contact 3D surface imaging technology is inline surface roughness measurement. A system with a stationary LCI sensor, designed to automatically monitor surface roughness of fast-moving continuous narrow products, covers applications with a line speed of up to 150 m/min.**
Another unit is equipped with a traversing sensor mount, and it measures surface roughness of sheets and coatings across the width of narrow or wide webs.† It was originally developed for measuring surface roughness of PVB interlayer films used in laminated safety glass structures, but it has a variety of other applications as well. Both systems monitor product surface continuously in real-time and can be used to minimize roughness or to keep it at a desired level.
*MCP100, from FocalSpec. **MicroProfiler MP900, from FocalSpec. †MicroProfiler MP9000, from FocalSpec.