
Optics: Lens Types
Fixed focus lenses
For the vast majority of industrial applications with a fixed object size and a fixed working distance, the most common type of lens used is a fixed focus model. Because they are designed for a fixed set-up, they offer the best optimised image quality. Adjustment on these lenses is usually limited to the focus and/or iris.
Zoom lenses
Zoom lenses change their focal length by changing the position of the lens elements in relation to each other. This allows various magnifications to be achieved, making them ideal for use in dynamic environments. For precise measurements or applications with high repeatability they tend not to be ideal due to their flexible construction and difficulty in repeating exact settings.
Some zoom lenses can come with presets which are resistors which provide feedback to monitor the position. They aid repeated set-ups but are not precise, while some precision motorised macro lenses use stepper motors to deliver a precise setting of the magnification.
One feature of a zoom lens when set-up correctly is that they stay in focus as the zoom (magnification) is changed. You may also see the term varifocal lenses. These are similar to zoom lenses but change focus when changing zoom. This makes them suitable for fixed applications where the actual focal length needed might not be known before installation, but will not change after installation. As this is not normally the case in machine vision they tend to be rarely used.
Lenses for multi-chip cameras
Multi-chip lenses are specially designed for colour cameras with 2, 3, 4 or more sensors where light is transmitted onto the sensors through a prism, which require lenses that correct for the optical effects of the prism. In order to avoid mechanical damage of the prism, rear protrusion of the lens should be kept very low.
The output from the standard lens shows distinct chromatic aberration, which appears as coloured fringes around the image due to the difference in the way the R, G and B components are transmitted through the lens elements. The output from the colour corrected lens does not suffer from these problems, making it superior for use in colour vision applications where accuracy and colour fidelity is important.
Some CCD or CMOS cameras can only be used with image-side telecentric lenses due to microlens arrays in front of the sensors. These require a perpendicular incidence of light to avoid shading effects. This is also true for cameras with beam splitter prisms, which seperate colours to severals sensors. A telecentric light incidence should be favoured for multi-chip cameras.
Example 2
Another application where telecentric lenses have proved very useful is in the inspection of drawn wire. The gauge of the wire must be checked very accurately as it leaves the die. However, due to the nature of the process, a resonance often occurs in the wire which causes its position to fluctuate and this makes conventional lensing insufficiently accurate.
If a standard lens is used, the distance from the wire to the lens is constantly changing and hence the apparent width or gauge of the wire. An important feature of telecentric imaging is that, as the target moves closer or further away, the size of the image projected onto the sensor remains the same. This means that no matter where the wire is in relation to the lens, the width remains the same. In this application, using a telecentric backlight will increase the accuracy of the inspection.
In general, the ideal configuration for optimum results is to adopt both a telecentric lens and a telecentric illuminator. Using such a setup, any light is parrallel and removes errors due to light diffraction.
The next diagram illustrates the reason why telecentric lenses are often so large. Using the same component as a target for both lens types, the "endocentric" lens collects the light rays from the object in an angle. The telecentric lens, however, only collects parallel or collimated rays that originate from the surface of the target and so the front aperture must be at least the size of the object. As lenses can only be produced up to a certain size, there are limitations for applications where larger objects have to be imaged.
Electrically tuneable lenses enable vision systems to focus (within milliseconds) over a large working range maintaining high optical quality. They are normally mounted either between the lens and camera or on the end of the lens depending on the required working distance. By applying pressure to a ring on the outer part of lens, the liquid changes shape and adjusts the focus.
This technology delivers a number of key application advantages including the ability to image across very large working distances, accurately focus images under computer control, all with response times of between 1.4 and 15 ms with an impressive MTBF in excess of 1 billion movements.
Applications include optical surface inspection, code reading on random distance packages, surveillance and 3D microscopy.
The lens shape is adjusted by applying a current to an electromechanical actuator which changes the shape of the liquid lens. Telecentric lenses require special lens designs to work with liquid lenses. SILL and Optotune have developed a full series of such lenses to be available off the shelf
Telecentric lenses with tunable working distance
Telecentric lenses are indispensable when measuring objects with different heights. Unfortunately, there are limitations with respect to the depth of field.
In order to achieve a significantly greater variation of the working distance while maintaining the necessary resolution, variable focusing is required.
These new products with temporally variable working distance allow telecentric measurements with a frame rate of about 40 fps. The working distance depends almost linearly on the refractive power of the variable focus lens.
Because of the influence of the focal length, the magnification of the lens is not constant. However since this behaviour is linear, it can be corrected by calibrating the setup.
Due to the telecentricity condition, the aperture diaphragm is projected to infinity on the telecentric side. This means that for an object-side telecentric image, the variable-focus lens must be positioned behind the the aperture diaphragm. With such a configuration, it is however no longer possible to achieve image-side telecentricity.
Liquid lenses and computational imaging
Fast stepping through varying focal lengths with a liquid lens while subsequently acquiring images opens up many interesting applications. This technique provides image data that can be merged afterwards, in order to create extended focus images or 3D data (depth from focus).
More know-how about lenses & optics:
Related products