Unified Optical Scanning Technology

Chapter 4.5.5 - Scanner Devices and Techniques: The Fiber Optic Scanner

4.5.5 The Fiber Optic Scanner

A novel method of resonant optical scanning, introduced initially for endoscope application [Sei], was expanded recently [Fau] for more general utilization. It consists simply of a single-mode optical fiber, suspended cantilever for a short distance, and driven at its (bending mode) resonant frequency by a piezoelectric (bimorph) actuator. The other (fixed) end of the optical fiber is coupled appropriately to a light source.

When the fiber is driven as above by a single actuator, the cantilevered output end executes an harmonic scan that is arced from an effective center of curvature, while light propagates outward from the oscillating end. In this one-dimensional mode, it has demonstrated reasonable freedom from perturbing crosscoupling, maintaining high conformity to movement in a single (x-z) plane. When driven by a pair of (x-y) actuators, it forms a proper Lissajous pattern, notably either circular or spiral for information scanning. Both speed and scan angle are high. One-dimensional scan tests have attained speeds above 20KHz and full scan angles of over 70°, with a displacement of the fiber tip of approximately 1 mm.

Processing of the fiber entails two operations, one more common to all applications and one dependent on its utilization. Starting with a standard optical fiber, the common process entails the narrowing or tapering of its diameter by 'pulling,' etching, or in general micro-machining, as conducted for near-field microscopy [Mur]. This allows the attained high resonant frequencies. The use-dependent process entails the formation and coupling of a microball lens to the output tip of the fiber to converge-to 'collimate' and/or to focus the initially widely diffracted output beam for subsequent scanning of an information field.

The ball lens is formed by controlled melting [L&B] of the tip of the cantilevered end of the fiber with a CO2 laser system. The fiber is mounted in an x-y positioning stage with the tip facing downward, centered along an axis coincident with that of the CO2 laser facing upward. An axial vibration technique is implemented whereby the fiber end is accelerated along the axis (by the armature of an audio speaker) driven by a sine wave generator. A pulse generator, synchronized with the sine wave generator, triggers the CO2 laser to pulse at the same point of acceleration of the fiber along the axis. This allows, under control of these time and laser energy parameters, adjustment of the lens melt to vary the size, shape, and focal length of the added lens. With a ball microlens mass added to the oscillating fiber tip, the resonant frequency is lowered, reducing the 20-KHz (unloaded) scan frequency to one greater than 17 KHZ-still respectable.

When the added ball lens focuses to a sampling point of size δ' (i.e., non-diffraction-limited actual spot size), the device may be considered to function in one dimension as an angular scanning objective lens (arced objective scan per Section In this form, it renders optical scan along a correspondingly arced information surface (such as a miniature drum scanner). When the ball lens propagates a con-trolled beam characteristic into a subsequent flat field-type lens, this is considered preobjective scanning, for the subsequent objective lens forms the final focusing of the spot δ' across the information surface.

Whereas the scanned resolution of this novel deflection technique may be determined experimentally as N= W/δ', the ratio of the scanned image length W to the spot size δ' at an appropriate spot overlap criterion, there is an informative interpretation of the intervening process. As in Section 3.4, this system falls into the class of scanners providing augmented resolution, that is, rendering resolution not only due to the angular deflection of a focused beam NΘ, but also due to the 'translation" of the source, Ns. For diffraction-limited systems, NΘ is the familiar resolution expressed by Equation 3-5 as NΘ = ΘD/aλ, which, when augmented with the Ns component, forms Equation 3-10 as N = (ΘD/aλ) • (1 + r/f). Referring to Figure 3.5, the parameters of the system are represented by the fulcrum at point o of the arced scan locus, the distance r to the output surface of the ball lens having a beam width D, its focal distance to P (whether imaging directly with spots δ' on to an arced surface or launched into a subsequent lens with a beam that would focus at P if the lens were not there), and the total scan angle Θ. Proper handling of the focal distance f is expressed at the end of Section 3.4.1, taken as positive for a converging beam and negative for a diverging beam, which is quite possible in this configuration when launched from a small D into a subsequent lens. If the beam remains reasonably collimated between the two lenses, f is taken as ∞. As in all calculations of scanned resolution, appropriate correction is instituted for known departure from diffraction-limited performance.



Already a GlobalSpec user? Log in.

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.

Customize Your GlobalSpec Experience

Category: Dimensional and Profile Scanners
Privacy Policy

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.