Technical Field This invention relates to optical fibers and more particularly to an optical fiber bulge.

Background Art Sensors for the measurement of various physical parameters such as pressure and temperature often rely on the transmission of strain from an elastic structure (e. g., a diaphragm, bellows, etc.) to a sensing element. In a pressure sensor, the sensing element may be bonded to the elastic structure with a suitable adhesive.

It is also known that the attachment of the sensing element to the elastic structure can be a large source of error if the attachment is not highly stable. In the case of sensors which measure static or very slowly changing parameters, the long term stability of the attachment to the structure is extremely important. A major source of such long term sensor instability is a phenomenon known as "creep", i. e., change in strain on the sensing element with no change in applied load on the elastic structure, which results in a DC shift or drift error in the sensor signal.

Certain types of fiber optic sensors for measuring static and/or quasi-static parameters require a highly stable, very low creep attachment of the optical fiber to the elastic structure. One example of a fiber optic based sensor is that described in U. S. Patent application Serial No. 08/925,598 entitled"High

Sensitivity Fiber Optic Pressure Sensor for Use in Harsh Environments"to Robert J. Maron, which is incorporated herein by reference in its entirety. In that case, an optical fiber is attached to a compressible bellows at one location along the fiber and to a rigid structure at a second location along the fiber with a Bragg grating embedded within the fiber between these two fiber attachment locations.

As the bellows is compressed due to an external pressure change, the strain on the fiber grating changes, which changes the wavelength of light reflected by the grating. If the attachment of the fiber to the structure is not stable, the fiber may move (or creep) relative to the structure it is attached to, and the aforementioned measurement inaccuracies occur.

One common technique for attaching the optical fiber to a structure is epoxy adhesives. It is common to restrict the use of epoxy adhesives to temperatures below the glass transition temperature of the epoxy. Above the glass transition temperature, the epoxy transitions to a soft state in which creep becomes significant and, thus, the epoxy becomes unusable for attachment of a sensing element in a precision transducer. Also, even below the glass transition temperature significant creep may occur.

Another technique is to solder the structure to a metal-coated fiber.

However, it is known that solders are susceptible to creep under certain conditions. In particular, some soft solders, such as common lead-tin (PbSn) solder, have a relatively low melting point temperature and are thus relatively unsuitable for use in transducers that are used at elevated temperatures and/or at high levels of stress in the solder attachment. The use of"hard"solders with higher melting temperatures, such as gold-germanium (AuGe) and gold-silicon (AuSi), can reduce the problem; however, at elevated temperatures and/or high stress at the solder attachment, these hard solders also exhibit creep. In addition, the high melting temperature of such solders may damage the metal coating and/or damage the bond between the metal coating and glass fiber.

Summary of the Invention Objects of the present invention include provision of a creep-resistant high-strength technique for attaching a structure to optical fiber.

According to the present invention, an optical waveguide comprises a cladding; a core within the cladding; and the cladding having a bulge of an outer dimension of said cladding.

According further to the present invention, the waveguide is an optical fiber. According still further to the present invention, a buffer layer is adjacent to the cladding.

The present invention provides a significant improvement over the prior art by providing an optical fiber (or waveguide) with a bulge which is easily and economically produced and which allows for many options for attachment of the optical fiber to a structure. The fiber exhibits low optical loss of light propagating along the core through the bulge and good mechanical strength. Also, more than one bulge may be provided along a given optical fiber.

The foregoing and other objects, features and advantages of the present invention will become more apparent in light of the following detailed description of exemplary embodiments thereof.

Brief Description of the Drawings Fig. 1 is a side view of an optical fiber showing a bulge, in accordance with the present invention.

Fig. 2 is a perspective view of a device that may be used to create the bulge of Fig. 1 in an optical fiber, in accordance with the present invention.

Fig. 3 is a blown-up perspective view of a heating filament used to heat an optical fiber, in accordance with the present invention.

Best Mode for Carrying Out the Invention Referring to Fig. 1, an optical waveguide 10, e. g., a known single mode optical fiber, has a cladding 12 having an outer diameter dl of about 125 microns and a core 14 having a diameter d2 of approximately 7-10 microns (e. g., 9

Fig. 4 is a side view cross-section of an optical fiber showing a decreased outer diameter region in an optical fiber, in accordance with the present invention.

Best Mode for Carrying Out the Invention Referring to Fig. 1, an optical waveguide 10, e. g., a known single mode optical fiber, having a cladding 12 with an outer diameter dl of about 125 microns and a core 14 having a diameter d2 of approximately 7-10 microns (e. g., 9 mircons), has a region 16 with an increased (or expanded) outer diameter (or dimension), in accordance with the present invention. The fiber 10 is designed to propagate light along the core 14 of the fiber 10. The cladding 12 and the core 14 are made of fused silica glass or doped silica glasses. Other materials for the optical fiber or waveguide may be used if desired. The region 16 has a length L of about 500 microns, and an outer diameter d3 of about 200 microns. Other dimensions of the cladding 12, the core 14, and the region 16 may be used if desired, provided the diameter d3 of the region 16 is greater than the diameter dl.

One technique for making the expanded region 16 is to use a fiber (or fiber section) which has an enlarged diameter d4 substantially equal to or greater than the diameter d3 of the region 16. The fiber section may be made using a suitable glass pre-form with a cladding/core diameter ratio that can be drawn down using conventional techniques to achieve the desired core size but has a cladding outer diameter d4 which is greater than the desired value for the final optical fiber. To create the expanded region 16, the diameter d4 of the fiber 10 is reduced to the desired diameter by eliminating an outer portion 15 of the cladding by conventional (or yet to be developed) glass manufacturing techniques, e. g., grinding, etching, polishing, etc. If desired, some of the outer diameter of the region 16 may also be removed. Using chemical etching (e. g., with hydrofluoric acid or other chemical etches), laser etching, or laser enhanced chemical etching are some techniques which reduce the fiber outer diameter without applying direct contact force as is required by grinding and

coating or buffer layer 18 used to protect the fiber 10 or bulge 16 and/or enhance attachment to the fiber (discussed more hereinafter).

Referring to Figs. 2 and 3, one technique for making the bulge 16 in the optical fiber 10 is to heat and compress the fiber 10 as follows. First, the fiber 10 is prepared by stripping any protective overcoating or buffer layers from the fiber 10 to expose the cladding 12 of the fiber 10 in at least the area where the bulge 16 is to be made. This may be done by chemical or thermal techniques, such as dipping the desired section of the fiber in a hot bath of sulfuric acid. Then, the fiber is cleaned using well known procedures in the field of optical splicing, such as dipping in deionized water and then in isopropyl alcohol. Other stripping and/or cleaning techniques may be used if desired, provided they do not damage the fiber.

Referring to Figs. 2 and 3, a device 20 that may be used to make the bulge 16 is a Model FFS-1000 Filament Fusion Splicing System, made by Vytran Corp.

The device 20 comprises a pair of movable fiber holding blocks 23, a pair of vacuum V-groove fiber holders 22, a movable splice head 25 and a hinge- mounted splice top 24 with a filament port hole 26. The fiber holding blocks 23 comprise a U-shaped frame and a center, spring-loaded block that contains a vacuum V-groove insert, in which the fiber is inserted. The components 22,23 are aligned such that the fiber 10 lies substantially along a straight line. Within each of the fiber holding blocks 23, a stepper motor-driven worm-gear rotary mechanism (not shown) allows for movement of the blocks 23 (and thus the fiber 10) along the longitudinal axis of the fiber 10. The parts 22-26 are supported by a transfer jig or housing 27. The splice head 25 comprises a heat source, e. g., a resistive heating element (such as a Tungsten filament ribbon) 29 (Fig. 3) having a width W of about 0.025 inches, which provides radiation heating evenly around the circumference of the fiber 10. Other heating techniques may be used if desired, e. g., a laser, a small oven, a torch, etc. Also, other devices and components for aligning and axially compressing the fiber 10 may be used if desired.

The fiber 10 is placed in the blocks 23 and the holders 22 (and across the splice head 25) which places the longitudinal axis of the fiber 10 substantially along a straight line, i. e., in axial alignment (along the longitudinal or Z-axis of the fiber). The vacuum in the vacuum V-groove fiber holders 22 is set strong enough to keep the fiber in axial alignment but not so strong as to cause surface defects on the fiber. Next, the fiber 10 is heated where the bulge is to be made by applying a predetermined amount of power to the filament 29, e. g., about 26 Watts power. The heating element reaches a temperature (approximately 2100°C), such that the glass is at about 2000°C (the melting or softening temperature of the glass fiber). The heat is applied to the fiber for a duration (pre-heat time) long enough to soften the fiber 10 enough to be compressed, e. g., approximately one second.

Then, while heat is still being applied to the fiber 10, the fiber 10 is compressed axially by translation of the blocks 23 toward each other as indicated by the arrows 21 by the motors within the blocks 23. The total translation of the blocks 23 (and thus compression of the fiber 10) is about 400 microns at a rate of 100 microns/sec for about 4 seconds. Other compression amounts, rates, and times for the axial compression may be used if desired. Compression may be achieved by moving one or both blocks 23 provided the same total motion occurs. After the compression is complete, the heating of the fiber may be maintained for a predetermined post-compression time, approximately 0.25 seconds, to allow the bulge 16 to reach final form. Other pre-heat times and post- compression times may be used.

Next, the fiber 10 is again heated with the filament 29 (or"fire polished") to remove surface defects, at a power setting of about 21.5 Watts. During fire polishing, the filament (and the splice head 25) is moved back and forth (e. g., 2 full passes) across a predetermined length of the fiber (about 2500 microns) across where the bulge 16 was formed, as indicated by the arrows 19, for a duration of about 3 seconds. Other fire polishing power (temperature), number of passes, and time settings may be used if desired, provided the surface defects are removed and the bulge 16 is substantially not altered or deformed. The fire

polishing may be performed immediately after forming the bulge without stopping the heating of the fiber or the heating of the fiber may be stopped (filament turned off) for a predetermined period of time after compression is complete and then turned on to perform the fire polish.

Also, during heating, the area within the splice head 25 around the fiber 10 is purged with flowing high purity argon gas to keep the fiber clean and to prevent high temperature oxidation of the tungsten filament.

The parameter settings (times, powers, etc.) described above results in an acceptable combination of mechanical strength and low optical loss. However, other suitable parameter combinations may be used if desired to obtain a similar effect which may be determined by one skilled in the art in view of the teachings herein.

The process described above for making the bulge 16 may be performed with the longitudinal axis of the fiber 10 (and the device 20) aligned horizontally or vertically or with other orientations. One advantage to vertical orientation is that it minimizes axial distortions caused by gravitational effects of heating a fiber. Alternatively, the fiber may be rotated during heating and compression to minimize gravity effects.

After the bulge 16 is made, the cladding 12 may be re-coated with the protective overcoat or buffer layer 18 (Fig. 1), such as a metal, polymer, teflon, and/or carbon, or other materials. The bulge 16 allows the fiber 10 to be attached to a structure in many different ways for many different applications, by providing a mechanical stop to reduce or eliminate creep, such as is discussed in copending U. S. Patent Application, Serial No. (Cidra Docket No. CC-0080), filed contemporaneously herewith.

More than one bulge may be formed along a given optical fiber if desired.

It should be understood that the Figs. shown herein are not drawn to scale.