Everything You Ever Wanted to Know About dB But Were Afraid to Ask Telecommunications systems used electrical signals and copper wire to transmit voice messages long before lasers, lightwaves and fiber optic cable. To describe and measure power levels in electrical systems, telecommunication engineers used the standard unit of decibels to express gain or loss and relative power levels. Meanwhile, scientists working with fiber optic signals were using units of milliwatts (mW) to determine the amount of light traveling down a fiber or the amount of light coupling from one fiber to another (as would be expected with optical radiation). As the telecommunications industry began to use fiber it did not adopt the milliwatts terminology. Just the opposite happened, fiber optics adopted the traditional telecommunications language of decibels or dB.Decibel is defined as a unit used to express relative difference in power, usually between acoustic or electric signals, equal to negative ten times the common logarithm of the ratio of the two levels. The main reason dB is used is because it makes power levels more manageable. Thus, it’s easier to add up power losses in a system. For example, a system with 4.0dB of fiber loss, 2.5dB of connector loss,3.0dB of splitter loss and0.5B of splice loss results in a 10dB system loss or the sum of each component loss. Translating 10dB into a percentage based upon the formula given earlier results in a signal that is 10% of the original intensity. Because dB can be used to describe both gain and loss, it is important to carefully consider the given optical parameter. A decibel expressing loss is a negative unit. However, in the fiber optics industry, it is common practice to omit the negative sign and speak of a 3dB loss rather than-3dB. For example, a back reflection level of *-40dB and *40dB are generally taken to mean the same thing (reflected light *0.01% or light is reduced by 99.99%) and therefore one must keep in mind the overall context when making system calculations. An additional point of confusion is the difference between dB and dBm. dB is a comparison of a signal to a reference signal without any specified unit of measurement: dBm is used when 1 milliwatt is the reference signal level: For example, a value of -3dBm means that P is 3dB (50%) less than than 1mW or .5mW. Conversion tables for dB to percent and dB to mW are shown below:
GlasSolder TM Improves Coupler Performance
Gould Fiber Optics has developed a glass solder process (patent pending) for making a glass-to-glass bond between optical fibers and a silica substrate. This bond is much stronger than epoxy and is not susceptible to degradation from humidity. The significance of this packaging improvement is evident in outside plant applications where long term reliability requirements are stringent. Figure 1. Schematic diagram of packaging used by most coupler manufacturers (sans glass solder). The addition of the glass solder eliminates the epoxy as the primary bonding mechanism and greatly improves performance and long term reliability. Background To explain the GlasSolderTM technique, an understanding of the fused biconical taper (FBT) coupler process and packaging is necessary. The basic FBT process, which is inherently stable with low excess loss, consists of fusing together two adjacent fibers by heating and stretching them until the desired amount of coupling is achieved. The basic packaging approach has been to affix the fused fibers to a silica substrate using an adhesive (epoxy) and then placing the substrate into an enclosure (usually a tube). The tube is sealed with a material which also provides strain relief for the fibers. This packaging technique has provided good stability and reliability. By properly attaching the fibers to a silica substrate, the differential thermal expansion between the fibers and the substrate is minimized, producing good thermal stability during temperature cycling. The enclosure and booting material further aids the resistance to hot, humid environments. This resistance has been demonstrated over several thousand hours at accelerated conditions (e.g., 85oC/85%RH). This bond is much stronger than epoxy and is not susceptible to degradation from humidity. While the damp heat test results have been relatively good, this is also the test where component reliability is affected. With long-term exposure to humidity, epoxies tend to soften and swell and the bond between fibers and substrate is weakened. This rarely leads to catastrophic failure, but the resulting variations in insertion loss may exceed the performance specification of the splitter. To improve the long-term reliability of our components, Gould has developed the GlasSolderTM package which augments the epoxy with a glass solder material forming a true chemical bond with both the optical fiber and the substrate. See figure 1. Not only does this method provide a bond which is stronger than an epoxy bond and impervious to moisture, but it is also fast and inexpensive for use in a practical, fused coupler manufacturing process. Process Description Glass solders are inorganic compositions that are often used for making strong, insulating and sometimes hermetic joints or seals between different materials such as glass, ceramics and metals. Usually mixtures of silica and other metal oxides, glass solders form strong ionic bonds which are particularly impervious to moisture. To be used in coupler packaging, glass solders must be chemically and physically compatible with silica fibers and substrates. This means the glass solder must have a surface energy less than that of silica, so that upon the application of heat it softens and sufficiently wets the surfaces. This is essential for obtaining adhesion and bond strength. Additionally, the glass solder should exhibit a thermal coefficient which is similar to silica in order to prevent the formation of cracks. The glass solder is applied in a slurry form. The slurry is comprised of the glass powder, a binder and a carrier or vehicle. The binder, which is eventually burned away when the slurry is heated, provides dimensional stability to the powder after the vehicle has evaporated. The heat required to soften and fuse the glass solder is applied only where the glass solder has been deposited. This is accomplished by using a C02 laser operating at a wavelength of 10.6mm. Evaluation Tests Specific mechanical tests, including vibration, impact, and fiber retention, were performed to evaluate the strength of the bonds between the glass solder and the optical fibers and the substrate. For* the fiber retention tests, standard singlemode fibers were secured to silica substrates with a small (-2 mm diameter) bead of glass solder using the process described above. Tensile forces were then applied to the fiber leads by attaching weights of known mass. All five samples in this study were able to support tensile loads in excess of 5kgf (49N) without failure of the glass solder bonds. By comparison, similar samples prepared using epoxies typically failed at substantially lower pull-out forces.Couplers packaged using the GlasSolderTM technique proved to be quite robust when tested for both vibration and impact. The vibration tests were conducted over the frequency range of 10Hz to 55Hz, in accordance with the test conditions specified in Bellcore TR-NWT-001209. The samples were subjected to simple harmonic motion amplitude of 1.52 mm (0.060″) for a period of two hours of three mutually perpendicular axes. The average change in insertion loss following the vibration test was only -0. 1 dB. During the impact tests, conducted from a height of 1.8 meters, each coupler was dropped eight times along each of three mutually perpendicular axes. The components were packed in a container with sand in order to fully transmit the shock of the impact throughout the coupler package. The average change in insertion loss following the impact tests was only -0.1dB. Figure 2. Long term damp heat test (85C/85%RH) for couplers with epoxy alone and with the addition of the glass solder. Note that, to further accelerate the effects of the hot, humid environment, the test was conducted with an open substrate, ie., without the tube and boots shown in figure 1. In order to test the thermal compatibility of the glass solder with silica fibers and substrates, seventeen fully packaged couplers were subjected to temperature cycle tests between -40oC and 125oC. Each coupler was tested for five temperature cycles and actively monitored throughout the test. The couplers exhibited minimal sensitivity to changes in temperature. Furthermore, only two of the couplers exhibited changes in insertion loss of more than 0.2dB. In an effort to evaluate the ability of the glass solder to withstand prolonged exposure to hot, humid environments, ten partially packaged couplers were subjected to an 85oC/85%RH damp heat environment for approximately 2,500 hours. In order to accelerate the detrimental effects of humidity, the couplers were secured to silica substrates using the glass solder and inserted directly into an environmental chamber. The couplers were not housed inside a protective tube or sealed from the humidity in any way. Figure 3. The maximum peak to peak change in the insertion loss during the Bellcore TR-NWT-001209 temperature cycle test is shown for two groups of couplers. Those couplers packaged with the GlasSolderTM process generally showed a smaller change and were more consistent than those couplers packaged only with epoxy. A second group of five couplers, packaged in a similar manner using only an epoxy, was also included in the test for comparison. Both sets of couplers were monitored periodically during the test. The typical performance of the exposed couplers during the damp heat test is shown in figure 2. After approximately 600 hours of test time, the humidity began to weaken and breakdown the epoxy, causing the insertion loss of the epoxy-packaged couplers to change substantially. In comparison, no significant degradation in the performance of the GlasSolderTM couplers was observed. Further more, while the behavior of the epoxy-packaged couplers was completely unpredictable, all of the couplers assembled using the glass solder behaved in the manner shown in figure 2. It is interesting to note that the insertion loss of the coupled leg of the GlasSolderTM couplers slowly decreased during the test, indicating a slight increase in coupling ratio. We believe that this was due to the diffusion of moisture into the glass fibers of the coupling region which was completely unprotected from the hot, humid environment. To evaluate this hypothesis, the above test was continued beyond 2,500 hours without the high humidity. As expected, their insertion loss returned to its original value within a short time. Qualification Tests A number of couplers were built for qualification testing in accordance with the test criteria specified in Bellcore TR-NWT-001209. Three different groups of couplers were assembled: dual window (1310/1550 nm) 50/50 couplers in the GlasSolder TM package, 10% tap couplers in the GlasSolder TM package and 10% tap couplers in an epoxy package. In order to compare the performance of the two packaging processes the 10% tap couplers were built using both the epoxy and glass solder. Despite the fact that these qualification tests are relatively short term, the GlasSolderTM packaged tap couplers showed superior performance in every test category. As one example, figure 3 shows the results obtained for the temperature cycle test for both the epoxy package and the GlasSolder TM package. Not only were the changes with the GlasSolderTM package in general, smaller, but the results for the entire lot showed greater consistency. Figure 4. The average peak to peak change in the insertion loss during the entire sequence of Bellcore TR-NWT-001209 tests are shown for two groups of couplers, i.e., 10% taps with and without the glass solder. While both groups performed well, a direct comparison between the two groups of tap couplers shows that the glass solder package out performed the epoxy package in every category. The average peak-to-peak change in the insertion loss for the 10% tap couplers, with and without glass solder, for all of the tests are shown in figure 4. While both package types performed well, the GlasSolderTM package clearly outperformed the epoxy package.Gould’s components are available now with the GlasSolder TM package in prototype quantities. To find out more about the GlasSolderTM package call a Gould sales engineer at 1-800-544-6853. Figure 5. Peak to peak insertion loss for a group of 50/50 wavelength independent couplers packaged with the GlasSolder TM technique which underwent the Bellcore TR-NWT-001209 test criteria.
webster dictionary p.13