Startup

Startup: It’s Not What You Think

This is how fibre optic cables start their life, rather unremarkable glass tubes. If you’re not familiar with the process, it’s easy to assume that fibre optics are just a simple thin strand of glass, but a single strand of cable is actually two strands. The actual fibre that transmits light is almost too small to see with the naked eye. The thin glass fibre that you see is actually the cladding, the part of the glass that keeps the light in. That part of the cable starts it’s life as a glass tube only a few feet long. Vaporized silica compounds are pumped inside heated tubes and let cool to form soot inside the tube, which is then re-heated so the soot also turns to glass. The tubes are eventually heated until they collapse in on themselves, and the whole thing can then be drawn out in a single thin strand. The former soot in the middle forms the actual core of the fibre which transmits light, wrapped inside the original glass from the raw tube. A tube only a few feet long can produce hundreds of yards of finished fibre optic cable.

  • kestrel131

    I did my graduate work in Materials Engineering, and one of my friends produced specialty fibers, where the core is doped with Rare Earth elements (Erbium, Praseodymium, etc). The Erbium compounds were deposited inside the pre-form at the same time as the silica core, and they can control the composition of the glass , first to create the refractive index difference between the core and cladding to form the wave guide, and to add the active elements.

    The Erbium doped fiber would be pumped with a 980 nm wavelength light source (usually an LED or laser Diode), and then emit light at 1.55 microns wavelength, the same wavelength that the long-haul signals are transmitted at. This means that as the optical signal is traveling across the ocean floor, the signal is optically amplified, prior to the Erbium Doped Fiber Amplifier (EDFA) the optical signal would have to be converted to an electrical signal, amplified and then re launched as an optical signal. That required a lot of electricity to power the amp deep in the ocean. Now they need just enough power to light up a few LEDs every 1000 kilometers or so.

    • B72

      So how does this amplifier thing work? You send a signal into one end of the doped fiber, shine a LED on the side, and somehow the wavelengths interact to send a stronger signal out the other end?

      • kestrel131

        The doped portion of the fiber will have 2 inputs, 1 from the main transmission line, with the signal, and 1 to attach the pumping light source (an LED or Laser Diode). This is done so all the light is confined to the core of the fiber. Trying to pump by side illumination is very inefficient. The Pump light (980 nm) excites electrons in the erbium out of the valence band to an excited state. Without getting too in depth, there is a non-radiative decay to a second slightly lower energy state, where you can maintain a state called population inversion (more electrons up there, than down in the valence band). While population inversion is maintained, when the signal being transmitted at 1550 nm wavelength passes through the erbium doped region, the excited electrons decay back to the valence band emitting a photon. Importantly, this photon is 1550 nm in wavelength just like the signal, is coherent and in phase with the signal, so it constructively amplifies the original 1550 nm signal.

        To maximize efficiency of the pumping light source, the active fiber (with Erbium in it) has a Bragg grating at either end, so the 1550 nm light can pass through, but the 980 nm light is reflected and is confined to the area where the Erbium is present.
        A similar technique is used to create fiber lasers.

        One of the coolest things is when the erbium is being introduced to the preform. The preform is spinning on a special lathe with chucks that allow the gasses to flow in and out of either end of the preform, while it is spinning and being heated by a Hydrogen-Oxygen torch. The torch helps the chemical reactions to form the soot described in the original article…but it also excites the erbium, so the entire 5 fool long tube is glowing at a brilliant green, It looks like kryptonite glowing. and once it is cool, the glass is transparent. Praseodymium on the other hand, keeps a pretty neat purple color.

        • B72

          That is some pretty cool technology. Thanks for sharing with us.This is why I come here.

    • The Professor

      How marvelous. I love reading about things like that. Someone did some finely crafted thinking there, and it's wonderful to see.

  • Alff

    So what is it about the heating and drawing process that imparts flexibility on what I assume is rigid glass?

    • kestrel131

      First, glass has a very high intrinsic strength due to the disordered network of Silicon-Oxygen bonds. Glass traditionally fails at defects in the surface (think about cutting glass with a diamond scribe – you nick the glass where you want it to break). When a pre-form is drawn into fiber the viscous glass is pulled to a desired diameter, and cooled as the fiber descends the draw tower, by helium cooling. The cooled fiber is then quickly drawn through a cup of UV curing resin that prevents exposure to moisture and humidity. Water will form small pits in a glass surface and cause breakage. Once this resin is on, the glass is protected. and when you bend the fiber and pass a critical radius of curvature you will break the fiber at the intrinsic strength of the atomic bonds. The draw towers have local environments near clean-room quality, so they will only have one defect per 100Km of fiber

      • Alff

        Thanks!

      • The Professor

        I've watched the drawing process before, and it's fascinating. A lot of good engineering was involved in developing today's glass fiber optics. The part where the silica 'soot' is placed in the tube before heating and drawing is genius. I wonder who thought of that?

        • kestrel131

          the Soot is actually formed inside the glass tube. The glass precursors are gas-phase compounds that flow into the tube and react once exposed to the heat of the hydrogen-oxygen torch. and the torch rasters the length of the tube as is spins in a lathe, to ensure the soot coats the inside uniformly

-->