fibre splicing

• Cut the fiber and trim back the woven Kevlar sleeve on both ends with a good quality scissor.
• Slide on a piece of shrink tube over the Kevlar sleeve.
• Slide on a splice protector over one of the fiber ends.
• Cleave the fiber ends to get a fresh, clean cut end on both fibers with a specialized cleaver.
• Strip off about 3/8 in of the fibers coating with a fiber stripper
• Inspect the ends with a fiberscope. Ensure a PERFECT cleaved surface. Repeat procedure if not.
• Place both ends in a fusion splicer that had been adjusted relative to the many parameters involved (temp.,lossy splice (offset), etc.) as required.
• Punch the button. Have a sip of a lovely beverage.
• Inspect the fused joint visually and photonically with a signal meter for amount of signal loss.
• If acceptable, slide splice protector over fused joint and process (shrink) in fusion splicer.
• Slide shrink tube to cover both ends of the Kevlar sleeve and shrink.
• Retest for acceptable signal loss.

That's how I used to build fiberoptic amplifiers in a manufacturing setting for seven months before the bottom dropped out of the industry in 2001. There are also gel type super-glue methods for in field splicing. I have not used those. There may be others since 2001.

You can understand why "fiber to the home" (FTH) has taken so long ("fiber to the pole" (FTP) has been around since the early 90"s). Mostly a durability issue due to the rigors of in home use (pets, vacuum cleaners, foot traffic, etc.) and the difficulty of making clean connections and repairs. Besides, it really won't increase speed in the computer until they build computers that can transition from optical input signal to electrical signal at the same speed. At present, that isn't the case or necessary really for home and general use. More of an issue of bragging rights. Industrial, technical, financial and other high information transfer industries have the need. It's like a 36" water main feeding a 1/4" faucet.