Fiber Optic Services

All right; just what is a fiber splicer and why should you even be concerned with them. In many cases, the fiber splicer is an "in house" craft and if you need a splicer you will know what he is and what he does. If you don't know, you unlikely have need of one.  Unless you're just naturally curious...

I am a craftsman; and like most true craftsmen take a great deal of pride in my work. I want the world to see what I do and the elegant appearance of a job properly done.

Splicing, glass or copper, requires meticulous attention to detail and splicers are often required to do repetitive, boring work for hours on end. The photos presented here represent two full days work. The personality suited to this work tends to appear distant and sometimes decidedly unfriendly to most people.

One of recent history's more interesting people was Henry Kaiser. As an industrialist, he devised methods to accomplish what tradition said couldn't be done. I like to quote him when he says:

"When your work speaks for itself, don't interrupt."

On that note, I will explain what is taking place and stop interrupting.

At the technical end, the fibers most often used in communications systems are 125 micron; "single mode" being the most common for long distance runs. Well, that certainly sounds small. So, just how small is small? A micron is 0.001 milli-meter. That's 1/1000 mm; 1/1,000,000 meter. With a meter at a little over 3 feet, a milli-meter is about the thickness of a dime, ~0.040 inch.

So, 100 microns works out to be a tenth of a milli-meter, about 0.004 inch. That means 125 microns is about 0.005 inch. The glass fiber is slightly larger than the hair on my head. Slightly smaller than the hair in my mustasche.

In "single mode" fiber, the core is nominally 9 microns, about 1/12 the size of the fiber. With "multi mode" fiber, the core is much larger. Often as large as 2/3 the overall diameter.

The outside of the glass strand is covered with a coloured plastic coating that serves several purposes. For my use, the colour coding is how I know which fiber is which during the terminating process. It also helps to offset the brittle fragility of the glass fiber. Helps, is about it. I have seen fiber broken inside the plastic coating with no external indication.

The OTDR said it was broken. It looked good in the tray. The splice had light on both sides. How to find it? Using a test light in the visible range, such a break will appear as the "star-burst" pattern that I use as a logo.

Curiously, the glass is incredibly strong in a linear pull. On the order of several hundred kilos. However, bent over a sharp edge, it can only withstand a very few grams.  It's also sensitive to vibration and shock loads.  When a cable is damaged or cut, I prefer to cut back several meters, when the slack is available.  

Most often, office computers and other local area networks are interconnected with multi-mode fiber. The size of the core is so large as to be forgiving of some mis-alignment. That makes it easier to apply mechanical connectors to the larger fiber. Many times, the in-house " I-T " staff will apply such mechanical connectors. Multi-mode glass is also more versatile, signals can readily travel both directions concurrently.

There is a caveat, though, as always. "There ain't no free lunch", as Goodman Long says. Multi-mode is "lossy". It is not suited for long distance transmissions. Most often chosen for short runs of a few hundred meters, it is, as stated, easily terminated, and bi-directional.

Actually, the current state of the art is such that "single mode" fiber can carry bi-directional signals. However, the electronics necessary to accomplish this are prohibitively costly. Such technology is usually reserved for existing installations where there are no spare fibers and the cost of new cable is too great.

Single-mode fiber, on the other hand, has extremely low loss. Unless specified differently by the customer, I will break and re-do any splice over 0.05 db. The norm is on the order of 0.02-0.01db.  Many telephony applications specify 0.15db loss for the entire circuit. Often 30 or more kilometers in length, such circuits may contain numerous splices. The losses of multi-mode fiber in such an application are virtually unthinkable.

A link into a vendor site giving excellent detail of fiber strand optical characteristics is at:

At this point, I need to bring up a very important matter;

Never, ever, for any reason, look down the end of a fiber or at the end of a connector!

During testing, we use a visible light source to locate faults. In normal operation, the light is outside the visible spectrum. You cannot see it. But, it is strong enough to cause blindness. The only safe fiber is when both ends are in your hand.

Down the inside of the glass fiber is a portion called the "core". The core is of very high purity glass and is where the light signal is actually transmitted. The outer portion, around the core, is of lower purity and exists primarily to reinforce or strengthen the fiber. The boundary layer between the core and the body is what keeps the light contained and prevents scatter losses.

During the splicing process, the core is what is actually aligned. Light is injected into one fiber and measured in the other. Alignment is adjusted until the light is at it's greatest intensity. The splice is then made up.

Normally, a single mode splice is made by "fusion". The glass is heated until it is plastic, then pressed together fusing the two ends into one contiguous strand.  The machine that does this is a rather precision device.  I use a Fujikura 50S.  Say!, do I get paid for plugging the manufacturer?

There are mechanical splices suitable for use with single-mode glass.  And fusion splicing is becoming more common with multi-mode.  Mechanical splices on single mode glass are usually reserved for restorations where time is of the essence.  Although, a good production splicer can burn glass faster than a tech can make up mechanical splices.   And, mechanical terminations don't have the long term reliability of a fusion splice.  Since multi-mode is most often used in protected environments, mechanical splices are not subjected to the harsh conditions that will degrade them over time.

As the technology matures, it is becoming more common to see job specs calling for "factory terminated" connections.  This is often accomplished by splicing on factory terminated "pigtails" at each end of the cable.  Splice losses are insignificant compared to connector losses;  this practice can significantly reduce field time and costs while still providing maximum connector reliability.

There are also other types of fiber.  They are really an exception rather than a rule.  I don't discuss them here.  Working with them sends my vocabulary into the gutter, nay, down the storm drain.  Words I prefer to forget I knew.  And usually at a high volume.

Preparation of the cable into the splice enclosure is often more time consuming than the actual splicing.  It is also where the physical work is.  The "bull" work, as is said in some crafts. 
The splice case is water proof, air tight, and relatively shock proof. They are made for direct submersion in water, direct burial in earth, and exposure to the elements on pole lines. 

For "High Reliability" applications, they are sometimes potted with a waterproof gel.  At first thought, it would seem that moisture shouldn't effect the fiber strand. But, when water freezes and forms a crystaline structure, there is some motion;  very little but enough to break a fiber.

Since fiber falls under the electrical codes and most cables have a metallic sheath, ground bonding on long runs is also an issue. The can must withstand lightning running the sheath to ground from 20 miles away.

Photo 1 is of an "end plate" where the (four) cables enter the closure. Cables can be brought in both ends of the closure. However, this practice is avoided as it makes the fibers cumbersome to dress and the case difficult to mount on a strand or bulky to bury.

Close examination will reveal the sheath bonding clamps attached to the clamping brackets. The brackets are in turn attached to metal studs that penetrate to the outside of the end plate.  The studs are 1/4 inch diameter.  Puts a perspective to the size of the tubes. Electrical bonding jumpers can be seen on the end plate in Photo 5

Photo 2 is of the last group of fibers just before splicing. This bundle is a breakout into a secondary cable. Close examination will show two groups of six (by colour). These will be spliced to a single group of twelve, out of the photo to the right. The big gizmo with the screen is the "splicer".

This is a Siecor / Corning machine.  I get my best speed with the Fujikura.  Many communications companies require there be at least two splicing machines available during a window.   I wanted the time on the Seicor as practice.

The smaller contraption is a "cleaver". The glass cannot just be broken, it would be splintery.The cleaver provides a clean, sharp, square break on the end of the fiber. The value of the equipment in this photo is somewhat higher than my house and outbuildings.  That's why splicers don't come cheap.

Photo 3 shows a completed "tray", ready for mounting in the enclosure. This is the target appearance of a completed tray. The coin is a US dime to indicate the size of the splice tubes. Notice that all the bends are of fairly large radius. Also that there is sufficient excess length to facilitate both handling and future changes. Not all splices are straight through.

Depending on the routing of the circuits, there are nominally 36 splices to a tray. I try for 48 when I can, like this one; sometimes it's only 24. Depends on the situation.

Photo 4 shows two of the trays mounted with a third in the background. The large tray on the bottom is for storage of excess length. Fibers are bundled together as a group. Usually six or twelve, sometimes 2, 4, or 8. Each bundle is then surrounded by a coloured thread, a coloured tube or taped into a lettered ribbon. There is some of each in this enclosure.

If a single fiber should break too close to the tube for splicing, the entire tube must be opened up further back. There is about three feet of slack from each cable wrapped in the bottom tray. Just in case.... Simply leaning on your elbow with a fiber underneath can break it. We won't mention, of course, dropped tools.

Photo 5 and 6 show enclosures (2) closed and ready to go in the ground. Note, again, the bonding jumpers connected to the through studs. The can on the left must be re-entered and two more cables added in a month or so. I placed waxed paper in the upper joint to facilitate that re-entry. The rubber tape used in these end plates will fuse to itself, under pressure, in about two weeks. Once it is fused, it cannot be re-entered. It must be replaced. Takes about three hours to make that repair. I snarfed the waxed paper from a diner. It's the simple things that make life easier.

The can looks pretty stout.  It is.  I could drive my truck over it without harming it. Might cut up my tires though.  But, it won't stand up to a trencher.  Trust me.  Times like this is when I get to put meat on the table for supper.  Service outages seldom wait for normal working hours.


Or don't; restoration pays quite well.

There were 144 splices in this case. I get paid on piece-work, per splice..... you do the math.

Fiber Optic Services

Now, for all you mud-hogs out there that like playing with trucks.... I have to, like it or not.  So, why is it that splicers have such fancy, gussied up 4X4 pickups?  In this case, the cable plowing crew left me the keys for the back-hoe.  I didn't need it. Managed to get out of this hole un-assisted.  Didn't even have to break out the winch.  The ground under the trailer is level.  And, keep in mind, my combined weight here is well over 12,000 pounds.  I have been dug in so deep the wrecker slid sideways when he took up slack.  Had to get a second one to hold the first.  The trailer has to go to the work; the cables can't come to me.

A Brief History
of Communications Splicing

Albeit somewhat tongue in cheek

The earliest known splice, as developed by the Phoenecians about a zillion years before Christ. This splice was devised for use with a primitive form of communications that still is seen, although rarely, today. A metallic vessel of convenient size is attached to each end of a length of cordage and pulled taut. Should the cordage be of insufficient length, a splice is in order. The splice shown here was quite common for such line extensions.

Then, of course, came the Cell Phone.  Most everyone has one, few will admit to where . . .
(With due regard to the artist, of course; this is a copyrighted work)

A recent development, specifically for "Multi-Mode" LAN systems, is the "Fiber Nut". In appearance, it closely resembles a device widely used on copper systems. The most noticable difference is the presence of a small parabolic mirror inside the apex. This mirror purportedly passes light at virtually unity.

The intent was to reduce the service calls that come about from rough handling. Reports, however, indicate this device to be somewhat less than reliable.

My eMail address is:

Bill Hudson
P O Box 101832
Irondale, Alabama, 35210

Website Builder