There’s strength in members
Many electricians have their first encounters with fiber optic cabling when asked to install the cable for others who terminate and test it. It usually doesn’t take long for the electrician to realize that he or she is losing a valuable source of revenue, so they too learn how to install fiber optics.
But along the way, they may find many trainers and installers spend little time on the actual process of installation. We tend to think that fiber optic cable, which is much stronger than Category 5e/6 for instance, needs little care in installation. The weakest fiber optic cable can withstand pulling tension of several hundred pounds, many times the 25-pound limit of UTP cabling.
But fiber optic cable can still be damaged during installation, causing temporary degradation if you are lucky or permanent damage if you’re not. Let’s look at some of the problems encountered when fiber is not handled properly.
The first thing to remember is that—if you are pulling fiber optic cable—it must be pulled only with the strength members. The silky, golden strands of aramid fiber used in most cables are unusual in that they are not only strong, but they do not stretch. (You may be more familiar with the duPont trade name for aramid fiber, Kevlar.) By winding them into cable with about 2 percent excess length, you can pull the cable with them without harming the fibers.
Before pulling the cable, you need to strip off the jacket about 6 inches, cut off the protruding fibers and attach a swivel pulling eye. The swivel eye allows the aramid fiber to unwind when it is pulled without twisting the cable. You cut off the fibers to insure no tension is placed on the fibers themselves, which can cause irreparable damage.
Only a few types of fiber optic cables can be pulled by the jacket itself, and those are mostly outdoor cables with armor or double jackets with strength members captured between the jackets. If you pull an indoor cable by the jacket, you will stretch the jacket, and when it is released from tension, it will shrink back and bunch up fiber and Kevlar, causing high stress losses in the fibers.
Even if you are pulling a cable by the strength members in conduit, you can stretch the jacket by friction unless you use proper lubrication. And if you pull through lots of bends, you can create tension at every bend, which can cause loss. Consider the needs for lubrication whenever pulling in conduit.
Tension is not the only problem for installing cable. Fiber is sensitive to many stresses, so if you have a lightweight fiber optic cable in a cable tray under a bundle of heavy copper cables, the weight of the copper cables can cause stress loss in the fibers. Fiber optic cable should always be on top or suspended underneath when installed in trays to prevent this problem. Alternatively, you can place the fiber optic cable in innerduct, which not only protects the fiber but will speed installation.
And sometimes, even neatness can cause problems. It’s common to bundle cables and tie them together with a cable tie. But the guns used with cable ties are too strong for fiber optic cable. The pressure of a cable tie on a bunch of fiber optic cables can cause significant loss. It’s better to use hook-and-loop fastener cable straps. If you do use cable ties, put them on by hand, tighten only enough for neatness and leave them loose enough so the tie can easily be rotated on the bundle of cable.
How do you know if you have overstressed a cable? You will find out in testing. Stress causes higher loss at longer wavelengths. So if you calculate the power budget for a multimode cable at 850nm and 1,300nm, the connector losses should be about the same, but the fiber loss should be much higher at 850nm (about 3 dB/km @ 850nm vs. 1 dB/km at 1,300nm). If the losses are about right at 850nm when you test with a optical loss test set, but the loss at 1,300nm is higher than expected, the likely culprit is stress-induced losses somewhere along the pathway used for the fiber optic cable.
If you suspect a stress problem, how do you find it? The most obvious solution is to use an OTDR, which can give you a picture of the losses along the length of the cable. But OTDRs are not very good at resolving close events or cables that are short, typical conditions in premises installations. A visual fault locator (VFL, a red laser test source) can be used on some cables if the jacket is transparent to the red light of the VFL, but few premises multimode cables are.
Often the best—sometimes the only—way to find the problem is to go looking for it. Look for cable ties, tight bends or too many of them, or fiber optic cables buried under bundles of other cables. There is where you are likely to find your problem. EC
HAYES is a VDV writer and trainer and the president of The Fiber Optic Association. Find him at www.JimHayes.com.