Fiber optics is not a new technology, having been available for communications for more than two decades now, but fiber still seems to have the “rocket science” aura. (In the interest of full disclosure, I once was a rocket scientist. I trained as a physicist and astronomer in the 1960s, pioneered the use of computers to analyze astronomical data and did simultaneous observations with NASA on rocket flights.) Unfortunately, some people—including instructors teaching installers—seem to feel fiber optics needs special treatment, and do not provide the simple skills development that is more appropriate.

As part of my role working with The Fiber Optic Association (FOA), I review curriculum from many training schools seeking FOA approval, and I see some programs that include materials irrelevant to most technicians and installers. Within the FOA, we can call on some of the best trainers in the business, and we have created standards for training technicians for fiber optic network design, installation, testing and troubleshooting. These are posted on the FOA Web site at This group of experts strongly disagrees with the rocket science approach.

For example, many fiber optic courses cover classical optics before ever talking about fiber optics itself; some taking a full day to explain how refraction of light makes lenses work or how reflection occurs at material boundaries. Some instructors feel obligated to tell you all about lasers, LEDs and photodetectors and how to design transmitters, receivers and data links. By noontime, most techs in these courses are tuned-out (or snoring away) because they don’t see the relevance of the materials or figure it’s way over their heads technically.

Training programs for technicians should instead focus on fiber optic basics, jargon, components, hardware and their installation. Practical information, directly related to their work, is what techs need and want. Most programs would be better if theory was replaced with cable handling. Too many courses focus on termination and splicing, but never get into cable pulling and preparation. It is impossible to be successful as an installer without fully understanding cables.

You need to know how to identify all types of cables. Premises cables usually are tight buffer simplex, zipcord, distribution or breakout, each of which has particular applications and requirements for pulling and stripping. Outside plant cables are generally loose tube, but may have single or ribbon fibers, dry or gel water blocking and even metallic armor. A good training course will introduce the student to all these cable types, ensuring they will not be surprised by what they find in the field.

It is considered normal to teach termination and splicing, but even that needs to be flexible, depending on the students. Outside plant techs work mainly with single-mode fiber and never attach connectors directly to the fiber; instead they splice preterminated pigtails onto the fibers. Premises installers practically never splice fibers, but they install a lot of connectors directly on the fiber, mostly using adhesive/polish or prepolished/splice methods. Certainly, every tech needs to know about all termination processes, but a fiber optic course should specialize so the OSP techs get more splicing practice and the premises techs learn to polish better.

I used to believe that premises techs did not need OTDR training, but now OTDR manufacturers are pushing their use in premises networks. Unfortunately, the limitations of OTDRs in premises applications can cause extreme problems for those who are not aware of their limitations. So schools (and my Electrical ContraCtor columns) now must provide that essential information to every tech.

Not everything you don’t need to know deals with training. For example, you don’t always need instruments that read loss to two or three decimal places. OTDR manufacturers persist in providing readouts of loss to 0.001 dB, about 50 times the uncertainty of the measurement. Power meters and loss test sets usually read to 0.01 dB, but the uncertainty of multimode premises cable plant loss measurement is generally worse than 0.25 dB. Readings to 0.1 dB resolution is usually plenty, and your training needs to make sure you know why.

Then there are cabling standards. The truth about standards is most were never written for installers and end-users. They really are written for component manufacturers as “mutually agreed upon specifications for product development.” Often, it’s not worth paying a small fortune to get a copy and spending the time trying to translate them into comprehensible English. They were written for the manufacturers, and every one of them will gladly give you a copy of their translation of the standards, usually in the back of their catalog. EC

HAYES is a VDV writer and trainer and the president of The Fiber Optic Association. Find him at