top of page
  • andrewwolinski

The Evolution of the Physical Fiber Plant and the Future of Monetizing that Network

Are mechanical splices and mechanical connectors this generation's load coils?

Service providers try to build networks as quickly as possible while simultaneously fighting to find competent contractors to build those networks. I am coming across more mechanical splices or mechanical connectors in fiber optic networks, particularly in the areas in and around customer locations. The push for mechanical splices as a method used by installation technicians is typically a conscious choice made by a service provider hoping to improve the speed and cost of an individual customer installation. This choice can lead to a long-term problem once those mechanical splices fail due to poor installation, exposure to the elements, or an inability to handle next-generation network services.

Poor installation and exposure to the elements are two variables that only some networks can avoid. Methods like using pre-connectorized drops, fusion splicing pigtails, or fusion splicing connectors are three ways to give yourself a better opportunity not to have to reroll a truck back to the customer later because those are more robust ways to connect fiber to help with your fiber loss budget and the other critical physical layer indicators. A customer link passes and gets excellent service now. Still, the evolution of the network will mean having to replace these mechanical splices because of the insertion loss and, most critically, the reflectance associated with them.

In the same way that load coils were a great way to boost telephone service, they eventually became a massive issue once DSL technology became the network's main priority. Mechanical splices, incredibly close to the ONT or the splitter cabinet, create much more loss and reflectance on the network. When next-generation networks with more wavelengths and stricter network construction standards, all those customer drops with mechanical splices must be replaced, for any network equipment manufacturer will verify their link.

Installing a mechanical splice is the same step (not to mention the most skilled labor steps) as a fusion splice. But the fusion splice removes most of the loss and all the reflection that a mechanical splice does. Doing the job right the first time is the most important thing anyone can do. It also means you will need fewer workers and stress more money-making tasks like new customer installations and expanding their networks.

How do mechanical splices affect a network’s performance?

The best place to start discussing how they affect a network's performance is to understand why SC/APC (Green) connectors are internationally accepted as the connector of choice in any FTTH application. Because of the angle, a high-quality SC/APC connector reduces the amount of light reflected at the connector, leading to less insertion loss and less reflectance than a standard SC/UPC connector (Blue). The higher the reflectance, the more light reflected back, the worse the connection. The table below shows the typical difference between events.

Comparing Loss and Reflection of Common Events Seen on the Network

The table above shows that the reflectance of a mechanical splice typically is double that of an SC/APC connector and the equivalent of that same connector just going to open air. Regarding insertion loss, a mechanical splice can be like a connect, but a good fusion splice has 5x or sometimes even less insertion loss.

As ONTs and other network equipment produce more power to get even higher speeds, this reflectance will begin to hinder your network’s performance because of the fiber link’s inefficiencies due to the mechanical splice. You are potentially burning out optics, splitters, and ONT, which becomes more expensive than just building the network correctly the first time and training everyone involved on your company's installation standards.

Do these splices put your network at a long-term competitive disadvantage because of potential inefficiencies?

The concise answer is yes, it does. As more people work from home on VPN, trade stocks, sports bet, stream live TV, remote education, and medical care, and add even more connected smart devices on the network. The data that a network is monetized will be more than just your customers paying their monthly bills. Companies will start prioritizing and incentivizing service providers who can prove them on all the interest layers to get their information to the customer first. A whole market will evolve in ONTs, Access Points, and AI that might seem decades away but is developing faster than most people are prepared for.

But one way you can be ready for this evolution is to take out your mechanical splices, create birth certificates for all your customers at levels 1 through 3, and keep pushing your network equipment manufacturers onto new technologies. No matter the size of your network, there is much value in the data passing through it. The only way that data gets there is by ensuring layer one is built and maintained in a way that gives you the best chance to be successful, and mechanical splices in any form significantly lessen a network's long-term success.

29 views0 comments

Recent Posts

See All


bottom of page