Breaking away from the norm

In an increasingly connected world, networks must not only become more reliable and powerful, but installation and maintenance need to become ever faster and simpler.

Today’s economic market conditions require a tangible increase in efficiency. Planners, installers and users expect pragmatic, secure and easily implementable cabling solutions for a variety of industries and applications, including data centres, office buildings, airports, hospitals, malls, sports stadiums and public broadband networks. In essence, the more demanding the data transmission, the simpler the cabling solution must be.

The installation practice we employ today, from design through to commissioning, is derived from two decades of trial and error, gradual and progressive standardisation, and advancements in technology. The ICT design is now so ‘standardised’, in both practice and documentation, that a once skill-laden profession is little more than a cut and paste industry. We have eliminated most of the technical thought by repeatedly using methods that were once cutting edge.

Get your priorities straight

According to the Top 10 CIO Technology Priorities survey conducted by Gartner, networking, voice and data communications was a number 10 priority. Ironically, the nine preceding priorities were related to applications and systems that depend on networking, voice and data communications systems. So how have we found ourselves in this position?

Most of the structured cabling designs created today are no different to those created in the late 90s. The components and their capability may have notched up a bit since then, but the topology hasn’t. Even the active equipment in the switched environment is, for all intents and purposes, stagnant. If you take a look at other industries, such as lighting, AV, security and HVAC, you would be hard pressed to find the absolute resistance to change that we find in the ICT industry.

Despite ICT being the lifeblood of just about every industry and business, the communication network — as proven by the Gartner survey — is an afterthought, overshadowed by everything else. The communication network is often the first to be marginalised and squeezed on price. How important can it really be — as the cost of ICT is just around 7% of total CAPEX. While it is necessary to provide employees with an aesthetically pleasing, hi-tech and comfortable work environment, one could argue that the system that underpins the entire network is equally, if not more, important. It would be short-sighted and overly optimistic to expect the office layout to be altered to accommodate cabling changes and requirements.

Telecommunications spaces are often the first to take a hit in favour of a new requirement, such as a larger break area, restroom or meeting room. In the current day and age, ICT cabling should be designed and installed in a way that it requires less space and is fit for purpose. It should support the business, not hinder it. However, the current design processes and standards do not support such radical thinking.

Conformance

In any cabling specification, the words ‘must conform’ jump out of the page. But what does must conform mean? The Conformance section (Clause 4) in the Generic Cabling for Customer Premises (AS/NZS 3080) offers three ‘alternative’ ways to comply. They are: channel performance by design; channel performance which contains a CP link or other components as detailed in the Standard; and the reference model. The reference model is the source of some of those numbers we all live by: 2, 5, 15 and 90 m, plus a few others. So why are we confined by these numbers? The text above the table that contains those values in the Standards document states that the table (Table 32 of AS/NZS 3080:2013) contains length assumptions of the mathematical model used to validate channel performance using components listed within the same Standard.

“They do not represent absolute restrictions on the implementation of channels and permanent links, but may be used for guidance in reference implementations,” states the document. Yet these exact figures are adhered to vehemently in both design and implementation, even though length is not a Pass/Fail criterion (not including TIA).

Beyond the standard’s length restrictions

The standard’s philosophy is to have guidelines that cater for everyday installations and cover most technologies being used in a generic way. However, there are installation requirements that sometimes just fall out of the scope of the standard. One of those situations is when you need to connect a device to the network which has a connection length longer than 100 m.

So how to conform in this new age of discovery? Well, by using Clause 4 we have another option — channel performance by design. This simply says that if a link is installed and tested as a channel and it passes, then it conforms. Of course, the issue then becomes one of risk — there needs to be a way for a designer to know the extreme limits of the channel before it will no longer function as intended. The main testing issues when increasing the length are: insertion loss (attenuation of cable and connectors), delay skew and propagation delay. If any one of those fails, it will affect the performance of the transmission protocol, and in turn the network and user experience.

But why would there be a need to go over the limit? To deliberately design anything to the outer reaches of its capacity is not a wise move, but what if things change? For example, a permanent link that was once suitable is now on the very edge of current design practices because of the factors outside of our control. The floor distributor (FD) cannot be moved, and the idea of a second FD being introduced to provide for a handful of links would get laughed straight out at a project meeting. Fibre could be used, as long as there is space for additional power and media converters. But the rack layouts will have to be altered to allow for changes. Designers/installers can easily address the over-length issue by using quality components and cables and allowing for an extended channel. Interestingly, by using those same components the opposite is also true. The 15 m PL rule can become 2 or 5 m, thereby removing the surplus cable looms we are all used to seeing piled on top of pathways. These are explained in R&M’s Installation and Testing Guidelines. Something else worth noting within that same document is the one connector link.

Current design practice is to allow for a telecommunications outlet (TO) and work area cord at each WAP or camera location (TE in the above drawing), which limits the PL to 90 m under current design practice. The use of the increasingly popular field installable RJ45 plug allows for a couple of things: reduced on-site labour due to the removal of the need for a TO and more flexibility in regards to the TE’s final resting place. The extended links should be regarded as a ‘get out of jail free’ card. It is anticipated that in any given design, only a small percentage will need to use this solution, but the cost implications should be immediately evident. The short links and one connector method can be used in any design, provided you’re willing to embrace change and revisit existing practices.

Summary

As expectations grow, budgets shrink and opinions on ICT cabling remain indifferent, it is up to the ICT industry professionals to force the industry into the limelight by adapting processes and designing truly integrated, flexible and cost-effective networks. Despite our heavy reliance on ICT and the crucial role it plays in our lives, its value is significantly underestimated.

Image credit: ©iStockphoto.com/Photoslash