In the early 1980’s Robert Metcalfe posited that the value of a communications network increased with the square of the number of users in that network. Initially this ‘Metcalfe’s Law’ was used to sell more Ethernet network cards but it more recently was recognised to see how this network effect could be seen in telecoms networks, the internet, social media and even cryptocurrencies.

The basic premise? More users = more Value

How could this network effect be attributed to energy consumption for telecom towers? Well sharing of radio towers and even radio access equipment is not new. TowerCo’s business models are built on the ability to share infrastructure with multiple tenants.

But More users = more value = also more power required!

What is pushing up power at bad grid cell sites in Africa and Asia?

Energy efficiencies can be easier to attain with multiple tenants or tower sharing. Each tenant may require between 1kW-3kW depending on what technologies it’s using. The direction of travel of site power requirements is going only in one direction now…. Up. So, should sharing of power generation be considered outside the boundary of the telecom site itself?

We already know that there is no magic bullet to solve the energy challenges of balancing energy costs, reducing dependency on diesel and increasing use of renewable energy and energy storage. Using 100% renewable energy, if there is enough space on site, does NOT equal 100% availability, nor reliability, nor predictability compared to a generator. This challenge is especially acute in Africa with load shedding, aggressive rural network rollouts with no grid connections, and logistics challenges, not to mention asset theft (fuel and batteries alike).

According to Statista the share of renewables in primary energy consumption has only seen marginal change in the past decades. Between 2007 and 2018, figures increased from 12 percent to just 13.5 percent, far from the share needed to reach 2050 zero carbon goals as aspired to by many governments. The OPEC forecasts that daily renewable energy demand would grow to 77 million barrels of oil equivalent by 2045, still far below the demand for oil and natural gas, which is to see the greatest growth.

At what level of grid quality does increased power load start to stretch currently dimensioned sites? Is 20 hours a day power enough?

By 2030 65% of every car driven in Europe will be electric, massive change in computing power needed for 5G, edge computing, EV charging, rural telecom rollouts across Africa, Asia and even Europe are all coming at the same time – perfect storm of energy demand from all sectors just at a time where media attention is on carbon emissions reduction. This time is different since the electricity grid wasn’t build, designed nor received the investment to see this massive transfer of energy from fossil fuels. Grid quality in Africa is already stretched to breaking point in many markets before these new pressures have emerged in volume. So, it will not be telecom networks on their own that will stretch or fracture grid quality. So distributed power generation is going to be a must have for sectors beyond just telecom.

Are cell sites which rely on solar and hybrid systems in danger of overtopping their energy supply with expanding 4G overlay?

The direction of travel of multitenant sites, shared networks, network densification is at odds with using 100% renewable energy in the form of solar PV, wind, and energy storage. So increased pressure on radio/BTS vendors is required to invest not just in broadband capacity and quality of service but squeeze every percentage efficiency out of the BTS and transmission equipment too. Higher temperature tolerances lead to eliminating the need for air-conditioning systems in some markets, but modular power systems are key. Starting with power modules that can grow in lock step with actual not predicted or forecast power consumption will control costs and make primary power generation more durable. Rectifiers, batteries, solar PV, 5G antennae all come in modular formats so why can’t power generators be modular too?

Do solarised hybrid sites need re-dimensioning and expanding or are there limits to what can be achieved through cleantech energy sources?

Bladon has seen many instances where rural sites start with a solar hybrid installation with enough capacity to serve the local population but very quickly becomes obsolete with people coming from extra rural areas to take advantage of new nearby mobile network coverage. Human behaviours can sometimes be left out when dimensioning sites, especially rural ones. Modular power that is silent and low emissions is also an important factor when installing near residential areas that may be new to this technology and to avoid noisy, dirty diesel generators are important to reduce vandalism, running costs and theft.

When 5G comes to bad grid areas of Africa, how much will this issue worsen?

The looming 5G energy crisis cannot be underestimated. Yes, it’s more efficient than 4G per MB but it’s hardly relevant if data traffic will increase 400% by 2025 (according to the GSMA). That will in turn give a network power consumption increase of almost double! This can also mean adding 5G to a tower will push an extra 2-5kW load onto the tower and bring the site outside the scope of renewable sources of power generation. So who pays for the difference? Well scalable standalone power independent of the grid will help insulate networks from huge swings in power demand, load shedding from utility networks or increased ‘carbon’ taxes that make grid power more expensive or difficult to factor and leverage efficiencies.

Bladon see increased popularity in minigrids where Horizon Power, Australia for example where geography prohibits widespread deployment of grid connections. Even where there are grid connections, recent and more frequent bush fires has melted thousands of kilometers of grid lines. Customers seriously reconsider whether to rebuild or simply generate their own power.

What would an ideal energy system look like that is more robust to uncertainty over current and future maximum site loads?

Using the word ideal is always dangerous since there are a million plus towers out there with no grid connection or bad grid connection. Each tower can bring slight changes in site power design and setup based on available space, access, security of fuel supply not to mention increases in extreme weather which can damage solar PV. It’s clear that the more dependent we are on renewable sources of energy the more we need durable, reliable, and modular ways to generating power. Adaptability and agility built into to how power is delivered is key and being able to mould into the power needs of sites in the medium to long term.

If the solution is a partial return to diesel generators for primary power, how can hydrocarbon fuels be made cleaner?

Modular power using technologies like Micro Gas Turbine engines to deliver Micro or Mini Grids allows high levels of adaptability which by their very nature can handle running with multiple types of fuels including use of low and no carbon fuels. HVO or hydrotreated vegetable oil has the same energy density as diesel but emits 90% less CO2, 40% less NOx and zero particulate matter. Using this type of fuel bridges paradox of using energy dense fuel but also tackle carbon emissions.

Furthermore, using hydrogen-based fuels can also be used in turbines without modification so once logistics and storage challenges for hydrogen derived fuels are solved at scale then moving progressively out of diesel to low/zero carbon fuels becomes a more realistic endeavour.

Hydrogen is currently seen as the lesser alternative compared to energy storage technology which is coveting most of the PR right now. But over $500bn is being invested by governments all over the world to embrace hydrogen based production, transport and power generation technologies. Networ operators and TowerCo’s also need to ready to future proof their power generation investments.