How ready are your towers for 5G technology rollout?

Recent TowerXchange research revealed that in excess of 50,000 new towers will be built in Europe over the next five years. During the course of that research it was also discovered that many tower owners are also considering embarking on a programme of tower upgrades and strengthening projects in order to prepare their infrastructure for the demands of 5G rollout. TowerXchange spoke to Spencer Crawford-White, CTO at Delmec, who estimates that as many as 180,000 existing European structures will need to be reviewed for upgrade as 5G rolls out. 

TowerXchange: What do we know about the next generation of active equipment? What will the implications be for tower loading? 

There are two key areas to consider in this. The first is that 5G equipment will have to be much bigger than current antennas. Just like with phone handsets themselves, the first evolution of any technology tends to be quite large as manufacturers are working out the optimisation of the electronics, cabling etcetera.

Any OEM will naturally want to make the antenna small as possible eventually due to cost but they need space to make the technology work, therefore the first generation will be larger than current panel antenna. This will apply purely to the panel antenna and not the microwave transmission technology. 

From talking to OEMs like Nokia and Ericsson about new technical equipment, no one has yet decided how big the antenna will be or what it will look like, but we know it will need to be bigger and heavier too, as there will be more electronics doing more work. Also 5G sites will be more densely packed – towers 10km apart won’t be sufficient for 5G, so new build to suit will be needed which will need to be purpose designed to take 5G equipment.

The second issue is that the 4G antennas will have to remain in place, as well as 3G in many cases, while 5G is installed as they won’t turn those networks off until much later in the evolution of the network. In developing countries you’ll also need 2G in rural environments, as rural communities won’t survive without 2G. So you might end up with 2G, 3G, 4G and 5G technology all on one tower. Of course, it will be rolled out in urban areas first but you still might have three technologies at any point in time before it drops back to two. 

So you have two issues: a) the antenna will be bigger, and will increase loading even if it was just a like for like swap and b) it won’t be like for like as the 5G antenna will be in addition to whatever is currently on the towers, so you’re increasing the load by order of magnitude. This hasn’t been factored in by a lot of tower owners, who will just know their tower takes ‘three’ operators, without worrying about the equipment being hung on the tower. 

TowerXchange: What are the unknowns about 5G equipment and how accurately can we speculate at this stage vs having to ‘wait and see’ what the outcomes are? Who are we relying on for this information? 

We know that certain technology providers have been deploying and testing 5G equipment with MNOs, such as the Nokia collaboration with Docomo.  We know what the technology has to deliver, we are testing bandwidths, power levels and the distance to cover then working out optimum size they can squeeze that technology into. The question is who will come to market with cheapest, lightest product that meets that requirement?

We know what the bandwidths and requirements will be, that we will be looking at low and high band spectrum depending on what’s available in each market. We know we might have to decommission a technology to put 5G in the 700-800 band range or that high band might bump into other products in the market which might cause conflict. So we know the physics of 5G will work but we don’t yet know who will come to market first with the most economic plug and play solution. 

A lot of this will be determined by ICNIRP, the health organisation which sets radio emissions limits. Whilst the technology is clear and we know what it will have to deliver, the setting of these power limits will determine the health and safety impact and in turn what will be agreed with the World Health Organisation in terms of how close you can put these antennas to the public. 

OEMs will need to define the antenna, then they will need to get it checked as a public health issue – and as 5G will need more antennas and they will need to be closer to the streets, emissions will be increasingly critical to the public and will need thorough assessment, like antennas on rooftops did when they first rolled out. The traditional rule of thumb is that a panel antenna projects a cone of 120 degrees, with most of the power focussed in the 60 degrees in the middle of that, the acceptable distance for the worker is about 15m and for the public its 25m with the down angle being 15 degrees from the base of the antenna. If any person can get within 25m of that cone of influence, it will need to be assessed. Vodafone and a lot of the other operators use this calculation for ICNIRP compliance for sites, but if the power levels are higher those distances increase. 

Output for a standard mobile antenna is about one watt (relative to a microwave which puts out 800w) and the evidence shows that handsets put out much more. Broadcast sites are monstrous compared to cell sites as they send more over greater distances, but you can see how cellular networks now changing and replacing broadcasting in terms of OTT content so power levels are increasing. OEMs and MNOs rely on health organisations to give this the final go ahead. 

It will take some time to align what the operator wants, what the OEM can develop and what the health organisations will accept. 

TowerXchange: Let’s think about the average tower stock in Europe – how old is it, what type of towers are most common and what is their current capacity vs current load? 

A lot of European towers are still in the hands of the MNOs, whose perception can tend to be that they bought a multiple operator structure, and they will expect to get multiple operators on it. While Europeans are accustomed to relying on engineered infrastructure, in African markets operators tend to be more open to discussing the maximum loads for their towers, and are reaping the benefits in terms of health and safety by assessing and overhauling their structures. 

Towers don’t tend to fall down in Europe as MNO processes have been more robust from the off, maintenance is better executed and the initial designs were more rigorous, meaning they were stronger in the first place, creating a robust and stable platform for hosting antennas. In developing markets weaker structures have been put in, and towers falling over happens more regularly, but the higher incidence of independent towercos means there’s now a heavier focus on capacity and strength. 

The two factors to consider are security and safety, and we look at the design capacity of each tower – specifically at the statistical probability of a worst case event happening in a three second gust over a 50 year period. If you will only have additional loading in place for 18 months to two years before older technology is decommissioned, you will probably have adequate capacity already. It’s a case of proactively assessing your towers and putting some rules in place – as long as you reduce the amount of equipment over a reasonable timescale you probably won’t have to do any reengineering. 

In the UK, as an example, legally you can put up a 15m structure in two weeks, whereas larger towers take much longer and are more costly to construct – meaning that a lot of UK infrastructure is only suitable for one tenant anyway. As CTIL and MBNL rolled out there was a need for significant upgrades to support co-location, which now will be fit for purpose. 

About 70% of European tower stock will not need any work doing to it. A further 10% will need to be strengthened and there’s a further 20% which will need further assessment. The type, size, age and condition of these towers will all need to be assessed. Typically towers won’t fail because of one issue alone, it’s a combination of errors which creates the failure – a combination of age, condition, installation quality and load will ultimately cause the tower to fail. In developing markets in Africa or Asia, the number which will need to be upgraded is higher – you’re probably looking at a 50/50 split. 

TowerXchange: Comparing our projections for 5G equipment with our understanding of current macro towers, what upgrades and strengthening work would Delmec recommend tower owners undertake in order to prepare their portolfios? Can you go into a bit of detail based on tower type and/or tenancy ratio? 

I think full tower replacement numbers will be quite low and won’t have much impact in traditional European build to suit programmes and the natural attrition rate within any tower owner from things like notice to quit, etcetera – we’ll see a 1-2% turnover if that. Swapping out towers in Europe is so costly it won’t impact current build to suit budgets, we might have the odd one or two towers which will need replaced but shouldn’t impact on an ongoing replacement programme. 

Some people might panic about having to replace sites but there’s always a solution, you just have to look at whether it’s economically viable. There’s always a solution – you might need to replace every single member, which might seem like overkill but keeps the tower in place while you do those works. Building a new site adjacent means the cost of a new site, new equipment, a hot swap and then decommissioning, plus it can cause issues trying to stay online.

Traditional towers have angled steel legs joined together in a cruciform shape, these angular tower legs can have bits bolted on to make them stronger. Let’s say as a basic cost model a new tower might cost £100,000, plus with the cost of the operator putting new equipment on, the cost of decommissioning old equipment and the old tower it can reach as much as £250,000 to swap towers. When upgrading a big tower you might need 5000-6000kg of steel which could cost you £7,000, so a total of maybe £30,000-40,000 when including labour to reinforce to the level you need. Add to this the fact that you can often rectify conditional issues at the same point, meaning you can dip into another budget and the case for upgrading a tower is strong.

The most complicated problems come from tubular section leg structures which can’t support the load and need to be replaced, so you’re talking maybe £60,000 due to the complexity of the methods which is still much cheaper than swapping out. Guyed masts are more tricky still – their cost of upkeep is 5x what a lattice tower would cost and strengthening is 10x what a tower would be, but they’re not designed for upgrades and not the types of structures that the 5G will need to be on anyway in most cases. 

We’re helping tower owners strengthen their tower across the world. It’s not always triggered by 5G but also for LTE and network sharing. Tower owners also need to think longer term. It might cost £10,000 to strengthen a tower for an additional tenant, but for 10-20% more budget you could strengthen it for two as the labour and transport is already in place. We help towner owners to predict the potential in the area to define the strengthening model the need. The only way to do that is for clients to embrace a full network assessment and know what the towers are like at the beginning of the process. 

We have developed a risk profile calculator for tower owners, which will tell them which of their structures with x amount of loading can take x amount of equipment and increase capacity by x% to give them a rough idea which towers will need work. It will give a rough idea but then all the environmental and location factors have to be taken into account. A small tower in middle of nowhere which is protected from wind will be a lot less exposed than the same tower on a clifftop: the same structures will have very different capacities based on the environmental conditions. 

What’s the long term effect of higher loading? It’s still rare to see towers falling over because design codes and adherence are good and we use 60% capacity as a factor of safety on most towers anyway. The big change will be as more and more people start looking at outsourcing their structures, towercos like Arqiva or American Tower etcetera will have less of a headache in keeping on top of it as their asset register is up to date already. Tower stock is getting older and more tired and will start breaking down eventually. The more it goes on, the more equipment we put on, the more the risks and impact. It will give independent towercos the edge as their towers will be more stable and compliant. 

TowerXchange: What are your recommendations in terms of timescale? Do tower owners need to begin an assessment of their towers now in order to ensure they’re prepared, or would a ‘wait and see’ stance pay off in the long term? 

Someone in health and safety will be looking at desktop assessments in the short term, pushing the OEMs they’re working with for feedback on this. The focus from MNOs and towercos will be on what new sites they need, not on the existing ones. Ultimately there will be a reactive process, it will get to the point where towers need upgrading urgently but this will come at a higher cost than doing a full assessment and having modular upgrade kits which can be rolled out when needed. Tower owners are facing budget constrains now, but every £1 spent on pre-emptively assessing and preparing towers has been shown to pay back three times as much in the long term through risk reduction, continuity in service to customers and payback on an ROI model which has evidence of working.