Mega project to bring fibre to SA homes - TechCentral

Mega project to bring fibre to SA homes

i3 Africa CEO Cornelis Groesbeek

Want fibre to your home? You could soon have it. i3 Africa, a new company backed by the National Empowerment Fund, plans to build a high-speed fibre network connecting 2,5m homes within the next 4-5 years in a multibillion-rand project that could transform SA’s broadband landscape.

TechCentral can reveal exclusively that the new network, which is set to be built in six SA cities — Durban, Cape Town, Johannesburg, Port Elizabeth, Bloemfontein and Pretoria — will offer minimum connection speeds of 100Mbit/s to consumers and will be “competitively priced” next to Telkom’s fixed-line broadband digital subscriber lines (DSL). Speeds of up to 1Gbit/s will be available.

i3 Africa is headed by CEO Cornelis Groesbeek and chairman Andrew Mthembu, both of whom have years of experience in the telecommunications industry. Both men are nonexecutive directors on the board of Broadband Infraco. They have declared their plans to the company and don’t believe a conflict of interest exists as Infraco focuses exclusively on national long-distance and international connectivity.

i3 Africa has appointed Merrill Lynch as leader arranger for funding from capital providers for the project.

Groesbeek says the company will spend between R5bn and R6bn on the network — about a third of the cost normally associated with traditional fibre-to-the-home projects.

i3 Africa — its shareholders include the National Empowerment Fund (with 23% of the equity), Groesbeek, Mthembu, Martin Cele (former CEO of Durban business incubator SmartXchange) and Craig Carthy (an investment banker) — has chosen Durban as the first city that will get the infrastructure. It will build a pilot network north of Durban in the next few months to test the system before beginning a commercial roll-out from midyear.

The network will operate on “open-access principles”, meaning Internet service providers and broadcasters can sell services — broadband, video-on-demand and telephony — directly to consumers using the network. i3 Africa will provide infrastructure only. It will not sell services directly to consumers, leaving that to third-party service providers.

These providers will be able to buy access to the network through facilities such as Teraco’s data centres, where i3 Africa plans to establish points of presence.

Groesbeek says the network will be built using technology developed by the UK’s i3 Group. i3, which is expected to take a 15% stake in i3 Africa, has already deployed similar fibre networks in the US, the UK, Australia and New Zealand.

i3 Africa consists of two subsidiaries: H2O networks, which will build the physical network, and Fibre City, which will manage the infrastructure and billing systems on behalf of service providers.

Groesbeek claims i3 Africa will be able to build the network at a third to half of the cost normally associated with fibre deployments of this nature — and in a fraction of the time.

The company will do this by making extensive use of metropolitan sewerage and water networks, obviating the need for expensive civil works. It’s already in talks with the big metros about the project, and is on the verge of winning a contract to begin work in the Durban metro.

Sewerage pipes will be used mainly to build the metropolitan rings around the cities. Access to people’s homes will be provided through fibre laid in water pipes.

i3 Africa plans to offer fibre to 2,5m SA homes

“We hop out of the sewer, as it were, and do an access ring around a suburban block using a technology called micro trenching,” Groesbeek explains. “This involves cutting a 50mm-wide and 200mm-deep slice in the road using a cutting head, which can be attached to the front of a construction tractor.”

He says each machine will cut, lay fibre and seal 1,5km of road a day at “20% of the cost of conventional trenching”.
From the micro trench, the company will use an i3-patented technology called Atlantis, which will feed the fibre through the water supply system into people’s homes.

“We cut the water pipe on the pavement, next to the water meter, attach a fitting which is certified to last 100 years. Then, just before the water valve in your house, we cut the pipe again and feed through the fibre,” Groesbeek says.

He estimates it will take about 30 minutes to complete the installation in each home.

i3 Africa has already deployed 100km of fibre in Durban on a test basis. In the next few months, homes and businesses in the suburb of Somerset Park, next to Umhlanga, will be used as a test bed for the technology. The pilot will involve about 100 premises — 80 private homes, 20 businesses, one community centre and three schools.

“This pilot will allow us to test and verify everything before we scale up to 2,5m premises over four to five years.”

The company also plans to connect 225 of the Ethekwini (Durban) municipality’s offices and public facilities. It hopes to work with other metros on similar projects.

i3 Africa has no intention of playing in the long-distance and submarine fibre market, which Groesbeek says is already adequately served by other companies and projects.

He says i3 Africa is already busy with site surveys and network design and is in talks with some of the bigger metros, including Johannesburg, about its plans. “We’ve decided to focus first on Durban, which has a history of being an early adopter of innovative technologies, but we are engaging with everyone.”

The Durban leg will consist of 2 500km of access fibre and another 7 500km to 10 000km of fibre into people’s homes and into business premises. “We will have all that done during 2012, and sometime next year we will start on the core network in a second city.”

Subsidiary Fibre Networks already has the necessary service and network licences it needs to operate the network, which it acquired from a shelf company.

Groesbeek cautions that the roll-out plans could change as construction gets underway and the project advances, but it has identified six cities in its business plan. “A lot is dependent on the reception we get from the municipalities and how quickly we work through the whole process.”

Asked whether he regards plans by the Independent Communications Authority to unbundle Telkom’s copper cable access network, allowing rival operators and Internet service providers access to this “last-mile” infrastructure, as a threat to i3’s plans, Groesbeek says copper can’t compete with fibre.

“If all consumers wanted was low-definition YouTube, then, yes, it would be a threat. But with our network you’ll be able to watch high-definition and 3D television and get video on demand. Our baseline planning reference is 100Mbit/s to the home.”

He says it’s too early to say how much consumers will be expected to pay for access to the system. But, he says, “we wouldn’t do it if it wasn’t going to be attractive to the market”. He expects service providers will offer a spread of products, including capped and uncapped broadband offerings, as well as triple-play options of television, broadband and voice telephony.

“It will be competitively priced next to DSL. It has to be.”  — Duncan McLeod, TechCentral


  1. Wow – ambitious plans. It’ll be interesting to see how this is playing on several fronts. Firstly, is there any reaction from or opinion on Neotel? How would this impact their plans? Secondly, I believe VOD & IPTV services would be an exciting addition to the SA television landscape. Is that ICASA regulated as well and what is their involvement in this project? I guess regulations would be between ICASA and the potential 3rd party operators. Lastly, would there be potential push back from DTH operators Multichoice and TopTV?
    Either way, if the government & 3rd parties are on board – and if there is sufficient funding – this is an exciting development.

  2. murraybiscuit on

    they can start trialling in bellairs drive. we’ve dug up half of it already…

  3. I’m in the western suburbs of Ethekwini, and I suggest they trial here instead of the Northern suburbs because … er … I really want a 100mbit fiber connection.

    It’s a real pity they didn’t start 2 years ago, when Ethekwini embarked on its water system revamp – literally every water pipe mains and feeder to every home was replaced, they could probably have dropped fiber next to the water pipes before it was covered up.

    The water project was really well done. Minimum disruption, quick cover-up and good repair of the trenches/verges. I’m happy they’re working with those guys, out of all the large-scale infrastructure guys, they seem the most competent.

  4. Nice to see this sort of thing happening. It will be a tough “sell” though and as the article mentions, the reception from the various muncipalities will be key – it is therefore rightly seen as a civils operation more than a communications business. More particularly since it will operate on an open access basis. Although Mtembu and Groesbeek are non-execs, Broadband Infraco has hardly knocked the lights out. Maybe they leant something on how not to do things there?

    The point about Telkom’s existing copper and the unbundling of the local loop appears not to be that it provides direct compeitition but the pricing per Meg on local access connectivity (from about 1Mbps up in speed) puts a ceiling on what customers will pay. Copper pair bonding will also increase the performance of existing copper infrastructure. High definition video streaming will be limited by the backbone network to which FTH services are linked as well as the attitude of your local ISP – FTH subscribers will need to budget for big ISP fee increases if they are to access HD video.

    It appears that the provisions of broadband connectivity is fast becoming a low margin commodity play. If this is so, then services or content will become inportant differentiators. Unless you can persuade Naspers’ Supersport to use the channel, the triple play idea could be hard to set up – the target market already has ADSL (installed), satellite TV dish and PVR, is or will be a Push Play subscriber and is generally prevents by existing exchange controls to buy music from iTunes.

  5. Sounds interesting – the old Telkom copper infrastructure in our neighbourhood (northern suburbs Cape Town) prevents me from getting 4Mbit/s ADSL. Im on neotel wimax now, its more reliable but not lightning fast.

    Im willing to give these oaks a chance, wonder what the pricing will be?

  6. If they can indeed roll out to 2.5mil homes for R6bil, that only works out at R2400/connection to get up and running; say it has a lifespan of 10 years, that suddenly becomes R20/month they have to charge to recover costs; R50/month would probably give them enough to maintain it and run it at a profit. If that’s the kind of numbers we’re looking at, there is no reason for someone that wants internet would consider anything else.

    Hell.. I’d pay R10k installation fee tomorrow to get onto a fiber backbone that offers a good monthly uncapped package.

  7. While i do welcome this from them, how can they say 100MB is their baseline yet say there will be capped options in addition to uncapped options. For me i feel an uncapped offering is the only affordable option in comparison to capped. In other words , they cost the same so just give us uncapped.

    Seriously, 100MB capped??? unheard of.

    Again well done for atleast i will get to test drive it before i pass on 😛

  8. So Telkom tell me that they are not replacing the copper in my street because it gets stolen too often. Charming. But they insist on billing me. Then I see optical being trenched not five blocks away. Granted, that’s a few km’s in my are but that’s not the point. So I phone them. I get told if I dig the trench they’ll lay it. WTF

  9. Good luck to the project, I hope you get what is promised by i3 Group and thier management team- they left a trail of distruction, huge debts and broke sub contractors and suppliers with thier FTTH projects in Bournemouth and Dundee in the UK.

  10. @Jonas, you right, a quick google looks like it didn’t work well over in the UK for them.

    “The i3 Group has reportedly been forced to scrap Fibrecity plans for using the underground sewage system in Bournemouth to spread its “super-fast” 100Mbps Fibre-to-the-Home ( FTTH ) fibre optic broadband ISP network around the city. It’s understood that Wessex Water, which is responsible for the sewers, at some point over the past few months simply decided not to go ahead with the plan on both technical and business-related grounds.”

    I guess if it didn’t work in the UK, lets try SA, they much more gullible and desperate for high speed internet

  11. The_Librarian on

    At last one company doing what NeeTel’s moaning about.

    NeeTel just doesn’t want to try, but loves to dish out excuses. Companies who do that sort of thing, deserve to die.

    Companies who’ll innovate, or try out things, deserve growth and profits.

  12. 100MB Lines in South Africa?

    People from Telkom’s 10MB lines requested to drop to 4MB, since 10MB was extremely unstable. To combat this, Telkom are now offering 20MB Lines (Which, following the trend, will be even LESS Stable than 10MB Lines)

    And @Bob (Assuming not a Troll)

    A 100MB Line will have a maximum download speed of approx 12.5MB/s (12,800kb/s) following the constant trend of [Line Speed / 8]

    The same way a 4MB Line has a maximum download speed of approx 500kb/s (I have personally never seen above 400kb/s), and NOT 4096kb/s

    Although, downloading anywhere ABOVE 5,000kb/s is currently a foolish hope of any South-African citizen.

    Seeing as how the article specifically states that it will only be laying down the infrastructure, in the rare case that this actually becomes a possibility, Telkom will most likely buy the majority, charging such a ludicrous fee (Seeing as how this is 30 times faster than their current stable “fastest”)

    So, to conclude:


    a.) It will never happen
    b.) It will be too expensive to afford for any non-business user.

  13. Aren’t these the same people running the government funded Broadband Infraco? What is the connection between their new venture and Broadband Infraco?

  14. wishfull thinker on

    Lets see which municipality cares the most for its residents and secure the first deals.

  15. Very interesting. Until now, providers have explained how its not feasible to bring fibre to the home and the focus has been on LLU.

    This technology not only sounds eminently sensible (through existing sewer & water pipes) and cost effective to deploy, but it sounds like it even has the potential to remove the need to unbundle the local loop.

    As they say, why use copper if you can have fibre?

  16. If the fibre optic cable is threaded through a sewer pipe, what happens to the fibre optic cable when the pipe becomes blocked and the maintenance services uses rugged sewer pipe cleaning equipment.

  17. Shopper Pete on

    How does i3 ensure that there is no contamination of our water supply as “We hop out of the sewer”, as well as during the 30 minute installation in residences?

  18. Interesting idea – the sewers are already struggling with removing the sewerage, so as long as my broadband doesn’t come with an unidentifable, but disgusting smell, then cool. Imagine having an account with ‘Stinky ISP’ 😉

  19. I really like the fact that i3 are aiming, from the start, at connecting premises. We’ve seen a number of companies punt big projects to lay fibre and from past experience, this seems to be limited to backbone fibre – leaving the last mile to Telkom and wireless providers.

    I’d also be interested to know what i3 expect their uptake rate to be, i.e. what number of the 2.5 million premises will actually use the connection?

    Great exclusive Tech Central!

  20. Is any of Broadband Infraco’s infrastructure being used by i3? If so, how precisely is there no conflict of interest?

  21. @Relix It’s fibre TO THE HOME. You can’t really compare it to the speed Telkom offers on its old twisted pair copper that needs to run from the closest distribution point/exchange.

    Also, I doubt this project would even commence if there wasn’t a solid business case supporting it, so your 2 points that it will never happen, or won’t be affordable don’t make too much sense.

    Personally, I think this initiative is fantastic, altho slightly overdue!

  22. was speaking to my nephew in the UK just the other day, and they have had fibre installed to their home and he was reporting speeds of close to 45mbps, I was ashamed of my purported 4Mbps line that barely gives anything over 500kbps on a good day, if this I3 thing flies, I will be there with my catapult ready to gather it to my cave.

  23. The beauty about this is that i3 will not be digging our roads, they’ll be putting the fibre in sewers…*Out of this world i tell you*. Well done guys

  24. Fibre in sewers? Easier said than done. Especially when sewer maintenance is required. Call ROTO ROUTER?

  25. Hi PH
    Broadband Infraco is a mess. They can not and will not be involved in access networks. They can’t even handle inter-city connectivity for the government.

  26. It will be interesting to see if this actually happens. i3 Group recently sold its UK fibre networks as they failed to get their business model to work. Three years in Bournemouth, tens of millions spent and just a handful of customers paying for services. Brisbane City Council also pulled out of a recent deployment with i3 Group due to a lack of satisfactory progress. So to say that i3 Group has successfully deployed FTTH is generous to say the least. TechCentral needs to do its research. Might I suggest Google?

  27. Its a great idea, but a sewer system does not only have water flowing through it…. the solids in that system could pack up against there lines and eventually cause some serious blockages….

    But if they can make it work… then HELLL YEAH!!!

  28. @Techcentral Glad to hear it – plenty of people have already been burnt by ill thought through plans. Hope it doesn’t happen to Durban.

    The sewer technology actually works very well from a technical perspective and is great for metro rings and point-to-point, but it doesn’t have the same cost advantage on FTTH, because of the number of break outs that are needed from the sewers. The wholesale open access fiber model is also notoriously difficult to make profitable – I only know one project worldwide that makes money (in Scandinavia).

  29. Another Andrew on

    @Dirk de Vos. The reason that Naspers aren’t more active in the triple-play space is because of lack of infrastructure… no other reason.

© 2009 – 2020 NewsCentral Media