Medupi Power Station has proved to be a financial and logistical nightmare for Eskom.
When it does come online, Medupi will produce massive power for the country. At the moment, however, big power just means big problems and stymies economic growth.
The mega power station in Limpopo will have six units, each with a 800MW capacity, and will be the largest dry-cooled coal-fired power station in the world.
But on July 8, two days before the release of its annual results, Eskom called a press briefing to announce that Medupi would not come on stream by the end of December as repeatedly promised, but rather in the second half of 2014 — and Eskom would have to resolve a potential 700MW gap in the interim.
Eskom chief executive, Brian Dames, said the new construction programme is "crucially important to stimulate economic growth, not just in South Africa but the Southern African Development Community region as a whole".
He noted that the three critical challenges leading up to the delay were months in work days lost to labour unrest, faulty welding on boilers and problems with the control and instrumentation system.
Eskom finance director Paul O'Flaherty said: "We are not running away from accountability … and, yes, we underestimated what is required."
Medupi problems deeply rooted
But analysts and industry experts say the problems with Medupi are deeply rooted, and Eskom's decision to go with a megastation could only ever have meant disaster.
Doug Kuni, managing director at the South African Independent Power Producers Association, said Eskom's choice to build a megastation was a poor one and threatens the entire economy.
"If you are in a power shortage situation, you don't place an order for custom-made megastations [referring to both Medupi and Kusile — another megastation under construction in Mpumalanga]. It is courting disaster."
He said the timeline for Medupi had also been "much too optimistic".
Pieter van Dam, a member of the Free Market Foundation's energy policy unit and director at Energy Research Associates, agreed that opting for a bespoke model was a failure in strategy.
"For some reason Eskom lost sight of the fact that speed was most important," he said.
"Eskom proudly boasts that they are building the biggest power stations in the world when their boast should have been 'we are going to do the quickest build possible'."
Tried-and-tested is the way to go
A number of analysts agree that choosing "off-the-shelf" standard technology and tried-and-tested modular 6x600MW plants would have been the better way to go — as seen in countries like India and China where modern power stations are regularly built on time and within budget.
This delay marks the second time Eskom has pushed back the timeline. It also said costs have increased from R91-billion to R105-billion, excluding interest.
Eskom said the biggest reasons for Medupi's delay were the issues with the control and instrumentation system. On top of that, post-weld heat treatment had been inadequate, meaning multiple welds needed to be retested and fixed.
And welds made using unqualified procedures also had to be replaced.
Eskom said significant progress has been made in its labour relations and a new partnering agreement has been signed with unions. But the site has a lengthy history of unrest and the project has lost up to nine months in days of production due to strike action.
Most recently, there was a work stoppage of more than three months — from January to April — which contributed to putting the December deadline at risk. At the press briefing on the Medupi delay, Eskom made it clear that it is not the main contractor.
Whose responsibility is it anyway?
O'Flaherty said Eskom is only now driving the management, development and supervision of labour — which is actually not its responsibility.
Kuni said because there is no single contractor who has accountability to commission the unit at Medupi, it automatically makes Eskom responsible for the unit commissioning. Unfortunately, this means that any delays are risks passed directly to the taxpayers.
In a statement following the announcement of the delay, Alstom confirmed that tests performed on the boiler protection system, a part of the control and instrumentation system package, were not satisfactory.
It noted that: "The boiler protection system is a complex system and the configuration requested by Eskom is demanding. No other coal plant has ever been built on this scale."
It added that other contractors were also facing technical problems.
Dames said that Eskom has always acknowledged that the new programme is a mammoth challenge and that there are risks to it.
"Medupi is large and complex," he said. "It is bigger than most things we have done in the country … there are massive logistical and transport issues."
The issue of time
Speaking to the Mail & Guardian, Dames denied the project was too ambitious but, rather, was necessary to close the energy gap.
However, he admitted that time had been an issue: "If the decision was made earlier … that would certainly have been better, but that was not the case."
He said it would not have made sense to split the construction over different sites given the logistical costs, but noted that, if Eskom were ever to do another project like Medupi, it would be done earlier, with the same design — but starting with just three units and adding more as they are needed.
But Kuni said economies of scale should no longer be the go-to defence: "What was supposed to be economies of scale has landed up as a disaster of a scale more than we possibly could have imagined."
Eskom should have instead gone to tender and given the contract to three independent power producers to knock back the power shortfall at the time, he said. This would have created competition, kept costs down and produced results a lot faster.
"They [Eskom] just did not have the competence and capacity to do an expansion like this," he said. "At the time the contracts were placed in 2006, not one member of Eskom's executive committee had built power stations."
Waterberg the right choice
But Xavier Prevost, consulting senior coal analyst at coal geology, mining, beneficiation, logistics and marketing consulting services company XMP, said going big in the Waterberg was the right choice.
The quality of coal differs in various parts of South Africa. In Witbank the coal is fairly clean, while in the Waterberg it is intercalated with rock, which can only be separated out at washing plants.
Thus mining operations yield far less saleable coal, but once washed it is of a very good quality.
For this reason "the ideal thing in the Waterberg is to have a power station", Prevost said.
"The Waterberg model has to be large simply because the Waterberg coal needs to be mined in large amounts [to make it worthwhile]."
However, the area's infrastructure isn't great and access to enough water for this kind of project remains a problem.
The issue of policy
Eskom noted in its annual report that it is working closely with the department of water affairs to develop funding models for water infrastructure in the Waterberg.
But before strategy and management came the issue of policy.
In the Energy Policy White Paper of 1998, the government agreed on a way forward, which was based on liberating the electricity supply industry by breaking up Eskom's monopoly, creating competition and introducing independent power producers.
This never came to fruition and Eskom was prevented from planning new projects in the interim until Alec Erwin secured its leading role in the new construction programme when he came into the public enterprises ministry in 2004.
This happened despite the 1998 paper predicting that new capacity would be required around 2007.