AFTER THE BELL
I was just explaining to a Martian my three-step plan to fix Eskom
If you were a Martian and you landed in South Africa today, what would you do to fix Eskom? The question is not who is to blame, or what we should have done years ago, but what we can do right now.
I’m not an electrical engineer, nor am I a management expert, and this could all be baloney, but for what it’s worth, this is what I would do to fix Eskom. Toss my advice, accept it, whatever. Everybody in South Africa has had their 10c worth, and this is mine.
Step 1: Decide what the problem is
There are four possible options here: this is a management problem, a grid problem, a fleet problem or a planning problem.
To isolate this issue, it’s worth looking at what exactly happened over the past few weeks, and thanks to Kyle Cowan from News24 for breaking this down so nicely.
Eskom has a total installed capacity of 46GW; of that, 6GW are peaking power, so it’s not always available for the grid. About 7GW are out on planned maintenance. About 10GW dropped out of the system over the past few weeks, and 4GW were taken up with partial load losses, which is just inefficiency that results from the production drop mentioned above. In total, 21GW dropped out of the system — almost half of the capacity. So that leaves you with 25GW, of which 6GW comes and goes.
So far, so good. So what happened over these past few weeks? (Thanks again to Cowan for this explanation.) First, on Saturday, 3 September, Unit 2 of Koeberg had a problem with the control systems, which lowered the cooling rods (this is truly scary, so I’m not thinking about it) and about 1GW dropped out of the system. Over the next weeks, 44 subsequent trips took place at power stations around the country.
Two days later, one of the units at Kusile tripped, shortly followed by nearby Matimba. Both were quickly returned to service, but both tripped again. Now we move to the electricity heartland in Mpumalanga, where in the next few days there were problems at one unit at Arnot, which was quickly returned to service. But then nine units at different power stations tripped.
Some of these units were old power stations, like Kendal and Majuba. But there were trips at Medupi too; the new power station that was going to solve all our problems. Each of these units was restored within about 10 hours, but 33 units then tripped over the next few days. Eskom at this point had eaten up all its peaking power. Eskom personnel rushed to fix these problems, but it wasn’t enough.
If you look from the outside at this picture, what becomes vaguely clear is that Eskom does not appear to have a management or staffing problem. Why? All the issues were addressed quickly and staff were on to them.
If it’s not a staffing issue, then it means, in essence, Eskom has a fleet problem. Small issues on the grid quickly explode into a total fleet breakdown, presumably because each problem places extra strain on the fleet, which can’t handle the load.
The sequence does not openly suggest management or staff have been at fault; they are just trying to make a car go faster than it has the capacity to go. Of course, it would be nice if they fixed the problems properly, but given the widespread nature of the issue, I would hazard a guess that unit management at each station is not at fault. I could be wrong, but there wasn’t one big problem at one plant — it was all over the place.
There is of course an underlying planning problem; you don’t want to be in this position. But I’m talking to a Martian now, so all that history is water under the bridge. The solution here is clear, and that is step two.
Step 2: Add more capacity to the grid, and do it quickly
Now, of course, the government has been trying to do this, so what is the problem here?
The notional solution has been to get the private sector to add capacity as part of the REIPPP (Renewable Energy Independent Power Producer Programme). This is what the government had plumped for as a solution. But REIPPP Window 5 was announced a year ago, and not a single one of the 13 winners has yet to put a single watt of electricity into the grid. Many haven’t even got to financial closure. It’s worth noting that mining companies that I have spoken to typically put these systems in place in four months, once the paperwork is done.
So we have a big problem here, and the problem is bureaucracy and agendas. At one point, you couldn’t get an offtake licence from the energy regulator, Nersa, without a wheeling agreement from Eskom (wheeling is the way of getting electricity on to the grid). But you couldn’t get a wheeling agreement from Eskom without an offtake licence from Nersa. I am not making this up. Apparently, this is fixed now, but, you know, WTF?!
Visit Daily Maverick’s home page for more news, analysis and investigations
And then the Department of Trade, Industry and Competition put its oar into the works by demanding its little slice of the pie by requiring maximal local production, as it does with everything, whether SA produces the stuff or not. That was reduced, but not by enough. And then there is general incompetence and resistance at the Department of Mineral Resources.
Add to these problems that the bidding process happened before prices shot up because of Russia’s invasion of Ukraine — you know, the war that SA supports — and other issues. There was also Karpowership but, inevitably, corruption. Is there a solution here? Yes, there is, it’s step three.
Step 3: We could short-circuit all this bureaucracy if we just allow Eskom to build some emergency renewable power
Controversial? Possibly. But think about it.
Eskom, sadly, does not have a balance sheet on account of being, you know, bankrupt. But Eskom is at the moment spending about R15-billion a year on diesel to keep the system going. It now costs R6,500 for a 650W solar panel, which means this is a ridiculously easy calculation: the retail market price is now R10 a watt. Magic! Hey presto, 1.5GW of power! Effectively free, because you are diverting cash you are already spending.
Now, of course, the mega-power station lobby will object, because, they say, renewable energy is not baseload power. And the essence of their argument is — and I am not making this up — that solar panels don’t work at night on account of the fact that the sun goes down. Sadly, nobody in the solar industry ever thought of that (kidding). There are plenty of 2GW solar plants out there these days, and there are a gazillion ways to store power. They just want to build another big power plant because that’s what they know.
But why should Eskom do it? Didn’t Eskom bankrupt itself because it was unable to manage Medupi and Kusile? Yes, this is true, but since even I managed to wire up my house from the solar panels in my back garden, I think Eskom might just be capable of handling this kind of project. Eskom might not be the ideal manager, but it does have the enormous advantage of not having to ask itself whether it wants to buy electrical power from itself or not.
This might be all desperately naive; there may be much more complicated issues here. But, on the other hand, my three-step procedure took me an hour to write. The government has been dealing with this problem — or not dealing with it — for 15 years now; the least it could do is be absolutely crystal clear and honest about what the problem actually is, what they intend doing and when that will happen. Crystal clear, people. Please. DM/BM
Become an Insider
Great ideas. Problem is – TOO SIMPLE. Our leaders just don’t do KISS. And too little room for the greedy gravy-trainers.
Privatise it. Excess solar can be reserved in pumped hydro electric schemes, ultra efficient water turbines do not require squillions of metres of head. It can be done fast, efficiently and to a scale where reserve capacity can be measured in weeks. But, duh, the dullards in government can’t let go of control, because who else cluld be trusted to screw it up as well as they do? This was never an issue about anything other than government keeping control over the biggest pot of cash spend in the national (overall) budget. They have to have their hands on power, and they will never let go until it is pried forcibly from their hands. It may even come to that. And THAT is the real concern no one wants to speak about innit?
Not quite so. Mechanical Engineer here. There is no such thing as ultra efficient water turbines. The ones that don’t need squillions of meters of head requires squillions of liters of water to be mildly useful. SA is simply not a country with great Hydro storage potential, and our risk profile is not great going forward to have the stability of the grid dependent on water availability from pumped storage schemes.
Retief, you are wrong. In the 1990’s already Eskom identified seven locations for possible hydro-electric pump stations in the Lebombo & Drakensberg mountains and in Eastern Limpopo (at the escarpment between the high – and Lowveld) and in the Eastern Cape. One was actually built (Ingula). It has a capacity of 1.3 GW and, if the top dam is full at the beginning, it can provide that 1.3 GW for between 16 and 20 hours. But these hours can be extended if the dam capacity is increased. And if all seven are in place and the water pumped upwards until all the dams are full with excess energy, all of these would have been able to provide +-10GW of power for 20 hours, or more if the dams are made bigger. Or, if only 5 GW of power is needed, it would be able to run twice that long, 40 hours or more. That takes more than a day, so in the middle the sun and wind can be used to take over and replenish the top dams again. And – dare I say – the Drakensbergs are so big, and we have other mountains also, that we can even organise more of it. So I am not convinced that this commitment to battery storage is the best way to go; I mean, how many batteries will be needed to store 10 GW for 20 hours? It is 200 000 000 KWH. My supplier says that a 2.5 KWH battery cost R15000; 80 million batteries will thus cost R1.2 TRILLION. Then there is also the mammoth matter of the space needed to position 80 million batteries. The batteries don’t last for ever either; the dams will last many times longer.
. . . Do yourself a favour and google the Ingula power station. The detailed document with all the technical information is there.
I would suggest the problem is much more complex than stated here. If one looks at the breakdowns then many are the other older stations, Kendal, Thutuka, Komati, etc. BUT many of the brand new units at Medupi and Kusile are also tripping. Even more disconcerting is that some trips, such as Unit 2 of Koeberg, occurred after a major “outage” (Major maintenance project). This means that yes we have to deal with keeping older kit running, but we also have some design issues with the new units and when we fix things, we seem not get it right either. This may be the result of another major challenge, technical staff without sufficient training and experiencing who replaced the experienced Eskom colleagues who either retired or left for better working conditions in the private sector or abroad. Lastly, procurement is still a major challenge for Eskom making it virtually impossible to appoint well-experienced, competent contractors to do the work. So, it is indeed a “wicked” problem that need a completely “out of the box” solution.
The problem with Medupi and Kusile has been explained by experts that know what is going on there. The equipment procured is not suitable for the low quality coal that SA provides. That is why so many breakdowns occur and why even De Ruyter underestimated the problem. We are sitting with R360 billion of basically unusable power stations. I can imagine that plans can be made to refine the coal (sift the rocks and stones out, mill it to be finer, etc) but I hear that other design flaws are also present. That is the result of putting it in place on an emergency basis without having the necessary experience on the job, if you ask me.
My suggestion would be to privatise the existing power stations with an appropriate power purchase agreement and let the private sector take over running and maintaining these power stations. It will resolve many issues, including the current shenanigans with the poor quality coal supplied to these power stations. To do this will take enormous political will and capital but I see this a the only way out. Let’s hear if our “fearless leaders” have the courage to do this.
Lastly, I want to acknowldege the technical staff in Eskom who are competent and doing a stellar job.
This is the only real option and I believe it is inevitable. Eskom is nowhere near sustainable anymore, and the political interference at all levels (unions, procurement, transformation and BEE requirements etc, nevermind flat out corruption – DMRE etc) means it will never be sustainable. The South African fiscus cannot afford to prop up all the inefficiencies mentioned above, so inevitably Eskom will collapse which will mean that it’s assets get sold to offset it’s liabilities (and us taxpayers will have to fork out for the difference.. yay.. thanks ANC and ANC voters!).
Eskom has been in a utility death spiral for a few years now, and with the major metros all making moves to get off Eskom, this will just hasten it.
The “good” news is that the sooner Eskom collapses the sooner we can start rebuilding.
i SO enjoyed reading this. thank you for taking the time to write it. some food for thought and a good chuckle interspersed during an otherwise dark time.
Yip, good use of an hour writing this.
Some technical perspective. Due to a minor maintenance error, Eskom managed to blow up a turbine at Medupi, mere months after handover. 600MW+ gone in a flash. BILLIONS down the drain. This is the engineering equivalent of a navy sinking a brand new world class submarine on it’s maiden journey. It should be impossible. Responsible people should die of shame. You are the absolute laughing stock of the engineering world. But Eskom did it. And they’ve done it before. The fact that this whole incident is basically ignored shows the deep pathology of SA’s collective Stockholm Syndrome. But as a journalist, you and your colleagues should cover it every time the “problems” of Eskom are analyzed.
You have hit the nail on its head. The question arises what happens to the people who allow such a damage to happen? Is the buck not stopping with de Ruyter? When I listened to what you are saying many questions raged in my mid including sabotage but there is supposed to be risk management in such a huge and expensive entity. Hitachi has been fined for their corruption in the US and had paid US $150 million to the US Treasury when people who were screwed are South Africans. The Department of Justice, NPA and Hawks know that there is an FBI office here and are clear of procedures to bring a criminal case against Hitachi, Chancellor House, ANC and Eskom Executives that were responsible but for political reasons and nothing else are not acting!
De Ruyter says that the law enforcement agencies did not give co-operation with the fighting of the criminality. Then of course, neither did other government departments; De Ruyter has also been pleading for more renewables; he has even started to repurpose some of the old plants with renewables. But the one thing that is still falling short is increasing the capacity of the grid from the northern, southern and eastern cape which is where most of the renewable energy will be generated. It seems that may take up to 10 years to organise; I wonder if that process has even started.
Of course Sasol 2 & 3 was also built on an emergency basis, and it has a lot in common with a power station. But it started in 1975 if I remember correctly, with the identification of the land, and by 1982 Sasol 2 was running (ie within 7 years). And there were a number of design flaws as a result of the speed with which it was put together; yet since then it has been providing fuel to SA reliably. I know, because I started working there in 1981 and retired in 2015. And it has to be kept in mind that before 1975 the only expertise Sasol had was with the relatively small Sasol 1 plant, while Eskom by 2007 had more than enough experience with big power generation. So I have to conclude that the politics of transformation had a lot to do with the mess that followed when Eskom had to build new capacity on an emergency basis.
Nothing like whittling the facts to the bone.
1950’s socialist ideals to control every aspect of the state is outdated, much like the collapsing coal fired power plants. New tech and solutions to generate and distribute electricity equitable shouldn’t be rocket science. Every South African should be allowed to contribute energy to the grid, managed by blockchain tech and partially subsidised through Eksdom’s diesel budget. Then we are all contributing towards our country bringing the power back to the people.
(Yes, very simplified… but it’s Martians we’re speaking to.)
REIPPPP bid window 5, issued in 2016, finally awarded bids in 2021. Thanks, Gwede. But now costs have increased so much that the projects aren’t viable. Most if not all cannot reach financial close.
My simple solution is for Government to offer a fair escalation on the bid prices. This will cover project cost increases and enable financial close.
There is no point in re-issuing the bid. The bid prices will go up in any event and all we achieve is a waste of time.
These projects are all ready to go. Let’s make an easy decision and get it done – and add the 2,600MW quickly.
Great analysing article, Tim – I finally have a better inkling of the problems within that opaque, mega-giant monstrosity known as Eskom. Can it really be this simple? But given that burning through diesel at the rate we have been doing is absolutely no solution, if, at the very least, it does away with this problem, it at least buys the engineers a bit of time to address the underlying maintenance issues, and most particularly enables our economy yo start to inch ahead based on better certainty of energy supply.
That same Martian would ask a key question; how is it humanly possible that the current President was appointed some 5-7 years ago to head up an “Eskom War Room” to fix the problem and achieved NIL. That same President simply has no idea or capacity to fix anything along with a bunch of incompetent “Ministers” sharing the table. The Martian would have figured this out in 5 seconds by replacing incompetence with people who understand and get things done!
Nobody denies that there has to be just transition from fossil fuels but you cannot transition from what does not exist. This is the core problem. You need to first have the coal and nuclear fleet functional and Ted Blom questions the notion of taking bicycle parts to fix a 50 ton truck which is what is happening at Eskom. We have a situation where basic principles enterprise wide risk management and operational risk management seem not to exist at Eskom and I have yet to hear about a Chief Risk Officer (CRO) in that organisation that procures very expensive parts and has critical parts of the power stations whose damage leads to breakdowns. It is for this reason that some of the reasons for the breakdowns are a legal basis to fire de Ruyter! When you listen to their press conferences any person with a coefficient of thinking would simply say for the reasons you are giving for the breakdowns you are fired for gross dereliction of duty! Eskom is not an IPP but a power utility to generate, transmit and distribute electricity from its coal and nuclear power stations and they are the primary responsibility of the board and the CEO. Any other garbage will not wash!
Essentially, the problem is political ideological, with vested interests putting a spin on it. If the government could let go of the reigns, just a little, this problem could probably be sorted out within a year.
Your pricing on solar panels is totally out – we currently pay R3,500 for a 540W panel – a lot less than you quoted