Never mind winter, load shedding is likely to get much worse from September

And Eskom’s ‘mid’ scenario sees 37 days of load shedding until the end of August …
Eskom’s CEO maintains that ‘we should not accept load shedding as the new normal’. Image: Shutterstock

Despite the nation’s mood being one of “misery”, as one Eskom executive termed it yesterday, the utility’s top management is confident that load shedding will be limited in the coming winter months.

This is mainly due to the fact that three large generators will return from maintenance outages between now and the end of next month.

Read: Eskom is burning  more diesel than ever to keep the lights on

Two units at Kusile Power Station (1 600MW) are expected back online in the “next few weeks” while Koeberg Unit 2 (920MW) is scheduled to return from its refuelling and maintenance outage by the end of June (although Eskom Group CEO André De Ruyter said “late June/early July” when referencing this in his opening remarks).

It also generally sees better performance from its coal fleet in winter months when temperatures are lower. This will help.

And because the demand profile looks quite different in winter with far steeper morning and evening peaks, when there is a generation shortfall, load shedding will likely only be required during those times (this is consistent with what we’ve seen this week, with load shedding implemented between 5pm and 10pm daily since Monday).

De Ruyter was defiant on Wednesday, saying “We should not accept load shedding as the new normal” and that he and his executives “are seized with the matter”.

New capacity

Key to turning around the coal fleet is the 4 000MW to 6 000MW of new generation capacity that needs to be brought on stream “yesterday”.

COO Jan Oberholzer highlighted that this was his answer to President Cyril Rampahosa in their first meeting on December 11 2019.

Not a single megawatt has been added to the grid under Ramaphosa’s administration.

While there has been some movement, no new projects have reached financial close, and some never will (as the investment cases no longer make sense). This new capacity will give Eskom the headroom it needs to do proper maintenance on its ageing fleet. Until then, it can’t as it is forced to operate units that should not be running just to keep the lights on.

The outlook is not positive.

Base, mid and worst-case scenarios for winter

The winter base case scenario of the utility’s transmission unit (which is in its final stages of separation in the unbundling process) saw no load shedding with only R1 billion spent on diesel for the open cycle gas turbines (OCGTs) until August.

Segomoco Scheppers, Eskom’s managing director for Transmission, admitted yesterday afternoon that clearly this is no longer the case. Since April 1, the country  has seen more than 10 days of load shedding.

This base scenario is predicated on up to 12 000MW of capacity being unavailable (due to outages). On Monday as it announced Stage 2 load shedding in the evening peak, Eskom said 15 943MW was unavailable due to breakdowns, on Tuesday this was 15 762MW, and on Wednesday it said this was 15 988MW.

Load shedding crisis as Eskom breakdowns hit record levels
Winter worries: Eskom defers Koeberg steam generator replacement to August 2023

While outages ought to improve in the coming months and planned maintenance is deliberately reduced to levels below 3 000MW, it is not likely that the base case is realistic for the whole of winter.

Scheppers said the transmission unit’s plan is “tight” and that there is inherent uncertainty.

He highlighted that it requires OCGT usage over weekdays and that the failure of Medupi 4 (which exploded last year) has increased the dependency on diesel generation to manage the power system.

In other words, the Medupi explosion = more diesel needed.

More likely as a ‘base’ scenario – certainly based on the last six weeks – is the operator’s second scenario which sees (on average) up to 13 500MW being offline due to breakdowns until the end of August. Here, load shedding up to Stage 2 will be implemented on 37 days. Diesel costs will be above R3 billion.

The current worst scenario as per the transmission unit’s plan is predicated on 15 000MW of generation being offline due to unplanned outages across winter. Here’s load shedding will be required on 104 days (the plan covers 153 days between April 1 and August 31) and the highest stage of load shedding will exceed Stage 3. Diesel costs will total more than R7 billion.

In the mid scenario diesel spend is between R500 million and R600 million a month, with load shedding on between six to nine days each month. In the worst case, this jumps to at least R1.2 billion a month (and as much as R1.8 billion) with load shedding on 19 to 22 days each month.

Brace yourself for a rough summer

Scheppers would not comment on the published outlook for summer saying that it would be revised based on the performance in winter. But the current base plan for summer (September 2022 to March 2023) sees 16 days of Stage 1 load shedding with total diesel spend of R5.2 billion. This is predicated on the level of unplanned outages being 13 000MW.

At 14 500MW of breakdowns, perhaps a more realistic number given performance in the recent past, the plan sees 132 days of load shedding (in a 212-day period). The highest stage will be higher than Stage 3 and diesel costs will total R16 billion.

But Eskom cannot physically take delivery of this much diesel to burn. It says it is not possible to use more than about R1.2 billion in diesel a month. With the recent spike in global prices, one may assume this figure is now R1.5 billion.

The problem is that the transmission unit’s plan in the mid scenario requires 23 days of Stage 2 load shedding in November at a cost of R2.5 billion and a further 15 in December (effectively every day until the country shuts down mid-month) at the same cost.

But this is not possible, which means an almost guaranteed diesel bill of R1.5 billion in each of those months and higher stages of load shedding (Stage 4, perhaps?).

In January, the plan requires R3 billion for diesel (twice as much as is physically possible) and the scenarios in both February and March are similar to November.

The worst-case scenario (modelled on unplanned outages being at the 16 000MW level) is, well, much worse. Here the numbers stop making sense from September. A total of 191 days of load shedding out of 212 with upwards of R4 billion required for diesel each month from November until March.

That is simply unworkable – even if Eskom had the budget to pay for as much diesel as it wanted (it doesn’t).

Unless its OCGT plants are switched to LNG this year still (which, in theory, means it can burn cheaper gas and more of it) or there is a miraculous turnaround in the performance of its coal fleet, we are all but guaranteed that load shedding will be a near-daily occurrence from November.

De Ruyter had to leave Wednesday’s state of the system briefing early due to an appointment with National Treasury and the Department of Public Enterprises. Those present at the virtual briefing were told any announcements from those engagements would be made when appropriate. Could there finally be some movement on the $8.5 billion in funding secured by the country at last year’s COP26 talks?

Still, those projects are going to take time … and September is around the corner.


Sort by:
  • Oldest first
  • Newest first
  • Top voted

You must be signed in and an Insider Gold subscriber to comment.


Not a single megawatt has been added to the grid under Ramaphosa’s administration.

So much for all the Vulindlela and other “plans” they always have !!!!!!!!!!!!!

Dear Mr Casper1:

You are constantly missing the point of ANC campaigns like Vulindlela, The New Growth Path, Youth Build, Mine Rehabilitation Programme, the Reconstruction & Development Programme etc etc.

These are never intended to create any change but are simply catering opportunities for wives of cadres to make some money on the side by employing F&F – Friends & Family at the lavish conventions announcing these grandiose schemes.

Like the plan to spend R22 million on 100-metre tall South African flag.

Sigh ….

“Not a single megawatt has been added to the grid under Ramaphosa’s administration.”
Actually factually incorrect,several heavily delayed projects of BW 4 of REIPP projects have come online over the last few years. They are not massive, in the greater scheme of things, PV and wind, without storage. These never started in 2015 or 2016 because of the halt brought to it by the Rosatom nuclear deal. Only restarted in 2018.

*under* not *during*
Those BW 4 projects were announced in April 2015.
Ramaphosa took office in Feb 2018. Today it was announced that the financial close deadline for the projects selected in BW 5 (announced in October 2021) would be extended to September. So from Feb 2018, it will have taken the Ramaphosa administration 55 months to get the BW 5 projects to commercial close (presuming they actually get there – unlikely).
At this pace, they might all be built by the end of his second term!

@ Hilton, indeed these REIPP projects are moving at a pace of a 95 year old snail, with Parkinson and long COVID.
CR should never be allowed a second term as Prez of SA, we need a complete regime change come 2024 elections.
We need at least two BWs of 5 GW each yearly to get anywhere. And the whole bureaucratic process should be immensely be simplified and not take more than 6 months, for permits, FC, EIA and all the rest.
All requirements for BEE, community projects and local content should be scrapped ASAP. Size should not be limited to 75 for PV, and 140 MW for wind. There should also be room for projects which include BESS, and CSP+TS, Concentrated Solar Power with Thermal Storage. Power produced during peak hrs should be priced higher. Eskom has to immensely hurry up spending R 100 B at least in extending the grid, especially to the NC. Maybe the R 40 B that will be spent on FGD at Kusile, is better used for improving the transmission system. Many countries intend to close coal fired power generation by 2030, SA should close even Kusile and Medupi not later than 2035.


100% agreed. Problem is the FGD at Kusile was part of the conditions of the World Bank financing. Contractually, the plant should already have it – that’s what a mess this is!

Mistake: mixed up Medupi and Kusile. Kusile is the only power station with FGD. Medupi was supposed to also get it, but for some reason it had to be fitted as a very expensive afterthought. Medupi, near Ellisras, Lephalale is in a dry area, and apparently the water supply for these wet scrubbers will already cost R 5 B.
I am all in favour of measures to reduce environmental damage by these old dinosaur power stations. Apparently R 300 B would be needed to equip all existing coal fired power stations with Flue Gas Desulphurisation and particulate filtering, the two legally required methods to clean emissions from coal fired units. Not even talking about CCS, carbon capture and storage. Clean coal doesn’t exist, and trying to achieve it will cost a fortune.
Too late off course for these measures, better to decommission these coal dinosaurs ASAP.

Why would Eskom be forecasting a GW more of unplanned outage in summer than in winter?

The entire notion of forecasting unplanned outages a year ahead is like something out of a Monty Python movie.

The ANC remains intransigent, refusing to implement the obvious solutions to remedy their disastrous policies. They ignore all the facts and doggedly cling to their failed ideas because those failed policies are keeping them in power. The overpaid unionized workers at Eskom, and other failed government entities, as well as the privileged BEE beneficiaries at coal mines, coal transport companies, Eskom supply lines, and Eskom contractors, form the lobby group that determines the ANC’s future.

Our national economic policy is determined by labor unions and BEE parasites. Both these parties are consumers and not generators of capital. Entrepreneurs who act according to the rules of the free market have to take risks and employ capital to finance the consumption habits of these parasites. Our economy has reached the stage where socialist policies and ANC parasites have destroyed the basis of economic activity. The implosion of the national grid constricts economic activity and marks the stage where the parasite has killed the host.

We hear all this talk about ESG investing. The first responsibility of any investor is to himself and his family. Responsible investors have to ask whether this country is investable at all. If we follow the high-net-worth individuals, the answer clearly is no.

At this rate, the parasites will soon parasitize themselves, as they do in Zimbabwe. The rising insurrection within the ANC and the escalating factionalism show that this process has already begun.

The devastation of SA’s SOE’s will take a long time to repair,if it at all possible.

I think it will be easier and cheaper to salvage the Titanic.

End of comments.




Instrument Details  

You do not have any portfolios, please create one here.
You do not have an alert portfolio, please create one here.

Follow us:

Search Articles:
Click a Company: