Is Eskom really doing more maintenance under De Ruyter?

André de Ruyter has prioritised the utility’s reliability maintenance recovery programme, but was Eskom really doing more maintenance during his tenure?

As the country reels under four days of constant Stage 6 load shedding (and counting), Eskom has been resolute in following its planned maintenance schedule, even though this removes additional generation capacity from service.

On Wednesday, it had 5 739MW of capacity offline due to maintenance when Stage 6 kicked in “until further notice”.

Eskom maintenance

It ran higher-than-normal maintenance throughout the festive period, with this peaking in the final week of the year at 17% of its installed capacity – equal to about 8 000MW.

While it is tempting to argue that maintenance ought to be deferred so that coal units can be run (even at reduced capacity) to keep the lights on, Eskom executives argue that this strategy will only create even bigger problems down the line.

It did however seem to – through all available coal units – in an effort to avoid load shedding on Christmas Day, as this article explains.

Last year, Eskom shared charts that show it cut overall maintenance of its coal fleet dramatically between 2016 and 2018.

Obviously, the energy availability factor (EAF; a measure of available generating capacity) of its power stations soared – to nearly 80% (from under 75% before this change in tack).

Plainly, this was unsustainable.

Impact

Some of the issues at Eskom’s fleet today can be traced back to the lack of maintenance five years ago.

Since his appointment in December 2019, outgoing Eskom CEO André de Ruyter has prioritised the utility’s reliability maintenance recovery programme, in an effort to turn around the performance of its ageing power stations.

This took some time to establish and implement; by September 2020, Eskom had approved all outage capex until the current financial year (FY2023).

Long lead times on spares (and the availability of funds) meant that certain projects could only be done years later.

Maintenance trend over time

Eskom’s own data from the last four calendar years shows that while there was a noticeable jump in the amount of maintenance in 2020 (following De Ruyter’s appointment), this has since slowly trended down.

In the year prior to him joining, the annual average for planned maintenance was 9.94% of installed capacity.

In 2020, this jumped to 11.24% before declining to 10.8% in 2021 and to 10.61% last year.

It must be noted that 2020 was impacted by the Covid-19 pandemic and despite Eskom being designated an essential service, certain maintenance projects were deferred.

Breakdowns

Breakdowns (the so-called unplanned outage factor) are up by almost 50% from around the 20% level in 2020 to nearly 30% last year.

This translated to a record year of load shedding, with power cuts on 208 days in 2022 (including near-continuous load shedding between early September and New Year’s Eve).

Eskom’s EAF (unaudited) was just 58% last year – the first time it has dropped below 60%. For the coal fleet, this number was below 50%.

Impressive, but …

On paper, the numbers for Eskom’s maintenance in the last few weeks of the year look impressive.

In the last week of November, only 9% of its fleet (on average) was offline due to maintenance. By the end of the year, this had nearly doubled to 17% (versus 13.8% in 2021).

In the first week of 2023, this above average level was sustained with approximately 14% of the fleet out of service because of maintenance.

Still, it is important to note that Koeberg Unit 1 began a long-term outage for its steam generator replacement on 10 December.

This removes around 900MW of capacity from the grid and, because of its size, will distort the planned outage factor; it is equal to between (nearly) one and four coal units, depending on the power station.

Koeberg maintenance

Also, Koeberg Unit 2 was offline for extended maintenance and refuelling for most of the year (between January and September), further inflating the maintenance figures for 2022.

Unit 1 will be out of service until midyear, whereafter Unit 2 will be taken offline for its steam generator replacement (a major project for its life extension, which was deferred by Eskom from last year).

Both units need to have this work done by 2024 in order for their operating licences to be renewed by the regulator. (This will also make the maintenance figure for 2023 appear higher than normal).

Medupi maintenance

Finally, a full 720MW from Medupi Unit 4 has been included in the planned maintenance figures from September 2021.

This unit was blown to smithereens in an accident in August that year.

Ordinarily, Eskom ought to count this as a breakdown but it shifted it to ‘planned maintenance’ as there is no prospect of this generating capacity being returned to service from what it euphemistically calls a “long term forced outage” before August 2024.

The distortion from Medupi 4 makes the maintenance picture look a lot better than it actually is.

One may contend that it is precisely because of the long-term outages at Koeberg that it didn’t have the headroom last year (and won’t in the next 18 months) to do as much maintenance on the coal fleet as it needs or wants to.

The figures however don’t lie…

Plan vs progress

In its last financial year (to 31 March 2022), Eskom completed 47 of the 84 outages under its reliability maintenance recovery programme.

A further seven were being executed at year-end, and it executed an additional 47 short-term outages.

One was cancelled and 29 deferred to this financial year due to “funding and capacity constraints as well as execution challenges”.

In the current financial year it had, by the end of September, completed 16 and was busy executing 13 of its 79 outages under the programme.

Seven were cancelled, 16 were deferred to within the fiscal, eight to FY2024, and 19 were remaining.

To turn around performance of its coal fleet, these numbers simply have to improve on those from prior years. We will know by May how well it has fared against its plan.

Regardless, because any gains from improved levels and quality of maintenance will only be seen in the years ahead, De Ruyter won’t be around to take the credit.

Perhaps the ‘new’ minister will.

This article first appeared on Moneyweb and was republished with permission.