A question about (currently) planned outages for Eraring Power Station through until (currently planned closure) in August 2027

Most readers here will remember that it was on 23rd May 2024 that the announcement was finally made that the Eraring Power Station would remain open until August 2027 (i.e. a 2 year extension on a stated closure date, to that point, of August 2025) … though perhaps fewer will remember the caveat that the extension might actually be for as long as 4 years, to August 2029.

… we noted this at that time in ‘Eraring Power Station closure delayed by 2 years (or maybe 4?)’.

That’s roughly 6 months ago today.

I’ve had a question that’s puzzled me for a while – and wonder if some more learned readers here can help me out, please?

(A)  A look forward, in MT PASA DUID Availability

In our article on 23rd May, the first image included was one of the ‘MT PASA DUID Availability’ widget inside of ez2view, showing (to that point) what had been the expected trend of availability across the four units at the Eraring Power Station.

Given that it relates to my question, I thought I’d start this article with a similar snapshot of this morning’s snapshot from 09:30 (NEM time) for the Eraring Power Station:

Note that this was too early for the first data update from the AEMO (notionally for 09:00 but often actually published after that point) for this data set, so the current run we see is the update for 18:00 on Saturday 16th November 2024.

I’ve set the ‘previous run’ to be 1st April 2024 so that the prior closure date of August 2025 can be clearly seen.

1)  Obviously there are near-term changes in availability (such as due to forced outage)

… but that’s not the focus of this article

2)  Rather, looking forward from August 2025 to August 2027 we can see the aggregate profile shows two drops (but only 2?!) in availability for planned outages over that additional 2 year period:

(a)  There’s a lengthy planned outage for Unit 4 from September 2025 to November 2025;

(b)  And there’s a shorter planned outage on Unit 1 through April 2026

… shorter, but still (at ~4 weeks) still pretty significant.

We wondered about the extent to which these current plans for planned outage might continue to evolve over time … and specifically if there’d need to be any planned outages on Units 2 and Units 3, given the closure date is a little under 3 years away from now?

Our interest in the above question is through several lenses, including:

Lens #1)  This is the first case that comes to mind for me where a coal-fired station’s had a clearly stated closure date (presumably around which has been built maintenance plans – including outage plans visible in the 3-year MT PASA DUID Availability time horizon), to find that service life significantly extended:

(a)  in that sense (for instance) the hush-hush ‘safety net’ agreement struck between the Victorian Government and Yallourn is different, as that might not have changed outage plans noticeably within the MT PASA DUID Availability time horizon.

(b)  but it may be that we see more of these types of things in future, if delays in development of sufficient firming capacity continue?

(c)  I wondered how long it might take (i.e. was 6 months since the extension announcement enough?) to put in place outage plans for the extended service life?

Lens #2)  With a 2 year extension (but notably also a further 2-year option), my rusty engineering brain is interested to see how this type of arrangement flows through to outage plans.

Lens #3)  We’re also curious how this decision (and what might(?) be still-to-be-finalised outage plans) impact on hedge contract prices and the like …

(B)  Backwards and forwards…

So, with the above in mind, let’s take a look at a 14-year time series of Availability at each unit at Eraring Power Station … back ~10 years of actuals and forward ~3 years of projections.

All four units

We’ll start with this trend of aggregate availability across all four units from 1st January 2014 to 1st January 2028:

One of the first thing that jumped out at us in the above was the period of about 4 weeks late in 2016 when all 4 Eraring units

… for those who are unfamiliar, that’s quite unusual!

My outsourced memory failed me on that one, so we fell into a bit of a rabbit hole in investigating and so have posted this (incomplete record) belatedly.

Eraring unit 1 (2014 to 2028)

Now we’ll unbundle the trend above, looking at each unit in turn – starting with ER01:

This unit had a major outage in the second half of 2022:

1)  122 days in total (including a failed RTS), with the unit back online from 19th December 2022

2)  a quick scan of WattClarity revealed no articles specifically tagged with the station from around that time that might have explored this outage

3)  but I don’t have time to fall into another rabbit hole today.

We noted above that this unit has an outage (roughly 4 weeks) planned from 3rd April 2026 … which would mean 3 years + 4 months between ‘longer’ outages.

Eraring unit 2 (2014 to 2028)

Here’s the same trend for Unit 2:

This unit’s last ‘major’ outage late in 2023:

1)  For 82 days in total

2)  Back online for 24th November 2023.

With no planned outages on the horizon out till the mooted closure date of August 2027, (and skipping over the 23-day outage to 25th May 2024 … which perhaps we should not?) that would be a duration approaching 4 years at closure.

Eraring unit 3 (2014 to 2028)

Here’s the same trend for Unit 3:

Note that this unit’s currently on outage:

1)  Since 26th August 2024; and

2)  With expected return to service Friday 29th November 2024.

There’s no further outages planned in the ~3 years from that point until mooted closure.

Eraring unit 4 (2014 to 2028)

Here’s the same trend for Unit 4:

We see that:

1)  the last outage of length was in April 2024:

(a)  41 days in total

(b)  Back online for 1st May 2024

2)  … and we noted (above) a planned outage (from September 2025 to November 2025)

3)  So, after that, a ~2 year period with no outage planned till closure in August 2027.

(C)  When were those outage plans added?

Utilising the ‘Forecast Convergence’ widget in ez2view we can see that the slated outages for ER01 and ER04 prior to closure in August 2027 were added into the

Ordinarily we’d note that this widget can be used to ‘look up a vertical’ in order to ‘see that other dimension of time’ … but, in this case, it’s the horizontal that’s more interesting.  As highlighted, it shows that the two outages added for the future were added into the AEMO data set on Friday 13th September 2024 :

1)  That’s approaching 4 months after the extension announcement.

2)  During that interim period, I imagine the generator (Origin Energy) would have been busily updating their asset management plans are arranging the necessary resources (people and plant) to complete the required outages.

3)  As noted above, I wonder if there are any other planned outages that are not yet finalised but which will pop up in the MT PASA DUID Availability data set in the weeks ahead?

With the above in mind, we’ll keep watch on how this changes in the future…