On Friday afternoon (7th November 2025), after seeing a startling long-range change in colour in our display copy of the ‘Generator Outages’ widget in ez2view, we asked the question ‘What’s up with Waratah BESS (unplanned outage till 3rd May 2026)?’.
Follow-on articles here on WattClarity®
Following from that initial article, we started conversations with a range of different people – including:
1) Some people we know at Akaysha Energy, the owner/operator of the battery;
2) But also with a range of other knowledgeable NEM (and battery) stakeholders we though might help us answer some of the questions that data posed for us.
From these conversations came more information (and many more questions as well), as a result of which we’ve published a few more articles separately (and will be collating them here) to follow on.
1) Over the weekend these were:
(a) An updated longer term trend (and tabulation) of operations at Waratah BESS
… but the short story was that many questions remained unanswered.
2) This article has been written today to reference some external reporting we’ve seen this morning (Tuesday 11th November 2025), as discussed below.
… but still questions remain
3) And so there will be further articles to come (time permitting).
What’s Angela say in the AFR?
Perhaps inspired by our questions on Friday and over the weekend, it’s great to see a journalist (Angela Macdonald-Smith) has taken up these questions and, through some independent investigations to find what’s not in the market data, is able to report more.
Angela’s article made the front page of the print edition of the AFR:
… but the title of the online edition of the article ‘‘Catastrophic’ failure delays massive $1b Waratah super battery’ published Monday evening 10th November 2025 (to me) speaks more to the significance:
Key points seem to be in Angela’s article:
1) The failure occurred towards the tail end of the Hold Point 5 testing:
(a) I’d illustrated the Hold Point 5 testing in this article,
(b) In that article, I’d noted that availability had dropped back to 350MW from Saturday 18th October 2025 (which I presume now was around when the failure happened)
(c) But had not seen anything specifically speaking to a failure
2) Angela notes that ‘Waratah had suffered “damage to the transformer, and it is beyond repair”’
3) Angela also quotes another Akaysha team member as noting ‘A second transformer has been taken offline to test whether it “can be re-energised safely or may need replacement”’.
4) So this means that the unit is operating with only 1 of its 3 transformers:
(a) The ‘damaged beyond repair’ transformer will be months to rectify … which explains the May 2026 return to service date
(b) Whilst the one taken off for testing will be one also to watch (the ‘may (also) need replacement’ is the most concerning bit) … to the extent that anything’s visible in the market data:
i. Given it’s one single unit
ii. And opacity is further complicated by the approach mandated by AER for reporting MT PASA Availability when units are part-contracted to provide SIPS;
iii. Even moreso, when I could not find clear details on the specific timings and sizes of required SIPS service provision.
(c) Obviously I would not be alone in being curious about the cause of the damage to 1 (or possibly 2) transformers that mean that 1 (or possibly 2) transformers need replacing … and whether there are any implications for the 1 that remains in service (e.g. some sort of design or manufacturing fault that’s common to all)?
So all of this, together, starts to explain some of the questions we have been pondering about ‘What’s up with Waratah BESS’.
Reporting in PV Magazine
Whilst we’re checking in on any other reporting from elsewhere about the battery, one of the team came across the article ‘‘Catastrophic failure’ at Waratah Super Battery in Australia’ by Tristan Rayner today in PV Magazine online (and in the international version).
This article:
1) References Angela’s earlier article
2) And points to an earlier article ‘‘Catastrophic failure’ at Waratah Super Battery in Australia’ dated yesterday (in sibling masthead) from Tristan Rayner.
That article also references Angela’s earlier article in the AFR, and also writes that:
‘Although few details have been made public, third-party tools show that in early November 2025, AEMO’s Medium Term PASA (Projected Assessment of System Adequacy) data for the battery’s dispatchable unit (coded as ‘WTAHB1’) was updated. The new data shows a long unplanned outage lasting through May 3, 2026.’
… which seems strangely like something that might have come from referencing ‘What’s up with Waratah BESS’?
Update from Akaysha Energy
Checking the ‘recent news’ section of the Akaysha Energy website revealed the article ‘Update regarding the Waratah Super Battery’ also published today:
Copying in the rest of the article here:
‘The battery continues to meet its System Integrity Protection Scheme (SIPS) service requirements of 350 MW, which is the interim commercial operating capacity.
During ongoing Hold Point testing to transition the battery to full capacity, issues were identified with two transformers. There is no impact to the NSW electricity grid as the WSB remains operational and continues to meet its SIPS service requirements of 350MW.
The battery’s remaining capacity is expected to come online during 2026.’
… so this speaks to issues with two transformers (i.e. not just that one has been taken offline for precautionary testing, as I had been inferring from the above).
Spare a thought for Eraring?
It’s worth remembering that 19th August 2025 (now almost 2 months ago) was the date at which Eraring Power Station had earlier been slated to close.
It’s likely that the cooler and wiser heads that made the rational decision in May 2024 to extend the service life of Eraring at least to 2027 (resisting the hype that had been promoted by some) are quite relieved that they made the right call … not just because of these delays with Waratah BESS.
Not often a Wilson transformer fails – especially not 2 out of 3. I wonder if they were wrongly specified (eg can’t handle the harmonics), or damaged during testing? Be interesting to get more “inside knowledge”.