Yesterday (Friday 10th April 2026) amongst other articles we wrote:

1) about ‘The curious case of a frequency spike (outside of the NOFB) on Friday 10th April 2026’, with the key bits being:

(a)  ‘The highest point was (at 50.174Hz) was at 10:30:41.5 NEM time on Friday 10th April 2026’ and

(b)  ‘the excursion was focused in the 10:30 (excursion) and 10:35 (recovery) dispatch intervals’.

2)  we then noted ‘Four big Non-Conformances at Wind Farms (coincidentally on Friday 10th April 2026 at the time of the frequency spike) … because surely it seemed that there were … Some gremlin in the works?’

Today (Saturday 11th April 2026) we can utilise ‘next day public data’ in the ‘Trends Engine’ of ez2view to quickly adjust the previously configured query to present this trend of ‘Dispatch Error’ by Fuel Type across all 288 x Dispatch Intervals across the day:

As per the annotations on the image:

1)  With such a massive collective over-performance across the Semi-Scheduled wind fleet in the 10:30 dispatch interval (NEM time) we surely have seen the ‘smoking gun’ pointing to the main reason why the frequency spiked in that dispatch interval

… but the more interesting question is why?

2)  Worth also noting that (as we have seen numerous times) it’s the BESS fleet and the Coal fleet that collectively do the most (via a combination of PFR and Regulation FCAS) to try to keep a lid on skyrocketing frequency.

Returning to the ‘Trends Engine’ of ez2view we then compute ‘Dispatch Error’ for each of the ~80 x Semi-Scheduled Wind Farms operational in the NEM on Friday 10th April 2026 … for each of the 288 dispatch intervals in the day – though it’s these two Dispatch Intervals of most interest …

The 10:30 dispatch interval

… with the frequency spiking above the NOFB

Here’s a view of the calculated  ‘Dispatch Error’  at 10:30 (NEM time):

Briefly:

1)  There are 7 x Semi-Scheduled Wind Farm units with levels of over-performance (relative to Target) greater than 50MW

… click on the image to open large-resolution view to read the DUIDs.

2)  Note that:

(a)   4 of these were the units that received Non-Conformance notices (communicated through Market Notices noted here) are all identified on the chart

(b)  but that clearly means there were some other contributors that did not receive Non-Conformance notices

… something to investigate another time.

3)  There are another 3 units with over-performance larger than 25MW (but under 50MW) …

(a)  which means 10 units in total over-performing by 25MW or more

(b)  and there’s not a single unit with an under-performance greater than 25MW

(c)  which is also notable, give (as Allan described well here) the one-sided nature of the Semi-Dispatch Cap means large under-performance (ordinarily!) is much more common.

… so we really wonder what’s happened here!

4)  Of these 10 x units, on quick scan:

(a)  Nine of them are in VIC, I believe

(b)  One in South Australia

… which makes me wonder, more, what has happened here (we have a theory).

The 10:35 dispatch interval

… with the frequency recovering back within the NOFB

With the frequency recovering in the 10:35 dispatch intervals, I’d not expected to see any units with large over-performance in this Dispatch Interval … but that’s not the case here:

These units are in South Australia.