Thread by @_richardblack

Initial reflections on this morning's second, more detailed @ng_eso report to @ofgem on the 9th August #powercut cut

Confirms many things in the initial report, including timeline, lighting strike as original cause and fact that three separate sets of generation went offline. Provides more interesting details in several key areas

Hornsea 1: confirms reason for it taking itself offline was unexpected reaction to voltage fluctuation on NG connection due to lightning strike. Ref to "insufficiently damped electrical resonance" & graph suggests control system somehow amplified fluctuations, causing switch off

Little Barford: confirms initial failure was on steam turbine due to apparent mismatch in speed sensor readings. This shouldn't have happened. But main report gives no reason for it. Suggests it wasn't disruption to station's own electricity supply... so, what was it?

New point: suggests the two gas turbines should have been able to keep working: '...for reasons presently unknown, after approximately 1 minute the first gas turbine tripped due to a high-pressure excursion in the steam bypass system.' Inspection due to discover reason

This is important because if the two gas turbines had stayed on, there probably wouldn't have been a #powercut (see graph)

Third loss was embedded generation. Lots detail here. 1) The very first set of generation to trip out was embedded generators 'protected' by vector shift (150MW) - remainder due to over-sensitive RoCoF. 2) suggest that an additional 200MW dropped out as frequency fell below 49Hz

"Protection operating at this frequency was not expected and has not previously been observed".

I discussed the fact that the RoCoF/VS problem has been known about for over 10 years and not been fixed in a blog yesterday eciu.net/blog/2019/powe…, and suggested the report might highlight it. And here's the report's third recommendation...

"Review the timescales for delivery of the Accelerated Loss of Mains Change Programme to reduce the risk of inadvertent tripping and disconnection of embedded generation, as GB moves to ever increasing levels of embedded generation"

Highly relevant, because... if this set of generators hadn't tripped out, again, we probably wouldn't have had a #powercut

There's more on trains that's worth reading but in terms of the root causes of the power cut, it seems to come back to this:

The problem at Hornsea 1 was new and unexpected and has been fixed. Little Barford should have stayed on but didn't, due to a fault that is been investigated. But the embedded generation issue has been known about for a decade and hasn't been fixed

Final point - as regulator, @ofgem has had oversight of this issue, the damage it could do and how it hasn't been fixed. But it is the one party in this whole issue whose role is not been investigated