The Broken Hill incident in Australia : More problems from renewables

By Jon C

In 2018, Broken Hill City Council announced its goal to become Australia’s first carbon-free city by 2030 (1).

In October of 2024, the isolated small city of Broken Hill and its surrounding communities in New South Wales, Australia suffered a grid failure when its main power line was brought down by a storm (2).

A week after the event, electrical supplies were still not reliably restored. (you can see the time-line in link 3 below). Reliable supply was only restored once Broken Hill was reconnected to the main New South Wales grid (with all its conventional machinery) by a series of temporary pylons.

Once Broken Hill was re-attached to the main grid, in November, the government of New South Wales released a statement in which it trumpeted the successful restoration of the power line and hence the grid (4).

What surprised many about this was that Broken hill was the site of vast “green energy” installations which generated many times the power consumption of Broken Hill.

A survey of multiple sources shows that Broken Hill had a maximum demand (also called load) of 36MW (including the large nearby mines), but the local grid could not be restored despite the presence of 200 MW of wind (5) , a 53 MW solar array (6), significant residential solar, and a large “grid scale” battery with a capacity of 50MWh capable of delivering 50MW (for one hour) (7). This battery was running by the time of the 2024 outage and was claimed by the installer, AGL, to be a reliable backup power source for 10,000 homes in the event of an outage. Unfortunately, at the time of the outage it was also discharged (and thus not being used for its “back-up” function) and needed to be recharged.

Note that the nominal generation capacity of 200 + 53 = 253MW is seven times the 36MW load of the town and mines and this does not take account of the roof-top solar which amounted to yet more “generation capacity”.

To add insult to injury Broken Hill also had a pair of old gas-turbine generators (8). According to reports, these were poorly maintained (being considered redundant) and at least one was inoperable at the time of the outage and it appears that for “technical reasons” (probably poor maintenance) the other could not be used either. So the only restart possible was from renewable generators. Had the gas-turbine generators (total 50 MW) at Broken Hill been properly maintained, these, being conventional generators, would have been able to re-start the grid.

A common part of the renewables narrative is that wind, solar and batteries have been shown to be able to support the grid and even to restart a grid by the use of “grid-forming inverters” (GFIs) (more on these later) as are used at the Broken Hill battery . The inclusion of advanced GFIs (AGFIs) at Broken Hill was an acknowledgement of the towns vulnerability to storms and the intent was clearly that such AGFIs would enable the formation of a “micro-grid” or “energy Island” to power Broken Hill and its communities in just such an event.

In reality, wind, solar and batteries all connect to grids that are supported by conventional generators (nuclear, gas, coal, hydro, geothermal) which use spinning machinery (the turbine-generators). These types of generators, which weigh many tonnes, have a lot of inertia (or momentum) and provide support to the grid in terms of frequency and phase stability (I’ll deal with this in a later post, for now please accept that this is really, really important), ramping and balancing (responding to changes in demand – and fluctuations in renewable supply) and voltage support (to prevent brown-outs, where many devices stop working even though there is still some – lower voltage – supply).

The 2024 Broken Hill outages (and there were repeated outages as attempts to re-start the grid repeatedly failed) demonstrates the inability of renewables, the total investment in which around Broken Hill was in excess of $A650 million (Australian dollars) for the 200 MW wind plant, a 53 MW solar farm and the 50MWh battery, the latter built with AGFIs, to restart a grid.

As stated above, the inclusion of the (expensive) AGFIs in the battery plant was clearly intended to allow a grid re-start, initially on the domestic load, and then as other renewable generators joined in, the mines could also be reconnected and all would be well.

The function of inverters, of which GFIs or AGFIs are a sophisticated (complex) type, is to change direct current, DC – the sort of electricity you get coming out of batteries, into alternating current, AC – the sort you have for your “mains” supply. GFIs are, in theory, capable of operating to establish a steady frequency and phase without external electrical supply. (This is currently not the case for many of the inverters used in most renewable installations). In the case of Broken Hill, these AGFIs failed to function as advertised.

One contributory factor to the problem was the large number of roof-top photovoltaic arrays on peoples’ houses. These continued to (try to) feed electricity into the local grid and, according to reports, impeded the restitution of the grid. These had to be shut down (a house-by-house) exercise before the later attempts to re-start the grid.

The Australian (behind a paywall again) ran an article entitled: Broken Hill: Powerless and left to live like mushrooms where it described the situation as expressed by a resident of Broken Hill: “The power comes on from time to time, but goes out just as quickly. It gives us just enough time to power our phones and read emails from energy providers sent the day before, alerting us to the fact the power was about to go out. They also warn we don’t have much time, and to avoid using unnecessary electrical devices – air conditioners, fridges or fans that need a power point.” The consequences were severe. Pharmacies lost their medicines due to refrigerators failing, peoples’ home fridges and freezers failed as well, so all frozen and chilled food had to disposed of, and vital medical equipment could not be used. The situation became so bad that emergency generators and food had to be trucked in (9).

The Australian also called the blackout a “green power warning”. The article cautioned that policy makers should learn from this experience that renewable resources are “almost useless” without conventional power. Broken Hill’s Solar panels were not just useless, they actually hindered efforts to establish reliability and customers were urged to turn them off. “(Wind and solar) are worse than useless (in a crisis like this), because it’s detrimental to having a consistent power supply”.

A quick search of the internet shows examples where various groups state that renewable electricity supplies all of an area’s power (10). These descriptions, whilst accurate (when figures are given) to the extent of how many TWh (TeraWattHours) etc. were generated and consumed are, in the kindest interpretation, incomplete because they ignore the fact that this was done when those renewable sources were linked to a grid containing conventional (spinning) generators that were and are essential to supply grid reliability services and maintain stability. In the UK, for example, the Scottish “all renewable” grid (10) was continually supported by nuclear, hydro and gas generators supplying those essential services.

To date (Jan 2025) no-one has demonstrated that renewables can fully support a grid, never mind re-start one, without the contribution of conventional generation.

Green energy advocates, whether lay or academic, don’t discuss, and I suspect many don’t even think about, the crucial point as to whether a grid can survive without rotating machines with their built in inertia which provide those essential grid support functions. Without this extra information, the claims about “green grids” really mean little or nothing.

Green energy advocates are aware of the intermittency issue (as is the general public of course which is why green advocates address the point). Typically they will claim is that batteries paired with these solar and wind sources can support the grid by providing energy when it was needed – as was stated by the builder of Broken Hill’s grid-scale battery (7).

In Broken Hill, the problems encountered in re-starting the grid were not related to intermittency – recall that residents were told to turn their roof-top photovoltaic arrays off. There was many times the load of Broken Hill available in renewable generation, but the energy from these disparate sources could not be reliably integrated into the grid, resulting in repeated collapses and renewed blackouts.

As the Broken Hill debacle showed, having vast renewable resources means nothing when they are not connected to a stable electrical grid.

Putting this another way: these events show that just having enough MW (or GW) of generation in place does not mean that demand can be reliably serviced.

The real problem is that wind, solar and batteries do not readily provide the essential reliability services mentioned above. Wind, solar and batteries provide energy through electronic inverters and to date even AGFIs are not capable of simulating those services.

Earlier I skipped over a few of the key points in grid operational Physics.

A key concern, at least to those tasked with ensuring reliability of supply, is how do you get solar and wind work better with the grid and at least maintain reliability and stability. As Broken Hill shows even when the system includes “advanced GFIs”, an “all renewable” grid is very unstable, prone to collapse and – at least in this case – impossible to re-establish without connection to a conventional grid thus the technical challenges of operating a grid on “green energy” alone appear insurmountable using current technology.

Renewable resources, whose outputs fluctuate continuously, are inherently variable and unstable, unlike conventional generators which have a lot of inertia (resistance to a change in their motion) due to all their rotating machinery (the turbo-generators). In addition, the power output of conventional generators is essentially controllable, unlike that of wind and solar which can be affected moment-by-moment by changes in wind speed or direction and cloud cover respectively. For a rough analogy, think about the last time you flew. A Jet Engine is a gas turbine. On take-off their power output is ramped up to their maximum (they get noisy). Once airborne the power of the engines is reduced to a “cruise” setting. In essence, something similar takes place with conventional generators. Turbo-generators are seldom run at maximum power. In turn this means they have a physical reserve to meet a sudden surge in grid demand and, even at maximum power, their inertia means that they can withstand an overload for a short duration without a significant change in speed which maintains grid frequency and phase (two vital factors for reliability).

I should add that there is a very good reason why conventional generators are not run at maximum power, when they are they tend to be much more likely to “trip out” (i.e. disconnect from the grid) precisely because they have no reserve to cope with changes in demand.

Renewable resources do not have these features. Consequently, as an ever greater percentage of power is generated using them grid stability and reliability inherently falls. And the problem gets exponentially worse as more and more of these unstable resources are tied to the grid. Many policy makers, academics and others seeking to increase wind and solar fail to recognise this problem and also do not realise the (current) limitations of GFIs and AGFIs, which, if Broken Hill is anything to go by, simply cannot do their advertised job. Instead the “green advocates” put their trust in technical and technological innovation. To be fair, these may indeed solve the problems at some time in the future, but at present these are either not working as advertised or simply not available.

As Russ Schlusser, an American writer with decades of experience in this field notes regarding the Texas grid: “(E)ngineers, academics and scientists jointly grapple with the critical such as providing synthetic or virtual inertia through inverter technology to aid the Texas grid. There is some hope that advanced computer controls can be developed so that asynchronous resources perform similarly enough to maintain the grid at higher penetration levels. It should be recognised that the talk is of possibilities not probabilities. Here the National Renewable Energy Laboratory concludes ‘Ongoing research points to the possibility of maintaining grid frequency even in systems with very low or no inertia’. The unsaid part of that statement is that it may not even be possible to maintain grid frequencies with low inertia. It’s also certainly in the mix at this point, based on the statement from National Renewable Laboratory, that in the next 20 years the best we may be able to do at higher penetration levels of asynchronous renewables is maintain frequency in a highly inferior manner with a boatload of reliability problems, with increasing blackouts at untenably high prices.” (11).

This fundamental problem has yet to be resolved (and may yet prove insoluble): How do you provide these essential reliability services when integrating high percentages of wind and solar power (and their “supporting” batteries) into a grid?

In the UK, the National Energy System Operator (NESO) has no idea as yet how to answer this question and it has six years, if the current Government has its way, to find and implement an answer.

Links. Correct as on 09/01/2025