Is affordable, 24/7-reliable, electricity ever possible again?

Current policy is for 100% renewables plus 100% battery back-up. That will deliver even more unaffordable and unreliable electricity.

Power bills are already soaring. These exclude extra costs unknowingly borne by taxpayers.

Even as reality starts to take hold, it's going to get a lot worse.

The rush to renewables generation is hitting installation and transmission/distribution headwinds. Base-load generation has been shut down well before alternative power supplies are in place. New plans to delay closure of those still operating is very expensive. The power cost trend is up.

In the NEM, renewables-heavy States depend on inter-State interconnector 'extension cords'. Other States' base-load and fossil fuelled 'peaker' generation are used to keep the lights on when their solar, wind, and hydro don't. South Australia, Tasmania, Victoria, and the ACT are net base-load importers. So far, Queensland and New South Wales are net base-load exporters.

Check out spot wholesale electricity prices over the recent past.

What happens when all NEM states are 100% reliant on renewables and battery storage?

For power storage, grid-scale batteries needed for anything remotely like 24/7 reliability would cost a fortune. See my opinion pieces (here, here, and here).

Can electricity become affordable and 24/7-reliable again? Consider these proposals.

Add more renewables. Power users' demand, whether continuous or intermittent, can never match renewables supply intermittency, seasonal or otherwise. Most extra power would need battery storage to be usable. Cost? Time to deliver? Beyond 2050? A renewables 'fail'?

Use EV batteries to avoid blackouts. How? Recharge EVs when sun and wind renewables allow? Redirect stored EV power to the grid when they don't? This would 'un-firm' EV owner investment expectations, at best. Time to deliver? Long (see below). Another renewables 'fail'?

Why? EV battery supply is paid for by EV owners themselves. That's their choice. Assume the average EV dispatches 100kWh from peak charge to zero. Isn't this wildly optimistic? But, if so, to dispatch 194,000kWh, the capacity of the SA Hornsdale 'big battery', would require 1,940 EVs.

In 2022-23, the NEM supplied 520,000,000kWh per day on average. Suppose Australian EVs grow to 1,000,000 by 2027 (an electric vehicle association target). Net of EV battery usage for owner-transport, assume, on average, 40% of EV battery storage is available for diversion to grid support. Way too optimistic? That's 40,000,000kWh, or about 7.7% of daily NEM power supply.

Not so fast.

If grid support is demanded of EV owners who have paid for their EV batteries, how will they react? One future buyer response is refusal to purchase EVs. That would undermine ambitious future EV targets. For those still buying or owning EVs, would they unplug them from their chargers when they expect their own stored EV power to be commandeered for grid support?

Like other lithium-ion batteries, do EV batteries last about 10 years? Must EVs be replaced every ten years? Who buys the old ones? Is resale value of old EVs now very low because replacing their batteries is costly? Isn't the second-hand EV market showing this?

Could very heavy restrictions on motor vehicle ownership and use be needed to make this idea work? Then Cuba-style ageing of fossil fuel cars as owners use them for longer to avoid EVs?

Chemical battery developments beyond lithium-ion. There's lots being spruiked in this area. I've already assumed battery storage costs for electricity fall to just 10% of SA's Hornsdale 'big battery'. Happy to react as new technical breakthroughs emerge. When? Time – possibly a long time – will tell. We can't, for sure.

Renewable solid gravity batteries. This notion is mechanically-intensive, physically large, and, I suspect, very energy-negative in net terms. Google it. Are proponents serious? In my opinion, this is a fanciful notion, with little chance of being a power storage device at any scale.

Real-world gravity battery options. We do have water-based gravity batteries operating today. The Guthega/downstream turbine generator facility in the Snowy Hydro system is one example.

Upstream water from the Snowy river is allowed to fill the Guthega Pondage during (human) off-peak demand periods. Excess inflows are allowed to flow unrestricted over the dam. Dam storage can be gravity-discharged, feeding downstream power turbines, during on-peak electricity demand (typically in the evening). That can be profitable for the Snowy scheme, via high peaking power prices. As a 'peaker' generator, it also softens prices downstream for users.

Win-win? In suitable locations, with reliable water flow, yes. Can be a long-lasting option too. Guthega Dam was completed in 1955.

Without enough free water, however, hydro power supply is cut off. As in Tasmania now.

Net zero gravity battery options. 'Pumped hydro' relies on releasing water from an upper dam (where potential energy is stored) to a lower dam (generating kinetic energy through power turbines as it does so). That water is then pumped up again to the upper dam, and the process is repeated as power is needed.

Too good to be true? What do the physics tell us?

There's no such thing in the real world as a perfectly efficient machine. Efficiency losses are always a practical reality. For 'pumped hydro' systems, where renewables pump the closed-cycle water uphill, these losses are large. Is the all-up net energy supplied less than 20-30%?

There are energy density realities here. Using USA measures, gasoline power is one billion times more energy dense than wind and water power. It's ten quadrillion times more energy dense than solar power. To store the energy contained in 1 gallon of gasoline requires over 55,000 gallons of water to be pumped up 726 feet (and that's assuming 90% recycling process efficiency).

Whatever the efficiency of a closed 'pumped hydro' power system, using renewables to pump water uphill will require massive extra renewables generation capacity. That means much more demand pressure on the already-stretched grid in the NEM.

And yet a major 'pumped hydro' scheme is being developed within the Snowy Hydro system. It's called 'Snowy 2.0'. Other such schemes have been proposed by politicians elsewhere.

Snowy 2.0 is already way over the initial budget, up from A$2 billion to A$12 billion and counting. Almost certainly, costs will rise further. It's also way behind schedule. First power to the grid was supposed to be produced in 2024. 2030 now seems very much a best-case scenario. The tunnel borer (named Florence) developing this system seems regularly to get bogged and hasn't got very far.

And what could Snowy 2.0 do if it eventually operates as planned? The claim is it can power 3 million homes for a week. Every single week all the time, in all seasons and all weather? I don't think so. If powered by renewables, it will need time off for intermittent power to pump water uphill again. Or would Snowy 2.0 need many, many more extra batteries to store intermittent renewables power?

Proponents call Snowy 2.0 nation-building. I don't agree. It's very, very high-cost power at best.

Demand response. This is a euphemism. It's electricity rationing. It does not make power more affordable or more reliable. It makes it unavailable. It can start today. It already has. This is renewables power supply failure in its clearest, arguably most painful, form.

'Green' hydrogen? Electrolysis to produce hydrogen requires lots of very pure water (9kg for every 1kg hydrogen). Water purification requires lots of energy. 'Green' electrolysis requires huge inputs of extra renewable energy. Yet more power demand pressure on the grid?

If 'green' hydrogen can be produced and used intermittently, very large increases in renewables generation capacity will still be needed, even if storage requirements are reduced.

If hydrogen use is for continuous manufacturing processes (eg, 'green' steel), many, many more batteries will be needed for storage, too.

In either case, downstream handling of hydrogen (compression, refrigeration, transport, distribution, etc) is complex and costly. Net energy delivered likely will be very negative. Time to deliver? Long at best. Beyond 2050? Ever?

Hydrogen is very flammable. It was when used for trans-Atlantic air travel. A Led Zeppelin?

'White' hydrogen? Naturally-occurring hydrogen has been found in various locations. Google it. In some locations, this might avoid the costs of electrolysis, however effected. But all the downstream problems of handling hydrogen noted above remain. Time to deliver? Long? As a widely available, grid-scale option, not likely.

Nuclear fusion here on Earth. This is a violent process conceptually akin to sustained tiny-H-bomb explosions. Extremely high, extremely concentrated, energy inputs are needed, preferably also in a very, very intense gravity well. Not available here on Earth? Workable in stars.

Research on terrestrial fusion has been going on for decades (eg, ITER since 1978). Including all energy inputs, little if any net energy has been produced so far, let alone at scale. Gravity-wise and temperature-wise, best supplied in the Sun's core and other stars. Unlikely on Earth?

Available in 30-50 years? Researchers have said so for decades. They still do today.

Nuclear fission here in Australia. Widely used. Relies on heat from slow radioactive decay. Long-lasting. Easily linked into current grid technology. Time to deliver? Relatively soon -especially compared with alternatives. Small Modular Reactors in future? Maybe.

Can low-emissions, long life, 24/7-reliable, lower-cost, electricity be delivered in Australia?

With renewables, I don't think so. Not by a very, very long shot.

'Firming' renewables at any scale needs fossil fuels today, not batteries. And at very high cost.

What's the known future alternative?

You be the judge.

On the evidence, of course.