Renewables remain cheapest, but cost reductions on hold | Hacker News Reader

Renewables remain cheapest, but cost reductions on hold

csiro.au

25 pointsstubish52 comments

I want to see a "solar dispenser" that can drive very slowly across open land, leaving behind a trail of solar panels.

Installation costs are a big part of current solar farms, and a big part of that is the labour to assemble frames, do up all the bolts, connect all the wires, etc.

If we could make a single truck drive along at 0.5 mph leaving a trail of installed panels, it would really make installation far cheaper. The truck could take as 'input' a container of solar panels with the necessary legs and supports, maybe even concrete bases, all ready to go, shipped direct from the factory.

Other components could be combined to save costs too - for example, the inverter could be built into the panels. The panel frame and the mounting frame could be combined to save weight and assembly time. In a big line of panels, only the end ones have high mechanical wind loads, so middle ones could use far more lightweight plastic supports.

Then creation of a 1 acre solar farm could be as little as a couple of days work for 2 people driving the truck.

Yes! this would be amazing!

Since I've seen these Images, I am kinda amazed by the potential of this idea.

https://www.artstation.com/artwork/dOkXJ1

It reminds me a little of those large coal excavators, but the similarity is probably just because these are big machines.

> I want to see a "solar dispenser" that can drive very slowly across open land

We gotta you covered (ha! pun!) [1].

[1] https://5b.co/web/image/website/1/social_default_image

I’d want the same but to place them on top of unused rail tracks.

This way we would have a solid base to have them installed and possibly a lot of land available. The machine would need to support different gaugages.

Place them on basically any commercially unusable and economically “worthless” space. Instead of mowing down forests to lay them 5 inches off the ground, put them on highway noise barriers, build ceilings over commercial walkways (Europe is full of these but without solar panels), spread them out and put them high over parks to provide shaded areas that allow for greenery to grow underneath, and so on.

The way solar farms are built now, or at least what I’ve seen, is wasteful. Solar has a great advantage in that it can be built alongside virtually anything, but a lot is currently in cleared out areas that can’t be used for anything else.

The problem here is that big, concentrated solar farms are far cheaper to install per kWh generated.

I think you'll be pleased to learn about Flugplatz Brandenburg-Briest.

I've been wondering for a decade or so, why building codes almost everywhere don't yet require the entire roof of new buildings (not counting ceiling windows) to be covered in PV?

Deutsche Bahn (German Rails) is doing this right now with tracks in use

Cabling in a solar farm is also a big part of the costs. If inverters were designed to output higher voltages - eg. a direct-to-10 kV output, then cables could be far thinner and cheaper. 3 phase silicone cable thinner than your baby finger rated for 20kV at just 20 amps could be run along (and carry all the energy from) a length of panels over a kilometer long.

Such a design generally isn't used today because you have to design for someone potentially breaking in and trying to use cable cutters on a wire - you aren't allowed to electrocute the thief! But higher voltage cables could be used with a metal 'detection shield' sitting at just 50 volts which can detect someone starting to cut a cable and de-energise everything.

Is this true? I wouldn't have thought that safety would be required there, is the same true for something like a substation plant? Those have warning signs all over.

Sounds like it’s a fundamental safety issue not just geared towards thieves. Even workers can trip and break a wire!

Part of it is worker safety too... But in general if someone chooses a substantially less safe design and then a thief breaks in and gets injured, yes they would probably end up paying out.

In the USA, that's normally what "up to code" means. Exceptions from code requirements are quite easy to get, but if you have an exception and someone gets injured, all the liability will be on you.

But that means the 20kV runs through all the panels?

Setting aside regulation and panel certification, is this really purely a safety issue?

Also, is it possible to convert the panel string 1KV DC output to 20KV DC? Are you familiar with this area?

Completely possible, yes.

The question is is it cheaper to do so, considering your silicon devices will be more expensive (specifically the diodes on the output stage), but you'll save tens of cents of copper in the wiring per panel.

I haven't done enough research to say for sure, but I strongly suspect it's cheaper.

While this is for Australia, the same is true in 1H2022 for every country Bloomberg NEF tracks except Russia (where NG is cheapest) and Japan, Malaysia and the Philippines (where Coal is cheapest): https://about.bnef.com/blog/cost-of-new-renewables-temporari...

* Does this study only look at generation costs? What about storage costs?

* Does this study look at the fact that the "traditional" grid and energy production technologies will always be necessary, because renewable energy production can drop to zero? This is what increases energy prices in countries that have both renewable and fossil fuel energy production. You are paying for two grids, so to speak

Did you read down to the 4th paragraph? It absolutely does.

I'd be curious to know if it also includes a "true" nuclear catastrophe insurance price or (as is more common) if theyre assuming the taxpayer will fund 99.5% of a Fukushima-type $800 billion cleanup bill like in US/Europe.

I read the original report, which still includes figures of >$100 per kWh for storage, projected for multiple decades (if not in this report, in others). For a higher-latitude nation, we need TWh, probably tens of, for even a small nation. Be assured I have crunched these numbers and confirmed against our current best wind+PV resource data.

In other words, if we run a nuclear plant and deliberately induce a, as you say, "Fukushima-style cleaup", we are about comparable.

That's how expensive proper storage is. "Storage will save us", is frankly, an irresponsible lie that has killed.

Australia is already building a lot of pumped storage. E.g. snowy 2 alone has 350 GWh of storage. Im skeptical of the idea that this + solar + wind costs more than nuclear + this even without free catastrophe insurance but id be curious to see your numbers.

(Obviously some form of storage would also be necessary for nuclear power which cant adjust output to match variable demand).

It is pretty cheap to build pumped storage, relatively. Pumping water uphill isnt too complicated. Until a few months ago though the cheapest "battery" was still natural gas.

Pumped storage is an unsexy 100 year old technology and it doesnt have a huge lobby doing public relations like the nuke industry so when big projects go ahead we tend not to hear about it but there are tons in the pipeline all over.

That's all is valid only if you have geography and water for such stunt.

This, along with wind turbine bird death thing, is one of the more bizarre objections Ive heard emanating from the carbon + nuclear public relations lobbies.

There isnt a shortage of up. The geography to make it work isnt rare:

https://www.sciencealert.com/scientists-spot-530-000-potenti...

Geography plus water plus land usage concerns make it rare. The western US has plenty of sites identified but little water to spare. There is one pumped hydro storage project in process in the entire USA and it is being challenged on environmental concerns. If we are going to make pumped hydro work these issues do need to be acknowledged.

The paucity of projects in the US is because about 50% of your electricity comes from natural gas. If it's very cheap (and it was) and you dont care about CO2 (& the US doesnt) it renders most kinds of storage fairly pointless.

Pretty much the only place without suitable geography in the US for pumped storage is Florida.

In Europe hydropower in the alps is already being used as a giant battery.

Significant new pumped storage capacity requires building large new dams, which is even more unpopular in populated areas than building nuclear.

Equally does it take into account all the costs of global warming?

They're specifically signalling an interest in Small Nuclear Power Reactors [1] which are less accident prone, deferring cost analysis until Dec 2024 [2], and looking for more of a global uptake to bring costs down [3]

[1] https://www.world-nuclear.org/information-library/nuclear-fu...

[2] https://www.iaea.org/projects/crp/i12007

[3] "Achieving the lower end of the nuclear SMR range requires that SMR is deployed globally in large enough numbers to bring down costs available to Australia." (linked GenCost Report, page 58)

> which are less accident prone.

All nuclear powerplants are cheap and free from accidents when they only exist as Powerpoint reactors. Reality has had a tendency to change that.

Ditto all technology - Bhopal disaster, collapsed dams, Texas power grid failures, etc. etc. etc.

Not all reactor designs are Chernobyl, Fukishima killed far fewer than the Tsunami that triggered it.

Small reactors have a long less than public reactor in submarines and elsewhere and have failure modes that are less destructive than larger scale early generation nuclear plants.

They're signaling open mindedness to what is currently vaporware.

If those small reactors are cheap and the industry agrees to assume full liability for whatever goes wrong Id be impressed but I could see nuclear fusion taking off before that ever happens.

Storage and transmission costs were included in this study as you could have easily found out yourself by reading down to the fourth sentence of this media release which says:

“The 2021-22 report confirms past years’ findings that wind and solar are the cheapest source of electricity generation and storage in Australia, even when considering additional integration costs arising due to the variable output of renewables, such as energy storage and transmission.”

There's an entire sub chapter and a bunch of tables on current storage technology costs.

Mainly focused on battery banks at scale, and pumped hydro, with a side note or more on hydrogen generation with peak excess for off peak turbines, etc.

You seem to presuppose that we would need to build new fossil fuel generating capacity. When in fact it's already installed. And I'm not sure they would always need to burn fossil fuels anyway. Coal fired power plants can burn wood, gas plants hydrogen or methane.

I don't really get the 2 grids thing either. Everything is being plugged into the same grid. Instead of replacing a coal fired power station, you just plug in a wind farm. We already have to maintain peaker plants now.

Yes if you had a grid that was as renewable heavy as we could make it, you would be factoring in the costs of these plants into running the grid. But currently most (all?) Countries aren't at that stage, and if/when it does get to that stage, those peaker plants will still have to compete against other forms of peak energy generation / storage.

I don't think there is enough wood to replace coal, so we'd probably need some new gas turbines. Remember that you need to be prepared for at least a couple of days of roughly zero renewable generation every couple of years. In many locations you have significantly reduced generation every winter. So you need to have sufficient backup capability available. I don't think today's battery tech scales to that amount of storage, so you'll likely have to rely on hydrogen or methane.

There's not enough wood to replace the coal we're using now, no.

But if we are only talking about a fraction of today's usage that is doable.

> because renewable energy production can drop to zero

On a small scale, yes, but on a large enough scale that's not the case. There's always wind or sunshine somewhere. You do need to have good transport and interconnections to take advantage of that, though.

Nobody has built to that scale, and you would need a global energy grid which all countries share. Sounds like a pipe dream.

Future cost reductions predicted to possibly be on hold for 12 months, before resuming. Actual real factual prices still reduced in the last 12 months.

I don't know who you are, but you brighten my day whenever I stumble upon your posts :D

> Actual real factual prices still reduced in the last 12 month

Ahhh the bright side of housing and cost of living doubling up!

>The status of nuclear SMR has not changed. Following extensive consultation with the Australian electricity industry, report findings do not see any prospect of domestic projects this decade, given the technology’s commercial immaturity and high cost.

Similar to Germany's assessment.

Yeah, the 400 billion which Germany invested in renewables really paid off. I'm pretty sure for the same sum you would get a nuclear reactor or two.

Great, we can stop all incentives for renewable energy now that it is the cheapest way to generate energy.

More seriously, where's the catch?

We still need to build quicker and build more. This just demonstrates which forms of power would make effective use of taxpayer subsidies.

We can definitely stop incentivize renewables when operators of fossil fuel plants pay for their externalities (including consequences of climate change).

Spoiler: that's impossible because they would immediately go bankrupt.

No need to get into climate change. The externality of pollution alone is much greater than any subsidy. Natural gas wouldn’t go bankrupt and it’s necessary to have natural gas peaker plants when you have renewables.

In this report they state that the cost of fossil fuels is 2.9 trillion in the US alone (this number is higher than the IRENA report and some other analyses). https://energyandcleanair.org/wp/wp-content/uploads/2020/02/...

But we should look at each fuel individually. According to one study, air pollution costs of coal are 32 cents/kWh, 13 cents/kWh for oil, and 2 cents/kWh for natural gas. Other studies have come up with lower amounts. https://www.ucsusa.org/resources/hidden-costs-fossil-fuels

The catch is that other forms of energy are also greatly incentivised. Arguably even more than renewables (sorry can't be bothered to find a source).

I live in a European country, and the first reaction by most governments here to rising gas prises was to subsidise gas [insert captain Picard facepalm here].

Setting prices in advance can still be good sense as it could be used to guarantee prices lower than wholesale prices when oil becomes expensive.

However I say the amount of storage would need to be capable of supply several days in a row (because of no renewable energy ) We have just come out of a rainy period where it rained or overcast every day for a few weeks in SE QLD region I'm skeptical that the amount of storage and transmission capacity has been included to handle such extended periods of no renewable generation.

Form Energy is a unicorn company that exists specifically to solve this problem with their multi-day iron air battery. I’m sure there are other companies working on this as well..

Oh no, it's another pie-in-the-sky battery company with no papers. I'm serious: running a search for their Chief Science Officer, Yet-Ming Chiang, with 'iron battery' provides no relevant results:

https://scholar.google.com/scholar?hl=en&as_sdt=0%2C47&q=yet...

And of course YM Chiang is the only remotely chemistry-related official on their Team page:

https://formenergy.com/team/

There's a serious plausibility problem with an "iron-air battery": the electrolytic reduction of iron oxide has been a holy grail of steelmaking roughly since we learned what electricity is. If you could reduce iron oxide at room temperature with electricity, you would revolutionize the steel industry and rapidly eliminate ~4% of global GHG emissions. See e.g.:

https://news.ycombinator.com/item?id=30348046

Oddly, this is the second scammy iron battery company I've heard about in less than two months:

https://news.ycombinator.com/favorites?id=scythe&comments=t

Meanwhile, real companies with real technology like Gelion [1] and EOS [2] are making real batteries and not getting attention they deserve. Gelion's lead researcher actually worked on the technology they're selling [3] and EOS mentions 21 patents on their technology page, and they're willing to tell you which ones [4]. This is what real science looks like: they want to show you how it works.

1: https://gelion.com/gelion-endure-battery/

2: https://eosenergystorage.com/products-technology/

3: https://www.sydney.edu.au/nano/industry-partners/case-studie...

4: https://eosenergystorage.com/patents/

* in Australia

Reply on news.ycombinator.com