How cheap does solar power need to get before it takes over the world? • Hacker News Board

Re: How cheap does solar power need to get before it takes over the world?

#11

Post by martinald » Tue, Apr 19, 2016, 1:30 AM UTC

I think the issue is that OK, say we managed to get to 100% electricity being solar when the sun is shining.

You'd still have the time the sun is not shining, which would likely be more expensive than before (a gas powered plant would only be running half the time - effectively doubling its capital/construction cost per kWh).

So you'd definitely have an improvement on some levels but I'm not sure that would really help that much.

The real big breakthrough would be somme big improvement on storage. I'm not convinced this will happen quickly (at least not in anyway similar to the declining cost of PV panels). The consumer electronics industry has been paying $$$ for decades for better batteries/battery r&d and has seen pretty poor gains.

I think Tesla and the electric car manufacturers will definitely rapidly get the price down by say 30% because of economies of scale but will then hit a brick wall as more of the cost becomes raw materials, transport, etc which will set a floor on the cost of batteries.

Happy to be proved wrong though :)

Re: How cheap does solar power need to get before it takes over the world?

#12

Post by Gibbon1 » Tue, Apr 19, 2016, 1:36 AM UTC

ZeroGravitas wrote: post #2

It says a lot about the progress of solar if the big worry is that it won't provide 100% of all energy by itself. Even cheaper solar would be nice but I'm not sure why it would be better than a radical battery breakthrough or some cheap and cheerful fusion thing or some mix of different solutions.

If you run the numbers pumped heat storage can likely do the job of time shifting electric power from daylight hours to night time.

The pricing structure will flip though, electricity at night costing probably 2-3 times peek daytime rates. One would expect industrial users would adjust and avoid using power at night[1]. That would reduce demand.

[1] Think of a mini-mill (steel) that currently runs its electric furnaces at night when rates are cheap. If rates are cheaper in the day they'll switch to daytime operation.

Re: How cheap does solar power need to get before it takes over the world?

#13

Post by oolongCat » Tue, Apr 19, 2016, 1:39 AM UTC

martinald wrote: post #11

I think the issue is that OK, say we managed to get to 100% electricity being solar when the sun is shining. You'd still have the time the sun is not shining, which would likely be more expensive than before (a gas powered plant would only be running half the time - effectively doubling its capital/construction cost per kWh). So you'd definitely have an improvement on some levels but I'm not sure that would really he…

If the whole world could get their s* together, we can perhaps have a worldwide grid.

Re: How cheap does solar power need to get before it takes over the world?

#14

Post by coredog64 » Tue, Apr 19, 2016, 1:57 AM UTC

There's another large effect that is left out of this story. As solar gets cheaper, and as we get more of it, it reduces the marginal pressure on fossil fuel production. That, in turn, makes fossil fuels cheaper and forces solar to get cheaper still to compete.

Re: How cheap does solar power need to get before it takes over the world?

#15

Post by strommen » Tue, Apr 19, 2016, 2:01 AM UTC

omginternets wrote: post #3

My biggest concern with solar power (which, by the way, may be based on outdated information) is that the production of photovoltaic cells produce more greenhouse gasses than will be spared in the cell's operational lifetime. To ask an actual question: is this still the case? (Was it ever?)

I'd like to thoroughly debunk this myth with a more illustrative reference[1].

Over its entire lifecycle, utility-scale photovoltaic power creates 48g of CO2-equivalent emissions per kWh. Natural gas emits 10x this; coal 17x.

Furthermore: the majority of PV emissions are dictated by manufacturing processes, which will probably get cleaner and more efficient over time. The emissions of coal and gas are dictated by chemistry, and will never change.

[1] https://en.wikipedia.org/wiki/Life-cycle_greenhouse-gas_emis...

Re: How cheap does solar power need to get before it takes over the world?

#16

Post by Retric » Tue, Apr 19, 2016, 2:09 AM UTC

Gibbon1 wrote: post #10

api wrote: post #4

Earlier quoted context omitted.

This is directly tied to EROEI. If the EROEI of a cell is greater than 1X, it will produce more energy that it took to manufacture it. In that case it's net-negative -- over time it replaces fossil fuels in a net sense. Lately I've read EROEI numbers for solar PV from 10-15X. So no, this is not true, and likely has not been true for a long time. It might have been true back when PV was kind of a lab curiosity and nic…

The interesting thing is the EROEI (energy returned on energy invested) for oil and gas has been falling for a century. I remember reading somewhere a claim that historical EROEI for crude oil. Year EROEI 1920 125 1970 30 2010 15 Things like tar sands are less than 10.

Some tar sands are below 1 once you include refining and transportation.

Re: How cheap does solar power need to get before it takes over the world?

#17

Post by losvedir » Tue, Apr 19, 2016, 2:10 AM UTC

jimrandomh wrote: post #9

There's an economics caveat to interpreting solar prices, which I think may have put a lot of people on the wrong track. Suppose solar panels were priced such that, after accounting for installation costs and subsidies and everything else, solar was much cheaper than grid power and paid off quickly. Then pretty much everyone would want one, and the manufacturer would have a huge backorder. So the manufacturers wouldn…

> Suppose solar panels were priced such that, after accounting for installation costs and subsidies and everything else, solar was much cheaper than grid power and paid off quickly. Then pretty much everyone would want one, and the manufacturer would have a huge backorder. So the manufacturers wouldn't do that; instead, they'd price the panels a little higher, sell everything they could produce but not have the backorder, and invest the extra money into expanding their production capacity.

That's what finance is for. I assume if this were actually the case, wouldn't the manufacturers _take_ the backorders and separately raise money for expanding their production capacity? I don't buy it.

Re: How cheap does solar power need to get before it takes over the world?

#18

Post by CapitalistCartr » Tue, Apr 19, 2016, 2:14 AM UTC

oolongCat wrote: post #13

martinald wrote: post #11

I think the issue is that OK, say we managed to get to 100% electricity being solar when the sun is shining. You'd still have the time the sun is not shining, which would likely be more expensive than before (a gas powered plant would only be running half the time - effectively doubling its capital/construction cost per kWh). So you'd definitely have an improvement on some levels but I'm not sure that would really he…

If the whole world could get their s * together, we can perhaps have a worldwide grid.

Electricity transmission is constrained by line loss, which is proportionate to amperage. To minimize it, voltage is raised to as much as over a million volts, keeping amperage down. This makes for very serious cross-country transmission lines, and electric companies still lose a quarter of the generated power to it. In my opinion, the two great strengths of solar are cleaner power and localizing power generating with its use.

Re: How cheap does solar power need to get before it takes over the world?

#19

Post by eru » Tue, Apr 19, 2016, 2:21 AM UTC

martinald wrote: post #11

I think the issue is that OK, say we managed to get to 100% electricity being solar when the sun is shining. You'd still have the time the sun is not shining, which would likely be more expensive than before (a gas powered plant would only be running half the time - effectively doubling its capital/construction cost per kWh). So you'd definitely have an improvement on some levels but I'm not sure that would really he…

> The real big breakthrough would be somme big improvement on storage. I'm not convinced this will happen quickly (at least not in anyway similar to the declining cost of PV panels). The consumer electronics industry has been paying $$$ for decades for better batteries/battery r&d and has seen pretty poor gains.

Actually, it's not too bad, just slower. Batteries improve around 8% per year, if I remember right.

That's less than semiconductor based stuff improved when Moore's law held, but pretty impressive compared to lots of other technologies.

Re: How cheap does solar power need to get before it takes over the world?

#20

Post by foota » Tue, Apr 19, 2016, 2:22 AM UTC

coredog64 wrote: post #14

There's another large effect that is left out of this story. As solar gets cheaper, and as we get more of it, it reduces the marginal pressure on fossil fuel production. That, in turn, makes fossil fuels cheaper and forces solar to get cheaper still to compete.

As less oil is produced though it will also likely decrease the returns to scale that I assume exist.