4 ms·

> 1: Commenters don’t compare like-for-like energy production, instead they often compare using the nameplate capacity e.g. compare construction cost of a 1GW n

> 1: Commenters don’t compare like-for-like energy production, instead they often compare using the nameplate capacity e.g. compare construction cost of a 1GW nuclear plant with a 1GW solar plant. The right thing to compare is GWh produced. Capacity factor converts GW to GWh, and the capacity factor for nuclear is about 90% and for solar is about 15%[1]. Therefore the equivalent of a 1GW nuclear plant is 6GW of installed solar This is captured by net capacity and LCOE. LCOE of new solar is rapidly approaching MCOE of existing nuclear. > Commenters ignore the cost of systems needed to provide power in winter on windless dark nights. The real cost of wind/solar needs to include the necessary energy storage (batteries or pumped hydro or stored chemical energy or whatever). Some of that can be overcome by installing overcapacity (e.g. install 3x as much GW solar to cover aircon usage in early evening in Texas). Nuclear is very good at providing power at all times. Storage is very expensive - usually doubles or triples the cost of the wind/solar installation. The need for storage goes up as our dependency on wind/solar power increases. This is captured by total system cost. While they're about on par (or favouring nuclear in europe) right now, today's optimistic renewable price is tomorrow's pessimistic. Storage is about to do what PV did in the 2010s, and anyone being forced to pay for a half working AP-1000 or EPR in 2030 rather than 2x the net production in PV panels made of sand and a battery array made of rust all hooked together with aluminum for their utility money is rightfully going to be extremely angry.

robocat 4y ago
> LCOE No, because that usually completely ignores point 2. For example the commonly used Lazard data on LCOE ignores storage and has zero cost component to allow for time shifting over short or long periods. You need more than LCOE to make a fair comparison of costs of different energy sources. Yes, there are solutions to correctly costing externalities, and delivering required power to load over time. My point is that too many commenters wilfully ignore the details, and they write specious cost comparisons. I feel your comment is not helping because it contains half-truths. LCOE is useful, but not valid. Optimism is nice, but just another opinion. For example, let’s look at https://www.lazard.com/perspective/levelized-cost-of-energy-levelized-cost-of-storage-and-levelized-cost-of-hydrogen/ https://www.lazard.com/perspective/levelized-cost-of-energy-... and compare the cost of solar they give, against a storage system that delivers the stored power over 10 hours of night (completely ignoring the bigger issues like seasonal requirements and peak power at noon). 1MWh of utility PV costs $28. Wholesale storage costs $131/MWh which is $1310 for 10 hours. The storage costs totally dominate the costs so tomorrow’s cost of wind or solar generation is completely irrelevant for a fair comparison. Currently, solar and wind can be cheap when comparing on LCOE MWh, because we have existing nuclear and gas plants providing power when they can’t. It is not intellectually honest to make dollar comparisons without trying to make the comparisons have some basis of equality. The cost of storage is dropping, but it has to drop a hell of a long way further over many years before it can compete with baseload generation. I am not pro-nuclear. I am anti-misinformation.
- Schroedingersat 4y ago
  > No, because that usually completely ignores point 2. For example the commonly used Lazard data on LCOE ignores storage and has zero cost component to allow for time shifting over short or long periods. You need more than LCOE to make a fair comparison of costs of different energy sources. Yes, you need to compare total system cost. Which, as I said before you attempted to derail the point again, currently favours nuclear for a 30 year timescale -- but only barely. You're also presently mixing up units of LCOS with units of capacity ($130 per MWh would be 0.05% of current costs). > The storage costs totally dominate the costs so tomorrow’s cost of wind or solar generation is completely irrelevant for a fair comparison Luckily I was comparing tomorrow'scost of storage. Storage is about to do what PV did in the 2010s. There are at least three separate battery technologies presently scaling production that need no scarce materials and have an extremely compelling case for a 5-fold reduction in price. But even ignoring storage entirely, each dollar on renewables can remove more emissions than a dollar of nuclear. It's only when you have funded maximum penetration with ~4h storage that you should consider adding more. Then there's the fuel issue. In all of the mines and all of the spent fuel and all of the enrichment tailings, there is not enough U235 and Pu for a single load of fuel to provide more than about 60% of primary energy. Adding more than ahout 200GW of nuclear is scifi.
  - robocat 4y ago
    Fair call. The figure had a terrible smell when I was writing it, and is orders of magnitude out - that is an embarrassing mistake. I am not denying that battery costs are falling drastically. I am decrying that numbers are thrown around with no meaning, and then I go and do that. Fuck. Thank you for your careful reply.
    - Schroedingersat 4y ago
      Thanks for that breath of calm and collaboration. I apologize for being a little hostile -- it gets a little frustrating constantly hearing people say they want to spend trillions pouring billions of tonnes of sulfuric acid into my country's ground water in order to try and manifest an all-PWR future before even considering funding the things that make a much larger impact. I guess it is probably just as frustrating seeing people claim a square mile of pv modules in a shipping container and nothing else is somehow a drop-in replacement for an AP-1000 for 1% of the price. Cheers.

robocat 4y ago

> LCOE No, because that usually completely ignores point 2. For example the commonly used Lazard data on LCOE ignores storage and has zero cost component to allow for time shifting over short or long periods. You need more than LCOE to make a fair comparison of costs of different energy sources. Yes, there are solutions to correctly costing externalities, and delivering required power to load over time. My point is that too many commenters wilfully ignore the details, and they write specious cost comparisons. I feel your comment is not helping because it contains half-truths. LCOE is useful, but not valid. Optimism is nice, but just another opinion. For example, let’s look at https://www.lazard.com/perspective/levelized-cost-of-energy-levelized-cost-of-storage-and-levelized-cost-of-hydrogen/ https://www.lazard.com/perspective/levelized-cost-of-energy-... and compare the cost of solar they give, against a storage system that delivers the stored power over 10 hours of night (completely ignoring the bigger issues like seasonal requirements and peak power at noon). 1MWh of utility PV costs $28. Wholesale storage costs $131/MWh which is $1310 for 10 hours. The storage costs totally dominate the costs so tomorrow’s cost of wind or solar generation is completely irrelevant for a fair comparison. Currently, solar and wind can be cheap when comparing on LCOE MWh, because we have existing nuclear and gas plants providing power when they can’t. It is not intellectually honest to make dollar comparisons without trying to make the comparisons have some basis of equality. The cost of storage is dropping, but it has to drop a hell of a long way further over many years before it can compete with baseload generation. I am not pro-nuclear. I am anti-misinformation.

Schroedingersat 4y ago
> No, because that usually completely ignores point 2. For example the commonly used Lazard data on LCOE ignores storage and has zero cost component to allow for time shifting over short or long periods. You need more than LCOE to make a fair comparison of costs of different energy sources. Yes, you need to compare total system cost. Which, as I said before you attempted to derail the point again, currently favours nuclear for a 30 year timescale -- but only barely. You're also presently mixing up units of LCOS with units of capacity ($130 per MWh would be 0.05% of current costs). > The storage costs totally dominate the costs so tomorrow’s cost of wind or solar generation is completely irrelevant for a fair comparison Luckily I was comparing tomorrow'scost of storage. Storage is about to do what PV did in the 2010s. There are at least three separate battery technologies presently scaling production that need no scarce materials and have an extremely compelling case for a 5-fold reduction in price. But even ignoring storage entirely, each dollar on renewables can remove more emissions than a dollar of nuclear. It's only when you have funded maximum penetration with ~4h storage that you should consider adding more. Then there's the fuel issue. In all of the mines and all of the spent fuel and all of the enrichment tailings, there is not enough U235 and Pu for a single load of fuel to provide more than about 60% of primary energy. Adding more than ahout 200GW of nuclear is scifi.
- robocat 4y ago
  Fair call. The figure had a terrible smell when I was writing it, and is orders of magnitude out - that is an embarrassing mistake. I am not denying that battery costs are falling drastically. I am decrying that numbers are thrown around with no meaning, and then I go and do that. Fuck. Thank you for your careful reply.
  - Schroedingersat 4y ago
    Thanks for that breath of calm and collaboration. I apologize for being a little hostile -- it gets a little frustrating constantly hearing people say they want to spend trillions pouring billions of tonnes of sulfuric acid into my country's ground water in order to try and manifest an all-PWR future before even considering funding the things that make a much larger impact. I guess it is probably just as frustrating seeing people claim a square mile of pv modules in a shipping container and nothing else is somehow a drop-in replacement for an AP-1000 for 1% of the price. Cheers.