3 ms·

We would have needed nuclear yesterday. Right now renewables are the most promising path. New nuclear probably wouldn't make a dent until it's already too late.

We would have needed nuclear yesterday. Right now renewables are the most promising path. New nuclear probably wouldn't make a dent until it's already too late. Of course we need storage and backup generators for renewables. It will be mostly pumped water with gas and oil. Nuclear does not fit as a backup. In the future, the pricing must include renewables PLUS their backup plan (e.g. a gas generator on standby) to achieve level pricing and availability. Having a large grid and moving to more flexible consumers in order to distribute and smoothen out supply and demand will also be a huge factor. In the end I'm positive that it can be done.

mjevans 10y ago
I think that it still does have a solid viable place. It makes the most sense for base load, as well as night time load (which is mostly the same as base load).
- jacobush 10y ago
  While I agree with you for the technical merits of nuclear, I think nuclear has lost for several reasons. One of them, a cultural shift towards decentralization. It is possible now, to build decentralized power grids with lots of local storage and real-time tariff changes on a micro level. This was not just possible before.
  - hx87 10y ago
    10MW passively fail-safe nuclear power plants are feasible and have been designed, so in countries where nuclear regulation is less-than-paranoid they might be an option in a decentralized grid.
jerven 10y ago
Just to point something out regarding nuclear. The announced cost overrun at Vogtle of 3 Billion USD, would pay for 7GwH of battery storage at tesla powerwall retail prices of 5500 USD per 14 KwH. 3 billion / 5500 = 545454 545454 * 14 KwH ~= 7.6 GwH. try at google ((3 billion / 5500) * 14kwh) in Gwh That is the current state of the Nuclear power construction world. What sane investor is ever going to invest in power generation that is online 12 years after the decision to go?
- drited 10y ago
  In addition to cost overruns on new plants, estimates on the cost of decommissioning are balooning on the other end. Sellafield: 2005 estimate: $59 billion. 2015 estimate: $155 billion. German utilities set aside $45 billion but an additional $26 billion may be needed The EU estimated recently that funds set aside to cover the cost of decommissioning 16 nuclear plants would fall short by $137 billion. Source: http://e360.yale.edu/feature/soaring_cost_german_nuclear_shutdown/3019/ http://e360.yale.edu/feature/soaring_cost_german_nuclear_shu...
- VLM 10y ago
  The biggest problem nuclear culture has is the costs are very aerospace or military or healthcare or higher education like, in that they'll cost exactly how much money is available, not a penny more or a penny less. How much should it cost in some abstract sense? How much should it cost to be safe? Who knows. But one thing is certain based on historical observation, it'll cost exactly and precisely every penny that is available, and not a penny more or less. There's a lot of profit to be made off spending each penny, and each penny will be spent. That mindset doesn't mesh well in an industry that has competitors like coal plants or natgas peaking plants or solar plants. What does "spend every penny you can" mean at a coal plant? Or what does "a penny not spent is a penny lost" mean at a solar plant? Meanwhile you're comparing a product that's been shipping for half a century, admittedly at various levels of technical and economic success, with vaporware under so far successful development. You can't compare the price of a commodity gallon of standard cow milk in 2016 to the price of a shipping in 2028 Intel CPU in a useful way.
- osmala 10y ago
  Yes. And for renewables with battery backup would require 375 GwH of storage to get a reasonable two week minimum to be considered reliable backup for a equal electicity producion. You can calculate easily 24 * 14 * 1.117 Its not about just how much long sun is up, its about snowstorms and ice.
  - jerven 10y ago
    That is an unreasonable backup requirement. It is not a backup requirement for any single nuclear power plant either. Those can be taken out by snow storms too (single pylon failure on a line). Or just a refueling stop, which are quite common for nuclear power. One assumes grid resiliency through Geo-distribution and recovery. Any snow-storm taking all solar down for two whole weeks would take down any other power source as that assumes frozen rain followed by massive snow over a huge area. Such a snow storm would kill anything generating power as well as taking down massive parts of the grid. i.e. not any snow storm seen in the last 100 years would be sufficient for this. In the end cleaning snow, while a pain in the ass, is not a real challenge. Much like desert sand cleaning, this is automatable (and small snow removing robots exist already) and would be worth the investment in northern (or extreme) southern climates. We are basically talking about plant equivalence. You need 3x nameplate capacity and about 12 hours of battery storage for equivalency between a solar and a nuclear solution. Assuming night load is equal to day load. That assumption does not hold for a grid, night load is normally half the day load. So a system must only survive the evening peak plus the normal night load. The quip about battery storage price basically showed that if you throw as much money at it as a single nuclear power plant then yes it is a solved problem today.

mjevans 10y ago

I think that it still does have a solid viable place. It makes the most sense for base load, as well as night time load (which is mostly the same as base load).

jacobush 10y ago
While I agree with you for the technical merits of nuclear, I think nuclear has lost for several reasons. One of them, a cultural shift towards decentralization. It is possible now, to build decentralized power grids with lots of local storage and real-time tariff changes on a micro level. This was not just possible before.
- hx87 10y ago
  10MW passively fail-safe nuclear power plants are feasible and have been designed, so in countries where nuclear regulation is less-than-paranoid they might be an option in a decentralized grid.

jerven 10y ago

Just to point something out regarding nuclear. The announced cost overrun at Vogtle of 3 Billion USD, would pay for 7GwH of battery storage at tesla powerwall retail prices of 5500 USD per 14 KwH. 3 billion / 5500 = 545454 545454 * 14 KwH ~= 7.6 GwH. try at google ((3 billion / 5500) * 14kwh) in Gwh That is the current state of the Nuclear power construction world. What sane investor is ever going to invest in power generation that is online 12 years after the decision to go?

drited 10y ago
In addition to cost overruns on new plants, estimates on the cost of decommissioning are balooning on the other end. Sellafield: 2005 estimate: $59 billion. 2015 estimate: $155 billion. German utilities set aside $45 billion but an additional $26 billion may be needed The EU estimated recently that funds set aside to cover the cost of decommissioning 16 nuclear plants would fall short by $137 billion. Source: http://e360.yale.edu/feature/soaring_cost_german_nuclear_shutdown/3019/ http://e360.yale.edu/feature/soaring_cost_german_nuclear_shu...
VLM 10y ago
The biggest problem nuclear culture has is the costs are very aerospace or military or healthcare or higher education like, in that they'll cost exactly how much money is available, not a penny more or a penny less. How much should it cost in some abstract sense? How much should it cost to be safe? Who knows. But one thing is certain based on historical observation, it'll cost exactly and precisely every penny that is available, and not a penny more or less. There's a lot of profit to be made off spending each penny, and each penny will be spent. That mindset doesn't mesh well in an industry that has competitors like coal plants or natgas peaking plants or solar plants. What does "spend every penny you can" mean at a coal plant? Or what does "a penny not spent is a penny lost" mean at a solar plant? Meanwhile you're comparing a product that's been shipping for half a century, admittedly at various levels of technical and economic success, with vaporware under so far successful development. You can't compare the price of a commodity gallon of standard cow milk in 2016 to the price of a shipping in 2028 Intel CPU in a useful way.
osmala 10y ago
Yes. And for renewables with battery backup would require 375 GwH of storage to get a reasonable two week minimum to be considered reliable backup for a equal electicity producion. You can calculate easily 24 * 14 * 1.117 Its not about just how much long sun is up, its about snowstorms and ice.
- jerven 10y ago
  That is an unreasonable backup requirement. It is not a backup requirement for any single nuclear power plant either. Those can be taken out by snow storms too (single pylon failure on a line). Or just a refueling stop, which are quite common for nuclear power. One assumes grid resiliency through Geo-distribution and recovery. Any snow-storm taking all solar down for two whole weeks would take down any other power source as that assumes frozen rain followed by massive snow over a huge area. Such a snow storm would kill anything generating power as well as taking down massive parts of the grid. i.e. not any snow storm seen in the last 100 years would be sufficient for this. In the end cleaning snow, while a pain in the ass, is not a real challenge. Much like desert sand cleaning, this is automatable (and small snow removing robots exist already) and would be worth the investment in northern (or extreme) southern climates. We are basically talking about plant equivalence. You need 3x nameplate capacity and about 12 hours of battery storage for equivalency between a solar and a nuclear solution. Assuming night load is equal to day load. That assumption does not hold for a grid, night load is normally half the day load. So a system must only survive the evening peak plus the normal night load. The quip about battery storage price basically showed that if you throw as much money at it as a single nuclear power plant then yes it is a solved problem today.