4 ms·
All you said is true, but I wouldn't still bet against the renewables because we're probably very early in the cost-curve of storage still. In the last 10 years
by pedrocr 6y ago
All you said is true, but I wouldn't still bet against the renewables because we're probably very early in the cost-curve of storage still. In the last 10 years lithium batteries have become 10x cheaper for example. Nuclear was by far our best technical option for the last 20 years. I'm not so sure about the next 20 though.
I ran the math this week on powering the house off-grid with solar and batteries. We even have a stream going through the property that could easily power all our needs throughout the year. The solar install is just so much simpler and lower maintenance though. With current available batteries going 100% solar already has a positive return on ~12k€ total investment. But it still makes more sense to only do a 1 to 3k€ investment in solar with no batteries, even giving away the rest of the power to the power company as we have no net-meetering. If we had net-meetering for the next 5 years it would boost the solar installs while the batteries keep going down in price. Within 10 years we can cover all the rooftops with solar. And with the distributed generation and storage we'd probably need less of a grid than we have now, not more.
- hairytrog 6y agoThe battery/solar/storage costs are near material costs already. And the material costs are artificially low. If you wanted to scale up production to match demand, prices will have to increase so that mines can extract worse ores in harder to reach places.
- pedrocr 6y agoFrom what I've seen there's enough known advancements coming to make the cost of the type of lithium battery I'd need to at least half of what it is right now. There's also the growing market for second-lives of EV batteries. So I expect to turn the house effectively off-grid within 5 years, keeping the traditional grid as backup and occasional filler. In rural areas with crappy power that's an improvement over the existing grid. Even today if I valued the insurance against power cuts enough to buy a generator just doing the battery install today would make sense instead. From what I've read lithium itself is plentiful, it's the other stuff that goes into batteries, and is actively being engineered out, that's a problem, like cobalt. So I wouldn't bet for a rise in the price of the needed raw materials, but insight into the cost curves of mining would be cool to see.
- erentz 6y agoI think the point he is making is that battery metals as a commodity have been unusually cheap. As have a lot of commodities for a while now. Commodity prices are expected by many to go up as demand increases. It’s why people are investing in companies that mine battery metals. As that happens the input costs of batteries will increase and given how massive the component of batteries is in the wind/solar green grid scenario that could change the economics considerably. FWIW uranium is also expected by many to undergo a big surge in price in the next few years with refueling contracts coming due in the US and the growth of nuclear overseas. But the fuel costs for a nuclear plant are a much smaller component compared to other sources, and I am guessing also compared to the battery metals component of the alternate solar/wind/battery/new grid system.
- pedrocr 6y agoYes, and I replied to that. The used materials are changing. The hard ones for lithium chemistries are being replaced and there are more storage technologies than lithium. Given all the ways to improve/grow both mining and the chemistries of batteries and new storage technologies I wouldn't bet against the reduction of cost of storing electrical energy as demand increases.
- p1mrx 6y ago> I ran the math this week on powering the house off-grid with solar and batteries. Does this include your vehicles and winter heat?
- pedrocr 6y agoIn this case it included only one of two vehicles and only part of the heat as we use wood as well. The impact of vehicles depends on where they are during the day. If they're at home you can put in more solar without more batteries as you'll just charge them directly. For winter heat the panel orientation is important so the winter drop isn't as severe when the sun is low. If you want 100% electric winter heat then sizing the array for that and having excess power in summer shouldn't be that much of a penalty as the battery is what is expensive and winter heat doesn't require any more of it. You'll just end up gifting the power company more power during the summer time, which will help with the electric load of all the air conditioning. Investing in insulation and a heat pump may be better though.