[mcv]

More panels don't help if you can't store it.

Well, I suppose if you have enough panels to generate power even on short, cloudy winter days, then that doesn't matter anymore; then you just need enough storage for the night. But then what are you going to do with all the surplus power on sunny summer days?

Some sort of cheap long term storage would really help a lot.

[hedora]

At least for our install, we spent about half the money on batteries and half on solar. It fairly reliably gets through the night unless we run the AC (but those are sunny days) or charge the car (car batteries are about the same size as the house batteries).

On a cloudy day, the panels provide 90% of our normal power usage. Anyway, to scale it up, we'd want to increase panels and also batteries. Increasing only one would leave us with no power at dawn or with a large battery that would never reach 100% in winter. One night of batteries with panels that reliably provide enough electricity to get the batteries to 100% is a good tradeoff for sunny climates. As it gets cloudier, batteries might have more incremental benefit, but multi-day storage probably doesn't make sense.

Also, you can tie a gas/propane generator to the battery to handle the "a few times a year" cases. That's probably less carbon intensive than 5x-ing the system for 1% of the days.

(Since the 1% days for us are in winter, we have a wood stove.)

[taneq]

> But then what are you going to do with all the surplus power on sunny summer days?

You don't have to do anything with it. Panels are dirt cheap these days, and if you want reliable off grid storage then you have to size them to keep up with baseload power under your target range of conditions anyway. Figure out how many days a year you're happy to run a generator or turn your fridge off, find stats on your local daily kWh/m^2 solar energy, size panels to cover baseload with that incoming energy.

I have 2kW of second hand panels hooked up to a 200AH 24V battery pack (again second-hand) to power a server rack, it uses about 4.5kWh of solar power per day without running the batteries down too far. The panels can generate 8kWh/day in summer, the rest is headroom for cloudy winter days. The last time the server saw mains power was... December, I think?

> Some sort of cheap long term storage would really help a lot.

I mean yeah, but so would Mr. Fusion.

[danans]

> But then what are you going to do with all the surplus power on sunny summer days?

> Some sort of cheap long term storage would really help a lot.

Run a still to make ethanol from waste biomass. Store it (don't drink it!!) and use it to fuel a generator in the winter.

I'm only half joking.

[mcv]

Generating some sort of fuel is probably the best idea indeed. Hydrogen is often suggested because it's easiest to create out of water, but it's also hard to store. Ethanol is definitely easier to store, but requiring a lot of biomass is definitely a bit of an obstacle.

[41b696ef1113]

Genuine question - how many engines can run on straight ethanol? Are there any negative consequences for running straight ethanol in an engine expecting gasoline?

[danans]

You can damage the engine if it isn't designed for ethanol:

https://www.car-engineer.com/adapting-an-engine-to-ethanol-f...

[koonsolo]

Yes, I get it, it's indeed a lot of surplus in the summer.

Long term storage would be the better solution, but batteries don't seems to be able to store a large amount of energy anyway. So in my opinion, it's really a tradeoff.

Maybe mine some bitcoins during summer? ;)

[hedora]

I looked into this. Even assuming PG&E's buy back rate is 50% of current numbers, the ASICs have an expected profitability horizon of over one year of uptime. I'm not convinced they'll be profitable much longer after that, since hash/kWh keeps improving.

Also, I bought the panels to reduce carbon emissions, and would rather sell the power back.