How big a solar battery do I need to store all my home's electricity? - Terence Eden
How big a solar battery do I need to store *all* my home's electricity?
Neat breakdown with data + some code.
We have a whole home generator that runs in natural gas. They’re not the quietest things. Been tossing around the idea of having batteries added so that when the power cuts we go to battery. Then when the battery gets low the generator cuts on just long enough to charge the batteries. Wash rinse repeat.
To be completely off grid you would ideally want to be able to go at least a week with minimal to no power generation. Personally that would mean I would need at least 100kWh of batteries.
I would also then want/need a petrol generator powerful enough to power everything that would usually run in a normal day, so that meant be a 15000W one which would be very expensive.
There was an article posted somewhere on Lemmy a few months back where someone tried to do similar calculations for the US as a whole. What I took from the result was 95% renewable was achievable and still cheaper than fossil fuels. However the over provisioning of renewables and over double the storage needed to reliably achievable 100% made that infeasible with today’s proving and technology. Basically you can install storage to cover when the sun is not shining but it’s much more difficult to cover weeks of gloominess
Solar isn't the only renewable choice, though. It's just the easiest to do on an individual level. Also, there are plenty of areas for which weeks of gloominess will never (on human timescales) be an issue.
Guessing it would be more practical to have enough solar panels to fulfill energy needs in winter.
Not really. As I say in my article, our roof is full. On a bad day in winter, we might generate 0.5kWh (assuming the panels aren't covered in snow). So we'd need 20x the panels - there's no room for that.
Hmm, I missed the part about being maxed-out on roof space. Great article and blog by the way!
How come we can't design energy storage that lifts something heavy when there's excess power, and lets it fall to generate electricity when needed?
It's usually called hydroelectric and implemented with dams and turbines.
Potential energy (in joules) is mass (in g) times height (in meters) times 9.8 m/s^2 .
So in order to store the 30 kWh per day that the typical American house uses, you'd need to convert the 30 kWh into 108,000,000 joules, and divide by 9.8, to determine how you'd want to store that energy. You'd need the height times mass to be about 11 million. So do you take a 1500 kg weight (about the weight of a Toyota Camry) and raise it about 7.3 meters (about 2 stories in a typical residential home)?
And if that's only one day's worth of energy, how would you store a month's worth? Or the 3800kwh (13.68 x 10^9 joules) discussed in the article?
At that point, we're talking about raising 10 Camrys 93 meters into the air, just for one household. Without accounting for the lost energy and inefficiencies in the charging/discharging cycle.
Chemical energy is way easier to store.
So do you take a 1500 kg weight (about the weight of a Toyota Camry) and raise it about 7.3 meters (about 2 stories in a typical residential home)?
Honestly that is way, way more reasonable than I was expecting. This isn't half as bad of an idea as I thought it would be
Actually, yes. Lifting the weight of a Toyota Camry 2 stories seems reasonable for a day's worth of energy storage for a house.
I'm not sure how expensive the lift and generator will be, but the weight itself can be anything that's sufficiently heavy.
You say chemical energy is way easier to store, but is it really easier and cheaper to store the energy needed for a home in a chemical battery?
It's an idea that's been played with a few times, but there are many energy loss points in such a system, as well as logistics for keeping the "stack" from falling over. The best so far is pumping water up to an artificial lake, but that's still not very efficient.
1 Watt is the equivalent of moving 1Kg 1 metre in 1 second.
If you want a kilowatt - you need to move 1,000Kg 1 metre in 1 second. Or, I guess, 1Kg a Km.
Plug the numbers together and you'll see that you need a massive physical load and a huge distance in order to store a useful amount of energy.
The energy math doesn't make sense for grid scale applications with solid objects.
However if you can get water between two places it can work quite well. You need to live close to a big change in altitude and do a bit of geoengineering to create the upper and lower reservoirs, which can be destructive to local ecology, but not as much as a dam.
https://en.m.wikipedia.org/wiki/Pumped-storage_hydroelectricity
You can also use pumped air underwater with higher energy losses than pumped storage hydro because of compatibility of air.
What a detailed and rigorous inquest into a question he admits from the outset is absurd and not applicable.
It's practical for someone with limited space for panels on a small room, but I ran these calculations by moving almost all loads to daytime, sizing the panel array to the (minimum daily usage + efficiency losses) * buffer factor for days long storms or equipment failure.
Start with the comparitively cheap panels if you have the space, move electrical loads to the daytime and design the house for thermal momentum, and size storage to the minimum inclusive efficiency losses times buffer. If you have the roof space the panels are the cheapest part and you should usually way, way over panel.
The most important thing is having thermal mass enough or living in a climate that allows your home to not need thermal input or extraction at night. Heat is expensive and exponentially moreso if you need to produce it from conventional storage.
It is possible that, not too long in the future, every home could also have a 1 MegaWatt-hour battery. They would be able to capture all the excess solar power generated in a year.
Braindead strategy, that most likely is discrete fossil fuel shilling, for purposes of making decision inpractical.
The cost of storage as a baselines is how much you can charge/discharge per day. Bonus for smaller (= cheaper) that can have more discharge/charge than its capacity per day. Plus the resilience/reserve capacity value which is a convenience factor. Resilience alternatives include fire places or gas generators (that are not expected to be used often) which tend to be cheap per kw. But noise, smell, variable costs, and startup effort are all inconveniences. Driving an EV to a public charger can be a similar inconvenience level to a generator for resilience value. If a 1mwh battery is used 10kwh/day it costs 100 times more per kwh than a 10kwh battery.
OP gives an example of 12kwh summer use (no AC?) which is very high for most people, but can include cooking and floodlights.
The braindead analysis parts are "because 100 days of 10kwh surpluses happen, I need 1mwh battery". Actual battery storage requirements are the lowest theoretical winter solar production over 1-2 weeks, together with running pumps for heat (stored mostly in fall) distribution. A 10kwh/day maximum deficit for 1 week straight, with 60 day average deficit of 5kwh/day (without requiring additional heat input), means that any consideration for a large static battery should stop at 70kwh. This is sharply reduced with 1 or 2 EVs where summer surpluses are free fuel, and EV provides backcharging at 3kw whenever needed. 30kwh battery is plenty to charge an EV overnight (300km range for small car) before next day's sunlight exceeds needs. Even less battery with 2nd lightly used EV, but 30kwh will be cheaper than un-needed EV.
Instead of relying on batteries for heat generation, which is where $100k 1mwh delusion proposition comes, heat generated from solar stored in under $1/kwh hot water and dirt storage. Outside of winter, this also provides completely unlimited showers and hot tub use, and a $10-20k heat pump and heating system (fossil fuel systems often cost the same) and insulation improvements is the the unquestionable non-distracting path.
(OP here) Sorry mate, are you accusing me of being in the pocket of Big Oil? Here's everything I've written about solar over the last decade - https://shkspr.mobi/blog/tag/solar/ - feel free to point out where I've said "yay fossil fuels!"
I didn't include AC because that's not a thing in the UK.
Oh, and I don't use electricity for primary heating. Solar thermal is pretty useless in my part of the world because you don't need much hot water in summer (mmmm! Cold showers!)
As I said in my post, this is a purely theoretical discussion about what future technology might look like. Your argument is like someone from 2001 going "a recordable CD can hold 650MB - so you only need two for a really long car trip. There's no way people in the future will have 1TB hard drives! For anything else, just use AM radio."
Basically, one of us is braindead - and I'm not so sure it is me!
I looked into one of these thermal systems for my own place but the outlay is just massive for the 11 weeks a year I really need heat, and the rest of the year it's just a stupidly oversized hot water heater that is cooking my glycol and DC pumps.
I ended up paneling up and putting a dumb 9kw resistive boiler for my hydronic floors. The house slab is the battery and although inefficient in terms of strict energy use, winter sun on my cheap pallet of panels dumps plenty into the slab all day. I do have to light the stove if we get a snow storm for a day or two though
Author's diagram is about summer. Fall, winter, spring is about heating-degree days. If you're heating your home with electricity, you'll not get there with batteries.
So, working towards a solution, there are other ways to store excess energy than in batteries. One example is sand, which can be heated to very high temperatures. Insulate a sand container well and its storage can do a lot of home-heating.
We'll need to put a lot of different methods into use. There are many practical ideas out there, and they'll need to be tried.
The sand storage is used for district heating. It's not much of a substitute for single homes that have electrical heating or are off-grid.
It's a great way to balance both the electrical and the heating grids so that more electricity from renewables can be used to offset other means of heat production, but it needs to be done by the district heating supplier. I doubt it makes sense for individual houses.
When I was a kid my parents had electric resistance heat with some very effective thermal storage.
Each room had a unit about the size of a typical radiator. The unit was basically an insulated box with a small circulation fan. I’m not sure what was inside but always assumed some form of brick - they weren’t expensive so it couldn’t be anything exotic. At night when electric rates were low, whatever was inside the units was heated up. During the day, the only power usage was a small circulation fan controlled by the thermostat.
I just got a heat pump installed and thought thermal storage would be worth considering since I was also looking into solar, but contractors acted like they never heard of it, and there really didn’t seem to be any consumer units available.
The solar panels are another story. I don’t see how such a scammy (in the us) industry even exists. They make it really hard to give them my money
Right, you really need scale for sand batteries to work. It would be difficult for individual people to do, especially in suburban London.
District heating also works better in denser housing. In other words, not suburban London.
Dunno what heat pumps are available in England, but that's probably the best option here.
Something very important that anti-nuclear but otherwise environmental minded people should realize is this sentence:
" There's no practical way to build domestic batteries with this capacity using the technology of 2025."
Also applies to grid storage. There does not exist a chemical energy storage solution that can substitute for "baseload" power. It's purely theoretical much like fusion power. Sure maybe in 50 years, but right now IT DOESN'T EXIST. Economically, practically, or even theoretically.
Why do I bring this up? Because I've seen too many people think that solar and wind can replace all traditional power plants. But if you are anti-nuclear, you are just advocating for more fossil fuels. Every megawatt of wind or solar, has a megawatt of coal or gas behind it and thus we are increasing our greenhouse gas emission everytime we build "green" generation unless we also build Nuclear power plants. /soapbox
It's very infuriating talking to people about this because they never really accept that nuclear power is necessary. They spend all their time complaining about how it's dangerous (it isn't) and how it's very expensive, and how you don't have a lot of control over its output capacity. And yeah, all of those are true, but so what, the only other option is to burn some dead trees which obviously we don't want to do.
Just because nuclear has downsides doesn't mean you can ignore it, unless of course you want to invent fusion just to spite me, in which case I'll be fine with that.
Well, unfortunately some people are using nuclear as an excuse to argue that we don't need any renewables at all and that they should be banned entirely. They do this because they know that nuclear faces extreme regulatory and societal challenges and it would allow coal, diesel and gas to continue unabated.
So it creates a backlash where renewable advocates feel they have to fight nuclear to survive.
The new tack is to conflate nuclear energy with fossil fuels. As in assuming that nuclear energy is "legacy" power generation, and that obviously we need to use modern gernation like solar and wind, and magical grid-level storage technologies that don't exist. Also ignore that baseload power is still required, and is currently fulfilled with Natural Gas and Coal.
This has been studied, and we don't need nuclear. All the solutions are sitting right there.
https://www.amazon.com/No-Miracles-Needed-Technology-Climate/dp/1009249541
In US, and EU is having similar nightmare, nuclear was last built at $15/watt. Installing solar is under $1/watt, and for 20 equivalent hours of nuclear per day (less demand at night means not full production even if available) equivalent to $5/watt-day. $1/watt capital costs is 2c/kwh for solar, and for full day production needs 10c/kwh. All before financing. Nuclear is 30c/kwh. It adds 10 extra years of construction financing, requires political bribes to suppress alternative supply whenever they decide to begin operations, uranium purchases/disposal, expensive skilled operations staff, security, disaster insurance.
Solar does need batteries for time shifting its daily supply. At current LFP prices of $100/kwh, 1c/kwh full cycle is prefinancing cost. and so 3c/kwh if triple the charging/discharging daily capacity. 6 hours of storage is a very high number in power systems. It will capture all energy from a northern summer. It will rarely fully discharge with any time shifting incentives to daytime (much higher convenience to consumers and industry) providing resilience to rainy days. A 2c/kwh value (before financing which is apples to apples comparison to nucclear) means a 5gw solar + 30gwh (much lower if enough private EVs are available for time shifting needs) battery costs 12c/kwh or $8B vs a $15B equivalent 1GW nuclear solution. Both last 60 years due to low battery charge/discharge rates and capacity cycle use, with much lower maintenance costs/downtime for life extension costs for solar/battery system vs keeping a nuclear reactor operational. No/minimal operations costs.
It’s very infuriating talking to people about this
Yes. Nuclear shills are frauds who should be frustrated in their theft of the commons.
I agree with this assessment of battery technology, I'm curious what your thoughts on storage through other means, such as dams, kinetic batteries, heat batteries, that style of thing? I understand that it'd be a massive undertaking, but if we really put our nose to the grindstone we might be able to pull off a good amount of power storage through methods that already exist.
A country like France would need around 20 truly massive STEPs like Grand’Maison to provide for a single winter night (~60GW for ~14h). That’s 100-200km² to put under water, a massive ecological disaster, and a massive hazard.
And you must find a way to produce enough energy and find enough water to recharge your STEPs in the next 10h before the next night.
And that’s with the current France needs, having only 25-30% of its energy being decarbonized electricity, it’s getting even worse if we go to electrical heating and transports.
Powering an entire country without hydro, geo, nuclear or fossils is just plain science fiction. And hydro and geo cannot be built everywhere, so realistically, you either go fossils, or nuclear to have clean electricity.
And you can verify it empirically: even with trillion invested in solar and wind, the only countries which have decarbonized their electricity have massive hydro/geo/nuclear.
Building a dam causes massive amounts of ecological damage, plus unless you're building it in the middle of nowhere you're probably going to be turning people out of their homes, out of their entire towns. We could never build enough dams to be able to meet demand so even trying would be pointless. You would be destroying huge amounts of landscape for no reason.
Kinetic batteries can only store power up to a point, the more power you want them to store the larger they need to be. Again to compensate for base load you would have to have a either a lot of kinetic batteries or a few enormous ones. Plus they are maintenance intensive since they are giant spinning things, or great big heavy falling things.
Heat batteries are a good idea and have relatively little in the way of downsides, but they only work where it's hot, not just sunny but hot. So the number of places you can build them is limited.
If only we could get hold of some astrophage or something.
Another myth is that hydroelectric is "green." It's absolutely not. The huge amount of land required to build something like the hoover dam or the three-gorges dam is massively destructive to the existing ecology. It's often overlooked, but land use has to be part of any environmentally sound analysis.
I would say that while the Hoover Dam, or the Three-gorges dam by themselves are acceptable, they are wholly impossible solutions for grid level storage for the entire united states/China. How practical do you think it would be to build thousands of hoover dams?
Other options like kinetic batteries etc, all come down to energy density. The highest energy density options that humans can harness are nuclear Isotopes like Uranium 238, or Plutonium 239 (what powers the voyager probes) After that is lithium batteries at ~<1% density of a nuclear battery. Everything else is fractions of a percent as efficient. Sure there are some specific use cases where a huge fly-wheel makes sense to build (data centers for example) but those cases are highly specific, and cannot be scaled out to "grid-level." The amount of resources required per kilowatt is way too high, and you'd be better off just building some more power-plants.
Pumped hydro exists.
Do some quick math. How much pumped hydro in terms of acre-feet would be required to power a hypothetical city like Chicago at night? Where would this theoretical reservoir be built?
But is extremely limited to specific areas with the right geography that are also relatively close to a population centre.
they generate about 3,800kWh per year. We also use about 3,800kWh of electricity each year
Obviously, we can't use all the power produced over summer and we need to buy power in winter. So here's my question: How big a battery would we need in order to be completely self-sufficient?
O, god, it's going to be huge. You really can't do the off-grid thing unless you have enough power production to satiate you over any given 3-day moving window. Trying to store power from summer until winter is going to be too expensive, instead buy more panel.
This isn't even going into the fact batteries lose charge slowly. So any power generated in summer will be much diminished by winter, even if you have big enough batteries.
Seems to me his panel capacity is to small anyway.
We have 11 kWh panels, and yes in the summer we routinely produce 4 times more than we use, and we have a 7.5 kWh battery
But November December and January it's not even close to enough.
In the Winter you can easily have a week with near zero production:
Our Import / export from grid last year:
November 215 / 59 kWh
December 300 15 kWh
January 268 / 34 kWh
Despite we have almost 3 times the capacity, and produce more than twice what we use per year, and we have a decent battery and believe it or not, even the shortest day we can produce enough power for a whole 24 hour day if it's a clear day! But we can also have clouds for 14 days!
But for those months we imported 783 kWh and exported 108 that could have been used with bigger battery.
But the net import was still 675 kWh!! For those 3 months, and that's the minimum size battery we could have managed with, and then we even need 10% extra to compensate for charge/discharge losses.
TLDR:
Minimum 740 kWh battery in our case, and that's without heating, because we use wood pellets.
That means it would require at least the equivalent of 10 high end fully electric car batteries. But also a very hefty inverter, which AFAIK ads about 50% the price of the battery.
PS: Already in February we exported more than we imported.
they generate about 3,800kWh per year. We also use about 3,800kWh of electricity each year.
Holy shit. I think we used that much last month, which is higher than average but not that high for August around here.
You leave 5,278 LED light bulbs on 24/7?
How ? Is it just AC ?
We oscillate between 300 and 800kwh per month and it's with an old water heater, an electric car charged at home, a dryer and electric oven.
glad I'm not the only one that noticed that.
last time I checked I was using around 4600-5800kwh from May to August. the rest of the year its 3300-4200.
I live in a dual zoned 5200sqft home and my average power bill is around $900.
I've had solar sales try to talk me into solar panels but once they see my consumption they stop answering my calls lol. could be because I told them I'll buy once I can get net zero.
You also lose some energy to heat while charging and discharging. And depending on load profiles, you might not be able to load all of your excess solar power at once (depends on how many Watts the battery can be charged at) or fulfill your power requirement with battery alone (depends on how many Watts your battery can deliver).
I’m a fan of small scale wind, if there’s climate and space for it. 20hrs a day of a (small) 500w adds up really quickly compared to more panels, especially in grey winter weather. The problem is that there’s a bigger difference between megawatt scale solar vs homeowner scale, and megawatt scale wind vs homeowner scale, so there’s limited investment.
Wind isn’t great small scale. You rarely can get high enough for constant wind energy. They are noisy. They don’t produce a lot. In many or even most cases solar will be better than wind.
I’d go so far as building both sun oriented and a solar “fence” line going north/south to get more non-peak solar before putting up small-scale wind.
As is mentioned in the article 😉 What is also mentioned is the fact that battery prices are going down. Soon it seems they’ll be down to $10/kWh!
There's also alot of new battery tech on the way.
There will be a market for batteries at home, and they will exist with the best suitable tech for it - and it's probably not lithium.
How many years, I dont know. What will it be, and who will do it, no clue. Otherwise my stock portfolio would look better if I knew these things haha.
I wish the second-hand battery market were more lively. Using half-worn car battery packs seems optimal for home use.
Sodium batteries?
BTW that's the wish for trend line, $10/kWh right?
How big a battery would we need in order to be completely self-sufficient?
Exactly. Haven't read all details of the link,so I react your comment, and have immersed myself a bit in this earlier.
You need to change your way of thinking and energy usage. Start with your daily energy supply and then change your energy consumption pattern to day time use Then, with for example a dynamic energy contract or if you can spare solar energy, buy or store cheap electricity in your storage ( battery ). The energy management system ( charge / uncharge and which cells) is very important.
Also, realize that battery life is tied to charge cycles and need replacing like every 10 years when talking about the better quality Lithium battery . Sodium systems could and maybe should be used in parallel, if you want more storage, safety and longevity (20 years).
It is yet all quite expensive, though imo having a half day reserve like 5 - 10 kwh, battery, would already create more independence (at around € 3K to € 10 K in Europe) .
With an EV you can have 80%-90% of days covered, and top up with EV. You also get to dump daily surpluses into EV, and you can think of covering winter heating with solar and a heat pump. Easier if you have a fireplace for extreme cold possibility.
Storing heat with fall surpluses is path to get winter heating covered. Heat pump can make hot water very efficiently, and resistance heating can make a pile of dirt 300+C. Radiant floor heating is most efficient because water is distributed around 30C. This means your 90C water volume is 60C effective heat storage that is generated at 600% efficiency in fall, and 300% efficiency in typical UK winter, and your dirt heat storage can be 5x more dense.
A 2nd EV even if not frequently used during the day can be an attractive option, especially if used, and tax credits will go away soon, or have gone away (makes used prices lower) can be much easier than home batteries, and much cheaper if it remains uninsured/unused, and resale value doesn't go down much because of few miles driven. Where utility service includes a high fixed monthly charge, ($50/month in Toronto), $12000 over 20 years savings creates high incentive to remove electric utility. Gas utility has similar fixed vs variable equation, but for Toronto, heat is somewhat reasonable from high supply on our continent.
What I want to do is find out what the maximum size battery I would need in order to store all of summer's electricity for use in winter.
I mean, I think that it's probably not a good idea for this guy to try to go fully off-grid if he has access to the grid, but for the sake of discussion, if one were honestly wanting to try it and one is in the UK, I'd think that one is probably rather better off adding a wind turbine, since some of the time that the sun isn't shining, the wind is blowing.
https://www.statista.com/statistics/322789/quarterly-wind-speed-average-in-the-united-kingdom-uk/
Wind speed averages in the United Kingdom are generally highest in the first and fourth quarters of each calendar year – the winter months.
The UK is one of the worst places in the world in terms of solar potential:
https://globalsolaratlas.info/
But it's one of the best in terms of wind potential:
I could probably get away with putting solar panels on my roof but I think my neighbours would have something to say about a wind turbine. They're pretty loud.
Ugh! Just tell your neighbors to shut up or at least keep it down.
WHAT? I CAN'T HEAR YOU OVER THE SOUND OF MY NEW WIND TURBINE. YOU SHOULD SEE MY ELECTRIC BILL!
Right but he’s not serious, he’s just doing a “in theory, what would it look like?”
Small wind turbines are really, really poor. You need to go high to access the good air-streams and wide to get useful efficiency out of the turbine. Any wind turbine you put on your roof will vastly under-perform for the cost spent on it.
I'd be pretty comfortable saying that buying enough battery storage to power-shift a year of power is more expensive.
Not true, a wind turbine is dirt cheap for the power it can generate compared to solar panels.
Here the problem is regulation that makes it impossible if you have neighbors within 500 m.
If it wasn't for regulation a wind turbine would be a clearly better investment than solar panels.
A huge advantage with turbines is also that it tend to generate power when you need it the most for heating your house.
Basically why the grid exists to begin with. You're not supposed to be solving these engineering problems on a household budget inside a single home.
You'd be better off simply reducing your consumption or finding alternative methods of power (nat gas or maybe wind or geothermal) during the longer winter nights.
If you really want to go crazy, you should consider investing in a bigger home with better insulation and roommates. An apartment/condo block can at least leverage economies of scale, if you're dead set on DIY. More people benefiting from the setup dilutes the cost per person.
I recently got a solar system and came to the conclusion that if you can sell power back to the grid (not everyone can) for some reasonable percentage of what it costs to buy it, then it will always be worth it to be connected (assuming you already are).
Quite simply, if you have enough solar capacity to get you through the winter (no house is going to have months of battery storage), then you will always be creating far more than you need in the summer. Selling this excess will easily cover any costs associated to being on the grid.
Also at current prices batteries are good for backup power only, it's always cheaper to sell excess power to the grid in the day and buy it back at night than it is to have battery capacity to get through the night. I worked out it would take 40 years for our battery to pay for itself (assuming the battery kept a constant battery capacity for 40 years...) but less than 10 years for the rest of the system to pay for itself.
I'm paying 50c per kWh for grid...its bad. And that's if I don't go over the limit. There's 4 teirs so it gets more expensive per tier.
We can't, but we can do net metering, meaning we can offset costs but not get paid. So the best investment is to pay nothing through Dec. 31 and keep costs manageable at the start of the year (net metering ends with the calendar year).
Basically why the grid exists to begin with
Agreed this is the best option. Economy of scales and our consumers wishes should dictate the Grids plan to incorporate cheap energy ( and emergency) storages.
And, also like you said, change your energy life style and insulate your house wherever you can.
I'm very ignorant on this subject, but couldn't you just sell excess to grid and get it back for a minimal markup? Seems like a good governmemt incentive to even supplement an even exchange program. Scaling things to everyone having their own giant batteries seems like a waste of the existing infrastructure.
Factorio has prepared me for þis challenge...
I don't need to get through winter, I just need to get from dusk to when the cheap energy is starts. Currently that's about 4kwh - or a small portion of my car battery before or recharges on the cheap rate.
battery and solar at the home level is what makes the most sense.
60% of the planet lives between the subtropics and tropics. There is way more than plenty of sunlight hitting our earth to support all of our energy demands, and any naysaying around battery technology is missing the forest for the trees.
You can plug your system into a free platform like opensolar, which allows you to play with the design to see what the effect of upgrades would be during the course of the year.
Why limit it to an electric battery rather than some subterranean storage where the excess electricity is turned into stored heat.