What viable options (other than batteries) exist for a home owner to be able to store energy from solar panels for use at night or when the sun isn't shining?

I realise there will be losses in the conversion.

  • 2
    Do you want to store it to reuse it as electricity, or would storing heat be ok too?
    – 410 gone
    Commented May 4, 2015 at 16:40
  • If lithium batteries are a concern because of the potential for overheating and combustion, you might want to consider vanadium redox flow batteries. They are larger than comparable lithium batteries & they do need maintenance, but they are safer & they last longer. They are mainly used for larger projects, such a large buildings or neighborhoods.
    – Fred
    Commented Jul 27, 2023 at 10:49

9 Answers 9


Thermal storage

Surprisingly to many, conversion of electrical energy from PV (photovoltaic) panels to heat energy and storage in hot water is a potentially excellent use of the energy - provided that you have a use for the hot water.

Water stores energy at the rate of about 1 Watt hour per litre degree or 1 kWh per 1000 litre degrees*. So eg a 100 litres tank (about 25 US gallons) heated by 10 degrees Celsius stores 1 kWh. So if water is heated from 10 C ambient to 60 C, a difference of 50 C, a 100 litre tank will store 50 x 100 = 5000 litre-degrees or 5 kWh of energy. A battery of that capacity would cost several thousand dollars and have a cycle life of under 1000 cycles - or cost several times as much and have a longer cycle life because it is not deep cycle.

  • *The actual conversion rate is about 850 Wh per litre degree, so 1000 Wh (1 kWh) is too high, but this allows 85% efficiency after losses for insulation. The actual effective recovered heat energy will depend on insulation, storage time and storage and ambient temperatures.

While it is usual to use thermal panels to heat water, using PV electricity for heating has some advantages. The panels are lighter, there is no liquid associated with the panel. no header tank, no plumbing., no pumps and no problems associated with corrosion and fluids. A good quality well installed solar system has a lifetime in the 20 to 30 year range and this can be higher. If direct heating is used with a low voltage resistive element a very simple MPPT controller can be used. There is no inverter to mains AC, and no "grid tie arrangements". Efficiency of storage as heat in the water is close to 100% (with wiring loses being the main loss). The tank will lose heat due to insulation not being perfect, but with good tank siting the lost heat can be useful. Any energy used to heat water replaces grid electricity which would have been paid for at full retail rates so you are "paying yourself" retail rates for energy generated as long as you can use the hot water. Stored energy can be used for standard hot water uses and also for central hearing.

Analysis of the prose and cons shows that using water for thermal energy storage is often a highly efficient use of the energy.

You can get elements for water tanks which have multiple "spears" on the one element with some being low voltage and some being mains voltage rated. These allow a mix of low voltage and mains heating as desired.


A possible storage system is to electrolyse eg water to produce Hydrogen & Oxygen and store the gas. As Hydrogen can be burned in air it is possible to store only the Hydrogen. Electrolysis can achieve efficiencies in the 80%-90% range and burning of Hydrogen can be highly efficient.

Hydrogen has the disadvantage of having very high volume per mass at a given pressure. Hydrogen could be stored over water in a simple 40 gallon drum gasometer at low pressure but mass per volume would be low. Storage in steel tanks at modest pressures may be viable. Storage in hydrides achieves reasonable densities for Hydrogen but poor by any other measure.

As an indication of the volumes required for gaseous storage, Hydrogen occupies about 20 litres per gram at STP. You need about 25 grams to produce 1 kWh thermally by burning at 100% efficiency so would need 20l x 25g = 500 litres per kWh. So a near atmospheric pressure gasometer using steel 40 gallon (imperial) drums would need about 4 drums per kWh. Fun as a home project but unlikely to be commercially viable in most cases. Storage at modest pressure brings the volume down - a single 40 gallon drum at 5 bar = 754 litres or about 1.5 kWh. Hydrogen will diffuse through steel and embrittle it as a bonus. At 5 bar it would probably not be a problem but joints are equally problematic. Surplus LPG or CNG tanks (much higher pressure rating) would be safer but still only allow modest pressures and certifications for Hydrogen are probably not available.
Storage in a bladder at essentially zero differential pressure may be viable for use within short periods.

With due design you could electrolyse a suitable quantity of water inside a Hydrogen rated say 3000 psi gas bottle such that the end pressure reached 3000 psi. You can produce mixed gas H2-O2 and even burn it safely with due design!. Without due design you can burn mixed H2-O2 but can expect a rocket or a bomb :-(.

Pumped storage - immense masses needed.

1 kg x 1 metre =~ 10 Joule.
1 kWh = 1000 Watt x 3600 s = 3,600,000 Joule
3.6 MJ/10 = 360,000 kg.m
For eg a 10m head that needs 36000 kg = 36 tons of water for 1 kWh at 100% efficient :-(.

Flywheels are doable but rapidly get large, massive and very very lethal.

Reconversion of thermal energy to mechanical and/or electrical:

Low temperature storage (as above) is viable for heating but not for reconversion to electrical energy.
Maximum conversion efficiency to mechanical is given by Carnot limit of Z (efficiency) max = (T_hot-T_cold)/T_hot.
Temperatures in absolute units.
30C is about 300K.
So a T_hot of 600K ~- 330C gives a Carnot efficiency of (600-300)/600 = 50% max.
Getting real world efficiencies of more than 50% of Carnot is doing well. Higher differentials help both Carnot efficiency and actual % of Carnot realisable.
At T_hot = 630 C Z_Carnot = 66%
At T_hot = 930C Zc = 75%

Building mechanical things to work reliably at 900C is 'a bit of an ask'.
600C is doable but falls in the "don't try this at home' category.

  • 2
    You're absolutely right when it comes to the use of said thermal energy. Highly underrated in home use. Yet it will not run your TV or refrigerator ;)
    – Ghanima
    Commented May 12, 2015 at 19:08
  • @Ghanima True as stated. Not necessarily so though. Many things will not do what they are not intended or designed to do. A key problem is that electrical to thermal conversion usually uses a reservoir temperature near ambient so max possible reconversion to electrical is limited Carnot limit of (Thot-Tcold)/Thot all in degrees K. IF you are willing to use a very high storage temperature then high efficincy reconversion to electricity is possible. eg at 330C with a 30C ambient Carnot Z != 50%. At 630C it's 66%. At 930C !!! it's 75%. Real world efficincies approaching 50% are known. Commented May 13, 2015 at 2:38
  • In Germany there is a hydrogen storage system on the market (+fuel cell) for seasonal home electricity storage. Prices start at "only" €85,000 for a small system since VAT has been reduced to 0%.
    – gerrit
    Commented Jul 14, 2023 at 6:50
  • You could add that 36 tons of water takes 36 m³, so a 3 m × 3 m × 4 m reservoir would do it. But doesn't bring one very far sadly.
    – gerrit
    Commented Jul 14, 2023 at 7:01

Ok, let's break this down. I will assume photovoltaic solar panels (as the question implies alternatives to batteries) and therefore skip all kinds of solar thermal collectors. Starting from electric energy coming from this PV panels we can convert to various forms of energy. For the purpose of storage this boils typically down to a few used forms only:

  • Mechanical, both kinetic and potential energies
  • Chemical, the bonding energy of molecules
  • Electromagnetic, the energy contained in electromagnetic fields
  • Thermal, the internal energy present in a system according to its temperature

The first and obvious choice for home owners are batteries that store chemical energy. They are common today, somewhat accepted and to be honest despite the high prices currently the only viable solution. Very benefitial to their use is the easy scalability (amount of stored energy) to the needs of the owner/user.

Potential energy could be used by lifting up things that if coming down again will release energy. While this principle is used at utility-scale with hydroelectric dams including pumped-storage hydroelectricity and is proposed for very small off-grid applications (mainly third world usage). Home-scale applications are actually rare.

Kinetic energy could be stored in so called flyweels, where rapidly rotating masses. Unfortunately last that I checked they seem to come in systems starting at 20k$. Some require supraconducting magnets to levitate the disc, operate within a vacumm and they all have rather low storage time (hours to days). As of today they are also not viable for home use.

Electromagnetic energy could be stored via the current in supraconducting coils (again not that home-use friendly) or via the electric field in capacitors. Those capacitors in the form of supercaps are investigated and could be one of the next big things in home use energy storage. Idealy the would also provide scalable systems like batteries.

Thermal energy storage including molten salts which can efficiently store and release very large quantities of heat energy. Unfortunately heat cannot be transferred back to electricity at high effiency. So this comes only into play if the energy will be needed to heat buildings or water for domestic use. Nice benefit is seasonal thermal energy storage to store the summer's solar energy for the use in winter time.

One last form that needs to combine mechanical and thermal energy: compressed air energy storage. Unfortunately (to my knowlegde) investigated only at utility-scale with only a few research facilities reported (in the US and in Germany).

This list is by no means comprehensive but reflects my understanding of the current situation.

  • 1
    Nice answer, but I think it would be better reformatted to focus separately (ie with 3 separate headings to keep it nice and clear) on options that 1. might be currently feasible for a home owner now, 2. might be feasible in the future, and 3. that really only work on a larger scale. Commented May 5, 2015 at 2:33

"Solar panel" includes both thermal panels and photovoltaic panels. A common way to store and use heat is in a domestic hot water storage tank. This is generally cheaper and more efficient than PV panels and batteries, but of course it only makes sense if you need the hot water!

Anyone with a heated pool or laundromat should probably be doing this. You'll also need a backup source of heat for cloudy winter days.

  • 1
    From what I've heard, in 2023 it's no longer true that thermal collectors are cheaper than photovoltaic + a modern air-to-water heat pump.
    – gerrit
    Commented Jul 14, 2023 at 7:12
  • Thanks, yes. 8 Years has made quite a difference in solar PV prices. At this point it's definitely not a sure thing for thermal solar, despite the efficiency differences. Commented Aug 1, 2023 at 18:57

I've been doing quite a bit of reading and other research in the area of using solar for hot water tanks. There are quite a few Youtube videos by people who have been using PV (Photo Voltaic solar panels or arrays) to either produce or supplement their hot water tanks with solar electricity.

I think this may currently be one of the more productive ways to use solar for both sustainability and economics. Much of this has been covered in some of the comments above, so I'll try not to repeat much of that. But here are some things that I have been researching and considering in this area that I hope might help anyone seek ways to use solar besides just to charge batteries or for a grid tied system:

If one wanted to heat their water off grid, temperature would be a very important consideration, because if the water is not heated to a high enough temperature there is a risk of biological organisms contaminating the water. This consideration would be especially important in an off grid situation.

Another idea for off-grid would be to use the hot water tank as a dump load for when your batteries are full. Again the water temperature would be the biggest concern, and I'm not sure exactly how high a tank would need to be heated and how often it would have to be heated to that temperature in order to prevent a dangerous level of biological contamination. If I were to guess I would say that it would have to be heated to at least 150 or 160 degrees F at least once per day in order to be safe. Biological contamination would be a concern even if you were only using the water for heating purposes. Perhaps this would be a good question for another post, and it's one I personally can't answer with any accuracy except to say that this would probably be one of the biggest concerns in using solar electric for heating hot water.

If one wanted to simply lower their electric bill without having to get into grid tie, Photo Voltaics for heating water may be one of the simplest and most cost effective ways to do that, but there is still a lot to consider in order for it to be effective.

One method would be to use a regular hot water tank off grid to preheat the water and then use small "point of use" water heaters (using the mains power) to bring the water to the desired temperatures when the solar isn't available or the water not quite hot enough for your needs. The temperature of the preheated water would still be an important factor here, as even if you killed any biological material with the point of use water heater, I'm sure it would still have a foul smell (at the very least) if the water sat too long in the preheating tank at too low a temperature.

Most household hot water tanks, at least where I live, have two heating elements, an upper and a lower. So, another idea might be to continue using mains power for the upper element and putting a DC element and DC regulator on the lower element power by solar PV. With careful calculation it's possible that no MPPT or PWM controller would even be required as long as the DC regulator and element you put in the tank can handle the array's maximum voltage and amperage (of course fuses would be a must in any case). And with a high temperature regulator on the upper element (the one powered by the mains) it's doubtful that one would have to worry about biological contamination of the tank. Of course this would only be a supplemental use of the stored heat from the solar panels, but such a small system would almost certainly pay for itself much sooner than a grid tied system as long as it was designed well. The whole system could probably be set up by an expert in less than a day and would probably lower the electric bill (in my case anyway) by $5 to $30 per month (or more) depending on usage and weather conditions. One problem I do see with this type of system is that with too many panels you would have to figure out a productive way of using the extra energy on really sunny days. On the other hand if you used 300 watts or less of solar, the panels would almost certainly be productive most of the time, but then I don't have enough information yet to determine whether or not such a small supplement would meet one's daily hot water needs.

This is something I'm considering trying myself and most of my research is done from a practical perspective, based on my own budget and the results I'm looking for. I realize that my comment here may raise more questions than give answers, but I'm certain that the questions I have raised (and couldn't answer) here are ones that must be considered by anyone looking to use a PV system to store hot water.

I've seen where quite a few people are using PV to heat water both off grid and as a grid supplement, and most of them say that it is (or at least can be with the right design) an economically practical and productive way to use solar electricity. But, all of these questions must be considered and calculated into the design of such a system in order to make such a project worthwhile from a practical perspective. Of course experimentation in this area has a value of it's own as long as one can afford it and can do it safely. There is a great need for such experimentation to help bring progress to alternative energy.

  • 1
    To save energy, you would want to use in air-to-water heat pump rather than direct heating.
    – gerrit
    Commented Jul 14, 2023 at 7:13

Adding Power-to-(natural-)Gas, since no one mentioned it yet:

By now, it's also possible to transform the energy from solar panels not only to hydrogen, but further into methane, 'the main component of natural gas'. Edit: This might not be possible to do for a single home yet, see below. Then you could either store the gas at home or feed it into the local gas grid. The PSI considers an efficiency of 60% a realistic benchmark (same article), though that would mean, of course, they do not currently reach this number.

If you feed gas into the local gas grid, of course you don't store it for yourself, but how I see it is: If you get money for it, you can later use that money to take gas back out of the grid. Big upside: You would not have to worry about building a gas container at your home.

I recently read about a house in Switzerland that is using this technology here (article in German only). But their electricity-to-gas transformation still happens externally. It seems the 'key' part of the system is city-owned and uses CO2 coming from a sewage plant, and energy from a waste-burning facility, as well as from this house. So I'm not sure how easy it would be to build such a system at home.

I also heard about bigger projects in the US doing something similar, so I hope there's going to be more progress soon.

  • Is it possible to feed natural gas back into the grid?
    – gerrit
    Commented Jul 14, 2023 at 7:16

I hate to sound glib, but have you considered abandoning solar panels in favor of biomass, storing solar energy in a natural form? Trees are a fantastic way to store solar energy; so is some sort of oilseed like rape. Of course, it takes long-term planning to get a sustainable cycle going, but you can figure out how much energy you use continuously (kWh/year), then plan on harvesting enough biomass in a given year and consuming it in some sort of generator (woodgas, in the case of trees; diesel in the case of oilseed), then plan on having the generator run more or less continuously (plan for maintenance), adjusting to consumption with a governor.

If you already have solar panels, which it seems you do, the cost of migration may be steep, but it's still probably the most sustainable solution for the long term.

  • 1
    Welcome to Sustainable Living! You are correct that trees store solar energy via photosynthesis, but the OP clearly mentions that he wants to store energy from solar panels in his question, so I'm afraid your answer doesn't really answer the question.
    – THelper
    Commented May 8, 2015 at 11:08
  • 2
    Blast! I guess I got caught up in reading the answers and forgot the OP. Rather than delete this answer, I think I'll edit it to offer an alternative.
    – J.D. Ray
    Commented May 8, 2015 at 14:40
  • 1
    Energy from biomass, done right, has some real advantages, but for many it will also have a giant disadvantage: land area. Most households in the developed world currently consume so much energy that (considering the inefficiencies of a woodgas generator) the idea only bears consideration if you have substantial unused land for a woodlot. Whereas what I have is my roof. :) It would be risky to take agricultural land out of production to go towards this solution at a utility scale. Whereas cows don't mind wind turbines! Commented Oct 6, 2021 at 18:44
  • 2
    A major problem is that to convert this stored energy back to electricity, you need to burn stuff. Burning anything causes air pollution. Also, most people can't wait 40 years before they've stored enough energy to heat their home, and if we burn things now before growing trees back we'll add CO₂ that will take 40 years to be absorbed again. We can't afford to have all that excess CO₂ linger in the atmosphere for 40 years. Biomass burning is therefore not a suitable form of domestic heating or electricity consumption.
    – gerrit
    Commented Jul 14, 2023 at 7:16

Very good idea of storing heat energy in water. Since I expect my biggest use of solar energy to be in the form of heat this might be a good idea for me. One can increase the storage capacity of water by adding Glauber's salt. It absorbs a lot of heat going into solution and gives it off going out of solution.

I myself planted a small patch of sunflower plants. They will grow 6' or more and produce seeds containing oil. The plants themselves will provide more cellulose than grass would on a similar area also. I mow my acre of grass and am in the process of pressing the clippings into large pellets and storing them in small paper bags. When ready to burn in my stove for heat I'll pour kerosene on them. I did the same with store bought wood pellets. They are inexpensive and work very well. This is true use of solar energy from the Summer! I will need to store the ground sunflowers in metal containers to keep mice etc from eating them.

  • 1
    This is a little sideways from the point. Here OP asks about ways to store energy gathered during the day for the night. Besides, be careful with that kerosene!
    – J. Chomel
    Commented Jun 5, 2017 at 8:08
  • @user4612 / user4616 / user4617 please do not create a new account for every single answer or comment.
    – THelper
    Commented Jun 6, 2017 at 7:07

Electricity grid-scale, the main options are:

  1. Pumped hydropower (also applicable for seasonal cycling)
  2. Compressed hydrogen stored underground (also applicable for seasonal cycling)
  3. Batteries (but only for daily cycling, not for seasonal cycling)

If you have a district heating grid, you can also add a fourth option:

  1. Thermal energy storage

The reason thermal storage isn't used to store electricity is very good: to have any reasonable efficiency, the temperature would be so hot heat would leak immediately away. But if what you want is heat not electricity, then it might make sense to create that heat at times energy is cheap, and then use it at time energy is expensive. Also in combination with steam power plants (where you anyway use heat to create electricity at a ridiculously low efficiency), it might make sense to use molten salt e.g. in a nuclear reactor that runs at 500 MW (reactor continuous power) but is able to provide 750 MW bursts from the molten salt storage pool to allow daily load-following: more power during daytime, less power during nighttime.

Also there's a fifth option that's mostly useful for grid stabilization:

  1. Flywheels

...but that isn't mainly used to store electricity, rather to provide rotational inertia to the grid so that once consumption exceeds production, there will be a clear frequency decrease signal that signals to the grid operator that new production must be added in few seconds, or some load must be removed from the grid in few seconds.

The reason flywheels aren't used for energy storage for long periods is the high self discharge rate.

However, at home none of the alternatives for batteries makes sense. Maybe if you have a refrigerator or a freezer that has a huge compressor, you could put a lot of water inside as thermal mass and then run it only whenever electricity is cheap. I have tried that with my refrigerator-freezers, but it doesn't work in my case because the compressor is so feeble it has to run at least 15 hours per day or else the temperature becomes unacceptably high. Where I live, there's only 6-8 hours of cheap electricity per day, not enough.

Also if you have to use cooling or heating, and have a system that's generously sized, you could cool or heat only whenever electricity is cheap, and then rest of the hours the temperature is in free increase or decline. With thermal mass such as a water tank in a system where water circulates, this becomes easier.

But apart from these cases where you want heating or cooling and use electricity to do that only whenever electricity is cheap, the only option is batteries. For example, I use 7 kWh per day. My house is 5 meters high. I would need to raise 5 million liters of water to my roof to store one day of electricity usage -- not feasible. However, a typical electric car has 70 kWh of batteries, and even if you only use 80% of that capacity, I could run my house from an electric car battery for 8 days.

Also hydrogen can only realistically be stored compressed at huge pressures or else the energy density sucks. Above the ground, pressure vessels are so damn expensive it really doesn't make any sense. Below the ground at considerable depth, it makes more sense, because you can use the ground as a pressure vessel and only need thin steel liners to contain that pressure without leaking, like an inner tube of a bicycle that bursts in free air but if it's surrounded by the outer tire, it can contain the pressure without leaking too fast.

  • none of the alternatives for batteries makes sensehome hydrogen storage exists, but the price (~€100,000) seems to confirm that it does not make much sense
    – gerrit
    Commented Jul 14, 2023 at 7:19

Seasonal electricity storage

A German company sells domestic hydrogen storage for seasonal storage of electricity at home.

The price is "only" in the range from 85.000 – 120.000€ since German VAT on domestic storage systems was reduced to 0% starting at 2023-01-01. If you use 10 kWh per night during the winter months when rooftop photovoltaic is insufficient for charging batteries and thermal storage for overnight use, and you assume electricity remains cheap at €0.50/kWh, and you ignore the costs of charging and several other factors, you might save €1000/year and the unit would be economical in only 100 years! Of course it would take much longer in reality due to operating costs and conversion losses. The service subscription is only €500/year, so if you factor in other costs you can probably expect the unit to pay for itself in the order of 200–300 years.

Not the answer you're looking for? Browse other questions tagged or ask your own question.