Wednesday, September 21, 2016

A heater battery

I've been thinking about home temperature management, and the problem as I see it is:
  • We have a huge supply of existing housing stock, with limited options and money for retrofitting improvements
  • People are unlikely to be satisfied with a 13°C house in winter and a 33°C house in summer
  • As we move to a more renewables-based energy grid, the importance of storing energy increases
  • heating/cooling is, for most people, the largest energy user
  • I believe that the burning of firewood in the city should be minimised
  • If we want to use batteries to supply electricity for heating/cooling, we will need an enormous supply, which will be very expensive (in terms of money and energy) putting it out of reach for most people.
So, we need a heating solution that is
  • Relatively cheap
  • Able to store a large amount of energy during times of plenty – I think in winter it needs to store 4 days' heat (during summer, it really only needs 1 or 2 days as solar PV is very productive)
  • Suitable for retrofitting
 I think I have come up with a possible solution.

Heater-cooler design

In Europe, it is possible to buy a storage heater. It is designed to run on off-peak electricity, which it uses to resistively heat a “bank” of iron “bricks” to about 700°C. Then, during the day when heat is wanted, it blows air over the bricks to warm the room.
My idea is a variant of this, except it uses phase change materials to store heat and “coolth” (the ability to cool air).

Phase change materials (PCMs)

You are familiar with phase change materials if you have ever put ice in a cool box. Ice keeps your food cold not (just) because it is cold, but because ice absorbs a lot of heat as it melts. To put it another way, to it takes a lot of heat to turn ice at 0°C to water at 0°C. If you add the same amount of heat again to the 0°C water, you will raise its temperature to 80°C.

My design uses this principle to store large amounts of heat and coolth.

It uses a PCM that melts at about 30°C. This is encased in small containers (eg. 1L soft-drink bottles) with a large surface area, within a tank of water surrounded by insulation.


In winter, heat is put into the tank, which melts the PCM. Because the maximum temperature of the tank only reaches about 50 – 80°C, there are more options for heating it. Instead of using a resistive heater (such as used in the European storage heater) we can use a bank of evacuated tubes or heat pump and get either free heat, or 4 kWh of heat for every 1 kWh of electricity (it might even be possible to use a roof-top solar pool heater). This melts the PCM and stores the heat. Later, when the building occupants want to warm the room, a small fan blows room air over the heated tank and warms the room air.


In summer, a heat pump can be used to pump heat out of the tank. This can freeze the PCM first (at 30°C), and then the water surrounding it (at 0°C). This stores “coolth” and in the evening when people come home from work, they can use the tank to cool the air in the room.


I have done some analysis and modelling in a spreadsheet that can be found here.
I have modelled the heater’s performance under various design considerations. At the moment, here is what I have settled on:
  • Heat/coolth storage volume: 350 L (total size: 1.2 x 0.9 x 0.5 m, allowing for 50mm of insulation all the way around)
  • Volume of hot-melt PCM: 250 L
  • Volume of water: 100 L
This design gives the following (theoretical) performance:
Heat storage (for use in winter): 31.8 kWh equivalent
Coolth storage (for use in summer): 12.5 kWh equivalent (I’ve deliberately designed it with less coolth storage, because solar PV is abundant most days in summer, but it could be adjusted. This also assumes that the heat pump can efficiently freeze the water.)


If the home occupant has solar PV and a reverse cycle system anyway, and if the marginal extra cost of my system is $1500 (my unit would replace the internal component of the split system), then even if the installation cost of an equivalent-performance battery system reduces to about $5000 (about ½ current prices) then my system is about 25% cheaper (this calculation is on the basis of heat/coolth delivered, and considers that it won’t be used in spring/autumn. It allocates batteries as having more value, but only at the same rate as heat/coolth – ie. when my heater is unused). In other words: if the price of batteries halves, my heater is still 25% cheaper.
Despite this, I doubt that this can be successfully produced as a commercial product, for the following reasons:
  • It’s a one trick pony – it can’t store energy for general use, only for heating cooling. Batteries store electricity, which is more generally useful. Batteries also have a lot of “public mindshare” and will be hard to compete against
  • The cost of the PCM is significant. To be sold for $1500, it would need to be made for 1/2 that, which I don’t think is possible. I think a bespoke heater could be made for $1500, but that would be using “scrounged” parts wherever possible, since the retail cost of PCM would be about $1500.
Despite this, I think my system has some real advantages over using batteries to power heating/cooling:
  • Its embodied-energy is very low in comparison with batteries
  • It can be expected to have a very long useful life and is repairable (the only moving part is a fan which is replaceable). I could imagine a unit lasting for decades if well-made.
  • It is (relatively) easy to integrate with evacuated tubes to collect extra heat in winter


Most Australian houses have poor thermal performance. This device allows one to store heat/coolth that can be released when it is wanted. Because there are low temperature gradients (the internal temperature range in the device might not go outside -4 to 40°C) it should be easy to insulate. It is a cheap and easy retrofit (compared with improving the building envelope) that allows people to store large amount of heat/coolth to use as they please and is inherently compatible with renewable energy systems.


  1. Interesting. You don't mention what chemical or compound the PCM would be. Any ideas?

  2. Hi dltrammel,

    There are lots of options, and it depends what temperature you want it to melt at. There are some subtleties with them, as you can get what is called "incongruent melting", which effectively damages the PCM with each freeze/thaw cycle. You don't want that!
    For the purpose of pricing this, I got a quote for this product:
    The wikipedia page is very useful for learning about the subject:

    Cheers, Angus

  3. This comment was made at another site, pasting here for reference:

    I looked into PCM storage a few years back while designing a "passive" solar off-grid house at 44 N latitude. My researches led me to conclude that calcium chloride hexahydrate (94.45%) with potassium chloride (4.75%) and sodium chloride (0.5%) stabilizers and barium chloride dihydrate (0.5%) nucleator (percentages by weight) was a strong candidate as a low cost, stable PCM mixture with melting point apx. 27.5 C. The melting point can be adjusted somewhat with NaCl being the critical component.

    I expected to contain the PCM in an array of re-used polycarbonate soda or pop bottles and store and retrieve heat and cooling with fans.

    The plan was to use apx. 3000 kg to triple the performance of an passive, insulated masonry house, to the point where it was self heating at winter design temperature.
    Cooling was less of a concern.

    You might want to look up expired US patent 4613444 (Dow Chemical) for information on this reversible phase change material.

  4. And how do you plan to distribute the heat/cold around the house? This will also be a big retrofit cost.

  5. Diego, I was thinking that this could replace the internal component of a split system, in which case the marginal extra cost would be limited to the heat/coolth storage components. It would transfer the heat/coolth to air which would blow into the room.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.