Earth, Air, Water, and Ice.

In my attempts at Ice Storage Heat Source popularization I have been facing one big challenge: How can you – succinctly, using pictures – answer questions like:

How much energy does the collector harvest?


What’s the contribution of ground?


Why do you need a collector if the monthly performance factor just drops a bit when you turned it off during the Ice Storage Challenge?

The short answer is that the collector (if properly sized in relation to tank and heat pump) provides for about 75% of the ambient energy needed by the heat pump in an average year. Before the ‘Challenge’ in 2015 performance did not drop because the energy in the tank had been filled up to the brim by the collector before. So the collector is not a nice add-on but an essential part of the heat source. The tank is needed to buffer energy for colder periods; otherwise the system would operate like an air heat pump without any storage.

I am calling Data Kraken for help to give me more diagrams.

There are two kinds of energy balances:

1) From the volume of ice and tank temperature the energy still stored in the tank can be calculated. Our tank ‘contains’ about 2.300 kWh of energy when ‘full’. Stored energy changes …

  • … because energy is extracted from the tank or released to it via the heat exchanger pipes traversing it.
  • … and because heat is exchanged with the surrounding ground through the walls and the floor of the tank.

Thus the contribution of ground can be determined by:

Change of stored energy(Ice, Water) =
Energy over ribbed pipe heat exchanger + Energy exchanged with ground

2) On the other hand, three heat exchangers are serially connected in the brine circuit: The heat pump’s evaporator, the solar air collector, and the heat exchanger in the tank. .

Both of these energy balances are shown in this diagram (The direction of arrows indicates energy > 0):

Energy sources, transfer, storage - sign conventions

The heat pump is using a combined heat source, made up of tank and collector, so …

Ambient Energy for Heat Pump = -(Collector Energy) + Tank Energy

The following diagrams show data for the season containing the Ice Storage Challenge:

Season 2014 - 2015: Monthly Energy Balances: Energy Sources, Transfer, Storage

From September to January more and more ambient energy is needed – but also the contribution of the collector increases! The longer the collector is on in parallel with the heat pump, the more energy can be harvested from air (as the temperature difference between air and brine is increased).

As long as there is no ice the temperature of the tank and the brine inlet temperature follow air temperature approximately. But if air temperature drops quickly (e.g. at the end of November 2014), the tank is still rather warm in relation to air and the collector cannot harvest much. Then the energy stored in the tank drops and energy starts to flow from ground to the tank.

2014-09-01 - 2015-05-15: Temperatures and ice formation

2014-09-01 - 2015-05-15: Daily Energy Balances: Energy Sources, Transfer, Storage

On Jan 10 an anomalous peak in collector energy is visible: Warm winter storm Felix gave us a record harvest exceeding the energy needed by the heat pump! In addition to high ambient temperatures and convection (wind) the tank temperature remained low while energy was used for melting ice.

On February 1, we turned off the collector – and now the stored energy started to decline. Since the collector energy in February is zero, the energy transferred via the heat exchanger is equal to the ambient energy used by the heat pump. Ground provided for about 1/3 of the ambient energy. Near the end of the Ice Storage Challenge (mid of March) the contribution of ground was increasing while the contribution of latent energy became smaller and smaller: Ice hardly grew anymore, allegedly after the ice cube has ‘touched ground’.

Mid of March the collector was turned on again: Again (as during the Felix episode) harvest is high because the tank remains at 0°C. The energy stored in the tank is replenished quickly. Heat transfer with ground is rather small, and thus the heat exchanger energy is about equal to the change in energy stored.

At the beginning of May, we switched to summer mode: The collector is turned off (by the control system) to keep tank temperature at 8°C as long as possible. This temperature is a trade-off between optimizing heat pump performance and keeping some energy for passive cooling. The energy available for cooling is reduced by the slow flow of heat from ground to the tank.


7 thoughts on “Earth, Air, Water, and Ice.

  1. Elkement. It all makes perfect sense. Why is it that so few domestic installations follow this design? Is there a high investment? Is maintenance an issue? When I read your articles, this looks perfectly scalable.

    • First, I think for the same reason that more air heat pumps are installed today than heat pumps using any other source that needs to be ‘built’ – geothermal, groundwater. As a home owner you don’t need to deal with another contractor (and interfaces / possibly difficult communications between contractors) and investment costs may be lower. Air heat pumps have a lower performance factor, so in the long run they are more expensive. But marketing of the combination of PV and air heat pumps lets people believe they can really heat their (typical = neither small nor passive) house in our climate in winter using such combined systems. Perceived innovations to come (Tesla, articles about translucent solar cells etc.) fuel hope that in the future better solar cells will be available so that more electrical power produced inhouse will outweigh the smaller coefficient of performance.

      Then the feasibility of this ‘unconventional’ source (… not covered by standards, standard software tools, or most heat pump vendors’ brochures …) is still questioned, and I still hear from clients that HVAC contractors confidently explain to them that ‘this can never work’ etc. Sensational articles about lighthouse projects using extremely large tanks – ‘It’s seems paradoxical and unbelievable but you can heat with ice!!!!’ – also do not help as in my opinion they lump these systems together with the many ‘ideas’ in renewable energy that really don’t work (like all these small wind turbines, that are marketed as if they break the law of physics and whose inventors don’t want to have their systems tested by neutral agencies…).

      Investment costs depend on possibly existing components and structures and DIY ambitions. The advantage of the system is that you can build it yourself from off-the-shelf components – and this is what our clients actually do: Some of them have an existing ‘cellar’, cesspit of the like they want to utilize, and all of them want to build it great parts of it themselves – a decision often based on prior frustrating experiences with contractors. So this is not a scalable business, but rather consulting and support for a niche market – one we enjoy a lot as we did not know such a ‘target audience’ did actually exist! 🙂

      We never had any ambitions to scale, depite (or because of :-)) numerous ‘offers’ by ‘potential business partners’ and unsoliticited lectures in How to Build A Business That Sells Standard Products. In my possibly naive view a very small business that wants to stay small – and spend time with technical tinkering, rather than with management – has to focus on the narrowest and best defined target group possible. I’ve also discarded ‘scaling’ options in IT before – so it is my personal preference. The less standardized a project is, the more interesting 🙂 Although I am often presenting the ‘standard hydraulic schematic’ here, in reality every project is very different and planning for integrating or gradually replacing existing systems or tweaking control systems is the most intriguing part – and fortunately this is exactly what other vendors of ‘standardized systems’ are not interested in anyway 🙂

      As a consequence, the option to do projects ‘remote only’ is a must as you don’t find such clients ‘just round the corner’. Actually the clients living far away found us – to our surprise this silly inside joke of planning heat pump systems as we work in IT projects became reality. So the target group is narrowed down to tech-savvy DIY enthusiasts who work in a very structured and well organized way.

      • This is a post in its own right and you hit the nail on the head. As you would. Although I am worried for the planet when I see fossil fuel product double in just a few decades, I’m encouraged that here are many alternatives that work perfectly fine. Sadly, miss-information and now with Trump, disinformation is a big problem.

        One of the first projects I worked on in 1984, straight out of university was a total power installation for a big garage. Very much like your installations this was a bespoke development that required a relatively small investment and easily paid back in a few years. But 30+ years later, I don’t see these installations scaled up either.

        It’s great to see the tinker-element to be central to your approach. In an era of maker’s that almost makes it mainstream.

        Thanks for your reply and great articles, Elkement. Super interesting!

        • Dear Ms Stangl
          About the Ice storage project, how big is the storage and how did you calculate this? We use about 1800m3 of gass equals to 63306MJ (18540 kWh). We think the heatpump can do COP 3.5 Can you help me out?

          Best regards

          • Hi David,

            the size of the water/ice tank is 27m3. See measurement data since 2012 and more details about the system here:
   . Your energy needs are very similar to ours, so tank and collector should have a similar size (Note that our collector is 24m2, but we use only half of it for research purposes since autumn 2014).

            We calculated this originally based on numerical simulations, then we derived rules of thumbs based on these simulations and our long-term measurements. If tank and collector are properly sized the seasonal performance factor (average COP) is above 4 – see our documentation for the past years.

            Are you going to build an ice storage system?

            • Thank you for your open answer. Looking about the waterreserve 25 kub meter is equal to 344*25=8600 kWh Ice energy. is it?
              I have some excel calculation but like to share it by email ok ?

              (We are calculating the ROI before we will build but we are very willing to do so.)

          • (Replying up here here as the answer below is already ‘nested too deep’ – you might not be notified otherwise.)

            Yes – you can contact me via e-mail:
            elke [at] punktwissen [dot] at

            No, the total latent energy store is lower – latent energy is ‘only’ 92,7kWh/m3 (Edited my comment: An issue with units perhaps: Latent energy is 333kJ/kg = 333MJ/m3 = 333/3,6 kWh/m3 = 92,5 kWh/m3).
            The latent energy of the tank to cover only extended periods of extreme frost, so this is sufficient (in our climate – but I think the Netherlands are not that different?)

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s