Photovoltaic Generator and Heat Pump: Daily Power Generation and Consumption

You can generate electrical power at home but you cannot manufacture your own natural gas, oil, or wood. (I exempt the minority of people owning forestry). This is often an argument for the combination of heat pump and photovoltaic generator.

Last year I blogged in detail about economics of solar power and batteries and on typical power consumption and usage patterns – and my obsession with tracking down every sucker for electrical energy. Bottom line: Despite related tinkering with control and my own ‘user behavior’ it is hard to raise self-consumption and self-sufficiency above statistical averages for homes without heat pumps.

In this post I will focus on load profiles and power generation during several selected days to illustrate these points, comparing…

  • electrical power provided by the PV generator (logged at Fronius Symo inverter).
  • input power needed by the heat pump (logged with energy meter connected to our control unit).
  • … power balances provided by the smart meter: Power is considered positive when fed into the grid is counted  (This meter is installed directly behind the utility’s meter)

A non-modulating, typical brine-water heat pump is always operating at full rated power: We have a 7kW heat pump – 7kW is about the design heat load of the building, as worst case estimate for the coldest day in years. On the coldest day in the last winter the heat pump was on 75% of the time.

Given a typical performance factor of 4 kWh/kWh), the heat pump needs 1/4 of its rated power as input. Thus the PV generator needs to provide about 1-2 kW when the heat pump is on. The rated power of our 18 panels is about 5kW – this is the output under optimum conditions.

Best result near winter solstice

When it is perfectly sunny in winter, the generator can produce enough energy to power the heat pump between 10:00 and 14:00 in the best case.

2015-12-31: Photovoltaics and Power Consumption, Heat Pump's Compressor

But such cloudless days are rare, and in the cold and long nights considerable electrical energy is needed, too.

Too much energy in summer

On a perfect summer day hot water could even be heated twice a day by solar power:

2015-07-01: Photovoltaics and Power Consumption, Heat Pump's Compressor

These peaks look more impressive than they are compared to the base load: The heat pump needs only 1-2kWh per day compared to 10-11kWh total consumption.

Harvesting energy in spring

On a sunny day in spring the PV output is higher than in summer due to lower ambient temperatures. As we still need space heating energy this energy can also be utilized better:

2016-04-29: Photovoltaics and Power Consumption, Heat Pump's Compressor

The heat pump’s input power is similar to the power of a water heater or an electrical stoves. At noon on a perfect day both the heat pump and one appliance could be run on solar power only.

The typical day: Bad timing

On typical days clouds pass and power output changes quickly. This is an example of a day when sunshine and hot water cycle did not overlap much:

2016-03-29: Photovoltaics and Power Consumption, Heat Pump's Compressor

At noon the negative peak (power consumption, blue) was about 3,5kW. Obviously craving coffee or tea was stronger than the obsession with energy efficiency. Even the smartest control system would not be able to predict such peaks in both solar radiation and in erratic user behavior. Therefore I am also a bit sceptical when it comes to triggering the heat pump’s heating cycle by a signal from the PV generator, based on current and ‘expected’ sunshine and weather data from internet services (unless you track individual clouds).

9 Comments Add yours

  1. Hmmmmm–tracking the clouds. That is, of course, a solvable problem. All you need is a clued in Ph.D student and some time. That said, another solution is to store your own excess energy rather than putting it back out there on the grid. I am sure that a battery-based system could do it but am far more interested in the ramifications of selling and buying power from the grid. Is that something you currently can do and how does the cost of selling compare with that of buying it back when needed?

    1. LOL–just read a more recent post that answered my question :-)

    2. elkement says:

      Re tracking the clouds: I had added a link to the last words of the post, an article about an Australian cloud tracking solution – … and it looks like a university spin-off or other PhD projects indeed ;-)

  2. I liked this very much Elke. Thank you. The design phase of our building project is just about complete. Even so, we’re pouring concrete but have yet to decide how to heat the structure! We’ve pretty much decided to go with heat pumps, having narrowly made the decision against a propane furnace. Being able to (someday) go off grid with PV was the final deciding factor in favor of heat pumps. Your graphical analysis is excellent and very instructive. What about energy storage and PV? Is Tesla’s technology good enough now to allow for energy storage of a magnitude to allow one to even out the blue, negative, spikes in all of your analyses? I just did a quick search … take a look …

    1. elkement says:

      Two things to consider:
      1) The total energy storage capacity of the battery would need to cover all electricity needed in the night – assuming that you can “fill it” during the day.
      2) The output of the battery (+ inverter) needs to meet current power demands.

      As for 1) Unfortunately hardly possible in winter in Northern latitudes: Tesla’s battery has 7kWh of energy (originally they also announced a 10kWh version, with the backup system, so I calculate using 10kWh, optimistically). 10kWh is what a typical household in Austria needs per day *excluding heating or cooling*, I have seen much higher numbers for the US (more like 30kWh). I would check old utility bills and calculate kWh per day. So you can expect to be autonomous in summer if you might get down to these 10kWh per day, but on a colder day in winter you might need several times this amount for heating.
      I have added figures of our daily consumption in this post – see especially the daily statistics for November (which was even extremely warm)
      Even if our building needed only half the heating energy (low energy standard) our total consumption would be much more than 10kWh per day.
      So 10kWh is hardly sufficient – but in addition to that is unlikely that the battery will be ever filled to 10kWh on winter evenings as you harvest less energy in winter (only a few kWh per day perhaps) and you already consume more PV energy during the day directly instead of charging the battery.

      2) These inverters typically deliver about 3kW output – so you would need to take precautions (in your electrical wiring) that several heating devices (like heat pump and water cooker) must not run in parallel. For off-grid systems such load shedding this is a must.

      1. Lots to think about. What if you dedicated your Tesla (or even two, in parallel) to run heat pumps only? Would it be able to keep up during winter nights? I wonder? 10kWh per day doesn’t seem like a lot. Can they be linked in parallel to provide > 10kWh per day?

        1. elkement says:

          Yes, batteries can be stacked, and Tesla had presented a system for commercial use last year that used stacked batteries. So in principle it should be possible. Details might depend on the whole solution – like the specifications of the inverter and fuse that limits the current from battery to inverter.

          On principle both heat pump and PV system could be connected to a dedicated circuit but then you miss the chance to use PV energy most efficiently when the heat pump is off and the battery is full (as feeding in to the grid is typically less economic than using the energy directly.) Sort of ‘priority’ for the heat pump but powering anything else nonetheless might be possible but expensive and involve switches that causes weird delays. For example, the backup / off-grid option is also rather complex: Normally, the PV system plus battery has to shut down when the power grid has an outage (otherwise electricians working on the power lines might be hurt). If you want to go off-grid in that case, your home network needs to physically disconnect from the public power grid which requires quite a big electrical cabinet (here in Austria. But to my surprise I once read that regulation is even stricter in the US, which is one of the reasons why PV systems are more expensive).

          Back of the envelope calculation:
          1) Do you know your building’s expected design heat load? This is the worst case heating power needed on a really cold day, depending on building size, insulation, and climate. Something between 5kW and 20kW.
          –> Take 75% of that for a more normal winter day –> multiply these kW by 24 hours –> heating energy per day –> divide by 3 (performance coefficient of air pump on cold day, might be too optimistic) –> this is the electrical input energy needed by the heat pump.
          2) Compare to typical PV energy per day in December or January – there should be free tools on the internet that allow for entering your location and size of your PV system, like 5kW. (I only know the European tools by heart, so I don’t have a link.)

          1. I’ll have to give this some thought. Currently, we’re only installing electrical infrastructure (conduits and boxes) for an eventual PV system. We’ll hold off on the arrays themselves for another few years and until the accounts recover from the building project. Hopefully, by then, Tesla will get things sorted out and we’ll be able to power the entire place with PV (or some other, as yet identified, power source). Thanks for all of your time today … it’s much appreciated. Joanna and I have lots to talk about over lunch now! D

            1. elkement says:

              One thing I forgot to mention is: When you double the battery you also need to oversize the generator – a lot. For comparison: We have a 5kW system (18 panels, typical medium-sized system), and in winter we harvest much less than 10kWh per day – sometimes only 1-3kWh per day for several days in a row. Planning a real off-grid system you would have to go for the worst case, that is no sunshine for several days … which means the system would have to be dramatically oversized to work in winter and cost a fortune (… and in summer the system is then so oversized that you have to ‘waste’ most of the energy by selling it cheaply to the utility as the battery is always full).

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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.