fub

One of the most interesting aspects of renovating the new house was the “heat loss calculation”. We wanted to see if we could go all-electric with a heat pump with only floor heating to keep the house warm. The temperature of the water pumped into the heating tubes by a heat pump is much lower than the hot water produced by a gas burner that’s pumped into radiators — so there were some calculations to be done. Based on the extra insulation we wanted to add, we should be able to calculate the amount of heat that is lost to the environment. The standard temperature difference taken is 30 degrees Celcius, with the idea that you should be able to keep your house at 20 degrees when it’s -10 outside.

Once you know the amount of heat lost, you also know how much heat you have to put in under those circumstances to keep your house warm if it’s already warm. So you want your heat pump to be able to output more than this number. But you don’t want to over-dimension the heat pump either, because the efficiency is low when it’s running at a low capacity compared to it’s peak output… So we went back and forth quite a bit.

This morning, it was -9 here. The heat pump was working at full capacity, but the indoor temperature dropped to 18.5 — half a degree lower than the set 19 degrees. But it was also heating the tap water at that time, and once that was done, it quickly regained the 19 degrees. It’s not been a sustained ‘middle freeze’ but it looks good.

But it does take a lot of electricity: the heat pump is less effective when the outdoor temperatures are lower, because there’s less heat in the air to pump (duh) and there are ‘thawing cycles’ needed because otherwise the outdoor condenser will just freeze shut. Yesterday was a peak day with 40 kWh used, and we’re on track to something similar today.

And that seems like a lot, but it’s just that now we have an app that gives us this information. In our previous, gas-heated house, we’d just burn gas and pump hot water around and we had no idea on how much we used.

Crossposted from my blog. Comment here or at the original post.

Our solar array is over-dimensioned for what we use ourselves. I don’t have all the numbers yet, but it seems like we produce more than we consume. Or at least: we feed more into the network than we get out. That will probably become costly from January 1st 2027, when “salderen” is ended. That allowed us to compensate our energy import with our energy export for the whole year. That’s great because in summer you have a lot of surplus, but in winter you don’t — and because we heat our house with a heatpump we use a lot of electricity in winter. But once that ends, we really want to maximise our use of electricity that we generate on our own.

One of the ways we could do that is by adding battery storage to the mix. There are now “plug-and-play” batteries available that you can just plug in that will charge and discharge from the same socket. Digital energy meters have a port that shows the current usage pattern, and you can hook the battery up to something that transmits that data so that it charges when there is a surplus of power generation and discharges when power is imported from the grid. They’re getting to a price point that I am seriously considering getting one, just a bit less than EUR 1400.

We have a “double tariff” for our energy: from 07:00 to 21:00 on weekdays we’re paying less for our energy. The idea being that on average, during the day, there’s lots of cheap solar energy available, so the price can be lower. This works out well for us, because we have a solar panel array on our roof as well. We want to use that electricity mostly ourselves, so we want to run heavy users (washing machine, dishwasher, baking and cooking) during the day as well. And we can: we both work from home. If we need more power than we generate ourselves, we import cheap(er) energy from the grid. So the battery would be best positioned for the energy consumption between 21:00 and 07:00.

And then the question becomes: how much surplus solar power do we generate on a day, and how much do we consume in the “expensive” hours? Because that’s the business case for such a battery for us.

And it just so happens that our Enphase solar array logs all the data. So I have for (most) of the year both our energy consumption and energy production per hour. (The integration with Home Assistant is not good and had lots of numbers that could not be right, so I downloaded the data through the Enphase app…)

With this data, I did the following calculations:

Let’s call the sum of the energy imported from the grid before 07:00 the “morning load” for a day;

Let’s call the sum of the energy imported from the grid between 21:00 and midnight the “evening load” for a day;

The evening load plus the morning load of the next day is the total “night load” for that day — that’s the total imported power from the grid that is the most expensive for us and thus the most interesting to get from the battery;

Let’s call the sum of the energy exported to the grid for a day the “surplus” of that day;

Because the battery has a 5kWh capacity, the “battery charge” for a day is the minimum between 5000 and the surplus of that day;

The power that we don’t have to import from the grid for a day, the “energy savings”, is the minimum between the night load and the battery charge;

The actual “money savings” is the price of the energy savings. For us, that is the energy savings (in kWh) times (the higher electricity price plus the feed-in tarriff (because we’re storing that in the battery, not exporting tot the grid) minus the price we get for a kWh from our supplier).

So now we have (almost) a full year of data to calculate with. Our savings up to date would be EUR 140. And it is all idealized, without conversion losses etc so the savings will be less. But they could also be more because the battery could also “dampen” our import from the grid during the day, even though the power has a lower price then.

But even then: with a price of EUR 1400, we’d break even in 10 years. That’s a long time, but with 6000 guaranteed charging cycles and assuming one cycle per day, that’s a lifetime of over 16 years.

Crossposted from my blog. Comment here or at the original post.