[modo_mario]

Isn't it better to heat the water up to a point (let's say 40 or 50 degrees) and to heat it the rest of the way with resistance heating?

>Versus resistance, which is exactly as efficient at 0°C and 1000°C

It isn't. The difference is smaller than for a heatpump tho obviously.

[coryrc]

1. No, absolutely not. Why would you settle for COP=1 when you can have COP>1?

2. The electrical to heat conversion efficiency is indeed 100% regardless of the temperature of the resistor. And if you're putting out 1000W, then all input losses are also identical. If you put a 1000W light bulb in the middle of your room, or 2 of them but run both at 500W, you'll get EXACTLY the same heat output in your room, but the single bulb is much hotter.

[modo_mario]

1. Are there not scenarios where it drops below 1? The stats for them always seem optimistic scenarios given what i hear from people during winter cold spell in houses that aren't absolutely perfectly insulated. If your water term needs to remain above 70C and it's starts freezing hard outside during winter when that COP starts to matter the most for example.

[ZeroGravitas]

Older heat pumps had max temperature limits and did often have resistance heaters to get that last push above 60C. Modern household heat pumps will reach 75C while staying above 100% efficient and can skip the resistance heater.

This is partly due to a change in the refrigerant used.

[modo_mario]

> Modern household heat pumps will reach 75C while staying above 100% efficient and can skip the resistance heater.

Is this adequately maintained even as temperatures drop? I was recently considering getting a heatpump in addition to my gas installation but I assume I need to go for more than a bit better than resistance heating during winter for that investment to make sense.

[leoedin]

> It isn't. The difference is smaller than for a heatpump tho obviously.

Where does the energy go then?

[modo_mario]

It mostly leaks and such. Limescale buildup is also a small issue for their efficiency and more so if they run hot. If we reduced it to a simple input output calculation that would never be an issue except for some speed of transfer.