It could be interesting in niches like say have some thermal storage (that say gets recharged at peak solar) for heating water on demand instead of draining battery for that, but that's a lot of extra complexity to save on some more batteries
My comment probably shouldn’t have even mentioned home radiators.
https://norwegianscitechnews.com/2021/04/developing-the-worl...
That's a bit short of the 1000°C that Rondo's system promises, so these approaches are useful for very different applications.
It's not nothing, but getting close to the theoretical limit tends to be incredibly annoying when it comes to thermodynamics. Typically it requires doing stuff extremely slowly (so you're always close to thermal equilibrium), which means it's nowhere near fast enough or you need almost perfect insulation etc.
Say, I have a heatpump that heats up water to 100°C at ambient temperature of 10°C. Couldn't I feed that hot water into another heat pump that heats up water to 200°C and so on?
You can make heat pumps 10x more efficient by using 10x smaller increments of temperature, but then you also need to pump each unit of heat 10 times so in the end it doesn't cost less energy to make a bigger temperature gradient.
The problem is at step two you have three units of heat, two of which came from ambient heat.
At step three you have 4.5 units of heat, two of which came from ambient heat.
Step five there are 6.75..and so on.
You're expending a lot of energy moving energy you put in so you get diminishing returns.
The theoretical limit is T_hot / T_change (in absolute units like kelvin). If your change is roughly the hot temperature, there's no benefit.
Have a big brick or water thing inside. Heat it while it is sunny with a heat pump.
Hell you could even melt a few tons of sodium acetate during summer and use it for december/january