Hacker News Reader

Comment by ChoGGi | Hacker News Reader

Seems nice, but I don't see mentions of drawbacks?

The article mentions using helium as the working gas... which there's a global shortage of.

Also, Helium is an escape artist, extra thought must be given to your design to prevent leaks.

Is there a specific reason helium has to be used? These things would be amazing if nitrogen, or some other cheap gas could be used instead.

I'm also wondering how heat is moved in to the unit, and out of the unit. Since normal heat pumps gain a lot from doing the compression/evaporation right where the heating/cooling is needed, but what I get from the article is that this focuses on heating water.

One of the key design parameters is the thermal diffusivity of the gas. The greater it is, the deeper the stack can be, and the better it performs. Helium is about 190, while nitrogen is about 22. Next best to helium is hydrogen, but that has a bunch of other tradeoffs in terms of use. Among other things, it's very hard to seal high pressure hydrogen, and it's notably explosive in almost all concentrations with air.

As for moving the heat, since it's using water as a heat medium, it just moves the water to the hot or cold spot on the internal stack to heat or cool as appropriate.

Good deal on helium here if you're buying in bulk: https://news.ycombinator.com/item?id=37391584

“can produce about 3 to 4 kW of heat for every kW of power drawn into the system”

Doesn’t that produce perpetual motion, given a cold reservoir?

That's most likely a misleading statement of the coefficient of performance... 3-4 kW of heat delivered into the conditioned space with 1 kW of electricity from the grid (or wherever) is pretty reasonable for a heatpump, given appropriate conditions.

The downside is probably cost of manufacturing. Ben & Jerry's prototyped thermoacoustic cooling in 2004 [1], but now their "Cleaner Greener Freezers" are refrigerant based, although they're using a hydrocarbon refrigerant rather than a traditional hydrofluorocarbon. I'm not sure which specific refrigerant they're using, but based on the timing, it might be R441a, which was the first hydrocarbon refrigerant approved for sale in the US and is a mixture of ethane, propane, butane and isobutane. [3]

[1] https://www.npr.org/2004/04/28/1861434/ben-jerrys-uses-sound...

[2] https://www.benjerry.com/values/how-we-do-business/cleaner-g...

[3] https://www.acr-news.com/hydrocarbon-refrigerant-r441a-gets-...

No, because 1 kW of heat would produce much less than 1 kW of electricity.

Is that so? I would have thought if you heated boiling water you’d turn the excess heat energy into steam pressure or something.

A steam turbine is (to some approximation) an inverse heat pump. You can use the temperature delta from an electric heat pump to run a turbine and generate electricity, but the laws of thermodynamics guarantee that the power output of the (heat pump + turbine) system will be less than its power input. The best you can hope for is to approach 100% efficiency.

But the unconverted energy is still in the form of heat, right? Sorry if I’m missing the point. I realize you can’t get 100 of the input, but the unused energy should still be there.

The key thing to understand is an energy source includes both an amount and a quality.

For heat energy the quality is related to temperature. And is basically the percentage of it that can be converted to 'work'. For electricity the quality is near 100%.

So a heat pump moves low quality heat energy inside you house say at 65 degrees and dumps outside where it's say 90 degrees. The energy you move needs to increase in the amount of quality. The increase in energy comes from the electricity used to drive the pump.

Yes, the unused energy is still there, but not "usable" since it is not possible to go from heat to any other energy form at 100% effeciency.

That's not the reason. Heat engines can reach an efficiency of ~50%, so a COP of 400% still gives you an overall efficiency of 200% which is a net energy gain.

The trick here is that the COP of both isn't static, but is usually a peak. Heat engines are more efficient the higher the temperature difference, while heat pumps are more efficient the smaller the difference. For any given heat gradient, you can't have a total efficiency of a heat pump + heat engine that exceeds 100%.

The heat gradient here is the big point. If you can use one kilowatt of power to pump 4 kilowatts of heat from one room to another, to turn some of that into electricity you need to use the temperature gradient between the two rooms, and therefore let that heat flow back into the room you're cooling. If 400% is peak, and you get 50% efficiency out of the heat pump, you can do it at peak at 200% efficiency, but that's only at peak, and at full efficiency of the heat pump. Unfortunately, the higher the temperature gradient in the direction you're pumping, the less efficient a heat pump is, so you won't get anything near peak efficiency. Not at peak and/or with too low a temperature gradient and you can't do it, you'll be not cooling the room you're trying to cool at all and spending electricity to accomplish nothing.

“One kg of feathers is lighter than one kg of lead…”

The thing to note is that heat pumps move heat energy, they don't generate it (other than as losses from the machinery). Your fridge moves heat energy from inside itself to the outside, leaving the insides cool. A heat pump moves heat energy from outside the house to inside of it, leaving the outside cooler. This allows it to be 3-4 times more efficient than resistive heating (which does generate heat energy), because it just moves energy around.

This heat pump will also have to move the heat energy from somewhere. It can't do that if there's none to begin with.

This is a common thread on heat pump threads, where numbers like 300% efficiency are quoted. The explanation, as I understand it, is that it’s not creating 3 or 4 kW of heat for each kW of power drawn: it’s moving those 3 or 4 kW from outside to inside (or vice versa when cooling instead of warming).

The point I think they're making is, if you could turn that heat from inside into useful energy, even with just 33% efficiency, you could run the thing off the heat it pumps out, no?

The text reads better if you substitute produce with provide. Heat pumps can provide x amount of heat for y amount of electricity, and x tends to be greater than y. The deficit x-y is made up by "pumping" heat from the cold side to the hot. Hence called a heat pump.

A 1kw heat pump providing 4kw heat is pumping 3kw heat from outside. Similarly it could be providing 4kw of cooling by pumping in reverse. Most heat pumps are reversible.

Reply on news.ycombinator.com