Hacker News new | past | comments | ask | show | jobs | submit login
New thermogalvanic tech paves way for more efficient fridges (cosmosmagazine.com)
91 points by westurner 3 months ago | hide | past | favorite | 51 comments



Something I've wondered about recently - often times in a kitchen you want hot water and a cold box for food - and lately there are heat pump water heaters available on the market. Is there a fridge/water heater combo yet?


There's not enough heat in the fridge to heat up any significant amount of water.

I would however like to see that waste heat being used for something like dishwashers or just being vented out (instead of into the room that might have to be actively cooled).


Many modern homes have a perforated panel in the ceiling behind the refrigerator to allow the heated air to escape the room.


Citation needed.


Any heat used is better than no heat used.

Water heaters that are also AC units do exist (a heat pump is an ac run backwards)


No it's not. Claiming heat from a fridge would be like grabbing a cup of water for a minute a few times a day. The extra shit that needs to be manufactured and transported is surely not worth the effort at all.


Mini-split HVAC systems create both heated and cooled air, and send them to just the rooms they are needed, while also redistributing existing disparities of hot and cold. The latter being the most efficient HVAC mode of all.

I could imagine at some point, all our heated and cooled air, water, and ice could be one system.

For fast temperature demands, such as hot water or ice, perhaps reservoirs of chemically stored heat and cold could be built up at useful locations, for very high speed release when needed.

Completely dispense with storing hot or cold air/water, except for fridges. But provide the cool air for them, as a byproduct of being one system, avoiding the local inefficiency of both cooling air in small quantities and generating heat at the same ___location.

It would make it cheaper to have more fridges and ice makers around a houses, if you just had to plug them into power and an HVAC controlled "inlet/outlet".


Correct! This is how some advanced systems already work in multifamily, where one room's excess heat is routed to another rooms lack of heat -- "Variable Refrigerant Flow" or VRF


> Claiming heat from a fridge would be like grabbing a cup of water for a minute a few times a day.

Did you know that people do this? The brits love making a cuppa tea. It's free energy and the heat pump in the fridge is already doing the work.


I wouldn’t be surprised if someone starts making these. The heat that comes off a typical refrigerator is enough to heat about 1/3 of a typical house’s hot water usage.


What?? Not even close.


To put numbers on this, you can buy a C rated fridge freezer in the UK which uses 176 kWh of energy per annum.

That's an average power consumption of 20W. But electric heaters are say 1200W or 2000W.


That includes the lights and all the rest. That gives you like 3m³ (790 gallons) of 60°C (140°F) water. Which is really not a lot.


Nope. If the cost to "reuse" the heat exceeds that of the status quo, it simply doesn't make sense.


A lot of people have these in their houses already. Go to Home Depot some time and you’ll see lots of heat pump water heaters.


Sorry, didn't see any "kitchen fridge/hot water tank combo units" on their website. Care to provide a link?

Or perhaps you may have missed a nuance in the comments above yours? Nobody is claiming that heat pumps don't work, they certainly require a source of heat to pump though, and the air in a kitchen fridge ain't that.


Your comment in reply to: "Water heaters that are also AC units do exist"

https://www.homedepot.com/b/Plumbing-Water-Heaters-Tank-Wate...


Just a heads up, your reading comprehension kinda sucks.


If you look at the temperature deltas, we really don't do much to cool our food. You can verify this by comparing the power requirements of a fridge with just about anything that heats water or food.

You could save some power, but not very much.


Waste water heat recovery, which heat exchanges with inlet water, would probably be more effective than heat exchanging with a fridge. Realistically both options are too costly for the limited benefit.


I looked into waste water heat recovery a few months back and it is a thing you can buy. You buy a special section of pipe that has a straight pipe with another pipe coiled around it. Some of the heat from the waste water gets transferred to the water coming in as they flow past each other in the pipes. They cost a few hundred bucks. I have no idea if the long term savings justify the cost (if you're looking purely at money for benefits).


These exist for showers, shower heat exchangers let the drained hot water exchange heat with the cold water tube to the shower.


Even for a furnace or boiler, condensing the steam in the exhaust only makes it around 10% more efficient.


"Solvation entropy engineering of thermogalvanic electrolytes for efficient electrochemical refrigeration" (2025) https://www.cell.com/joule/fulltext/S2542-4351(25)00003-0


Thanks for that, it's great that its 10x better than the previous best effort. It's notable that it needs to get 20x better again before it starts to have useful applications :-).


Someday maybe!

Quantum thermal diodes: https://news.ycombinator.com/item?id=42537703

From https://news.ycombinator.com/item?id=38861468 :

> Laser cooling: https://en.wikipedia.org/wiki/Laser_cooling

> Cooling with LEDs in reverse: https://issuu.com/designinglighting/docs/dec_2022/s/17923182 :

> "Near-field photonic cooling through control of the chemical potential of photons" (2019) https://www.nature.com/articles/s41586-019-0918-8


High hopes, low expectations :-). That said, 'quantum thermal diode' sounds like something that breaks on your space ship and if you don't fix it you won't be able to get the engines online to outrun the aliens. Even after reading the article I think the unit of measure there should be milli-demons in honor of Maxwell's Demon.


If you're interested in going down a rabbit hole, other thermo tech that always seems to be right around the corner of practicality: thermoacoustic chillers, magnetocaloric fridges.


Magnetocaloric cooling is used in cryogenics as are thermoacoustic chillers. I think there are some aerospace applications.



If we cared about making refrigerators more efficient, we would just put the heat exchanger outside or better yet underground


Depends on where you are. Where I live, it's cold seven months out of the year so keeping the resultant heat inside the house lowers the heating bill somewhat.

I've always wished for a fridge that was mounted to an exterior wall. In the summer, it operates like a normal fridge but with the condenser coil outside of the house. In the winter, the compressor system turns off when the outside temp goes below a certain value and little door/shutters with fans regulate the temperature of the refrigerator either by letting in cold air from outside (if the interior needs to be colder) or from the inside (if the interior needs to be warmer to prevent food from freezing).

There's probably a few good reasons why this is not actually practical but I'm tempted to try something like it in retirement.


Aside the complexity and standardization needed for building it, which alone will absolutely not make it worth it, you'd also have the problem of temperature fluctuations. You probably don't want your milk to be frozen right? ;-)

So if you really need to push indoor air inside to avoid that, you are basically killing the efficiency. And not only that, you would push warm and comparibly humid air inside the fridge a lot. It will condense and now your fridge has to deal with a lot of condensation and probably get moldy.

But good to see that I'm not the only one making these thought experiments.

My personal conclusion is: eventually we will have a single refrigerant distribution system where other systems can be connected and then make use of it (similar to multi-split ACs, but extended to more devices). Then, your fridge would just be part of the system.

And, just to understand: transferring heat from inside the fridge to the outside in the case that the outside is colder is so energy efficient already, there's no need at all to optimize anything. A solar panel as big as your hands will be sufficient even in winter to take care of keeping the fridge cool. That's how little energy is needed.


I've had the same desire!

However, it probably doesn't pan out.

An energy star fridge uses about 400kWh/year. In pittsburgh, we have 5710 cooling degree days and 736 cooling degree days. Those cooling degree days are spread across about 4 months. So we only need to dump heat from the refrigerator about 1/3 of the time: 133kWh. With an A/C CoP of 5, that means that cooling the extra load from your fridge costs you about an extra 26kWh / year, or about $5.

On the heating side, it's actually better to heat the house from the fridge than to burn gas - you've basically got a heat pump system! Heat pumps are more efficient than direct heating. If you already have an efficient heat pump system this may not be as true but it probably comes out in the wash.

The physical modifications to the house would have to be very cheap to make this cost-effective -- and more penetrations through the house envelope create opportunities for leaks.


The very first household refrigerators had the condensing unit mounted outside.

You can see an example in this 1920s Frigidaire training video (which is also worth watching for other reasons): https://www.youtube.com/watch?v=W-t7DqOAMME


Oh man I just watched that whole video. What a ride. I loved that tossing your cigs in the driveway was the correct solution, instead of dip on the front porch. Also the maid taking the service guy’s hat and coat.


Did you notice he used a separate entrance, for the lower-class servants? And the gauge he adjusts is labeled "retard".


That's actually commonly done in industrial cooling applications with rapid turnover (think dockside food storage warehouses), but it's not practical for the heat output of domestic units.

Even in parts of Europe and Iceland with ready access to a common municipal heat exchange, household refrigerators don't have have enough heat output to justify the infrastructure.

The typical domestic fridge just needs to "get cold once" and the remainder of its duty-cycle (aside from the rare holiday feast or party) it's just trickle-cooling.


AFAIK, that's how lots of restaurants keep their stock chilled, at least in my area. They have relatively noisy exchangers outside connected to their fridges that are indoors.


Imagine heat pumps 400+% efficient with 60% heat to electric conversion.

https://en.m.wikipedia.org/wiki/Johnson_thermoelectric_energ...


Does it work?


Both work, seems funny to run in a loop.


It is going to be pretty hard to beat the efficiency of phase-change heat pumps. And I hope it is more powerful than merely being able to keep a home fridge cool, because they already don't use all that much power unless it is absolutely ancient.

This is great stuff to research, but trying to market them for potentially slightly more efficient refrigerators seems like going for the highest fruit on the tree.


What do you make of this guys? https://x.com/BrianRoemmele/status/1884477625084830006

It seems like older fridges worked really really well and last a long time.


Even if true, no one is going back to an R-12 refrigerant because CFCs contributed to a hole in the ozone layer that the entire world agreed was a bad thing. [1]

The "last a long time" bit might be survivorship bias. They might last a long time (or not) but having one old working fridge does not prove reliability.

As far as I know, there was no ban on using old fridges and I don't think people usually throw away major appliances that work.

[1] https://en.wikipedia.org/wiki/Montreal_Protocol


They did often use better refrigerant back in the day, but it is refrigerant we don't use now because it is either highly poisonous or hell on the ozone layer. The compressor itself wasn't often all that efficient mechanically, but the good refrigerant helps make up for it, but also there is only so much to gain out of compressor efficiency. The biggest problem with old fridges is they were insulated like absolute garbage. Not only is the insulation often super thin, but the steel shell on the inside was rarely separated from the steel shell on the outside leading to a lot of heat transfer. So even with it closed they ran constantly almost like you did have the door open all the time.

You also have to consider its cooling capacity, you only need a little bit of cooling 95% of the time, but when you put fresh groceries in it or water in the fridge, until those are cooled down they are heating up everything else in the fridge and so stuff will spoil faster/easier. Some older fridges might take a long time to cool everything down if you put in more than a single item at a time.

But really, most fridge breakdowns are not because of a compressor failure or anything major, 99% of the time it is some simple electronic component. A simple bad capacitor can prevent your fridge compressor from kicking in. Or a thermocouple could go bad and not tell it to turn on when it should. Both of those parts cost like $2 to produce, but of course they choose more proprietary designs and configurations to make a generic fix more difficult which lets them markup the parts on prices. Which has the side effect of convincing more people to just buy new because they don't know if the 1500% marked up proprietary part was the only part near failure, and they don't have the knowledge or confidence to start tinkering on their refrigerator before they lose all their potential savings on spoiled food, or use an appliance repairman who will add on $200+ onto the inflated parts prices because they don't want the liability taken upon themselves if they were not using direct replacement parts.


10x Improvements in this kind of effect don't happen very often. I'm wondering if a chain of these can "pump" coolth uphill, or if it can be used alongside other methods.


Yeah am I reading that? 1.7 degrees Celsius is what they've achieved? That paves the way for more efficient refrigerators?

I thought paving the way meant a clear path to the end goal, because you already know the route and all the necessary engineering to get there.


Promo gotta pro mo'

But upping the effect from 0.1c to 1.7c is pretty good. I doubt there's a lot more to come from this, so i doubt the path finding too.


I think we're using 100 year old technology. Compressors and stuff.

Whatever happened to acoustic cooling?

Did peltiers ever get efficient?


> Did peltiers ever get efficient?

Not compared to gas compression. For a relaxed take on it:

https://youtu.be/CnMRePtHMZY?si=jusDm81RWkDF_HJ0&t=900 (Thermoelectric cooling: it's not great)

Though part of the complaints about the mini fridge is due to lack of insulation. But he compares it to his usual fridge, which has a similar power draw as this 60W mini fridge, and there's a 50x difference in volume to cool.




Consider applying for YC's Summer 2025 batch! Applications are open till May 13

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: