Science desk
< June 24	<< May \| June \| Jul >>	June 26 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 25

A refrigerator that powers itself.

Is it possible to make a refrigerator that powers itself? Imagine a Carnot engine that uses the temperature gradient to power a refrigerator, maintaining the temperature gradient. 203.45.159.248 (talk) 09:58, 25 June 2014 (UTC)[reply]

What you have described is a perpetual motion device; it assumes greater than 100% efficiency, since the cold box itself could never be perfectly insulated. AlexTiefling (talk) 10:07, 25 June 2014 (UTC)[reply]

I just saw a promotional video the other day about a refrigerator that Coke has developed that uses sunlight and plants to keep things cool. It requires no electricity. Dismas|^(talk) 10:16, 25 June 2014 (UTC)[reply]

It would be using an external energy source, however unconventional. ←Baseball Bugs ^{What's up, Doc?} carrots→ 11:19, 25 June 2014 (UTC)[reply]

Lol, sunlight definitely counts as a source of energy! Zzubnik (talk) 12:18, 25 June 2014 (UTC)[reply]

Precisely. Purists may not regard the sun as a perpetual motion "machine", but for all practical purposes it's a perpetual energy source. ←Baseball Bugs ^{What's up, Doc?} carrots→ 12:22, 25 June 2014 (UTC)[reply]

The pot-in-pot refrigerator works without electricity, but of course someone has to expend energy by regularly adding water to it and the Sun's energy is required to evaporate the water. Richerman (talk) 12:37, 25 June 2014 (UTC)[reply]

You could (probably) build a solar-powered Einstein refrigerator (gas is the normal heat source for them). CS Miller (talk) 12:47, 25 June 2014 (UTC)[reply]

Or a Thermoelectric_cooling refrigerator (better known as a "Peltier cooler") which runs from a 12-volt power source could be powered by a solar array charging a 12-volt car battery or a stack of such batteries. Peltier coolers are cheap (<US$100) and plentiful under the "Koolatron" and other brand names; the Coleman sporting goods company has sold them for years. I own several. Harbor Freight Tools sells a portable solar array with charging circuitry for one or more 12-volt car batteries for US$150 or so. The batteries themselves cost less than US$80 each at auto parts stores new, or much less used from junkyards.

The mechanism here is using a bimetallic junction (a Peltier device) to transfer heat outside an insulated box by running electricity through the junction in the proper direction. Interestingly, these coolers can be made into warmers (for food, etc) simply by reversing the direction of electricity through the Peltier device, a simple matter of removing the power plug from the cooler and flipping it as you re-insert it, so the power wires are connected in reverse order than before. loupgarous (talk) 07:01, 26 June 2014 (UTC)[reply]

Thinking about it, though, it depends on what you mean by "refrigeration." Very hot food could be placed inside a Peltier junction "refrigerator," so that its heat was transferred through the junction by an electric fan... and NOT applying power to the junction, electricity could be tapped from that junction (by the Seebeck effect) to charge a very efficient battery... which would store power then used later to drive the Peltier junction and fan to cool the food further than it was cooled by losing heat initially by simple radiation into the insulated box to then drive the Peltier device as a passive heat-driven generator.

The food wouldn't get as cool as it would have just sitting on a table top, because convective transfer of heat to the open air is a much more efficient way of cooling food - it couples the hot food to a heatsink the size of the Earth's atmosphere. The food just never gets cool enough to be safely stored that way outside a REAL refrigerator or the outside of a home on a winter day. But technically, power could be generated by Seebeck effect from some very hot food radiating its heat to a Peltier device, then stored to cool the food further. It would just be a ruinously inefficient process. Joule heating would rob you of energy at every step in the process, to say nothing of the power consumed by the fan pushing the warm air past the Peltier device, which would have to be re-configured from its commercial state to be a more efficient Seebeck effect generator. Not a practical or even a reasonably effective refrigerator at all.

This takes what I assume is the original poster's point that the food has heat that could, theoretically, drive a Carnot cycle or other heat transfer device - but inefficiencies in operating such devices make that scheme much less efficient than letting the food cool in the open air, also a heat transfer device with the advantage of coupling ultimately to comparatively huge heatsinks like the air inside a building or the Earth's atmosphere without the use of outside power to effect the heat transfer. loupgarous (talk) 07:20, 26 June 2014 (UTC)[reply]

As a practical example, look at a microprocessor. A passive Peltier device is easy with one side at ambient and the other junction at the CPU temperature. The ambient side forces a cooling. But a heat pipe is much more efficient. The last time I looked the thermal conductivity of the wires contributed more than the passive Peltier junction. Active Peltier can force a temperature at the CPU but the heat generated is pretty large. It's not particularly effective at heat transfer. If it were, the CPU heatsink would be far way and submerged in liquid, connected only by wires. But in reality, most efficient CPU cooling brings the liquid to the CPU without any Peltier effects. Thermocouples use elements like platinum/rhodium junction at near 0 currents for temperature measurement. Forcing a temperature at the juncion incurs IR drup that is usually unacceptable. Not sure what the commercial cooler use or what their range is. I believe they need a fan at the cold end for IR drop waste heat. --DHeyward (talk) 09:36, 27 June 2014 (UTC)[reply]

You have a missconception of the Carnot engine and thermodynamics. It "feeds" on a thermic difference and you want to use it to poduce its own food. Its efficiency factor would have to be higher than 1 because it had to compensate the thermic loss by leakage of the refrigerator on top of being a perpetuum mobile, which in theory has an efficiency factor of exactly 1. Thus this is impossible². --Kharon (talk) 17:45, 27 June 2014 (UTC)[reply]

Electric Power vs Horse Powet

Is it true that an electric eel(by its elecyric shock) can paralysie a horse or it is just a misconception? — Preceding unsigned comment added by 106.212.123.52 (talk) 23:49, 25 June 2014 (UTC)[reply]

According to the National Geographic article here they have been known to knock a horse off its feet. Also see an alligator getting more than he bargained for. Richerman (talk) 00:44, 26 June 2014 (UTC)[reply]

The Electric eel has electric organs capable of producing potentially-lethal shocks that allow them to stun prey. However horses are not the eel's natural prey and the linked article notes that the shock's short duration (<2 ms) make it extremely unlikely to be deadly to an adult human. In 2011 two racehorses were fatally electrocuted in a freak accident where they may have contacted underground cables. Humans who generally wear insulating footwear, in contrast to conductive metal Horseshoes, noticed only slight tingles when walking with the affected horses. The newspaper article claims that "Horses are far more sensitive to electricity than humans...Their heavy muscles, thick tendons and large bony skeletons keep electrical currents in their body longer, causing more damage". Exposure of a horse to electric eel shock is very unlikely, though horses are routinely restrained by electric fences. 84.209.89.214 (talk) 01:07, 26 June 2014 (UTC)[reply]

A horse walking in a river could easily get a shock from an electric eel Richerman (talk) 01:53, 26 June 2014 (UTC)[reply]

Sounds like complete misconception. The energy from the eel is very small. The voltage and current are impressive but unlike utilities, the duration is finite. The most it could do would be to contract muscles in the leg but only for a very short duration. The duration of shock is much too short to affect heart muscle (I believe 25mS or longer is required where the eel is at about 2 mS). Longer muscle groups like legs and back are affected by shorter pulses (one of the reasons a Taser causes legs and back muscles to spasm without affecting heart muscle). At 2 mS, a 600V, 1A shock is 1.2 Joules of energy. A cell phone battery, I believe, has about 15,000 Joules. Can you stop a horse with a cell phone? --DHeyward (talk) 10:13, 27 June 2014 (UTC)[reply]