SoylentNews is people
SoylentNews
Ever since our ancestors mastered fire, humans have been able to warm themselves. Cooling down when it's hot has been more challenging.
The eccentric Roman emperor Elagabulus sent slaves to bring snow down from the mountains and pile it in his garden, where breezes would carry the cooler air inside.
[...] Needless to say, this was not a scalable solution. At least, not until the 19th century, when Boston entrepreneur Frederic Tudor amassed an unlikely fortune doing something similar.
He took blocks of ice from frozen New England lakes in winter, insulated them in sawdust, and shipped them to warmer climes for summer.
Until artificial ice-making took off, mild New England winters caused panic about an "ice famine".
Air conditioning as we know it began in 1902, but it had nothing to do with human comfort.
New York's Sackett & Wilhelms Lithographing and Printing Company became frustrated with varying humidity levels when trying to print in colour.
The same paper had to be printed four times in four colours, and if the humidity changed between print runs, the paper would slightly expand or contract. Even a millimetre's misalignment looked awful.
(Score: 2) by edIII on Tuesday June 06 2017, @10:04PM (6 children)
The Peltier effect was understood in 1834. With the application of electricity to it you have a solid state air condition circuit in 1834.
Probably horribly inefficient and way too expensive, and I think material sciences needed 100 years or more to catch up and make efficient Peltier/Seebeck devices. I know they use the Seebeck effect to generate electricity on some offshore drilling platforms now because they've become that efficient.
We were close though. An air conditioned White House in the 19th century :)
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 5, Interesting) by Grishnakh on Tuesday June 06 2017, @10:22PM (5 children)
You don't need the crappy Peltier effect for cooling air, you can use vapor-compression refrigeration just like we do now and like they did with the first A/C systems. Vapor-compression refrigeration goes back to the 1830s too [wikipedia.org], though it wasn't really used practically for another 15 years or so. So they could have had actual air-conditioning in the White House back then too. The problem was electricity; they didn't have a practical way of powering a large refrigeration system back then, and they probably hadn't come up with the idea of blowing air over the chilled coils either. A lot of technologies are like this: the fundamentals are discovered or invented decades or centuries before they become practical and commonplace. Look what happened with electric cars for instance; we had those back in the early 1900s, and only now are we getting them back, mainly because of limitations in battery technology.
(Score: 2) by edIII on Tuesday June 06 2017, @10:29PM (4 children)
Why do you think the Peltier effect is crappy? The efficiencies they have now make it work. I'm just curious what you think is wrong with it.
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 2) by The Mighty Buzzard on Tuesday June 06 2017, @11:10PM
There's nothing wrong with it but unless there have been massive improvements in efficiency in the past twenty years, phase-change cooling is more efficient.
My rights don't end where your fear begins.
(Score: 2) by bd on Tuesday June 06 2017, @11:59PM (2 children)
The peltier elements I worked with had something on the order of 0.6 W thermal for every 1 W electrical you put in, so the efficiency P_thermal/P_electrical would be something below 1. I think there was some fundamental physical reason for that efficiency being below 1, but not sure anymore.
That efficiency is not too bad. The problem is, even crappy chillers will move more thermal power than the amount of electrical power you put in. For a good one, the efficiency should be more like 3 or 5, I think. Add the price of high-power peltier modules and it quickly becomes non-economical.
I would rather use peltiers when you can't afford vibrations and complex plumbing. Even though, the nasty bit is that you have to keep the warm side of the peltier cool or P_thermal goes down and at some point it will suddenly start heating the cold side. In the application I used it for, that cooling was ironically done with a chiller.
If you know about higher efficiency modules, could you provide a link? I would actually be interested.
(Score: 2) by edIII on Wednesday June 07 2017, @01:20AM (1 child)
You know, I honestly just thought if they were capable of cooling a modern CPU, that at scale that must be capable of cooling a room. I really thought that they were more efficient than that these days. Perhaps I was confused with the Seebeck effect instead where you generate electricity.
Thanks for the technical descriptions. I think I may have learned something today...
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 2) by bd on Wednesday June 07 2017, @09:03AM
They are totally capable of cooling a room or being used as A/C in a car. Peltiers do not have particularily bad efficiency, the problem is just that closed cycle chillers are _much_ better.
Actually, disregard the thing about the coefficient of power for the peltier always being below 1. I read up a little bit and you can achieve something like 1.2. The problem is: that is generally only possible with a thermal load far lower than the maximum of the peltier. If you want to use it at its rated specs, prepare for inefficiency. And 1.2 is still less than chillers with up to 7, if you take an extremely good one. There are also peltier stacks, but that is a different can of worms.
Another drawback of peltiers is that they get less efficient the higher the temperature difference. The optimal application of a peltier is therefore to keep something that generates heat temperature stabilized close to room (or rather heatsink) temperature.