Can we air condition our way out of extreme heat? A primer on air conditioning from The Climate Brink
https://www.theclimatebrink.com/p/can-we-air-condition-our-way-out
Fun thermodynamics fact: if you want your house to be 75F and the outside temperature rises from 96F to 100F, your air conditioner will consume 42% more power.
@andrewdessler I enjoyed reading your article Andrew! You might wish to correct a minor mistake in footnote 1, where the 600F temperature should translate to 589K instead of 615K. This correction does not affect much the efficiency, which remains close to 50% (0.49) using T_h = 589K.
@andrewdessler I don't think so. (As your math shows quite clearly!) That's why we did Ep. 208 on vernacular architecture. I plan to explore this line of thought more in the coming years. https://xenetwork.org/ets/episodes/episode-208-vernacular-architecture/
@andrewdessler centigrades please, so the world can understand
@andrewdessler That's a wild stat. I wanted to see if I could plot some of that data against a popular central air unit (Full disclosure - ChatGPT is helping).
So, I took a look at what was the best-selling central air unit on Amazon, it was the "ROVSUN 11000 BTU". Looked up the specs for the efficiency curve. I had ChatGPT graph theoretical efficiency against actual efficiency at 96 to 100, as well as 92 to 96. The theoretical efficiency does track, but much more accurately, at 96 to 100.
@andrewdessler Efficiency data found here (based on furnace model number): https://qualifiedproducts.efficiencyvermont.com/evt/products/details/1191157
Choosing the central air unit was unscientifically done by searching "central air unit", sorting by "Best Sellers" (based on this month), and chosing the first non-sponsored listing.
@chris_hayes
wow, that's interesting. I'm guessing that these AC units are really optimized to run in the 90s, so this probably should not be super surprising.
/@andrewdessler@mastodon.world The explanation that comes with this fun fact is that the power goes with the square of ΔT, as the efficiency of a heat pump is proportional to ΔT/T, and the amount of energy pumped is proportional to ΔT, as well.
IMHO, this ignores that aircons also have to take out humidity, and the amount of water hot air can carry is exponentially growing with 7%/°C, or around 4%/F. So if your heat is humid, it is even worse than that.
@andrewdessler Sorry, but this can’t be true. A 4 F degree difference in outside temperature wouldn’t make such a huge difference in efficiency. I see now that you are using the Fahrenheit temperature in an equation that requires absolute temperature, such as Kelvin temperature.
@EricFielding @andrewdessler Make OP use the Rankine scale, as penance.
@Virginicus @andrewdessler I could not even remember the name of the Rankine temperature scale yesterday.
@EricFielding At a glance, it looks like the calculation is just combining Carnot efficiency with Newton's law of cooling. The result of that line of logic seems to be that in the two scenarios the ratio of the work required is W_2/W_1 = (dT_2/dT_1)^2, so in this case (100 F - 75 F)^2/(96 F - 75 F)^2 ~=1.4, so it's not a typo. @andrewdessler
@andrewdessler except it's not true for real heatpumps, more like 5%
@andrewdessler I missed your original post and I see you were referring not just to COP but to the whole energy budget of a building. But your assumption that the heat exchange would be dominated by Newtonian thermal conduction makes sense only in very special scenarios. In those cases, if you have a decently insulated house the conduction would be so small that you're actually increasing by 42% a VERY small power...
Anyway, the COP curves are not linear in DeltaT.
@andrewdessler The only case I see where your 42% number could be a decent approx is a very reflective car (not too much radiation involved) with little or no insulation. In this case the conduction would dominate especially due to turbulent exchange effects increasing hugely the energy transfer.
On the other hand, if the air conditioner was based on a ground source heatpump, or better, linked to a district thermal network, the COP would be much less sensitive to outside air temp. (Same applies to heating in the winter)
Why aren't we exploring how to implement district thermal networks in urban environments? Significant energy efficiency benefits
@andrewdessler what's the main c factor here? Thermopump having to work harder? Or a bigger difference between the house and outside which leads to faster heat transmission to the house? Or some other factor all together?
@andrewdessler ah... Just saw the article in the original toot. The answer is the pump having to work harder. There's probably the house getting warmer faster as an additional factor too.
@andrewdessler So, we’re gonna implement that in Texas here soon, yeah? 