But in any case, I finally found them and ran the numbers. The latent heat of water is ~4x that of air, but even on a really hot humid summer day water is only 3% of air by mass, which means in a worst-case common scenario, 11% of energy is going to condensing the water, while 89% is simply cooling the air and (non-condensing) humidity.
So if we round, we can say that a heat pump cooling humid summer air rather than drier winter air uses up to around 10% more energy (since winter air has a bit of humidity too), which means heating/cooling are still roughly the same for most purposes.
Whereas the commenter I was originally replying to seemed to be suggesting cooling takes 3.5x as much energy, which doesn't seem to match summer/winter energy bills at all.
Does that cover it, or is there something else my math is missing? Thanks again.
[1] https://en.wikipedia.org/wiki/Latent_heat#Specific_latent_he...
(Sorry for the use of “stupid units” here. It’s a lot easier to find the non-SI units for the HVAC applications of these and being an American, it’s easier for me to think in BTUs than calories, kWh, or Joules for HVAC.)
It takes (negative) 970 BTU to condense one pound of water from gas to liquid. (That’s the latent heat of vaporization.) It takes -0.24 BTU to cool one pound of mixed air by one degree Fahrenheit.
At 90°F, air can hold almost 5% of water vapor in suspension. If the task is to condense out about half the water (from 90% to 45% RH) and cool it from 90°F to 70°F, I think it’s:
For every 40 pounds of 90°F/90% RH air that you want to make into comfortable inside air:
You condense out 1 pound of water at a cost of sinking 970BTU. It’s slightly less than 1 pound, so generously call it 900BTU of latent heat removal.
You then cool 39 pounds of mixed air by delta-T of 20°F: 39 * 20 * .24: 195BTU of cooling required for sensible heat removal.
That makes it look to me like over 80% of the heat removal is latent and under 20% is sensible. (Phase change is expensive, but you’re glad for that when you’re sweating in air that’s not 100% RH.)
This is tapped out on a phone while traveling, so I’m not staking my life on it, but I think it’s right.
That's a worst case scenario, though.
For one, 90/90 would translate to a 87F dew point -- which is not a record, but it's pretty close, and there's a good chance it would be lethal to unprotected humans. So it does not happen too often, e.g. normal in a place like Phoenix Arizona might be 110F but 20% RH.
Second, air conditioning doesn't cool outside air, it cools inside air. So on average it isn't going to have nearly that much water to pull out of the air.
That would come as a great surprise to the millions of people who live in places like Houston, Texas or Miami, Florida, where 90/90 is common. (And that's only places in the US, since those are what I'm familiar with from doing HVAC testing years ago. Around the world there are many places that are even worse.)
> normal in a place like Phoenix Arizona might be 110F but 20% RH.
Yes, but Phoenix is a desert climate, not a hot, humid climate. There are many places that have hot, humid climates.
I’ll assume you were kidding about a dew point of 87°F being lethal to unprotected humans.
Make the assumptions such that you take 1 pound of water out of 200 pounds of air and (0.5% reduction in water content) and the latent and sensible heat are about equal. Dehumidifying is expensive energy-wise is the point.
That's a heat index of 122°F. You're literally able to die because you're unable to evaporate away body heat quickly enough if you don't take precautions.
On the NWS's heat index chart, that's color-coded "danger" while a single tick further up in either humidity or temperature is coded "extreme danger".
So I don't see this as kidding... it's some pretty extreme conditions you chose. :)
If you stay out in it without taking any precautions for long enough, it could be. But that doesn't mean that as soon as you step out of your front door in the summer in, say, Houston, Texas, you fall over and die. Millions of people in the US (quite possibly billions if you count the rest of the world) live in places where the outside conditions are 90/90 (or worse) frequently in the summer.
> it's some pretty extreme conditions you chose
As I have already posted elsewhere in this thread, 90/90 is a condition that is commonly tested for HVAC systems; it's not at all "extreme".
I can see it basically depends on a lot of factors.
I'm presenting temp/humidity numbers that are reasonable for where I live (NYC where the summers still are pretty hot and humid), but your numbers are much more extreme (and god forbid they should be daily averages!). And then, you're dehumidifying down to 45% RH -- my consumer-level heat pump doesn't even have that option. In the summer (like today) it never goes below 65% RH indoors on hot humid summer days, even if I put the indoor temperature as low as 68°F -- I can't control the humidity. I'm sure for office HVAC systems there's more control.
Also as another commenter mentioned, I'm also mostly cooling indoor air that's already had the humidity removed. You commented that typical houses have 0.5-2 air changes per hour, but my new-construction NYC apartment is nothing like that. It's maybe 0.5 air changes per day, it's so tight -- I know because I have a CO2 meter so I know when to open the windows and run a fan to force fresh air when levels double. Again, professional office buildings might bring in a lot more fresh air -- my apartment is new-construction so pretty airtight, and with a split heat pump so it can't intentionally bring in air from anywhere. It's not some massive HVAC.
Anyways, it seems like we're both right -- in a regular new-construction home in a climate where it makes sense to use a heat pump both for heating and cooling, heating and cooling costs are roughly the same for a temperature difference in either direction (as reflected by my electric bill). While in a humidity-controlled office building in an extremely humid tropical climate with tons of airflow where you can control the humidity level independently, the energy usage for cooling and removing humidity could significantly exceed the energy used for cooling air.
NYC, while it does get fairly hot in the summer, is actually not very humid compared to South Texas or South Florida.
> you're dehumidifying down to 45% RH -- my consumer-level heat pump doesn't even have that option
A typical home A/C system (not heat pump, just straight A/C) in South Florida or South Texas will output a dewpoint of around 50 to 55 degrees F. (I used 12 C = 53.6 F in my heat transfer calculations upthread.) That's significantly lower than the typical dewpoint output of a heat pump (different working fluids, hence different temperatures at the indoor heat exchanger).
That’s why I gave you the term, in an effort to show you politely that you were wrong enough that the HVAC industry has an entire term specific to this concept and that suggests that it’s more than a rounding error.
It appears that you at least now agree that water in gas form is condensing into liquid form (which requires the removal of the heat of vaporization [what science calls it] or the latent heat [what HVAC calls it]).