Humidity is definitely a huge factor. A/C acts as a dehumidifier. Another factor is the amount of IR being emitted by interior surfaces. If it's really hot outside (and the building isn't insulated to the latest top-tier building standards), the ceiling and walls can be radiating enough IR heat that a 78F air temperature feels quite uncomfortable.
I don't know much about how A/C works, whether the dehumidifier mechanism can be decoupled from the cooling mechanism. If it can't, then shouldn't the thermostat be coupled to a hygrostat so that that A/C runs when it is humid enough despite being cool in terms of temperature?
On the other hand, today I learned that there's something called heat index and humidex.
> I don't know much about how A/C works, whether the dehumidifier mechanism can be decoupled from the cooling mechanism.
It can't. When you cool air down to 50 degrees F (10 degrees C) or so (which is a typical outlet air temperature for a home A/C unit), it simply can't hold much water vapor, and the excess condenses out. That effectively dehumidifies the air--it lowers the dew point to whatever the outlet air temperature is.
> shouldn't the thermostat be coupled to a hygrostat so that that A/C runs when it is humid enough despite being cool in terms of temperature?
In principle it probably should, but in practice there is a close enough correlation between the two that it's not considered necessary.
If energy savings is the goal, the real fix would be to use variable compressors in home A/C systems: that way, instead of the A/C running full on for a while, lowering the temperature (and dehumidifying), then shutting completely off for a while (so humidity can build up again), the A/C could just be running all the time, and changing the thermostat setting would simply change how much cooling and dehumidifying the A/C was doing--what the actual outlet air temperature/dewpoint was. "Full on" would still be, say, 50 degrees F outlet air/dewpoint, but most of the time you wouldn't need that; "energy saving mode" could be, say, 68 degrees F outlet air/dewpoint, which would still be reasonable in terms of comfort but would use a lot less energy, so it could be kept on continuously. (There are already systems like this in cars.)
There are variable refrigerant flow heat pumps and variable speed blowers out there. And standalone dehumidifiers, where the hot and cold side of the heat pump are in the same air stream. The dehumidifier might have its own separate vents, or it can tie into the central vent system, or be part of an external vent system with an energy recovery ventilator.
This channel has some interesting videos about the subject:
> "Full on" would still be, say, 50 degrees F outlet air/dewpoint, but most of the time you wouldn't need that; "energy saving mode" could be, say, 68 degrees F outlet air/dewpoint
If you do that, then the outlet air will have twice as much water.
If you started with hot, humid air, that means you're still going to have quite a bit of humidity even as you drop the temperature through the 70s.
So you're going to need to cool the air more to reach the same level of comfort.
> If you do that, then the outlet air will have twice as much water.
As long as that dewpoint is reasonably comfortable, that's not necessarily an issue.
> If you started with hot, humid air, that means you're still going to have quite a bit of humidity even as you drop the temperature through the 70s.
So you're going to need to cool the air more to reach the same level of comfort.
You can't cool the air further without also further dehumidifying it. If you cool the air to, say, 60 F, the dewpoint can't be any higher than 60 F. There's no way to cool the air to 60 F but have a dewpoint of 68 F. That's physically impossible.
> Are you sure the savings outweigh the costs?
There is no such tradeoff: lower air temperature and more dehumidification always go together; they have to, as explained above.
The savings comes from running continuously at a lower level of both cooling and dehumidification, one that is just sufficient to maintain comfort, instead of switching between full on (which cools more than needed to maintain comfort--it has to, to compensate for the period when the system is going to be off again) and full off (which doesn't cool or dehumidify at all, so the air warms up and gets more humid again until it reaches the point where the system has to switch back to full on). The switching on and off is less efficient for a given level of comfort.
> You can't cool the air further without also further dehumidifying it. If you cool the air to, say, 60 F, the dewpoint can't be any higher than 60 F. There's no way to cool the air to 60 F but have a dewpoint of 68 F. That's physically impossible.
The other way around. It's not that the dew point will be above the temperature, it's that it could go below the temperature.
So as you suggested, let's have one AC that outputs air at 68F, and compare it to another that outputs at 50F.
So clearly we have to be cooling the room to a temperature no lower than 68F.
If we look at the initial cooling of hot wet air, there's not much difference. The 50F unit will cool a smaller volume of air and cool it to a lower temperature, and the resulting dew point is about the same for either unit.
But as heat keeps leaking in through the walls and windows, the 68F unit will never get the dew point below 68F. While the 50F unit will keep outputting air with a dew point of 50F.
If you had an airtight room with heat constantly leaking in, eventually the 50F unit would cycle all the air through it, and the dew point of the room would reach 50F, even though the temperature of the room might never go below 70.
All rooms leak some air, so you'll never get quite that far, but the 50F unit will get the dew point a lot lower than the 68F unit.
> The savings comes from running continuously at a lower level of both cooling and dehumidification, one that is just sufficient to maintain comfort, instead of switching between full on (which cools more than needed to maintain comfort--it has to, to compensate for the period when the system is going to be off again) and full off (which doesn't cool or dehumidify at all, so the air warms up and gets more humid again until it reaches the point where the system has to switch back to full on). The switching on and off is less efficient for a given level of comfort.
There's an efficiency penalty to letting the temperature wander. But if it only wanders a couple degrees, it's not a huge penalty.
On the other side, there's a penalty for keeping the fans and compressor going all the time, even if they're running at a lower power.
Neither one of those factors is enormous.
But if you could leave the room at 73 with a low dew point, instead of 70 with a high dew point, that could eventually save you a lot of energy.
So that's why I'd want to see real-life efficiency numbers and comfort ratings. It could go either way on base principles.
Your home's A/C unit can't just pick a particular volume of air and cool it. It takes in air from your entire house, cools it, and puts it back out into your house.
Moreover, your whole comparison is irrelevant to what I am saying. The objective is not to get the dewpoint (or temperature) as low as possible. The objective is to get the dewpoint (and temperature) just low enough for comfort (note that the target dewpoint might indeed be lower than the target temperature--see further comments below), and then keep it there by running the A/C continuously at a lower output, instead of running the A/C at full on for a while, which outputs air with a dewpoint (and temperature) lower than required, and then shutting off the A/C for a while, which lets the air's dewpoint (and temperature) rise to a level higher than desired. In other words, target the A/C's output to just where it needs to be, instead of cycling it on and off.
> There's an efficiency penalty to letting the temperature wander. But if it only wanders a couple degrees, it's not a huge penalty.
The main efficiency penalty is not from letting the temperature wander. It's from letting the dewpoint wander. Most of the work the A/C does is actually condensing extra water vapor out of the air, not cooling the air itself. If your room temperature rises from, say, 74 F to 76 F, that's not much more work for the A/C to do; but if the dewpoint of the air in the room rises from 50 F to 76 F (which it might well do on a hot, humid day during the off cycle of your A/C unit), that's a lot more work for the A/C to do to condense out all that extra water vapor.
> there's a penalty for keeping the fans and compressor going all the time, even if they're running at a lower power.
There's also a penalty for starting them up every time from off. For many such devices, given the number of times they cycle on and off during a hot, humid day, that penalty can easily average out to be larger than the cost of running the unit at lower power continuously.
> if you could leave the room at 73 with a low dew point
Which you can't if the system is off. The only way to control the dewpoint is to run the A/C continuously. As soon as you shut the A/C off, on a humid day, the dewpoint will start rising. And then your A/C will have to work harder during its next on cycle to lower it again.
If you run the A/C continously at lower power, then yes, you can adjust things so the heat coming into the house raises the temperature (from the A/C outlet temperature, which is basically the dew point) just enough to have the combination of temperature and dew point that you want. That's basically what I've been describing. The dewpoint of 68 F that I first threw out might be higher than optimal; the point was not to pick a specific optimal number but just to illustrate the concept.
> Your home's A/C unit can't just pick a particular volume of air and cool it. It takes in air from your entire house, cools it, and puts it back out into your house.
I'm talking about how much air goes through the unit. By the time it's done, it will have emitted fewer cubic feet of air.
> Moreover, your whole comparison is irrelevant to what I am saying. The objective is not to get the dewpoint (or temperature) as low as possible. The objective is to get the dewpoint (and temperature) just low enough for comfort (note that the target dewpoint might indeed be lower than the target temperature--see further comments below)
It's not irrelevant if overshooting one target lets you undershoot the other target.
> The main efficiency penalty is not from letting the temperature wander. It's from letting the dewpoint wander. Most of the work the A/C does is actually condensing extra water vapor out of the air, not cooling the air itself.
Same difference. If you cycle the AC on and off every few minutes, neither one will wander much.
> There's also a penalty for starting them up every time from off. For many such devices, given the number of times they cycle on and off during a hot, humid day, that penalty can easily average out to be larger than the cost of running the unit at lower power continuously.
It only takes a few seconds of power to spin up. If that is the dominating inefficiency in the system, then it barely matters what you do.
> Which you can't if the system is off. The only way to control the dewpoint is to run the A/C continuously. As soon as you shut the A/C off, on a humid day, the dewpoint will start rising. And then your A/C will have to work harder during its next on cycle to lower it again.
If you run it a little bit every ten minutes it will do almost as good a job at preventing drifting. It will work "harder" but for less time, so averaged over the entire 10 minutes it's probably about the same. Might be more, might be less. Depends on how efficient the fans and compressor are at different speeds. And how efficient the evaporator is at different flow rates.
> If you run the A/C continously at lower power, then yes, you can adjust things so the heat coming into the house raises the temperature (from the A/C outlet temperature, which is basically the dew point) just enough to have the combination of temperature and dew point that you want. That's basically what I've been describing. The dewpoint of 68 F that I first threw out might be higher than optimal; the point was not to pick a specific optimal number but just to illustrate the concept.
The different modes having different dew points was a core part of your original suggestion.
If you're changing that, and now we're talking about a system where "full on" is an optimal dew point but high air flow, and "energy saving mode" is the same dew point but with low air flow, then that sounds great. And if you had suggested that originally, I wouldn't have posted to question it.
> If you cycle the AC on and off every few minutes, neither one will wander much.
The dewpoint will, yes. There's a lot more room for dewpoint to rise than for temperature to rise.
> It will work "harder" but for less time, so averaged over the entire 10 minutes it's probably about the same.
I don't think so.
> The different modes having different dew points was a core part of your original suggestion.
> If you're changing that, and now we're talking about a system where "full on" is an optimal dew point but high air flow, and "energy saving mode" is the same dew point but with low air flow
Neither of those is what I'm talking about. I'm talking about a system whose only mode (unless you want to have an "initial cooldown" mode as well, see below) is what you called "energy saving mode": the dewpoint is set higher than a current A/C unit's "full on" mode, so that it's just low enough for comfort, ahd the A/C runs at much lower power (lower compressor power and lower airflow) continuously to maintain that dewpoint, instead of running "full on" (high compressor power, high airflow) for a while at a lower dewpoint, then cycling completely off for a while and letting the dewpoint get higher than the comfort level.
The point of running in the above "energy saving" mode is that you do it all the time; you never shut the system off. But there might be cases where the system turns off for a while unavoidably (say a power outage) and the house gets well above the temperature and dewpoint target. For that case, the system would have an "initial cooldown" mode, where it does run "full on" for a while (high compressor power, high airflow, dewpoint below optimal) until it senses that it's back to its usual setpoint, then it goes back into "energy saving" mode described above.
Running at the same (low) dewpoint but lower airflow would save some energy (less fan power), but much less than running at a higher dewpoint, since that reduces the amount of dehumidification, and dehumidification is most of the work that the A/C system does.
