Central A/C - turn off or leave on at higher temp?

Do you have your air-conditioning on all the time when you are home or as required?

I would probably be leaving it on when I'm home, sleeping, etc.

Keep it off while you are out. An app will recognize that you are returning home using the GPS on your phone and turn on the AC about 15-30 minutes before you return home. Using the timer would work too, just not as cool.

if so important for you that house is cool when you get in then consider investing in an 'intelligent' thermostat like Nest - it will know when you come in and leave and when you sleep etc then will tailor your cooling (and heating) to your patterns. Costs about $250 per unit though and you would need one in most rooms. Should ultimately save money compared to leaving AC running.

Too bad no one answered with a "it depends" because it depends.

Do you face sunlight? How many walls face outside? How big are your windows? Do you want to keep the shades open or closed (for plants or pets)? Is there Low-E glazing on your windows? Are you the top floor?

The less exposure you have to sunlight and/or the outdoors, the more likely it is that your apartment will stay cool with minimal effort. In such an apartment it would not take much at all to keep the temperature at a comfortable low-80's that can be brought down further quite quickly when you get home. On the other hand, if it is an apartment with a lot of exposure, it would take a lot of effort to maintain such a temperature and you would be much better off leaving the windows open, then blasting the AC when you get home (or set to do it a bit before you arrive).

(In my old apartment I had 8000-9000 BTU AC in the Living room even though it was only about 170-200 Sq. Ft., but there were two facings, the equivalent of 4-5 windows, the roof was above me, and no obstructions, so open to a lot of sun.)

The Energy Conservation Police are out to get you. Turn off that AC when yo are not home. Shame on you!

Turning it on when you return is definitely more energy-efficient. As far as putting it on a timer, depends on the strength of the cooling. My current building has centralized cooling, so you get massive cooling capacity (that is there for the full building) delivered to your place. If you crank the AC on full-blast, it'll be frigid in 5 minutes even on a hot day. My last building was woefully under-cooled, had to run AC 24/7. If you turned it off at night, temperature would rise above outside air temperature as the heat from the sun-baked building rose up to the apt.

Yeah, I figured it would theoretically make more sense to just turn it off when I'm out. But a few friends of mine had said that turning it of and on takes more energy because the system has to work harder to get to the desired temp than if you just left it (albeit at a slightly higher temp when you're out of the house).

AvUWS - The living room/kitchen area faces south and has 25" of window exposure. The bedrooms face the inner courtyard on the north side so they get little sunlight.

Your friends lack a basic understanding of physics and AC mechanics. Remember "energy = work * time".

ACs are either "on" or "off", the amount of cooling they do (and hence power they consume) cannot be modulated. Whenever the temp goes above the thermostat setting, they turn on full-blast. Once it falls below, they turn off. So the amount of "work" they can do is fixed. Less cooling involves doing that same amount of work for a smaller fraction of time. This idea is called a "duty cycle".

As far as energy usage, ACs spend about 1 unit of energy to remove 3 units of energy (aka heat) from a room. In a perfectly-insulated environment, you'd never need to cool a room once you take it down to the desired temp. However, rooms are not perfectly insulated, so they will heat up over time as the walls/etc. transfer in energy (i.e., heat) "leaking" from the outside. It is this sucked-up energy that the AC is removing. The rate at which the room sucks up energy (and hence the energy the AC must expend to remove it) is proportional to the temperature difference between the outside & the room. So the more time you keep the room cool, the larger the amount of energy that the AC must get rid of.

>ACs are either "on" or "off", the amount of cooling they do (and hence power they consume) cannot be modulated. Whenever the temp goes above the thermostat setting, they turn on full-blast. Once it falls below, they turn off. So the amount of "work" they can do is fixed. Less cooling involves doing that same amount of work for a smaller fraction of time. This idea is called a "duty cycle".

Interesting theories.

Inonada is smart

Inonada isn't very smart at all ... he forgot that while the compressor cycles on and off, the fan remains on constantly (unless the silly and annoying "energy saving" mode is chosen, which is rare).

So ACs are NOT generally either "on" or "off".

What a maroon.

"Interesting theories."

Thanks. But alanhart has had the last word, the shame.

The theory of evolution is just a theory, you know...

Inonada. What is the difference in your cooling bills with the centralized cooling system? Do you find you spend significantly less compared to a non-centralized system. Obviously you still have to drive your fans with electricity but how does it compare to enitre heater/cooling unit in your prior apt(s).

Obviously someone is paying for that centralized cooling - is it you or your landlord (i.e is rent higher to take account of this factor)?

It's complicated and not directly comparable.

In my old building, I split the energy bill with my LL 50/50. My unit was about twice as large as LL's and was on top, LL never ran AC (personal preference, and cooler to the tune of 10-20 degrees simply due to rising heat). Small building, shitty insulation, we were cooling for everyone. Summer bills would come in at 18 cents per sq ft per month for July / August (averaged over both units). Probably an average temp of 80 degrees in all floors, ours cooled by AC and LL getting residual cooling but cooler to begin with. Absolute-worst winter month was 35 cents per sq ft. Best spring/fall months, probably as low as 6 cents, so the difference should give you a sense for actual heating/cooling costs.

Current building is a big condo, so heating & cooling are in common charges. Common charges are on a per sq ft basis, but I don't know the building's ConEd charges. Without AC, apt would probably stay at 80-85 on worst days because hallways, other apts, etc. are cooled. I'm not sure what temp we average, probably 75. Electricity bill runs 3.5 cents per sq ft month in summer at worst, something like 2.5 cents in spring/fall when AC/heat is not run. So the fan for AC/heat is probably costing 1 cent per sq ft at worst. I pay electricity, consideration for common charges is implicit in rent.

I have central A/C, no building wide cooling plant. 1.2 cents/sq. ft - set to 75 at night, 80 otherwise. If it is below 75 at night, I turn it off and open windows. Have ceiling fans turned on in LR and bedrooms; use if it feels too hot. Turned off if we are away. 2 computers on; 1 large flat screen, refrigerator, wine fridge.

So for 1000 sq ft, a $12 electricity bill??? Using 25 cents per kwh, that works out to a 66 watt average usage per 1000 sq ft. Having a window-unit AC for an hour would blow the entire day's budget. You are either an eco-saint, or off by a factor of 10.

Typo. 12 cents.

IMHO, for some people walking into a room that's 80 degrees might not feel that good on a hot summer day. Same for sleeping in a room that's 75 degrees.

Winter is easier - turn off the heat and just pile on the down comforter(s).

This is an old question that has been addressed by HVAC specialists and physics geeks -- quick Google search finds summary answer:

Turn it off during the day. Basically it is easier for the air conditioner to remove heat the closer the inside temperature is to the outside temperature. Even ignoring the fact that your house will absorb less heat when the house is warmer, it is still more economical to turn the air conditioner off when you aren't home.

There are two reasons for turning it off:

1. The house absorbs less total heat energy once it warms up to the outside temp. So when you get home, the air conditioner has less total heat to move outside than had it been running all day.

2. The air conditioner is more efficient at moving heat from a hotter house to a hot outside than from a cooler house to a hot outside. Even if we ignore point 1, and assume that the total heat the A/C will have to move is the same if you leave it on or not, it requires less energy input to do this for a warm house than for a cooler house.

So, what inonada said.

I'm currently trying to convince my building to let me put an air conditioner where I want it -- they don't want me to put one in the bedroom; instead they want me to put the AC in the bigger room (which has two exposures and isn't suitable for sleeping in daylight) and either sleep in that room or buy a bigger and more powerful unit that will cool the entire 400-SF apartment.

I suspect they're going to get their way: yesterday's temperatures where I live were 88 F in my office, 81-94 F outside, and 93-103 F in my apartment. I'm going to collapse from heatstroke one of these days.

People in hot desert places insist that the upholstered furniture absorbs and fiercely holds so much heat that you really need to run the a/c while you're out or it'll take hours to cool the place down.

nyc10023, you are lying ... that was not a typo, it was an ERROR. You must issue a formal statement of apology.

Triple_Zero, don't you live in the Japanese Empire? I thought they use split-ductless there, in which case you can put the outside unit wherever your building says, and the inside unit wherever you want. No? Or can you use a (very inefficient) portable unit in the bedroom?

I'm opposite to most people in terms of heat tolerance and sleep habits. I would rather sleep on top of sheets and/or duvet in the summer in an 80F room with fan on (find it soothing) than air-condition room to 68F and sleep under duvet.

West34 - your physics isn't completely right.

There are multiple variables that make it "easier" to reduce a higher temperature than a lower temperature, but it isn't because less energy input is required.

First lesson is that the energy input in all these situations is ALWAYS THE SAME. This is what Inonada was trying to explain. Any AC unit, when it is trying to cool, will always "try" at the same rate. This is one of the biggest areas where the public "lacks knowledge" (ignorance). Your AC will not work harder if the thermostat is lowered more. You oven will not get to a heat faster if you raise it more. Thermostats (unless you have a complicated thermocontroler and complicated heating systems, which unless you are in some sort of industry, you don't) only turn things on or off. That's it. The AC's only have one setting for their compressors, on or off.

Any other variation is a factor of the rest of the variables. the amount of energy (heat) that leaks in from the outside, from warm parts of the building around you. From unblocked sunlight. From poorly insulated windows. From normal air-exchange in the apartment (the cracks under the door and the vents in the bathroom. They are acting to raise the temperature of the apartment to some ambient temperature. The AC has to counteract that. It will have to be "on" for longer the more of this work it has to do.

That said, there is slightly more energy is used on starting up as motors spin up fans, compressors build up pressure, etc. But if the unit is not turning on and off quite often then this isn't much of an issue. Here is where the size and power of the unit matters. The most efficient unit is the smallest one that can maintain the desired temperature. And the smaller it is, the cheaper it is and the cheaper to run. But it will have a harder time getting to that temperature.

All of these will have a greater effect on your costs than the fact that it is harder for 30 deg refrigerant to change the temperature of 80 degree air by 20 degrees than it is for it to change 90 degree air.

So if you have a well insulated apartment, a smaller unit intermittently working to hold down the temperature might be more efficient than a bigger one that can bring it down quickly when you get home. (Remember, if your apartment was hot your unit has to both cool all the air and also fight the tendency of all the mass in the room -furniture, walls, etc.- to radiate their own now-excess heat into the air). And a bigger unit will cost more to operate, so it is best to use it less.

(I didn't really even go into the fact that for the same energy use, the same capacity units can be more or less efficient depending on the construction, in other words the units actual efficiency.)

http://www.slate.com/articles/health_and_science/science/2012/08/air_conditioning_haters_it_s_not_as_bad_for_the_environment_as_heating_.html

There're now two-stage or variable-speed compressors that could change the calculations.

West34/AvUWS, I can believe that an AC's efficiency can depend on both the temperature differential between inside/outside and the general level of temperature. However, I would have thought the efficiency curve is pretty flat in the temp ranges of interest around the EER / SEER standard conditions (e.g., within 10%). Any pointers to actual data on this?

They probably are similar because the technologies and market expectations are similar. Most vary 10%, some as much as 20%. There is only so much leeway each technology gives you. The big efficiency gain in a given size is bigger coils, but bigger coils means a lot more copper, and copper is today one of the biggest costs. And no one wants to be the guy selling a $350 AC window unit when the other guy is $250 or 300. (I went with a Friedrich quietmaster because it was 20% more efficient, but mostly because it was almost silent, but it was $850 instead of more common $250-300 for similar BTU units. I also guess they sell a small fraction of what is sold in the other units.)

Quiet is also a factor of cost. Bigger fans can move the same amount of air with less work (fewer rotations) and less work means less noise. But bigger fans mean more material, bigger housings, larger bearings, etc. and hence more cost. I was never able to find the AC I was replacing because LG no longer made them the same. Copper was cheaper 15 years ago, as were the Koreans that built it.

You also get better efficiency with bigger units, and even more with split units, but the price and complexity of installation goes up.

AvUWS -- I was actually just cutting and pasting since I had seen the question asked somewhere else awhile back. But I did take a lot of physics and math back in the day and I just thought about it and I think my original post is generally right:

- It's all about how much time does the compressor/fan need to run to transfer a given amount of heat from in to out.
- Leave the ac on - totally made up example - - let's say compressor kicks in once every 10 minutes for 1 minute to lower room one degree. ac turns on 6 times an hour for total 6 minutes, 48 minutes in an 8 hour day - theoretically temp keeps going from 70 to 71 back and forth all day every 10 minutes.
- leave the ac off - room warms lets say 10 degrees total during the day up to 80 degrees - ac needs to only run 10 minutes (1 minute per degree). Ha, highly idealized example but you get the idea.

AvUWS, that's not what I wasn't talking about the relative efficiency of one AC against another. Rather, EER and SEER have specifications. I think the former is 80 degrees interior, 95 exterior. The latter is 80 interior, 82 exterior. Or something like that.

If you put different temps, efficiency changes. I was talking about curves like the one found in http://www.fire.nist.gov/bfrlpubs/build02/PDF/b02186.pdf, figures 3 & 4. Their setup was 80 degrees interior, and based on outside temperature they measured the efficiency. It seems for those setups, as outside temp was taken from 82 to 95, efficiency dropped 15-20%.

If it's a very dry night (<50% humidity), you can just drape cloth all around your porch, wet it, and go to sleep out there. Cool. Some nonsense about the latent energy of evaporation, or something or what.ev.errrrr.

West34 - I get your point. Of course the situations don't exactly hold that way. In the warm room the AC has to cool the air and all the heat trapped by the mass in the room. If on all day the mass would remain at the 70 degrees. But you also need a unit powerful enough (in that example) to keep the temperature while working only 10% of the time. A unit that worked 100% of the time would waste less energy in turning on and would be cheaper because it is 1/10 the power, but it would not be able to cool the room in 10 minutes. (Life is full of compromises and choices.)

OTOH, if the room is leaking in a lot of heat, it is expensive to keep it cooled even to a moderate 80 degrees.

That is like the AC's in our cars. They are designed to cool a 100 deg. car in 3 minutes at full blast, but that means that they are bigger than necessary for regular use and use a lot of MPG when they are on.

So then what's more fuel efficient, with the average typical car on an average typical hot summer day in full sun at highway speed with 8-track on: a/c off, windows fully open (wind drag) ... or windows fully closed but a/c on?

At highway speeds, AC is better.

AC is better. Just check out Mythbusters. Open windows create a lot of drag.

Thanks, i & A ... I had searched for the answer to this many years ago, probably before the Mythbusters tests. Their conclusion (after revisiting the problem) seemed to be that below 50mph, drag is small enough that windows-open is more efficient. Above 50mph, AC is better.

Of course, on a really hot day you'll want the AC on when you're driving at slower speeds, not when you can have the wind on your face at highway speeds. Sidecars in spillproof sippycups help mitigate that problem, though.