Minisplits are awesome. I really wish they had a diagram of how the new style works, because "second coil" is so incredibly vague. Is it just a double-layered coil with a better layout? I will be...
Minisplits are awesome. I really wish they had a diagram of how the new style works, because "second coil" is so incredibly vague. Is it just a double-layered coil with a better layout? I will be in the market for a second minisplit system soon enough and the majority of summer has humidity levels over 80%, so this is very relevant to my interests.
Generally, it's actively worse to be turning the unit on and off throughout the day, unless the room is impeccably sealed. Exactly because of that humidity rise. Turn it down a degree or two at night, when it'll cool more efficiently, but then dial it back in the morning. Maybe let it go up an extra degree or two during peak, but never turn it all the way off.
Do you have clear evidence supporting this? I’ve always thought the same, but my buddy has always disagreed and I tell him all the time he doesn’t need to turn his central AC off when he leaves...
Do you have clear evidence supporting this?
I’ve always thought the same, but my buddy has always disagreed and I tell him all the time he doesn’t need to turn his central AC off when he leaves his house for the day or weekend
If vord doesn't fill you in (he's got a solid grasp of HVAC for not being a technician) I'll swing by tomorrow to give an explanation, I've got a wedding that's going to take up my whole day. But...
If vord doesn't fill you in (he's got a solid grasp of HVAC for not being a technician) I'll swing by tomorrow to give an explanation, I've got a wedding that's going to take up my whole day.
But spoilers: you're both right, your friend is wrong
Thanks! I'd like to thank my intro to engineering classes, too many books on energy efficient home design, and chitchating with every HVAC tech that comes my way for a quote.
he's got a solid grasp of HVAC for not being a technician
Thanks! I'd like to thank my intro to engineering classes, too many books on energy efficient home design, and chitchating with every HVAC tech that comes my way for a quote.
Sanity check: I live in a very drafty 1800s New England house with some imperfect insulation tacked on in the 2010s. I don't have central air, but I do use one giant 10,000 BTU (iirc) dual pipe...
Sanity check: I live in a very drafty 1800s New England house with some imperfect insulation tacked on in the 2010s. I don't have central air, but I do use one giant 10,000 BTU (iirc) dual pipe floor unit to cool the entire upstairs.
Generally, I run it June thru September, except when we leave for an entire weekend or more. Is that reasonable, or am I missing some optimisation?
It depends on the climate you're in. If your climate gets humid, like 70% relative humidity or higher, I don't recommend turning it off entirely due to long term mold issues. You can turn it up to...
It depends on the climate you're in. If your climate gets humid, like 70% relative humidity or higher, I don't recommend turning it off entirely due to long term mold issues. You can turn it up to like 78 or so to cut down on morning and evening runtimes.
The fact that you have a 10,000 BTU unit for just the upstairs and nothing downstairs, tells me it doesn't get super hot or humid where you live. You're probably fine turning it off, but you'll have to be the judge on that.
New England can certainly get quite humid, and hotter with these global warming summers, and I dropped another comment to highlight how that imperfect tack-on insulation and and AC system might...
New England can certainly get quite humid, and hotter with these global warming summers, and I dropped another comment to highlight how that imperfect tack-on insulation and and AC system might well be causing their frame to rot out.
When you say "dual pipe floor unit" Are you referring to a portable AC unit? Those are fine for temporary jobs but horribly inefficient (even though the duel vent ones are much better than the...
When you say "dual pipe floor unit" Are you referring to a portable AC unit? Those are fine for temporary jobs but horribly inefficient (even though the duel vent ones are much better than the single vent ones)
Do you own your home? Are you comfortable doing a tiny bit of DIY? A box window unit will be much better if its installed well for both efficiency and performance. Those portable units work fine for a room or two, but they really aren't meant to cool an entire floor, especially one with bad insulation. I encourage you to get a Kill-A-Watt and check how much electricity you end up burning with that unit in a day and do some calculations based on that to decide if investing in a better system will be worth it. New England electricity costs are usually fairly high, meaning you might be throwing away a fat chunk of change every year by not getting a better system worked out.
Modern window AC units are cheap, fairly efficient, and quieter than those big clunky portable units. They even make some versions which don't block the windows by hanging down over the sill, but I don't have any experience with them so I can't say for sure they're better. If you're willing to do some work you can even use foam board insulation and spray foam to create a an airtight seal around the unit which will further increase efficiency.
If you can't/won't ditch the portable AC unit, look into wrapping insulation around the vent pipes as they're usually pretty thin and leak heat back out into the air around them, also make sure that the entrance/exhaust vents are as far apart as they can be and there is no air gap where they meet the window frame. The weird plastic liner they come with does not do a good job at this, so you'll probably want to do your own solution.
Speaking as someone that needed to use a portable unit, that last sentence is spot on. I found that I was able to combine the original PVC window exhaust adapter, foam board, and DIY gasket made...
Speaking as someone that needed to use a portable unit, that last sentence is spot on. I found that I was able to combine the original PVC window exhaust adapter, foam board, and DIY gasket made of foam rubber to create a pretty efficient seal that helped a lot. Also, use something to insulate the exhaust tubing.
The very-drafty is actually working in your favor for controlling humidity. In fact, there's a major problem with blowing insulation into and sealing off these old homes too much without proper...
The very-drafty is actually working in your favor for controlling humidity. In fact, there's a major problem with blowing insulation into and sealing off these old homes too much without proper work, causing the frames to rot out.
Insulation placed between the studs of pre-World War II homes is the single most dangerous element in the wall assembly. This may not be what the energy-efficiency community wants to hear, but physics is physics. Uninsulated, unsealed walls dry out because they “breathe.” But adding insulation—and doing nothing else to manage bulk water, vapor, or ventilation—destroys this cycle.
I'm much more interested in ACEmat's explaination as he'll do it better justice. But in short, it takes a lot more energy to change temperature than to maintain it. IIRC more accurately a...
I'm much more interested in ACEmat's explaination as he'll do it better justice.
But in short, it takes a lot more energy to change temperature than to maintain it. IIRC more accurately a temperature differential rather than exact temp: Takes about the same energy to keep a room at 75F when it's 80F outside, and 85F when its 90F outside. (Edit: See his post further in, this mostly only applies if this is a constant temperature setting, not a changing one). Most of the complaints I see about utilities ripping people off this summer are due to AC units struggling to maintain an extra 5 degree temperature differential, especially if their home has no shade trees causing massive temp spikes in full sun.
And humidity makes it more expensive to cool for the same reason freezing a gallon of water takes longer than freezing a thimble: More humidity == more water in air == more thermal mass to cool.
I think a few things could be cleared up here. First, the amount of energy it takes to change temperature is always less than the energy it takes to maintain temperature. This is easily proven....
it takes a lot more energy to change temperature than to maintain it.
I think a few things could be cleared up here. First, the amount of energy it takes to change temperature is always less than the energy it takes to maintain temperature. This is easily proven. For the case of lowering the temperature in the summer, consider the energy needed to maintain an indoor temperature at the outdoor temperature.† This is zero, of course. Suppose that you would like it to be cooler, so you turn on the A/C. Well, the energy needed to maintain a 1°∆ is less than a 2°∆, and so on. So the energy to maintain the desired 5°∆ has to be greater than the energy to maintain a warmer temperature. For the case of raising a lowered temperature in the Summer, the energy needed is also lower: just turn off the thermostat and let thermal diffusion work it's magic. The thermal mass of the house will lend some hysteresis to this curve, but its not a source of heat, only a store of it.
Second, the energy seems greater because of the way an ON-OFF thermostat works. In the beginning, the house is hot and the A/C just turns on and stays on until it hits the target temperature. Then it shuts off and starts to thermo-regulate. During this part, the power consumption is very high, but the duration is short. Energy is just the time integral of power, so the amount of actual kWh consumed during the cooldown is (hopefully) limited by its brevity.
Suppose I were patient and knew that my A/C with a 10% duty cycle would eventually give me 1°∆, a 20% duty cycle would give a 2°∆, and so on. I could set my A/C at 50% duty cycle and I would eventually get my 5°∆ without any additional energy consumed during the initial cool off compared to the constant temperature phase. It would just take longer to get there. But if it takes the same amount of energy to cool off, fast or slow, I may as well set it at 100% in the beginning and enjoy more time in comfort.
† "Outdoor temperature" in this context refers to all the various conductive, convective, radiative, and internal heating processes that combine to establish whatever the temperature is inside the house that it would be in the absence of mechanical cooling.
There is...A lot to address here. Frankly you've disregarded a ton of variables, like AC performance, heat load, variable units, and real world application / everyday usage. In, well, basically...
Exemplary
There is...A lot to address here. Frankly you've disregarded a ton of variables, like AC performance, heat load, variable units, and real world application / everyday usage.
In, well, basically all AC contexts, we're talking about people maintaining a temperature lower than the outdoor temperature, and varying that setpoint, but still below the outdoor temperature. We're not talking about maintaining 85 degrees when it's 85 degrees and working down from there.
Most people have a pattern with their setpoints, so we'll just use some anecdotal averages. If we take average Joe trying to save money on electric but be comfortable, they will sleep with their setpoint at 69 degrees, set it up to maybe 75 during the day while they're at work, then bring it down to 72 when they get home.
But you have some people that do something crazy like have a setpoint of 80 during the day when they're not home, or turn it off entirely, then bring it down to 72 when they get home. Sure, the system wasn't running during the day if it was off, but now the house has spent all day absorbing that heat. And this is where most people's understanding of heat from their high school chemistry class disappears. Heat isn't just the heat in the air. It's an energy, it gets absorbed into everything in the home: walls, carpets, furniture, the water in the pipes in the walls, etc. So now the setpoint gets set to 72 / the system is turned on, and now the AC has to remove all of that excess heat. Since most people are getting home around 5/6, and the sun's still out for a few hours, not only is it removing the excess heat that's been stored in the home, it's also actively trying to remove the heat entering the home.
Now it has to run longer into the evening than it originally needed to, and it has a much higher heat load running across it. When there's a high heat load, it raises the temperature of the whole system. A typically evaporator (the coil that gets cold) might run at 42-45 degrees, but that could rise up to 50 or even 54 degrees depending on how warm it got. When temperature increases, so does pressure, and when pressure increases, now the compressor has to work harder to compress that refrigerant vapor back into a liquid, which results in a higher amp draw, which also hurts the compressor in the long term.
We haven't even touched on humidity yet. When it's humid outside, an ACs cooling capacitance can be as much as halved, so a 3T system is now only putting out 1.5T of cooling, because the other 1.5T is just taking water out of the air. And by leaving the system set high or off entirely, the house is now sitting around 65% RH at minimum --
(which, although a separate issue, is still important, as most people don't realize 65 is the RH at which mold can begin developing depending on the dewpoint at the time, as the walls, floors, and wood in the house will begin absorbing the excess moisture in the air. This is a long term concern but it's one that a frightening number of people do not consider.)
-- anywhere where it gets remotely humid. If there are crappy dehumidifiers in the house, or if the thermostat has a dehumidification setting, now there's increased energy draw to bring that RH back down.
When we talk about maintaining temperature, we're talking about picking a temperature and keeping it there. An AC doesn't have to work harder to maintain 72 than it does 76 unless it's excessively hot and / or humid outside. When a temperature is maintained, all an AC has to do is remove the heat and humidity actively entering the building during the day, there isn't going to be excess stored heat or humidity in any of the furnishings or structure itself, and the amount entering the building doesn't change if the house is set to 72 or 76 (okay, technically the amount of humidity in the air will be higher in a building that's 76 vs 72, as warm air holds more humidity, but I'm not going to get into that level of minutia). This also means that setting an AC to just a high point during the day only means the system isn't running for a little bit longer while the house warms up, but then it will run just as it would have at 75 when it gets to 80 (again, more minutia, as a house gets to much higher temperatures, like 85 degrees and it's 95 outside, the amount of energy absorbed lessens marginally due to the decrease in temperature difference between the outdoors and indoors, but like the humidity above, it becomes insignificant for practical purposes.)
Which, yes, for any onlookers, that means you're not actually saving money at 78 vs 72 (again, unless it's excessively hot, I will get into that later). An ACs runtime is almost exclusively based on how hot it is outside. Yes, it will turn on later in the day, and turn off sooner in the evening, but assuming we're talking hot hot days, which I typically assume is what most people care about, it's not a matter of hours, and it's not going to impact the middle of the day when the heat is bearing down on the house.
This next part ties into your third paragraph as well a little, but the start up process of an AC is the hardest part on the system. If a particularly fiddly homeowner is constantly adjusting the temperature and turning the system on and off, they're putting a ton of wear and tear on their compressor (and thermal expansion valve (refrigerant regulator that actively adjusts) if the system has one, which most do nowadays). Amp draw on a compressor on start up can be 3-4x as high as its runtime amps, which on most residential system is 4-6A, so start up could be drawing 15-20A for a couple seconds. Best case is the capacitor dies early, which relatively speaking is a cheap fix, but if the AC is trying to run while the cap is dead, and the homeowner doesn't realize, because they're used to their home being warm af for their entire evening, this can damage the compressor and the condenser fan motor. Worst case is the start winding in the compressor gets burnt out and shorts (that's a new compressor by the way).
The long and short of that is: ACs are built to run constantly. They're not made to turn on and off all day. Frankly if an AC can achieve a 1 degree drop with a 10% cycle per hour, the AC is ridiculously oversized, or it is very cool outside and it would be better to just open a window. I understand that's an extreme exaggeration on your end, but cycle rates are super important for mechanical health and maintaining humidity. Cooling starts much quicker than dehumidification does, and if a system is only running for 10 - 15 minutes at a time, it's not running long enough to dehumidify. At that point we would call that short cycling. That's a great way to get mold growing around the ducts and registers by the way, as the house is humid because the moisture isn't being removed, but the cold air from the AC across the metal duct system brings the surface temperature of those parts below that of the now very high dew point and generates condensation.
Now for some niche caveats:
-- Some ACs are two stage, so they literally have a low cool and high cool mode. High end communicating systems that have variable capacity might have 20 stages of cooling. They can adjust their capacity based on runtime, how quickly the temperature increases, it depends on the system, and that gets absolutely wasted if the system is being turned off or set super high during the day. A variable capacity AC may be able to run in a lower stage and maintain temperature in the house fine, but if the house is super warm and now it's being asked to cool off 5-8 degrees, it's just going to run in high stage, and there goes the money invested into a more expensive system.
-- I also referenced a couple times that extreme outdoor temperatures can have an effect. ACs for example in the SE US are rated for 92 degrees, meaning anything above that and it's kind of hit and miss. Generally an AC can expect to get a serviced area (bunch of factors go into this, I won't get into all of them) about 20 degrees cooler than the outdoor, but if it's say 99 degrees outside, and the system is set to 72 during the day, it might only get to like 76. Well now the AC will need to cool a further 4 degrees once it's cooled off outside to reach the desired 72, so in that situation, keeping the setpoint higher would be advantageous, as one set to 76-78 would not have to run as long into the evening.
-- If a house is going to be vacant for a while, like a weekend or longer, like I said above, keeping it set higher will cut down on the morning and evening run times, but I never recommend turning it off unless the house is located in a drier climate, as that brings us back to humidity and potential mold issues. I've had some customers that had apartments over top their detached garage that nobody ever uses, and I would still recommend keeping the AC set to 78.
Forgive any typos or grammatical errors, I've been typing this up for about 2 hours and it's a little after 2AM at the moment for me. I probably forgot some stuff too. Obviously the climate a person lives in affects things a lot, this is just a general overview for most people who live in temperate climates with humidity and 85+ degree summers.
Thanks for all the info! Does this advice also apply to mini splits? As in, should we keep all of our mini splits on all the time in the summer, turning the set temperature up (but not off) during...
Thanks for all the info! Does this advice also apply to mini splits? As in, should we keep all of our mini splits on all the time in the summer, turning the set temperature up (but not off) during the day? We live in an area that probably gets upwards of 95 F and high 60s RH.
In a simple box view of thermodynamics, heat ingress into the home from the outside is proportional to the temperature difference. For example, it it's 80F outside, then a home at 72F would have...
An AC doesn't have to work harder to maintain 72 than it does 76 unless it's excessively hot and / or humid outside.
In a simple box view of thermodynamics, heat ingress into the home from the outside is proportional to the temperature difference. For example, it it's 80F outside, then a home at 72F would have about double the natural heating flux from the outdoors than a home at 76F.
I presume you're rejecting or at least heavily nuancing this view? I suppose this would be less true for heavily-insulated homes, where the largest share of heating comes from interior energy use (both electrical equipment and mammal metabolism) or temperature-independent fluxes like direct insolation.
I'm an HVAC tech, not an engineer, so I only speak from having checked thousands of units' refrigerant pressures and compressor amp draws at all varieties of outdoor and indoor conditions.
I'm an HVAC tech, not an engineer, so I only speak from having checked thousands of units' refrigerant pressures and compressor amp draws at all varieties of outdoor and indoor conditions.
In this vein, what's your thoughts on 'soft starters' which reduce the peak amperage draw? I've heard they're good for the compressor even if it doesn't draw less electricity, as well as being a...
they're putting a ton of wear and tear on their compressor
In this vein, what's your thoughts on 'soft starters' which reduce the peak amperage draw? I've heard they're good for the compressor even if it doesn't draw less electricity, as well as being a sacrificial device if there's an electrical surge.
I always say if you have the know how to do it yourself, go for it. A lot of companies will use it as an accessory sell and charge a ton for it. Some manufacturers are even having them factory...
I always say if you have the know how to do it yourself, go for it. A lot of companies will use it as an accessory sell and charge a ton for it. Some manufacturers are even having them factory installed now.
And this isn't directed at you, but anyone reading this, if your AC stops working, and a company comes out and installs a hard start kit, your compressor is very likely on its last leg. Preventatively they're great, but reactively, they're band aids to limp a compressor along for a bit longer.
This was my thinking as well. I admit there might be detail's we're overlooking, but from first principles this is where I arrived as well. My naive expectation would also be to that ACs are more...
This was my thinking as well. I admit there might be detail's we're overlooking, but from first principles this is where I arrived as well. My naive expectation would also be to that ACs are more efficient when the inside temperature is high, because thermodynamics passively helps the heat exchange along, but again, could be details I'm missing.
Its all about that delta, for exactly what you mention. Someone with fresher knowledge of thermodynamics equations could almost certainly work out some of these thresholds in a simplified example....
Its all about that delta, for exactly what you mention. Someone with fresher knowledge of thermodynamics equations could almost certainly work out some of these thresholds in a simplified example.
I believe the AC will run very efficiently if the indoor temps are higher than outdoor temps, but there's certainly a threshold of delta and humidity where it's worth just opening the windows and doing either passive or with an intake/exhaust fan, as those will use far less energy than the compressor.
One thing I'd like clarified but never see in explanations like this is the time function of running. I accept both yours and @ACEmat 's explanations that it takes way more energy to cool the room...
One thing I'd like clarified but never see in explanations like this is the time function of running. I accept both yours and @ACEmat 's explanations that it takes way more energy to cool the room than to maintain the temperature but let's say I only cool the room for sleeping, so it runs for ~9 hours per day is there more energy savings by the fact that the unit is turned off for the other 15 hours? At some point the time that the unit spends off would have to outweigh the need to completely cool the environment but at which point are we hitting that inflection point?
Honestly, it varies too much home-by-home to say with any level of certainty. If you're only ever cooling a single room there's almost certainly an inflection point, but the only way you'd really...
Honestly, it varies too much home-by-home to say with any level of certainty. If you're only ever cooling a single room there's almost certainly an inflection point, but the only way you'd really be able to figure that out is tracking the usage during a week with consistent weather. Do 3 days with it on continuously, do 3 days turning on/off as you describe, see which is using less energy.
A smart plug which is rated to sufficient amperage for a window unit (15 is usually enough, 20 is better to guard against spikes) will usually be able to keep track of this for you.
Minisplits are awesome. I really wish they had a diagram of how the new style works, because "second coil" is so incredibly vague. Is it just a double-layered coil with a better layout? I will be in the market for a second minisplit system soon enough and the majority of summer has humidity levels over 80%, so this is very relevant to my interests.
Generally, it's actively worse to be turning the unit on and off throughout the day, unless the room is impeccably sealed. Exactly because of that humidity rise. Turn it down a degree or two at night, when it'll cool more efficiently, but then dial it back in the morning. Maybe let it go up an extra degree or two during peak, but never turn it all the way off.
Do you have clear evidence supporting this?
I’ve always thought the same, but my buddy has always disagreed and I tell him all the time he doesn’t need to turn his central AC off when he leaves his house for the day or weekend
If vord doesn't fill you in (he's got a solid grasp of HVAC for not being a technician) I'll swing by tomorrow to give an explanation, I've got a wedding that's going to take up my whole day.
But spoilers: you're both right, your friend is wrong
Thanks! I'd like to thank my intro to engineering classes, too many books on energy efficient home design, and chitchating with every HVAC tech that comes my way for a quote.
Sanity check: I live in a very drafty 1800s New England house with some imperfect insulation tacked on in the 2010s. I don't have central air, but I do use one giant 10,000 BTU (iirc) dual pipe floor unit to cool the entire upstairs.
Generally, I run it June thru September, except when we leave for an entire weekend or more. Is that reasonable, or am I missing some optimisation?
It depends on the climate you're in. If your climate gets humid, like 70% relative humidity or higher, I don't recommend turning it off entirely due to long term mold issues. You can turn it up to like 78 or so to cut down on morning and evening runtimes.
The fact that you have a 10,000 BTU unit for just the upstairs and nothing downstairs, tells me it doesn't get super hot or humid where you live. You're probably fine turning it off, but you'll have to be the judge on that.
New England can certainly get quite humid, and hotter with these global warming summers, and I dropped another comment to highlight how that imperfect tack-on insulation and and AC system might well be causing their frame to rot out.
When you say "dual pipe floor unit" Are you referring to a portable AC unit? Those are fine for temporary jobs but horribly inefficient (even though the duel vent ones are much better than the single vent ones)
Do you own your home? Are you comfortable doing a tiny bit of DIY? A box window unit will be much better if its installed well for both efficiency and performance. Those portable units work fine for a room or two, but they really aren't meant to cool an entire floor, especially one with bad insulation. I encourage you to get a Kill-A-Watt and check how much electricity you end up burning with that unit in a day and do some calculations based on that to decide if investing in a better system will be worth it. New England electricity costs are usually fairly high, meaning you might be throwing away a fat chunk of change every year by not getting a better system worked out.
Modern window AC units are cheap, fairly efficient, and quieter than those big clunky portable units. They even make some versions which don't block the windows by hanging down over the sill, but I don't have any experience with them so I can't say for sure they're better. If you're willing to do some work you can even use foam board insulation and spray foam to create a an airtight seal around the unit which will further increase efficiency.
If you can't/won't ditch the portable AC unit, look into wrapping insulation around the vent pipes as they're usually pretty thin and leak heat back out into the air around them, also make sure that the entrance/exhaust vents are as far apart as they can be and there is no air gap where they meet the window frame. The weird plastic liner they come with does not do a good job at this, so you'll probably want to do your own solution.
Speaking as someone that needed to use a portable unit, that last sentence is spot on. I found that I was able to combine the original PVC window exhaust adapter, foam board, and DIY gasket made of foam rubber to create a pretty efficient seal that helped a lot. Also, use something to insulate the exhaust tubing.
The very-drafty is actually working in your favor for controlling humidity. In fact, there's a major problem with blowing insulation into and sealing off these old homes too much without proper work, causing the frames to rot out.
See How to Kill a House.
I'm much more interested in ACEmat's explaination as he'll do it better justice.
But in short, it takes a lot more energy to change temperature than to maintain it. IIRC more accurately a temperature differential rather than exact temp: Takes about the same energy to keep a room at 75F when it's 80F outside, and 85F when its 90F outside. (Edit: See his post further in, this mostly only applies if this is a constant temperature setting, not a changing one). Most of the complaints I see about utilities ripping people off this summer are due to AC units struggling to maintain an extra 5 degree temperature differential, especially if their home has no shade trees causing massive temp spikes in full sun.
And humidity makes it more expensive to cool for the same reason freezing a gallon of water takes longer than freezing a thimble: More humidity == more water in air == more thermal mass to cool.
I think a few things could be cleared up here. First, the amount of energy it takes to change temperature is always less than the energy it takes to maintain temperature. This is easily proven. For the case of lowering the temperature in the summer, consider the energy needed to maintain an indoor temperature at the outdoor temperature.† This is zero, of course. Suppose that you would like it to be cooler, so you turn on the A/C. Well, the energy needed to maintain a 1°∆ is less than a 2°∆, and so on. So the energy to maintain the desired 5°∆ has to be greater than the energy to maintain a warmer temperature. For the case of raising a lowered temperature in the Summer, the energy needed is also lower: just turn off the thermostat and let thermal diffusion work it's magic. The thermal mass of the house will lend some hysteresis to this curve, but its not a source of heat, only a store of it.
Second, the energy seems greater because of the way an ON-OFF thermostat works. In the beginning, the house is hot and the A/C just turns on and stays on until it hits the target temperature. Then it shuts off and starts to thermo-regulate. During this part, the power consumption is very high, but the duration is short. Energy is just the time integral of power, so the amount of actual kWh consumed during the cooldown is (hopefully) limited by its brevity.
Suppose I were patient and knew that my A/C with a 10% duty cycle would eventually give me 1°∆, a 20% duty cycle would give a 2°∆, and so on. I could set my A/C at 50% duty cycle and I would eventually get my 5°∆ without any additional energy consumed during the initial cool off compared to the constant temperature phase. It would just take longer to get there. But if it takes the same amount of energy to cool off, fast or slow, I may as well set it at 100% in the beginning and enjoy more time in comfort.
† "Outdoor temperature" in this context refers to all the various conductive, convective, radiative, and internal heating processes that combine to establish whatever the temperature is inside the house that it would be in the absence of mechanical cooling.
There is...A lot to address here. Frankly you've disregarded a ton of variables, like AC performance, heat load, variable units, and real world application / everyday usage.
In, well, basically all AC contexts, we're talking about people maintaining a temperature lower than the outdoor temperature, and varying that setpoint, but still below the outdoor temperature. We're not talking about maintaining 85 degrees when it's 85 degrees and working down from there.
Most people have a pattern with their setpoints, so we'll just use some anecdotal averages. If we take average Joe trying to save money on electric but be comfortable, they will sleep with their setpoint at 69 degrees, set it up to maybe 75 during the day while they're at work, then bring it down to 72 when they get home.
But you have some people that do something crazy like have a setpoint of 80 during the day when they're not home, or turn it off entirely, then bring it down to 72 when they get home. Sure, the system wasn't running during the day if it was off, but now the house has spent all day absorbing that heat. And this is where most people's understanding of heat from their high school chemistry class disappears. Heat isn't just the heat in the air. It's an energy, it gets absorbed into everything in the home: walls, carpets, furniture, the water in the pipes in the walls, etc. So now the setpoint gets set to 72 / the system is turned on, and now the AC has to remove all of that excess heat. Since most people are getting home around 5/6, and the sun's still out for a few hours, not only is it removing the excess heat that's been stored in the home, it's also actively trying to remove the heat entering the home.
Now it has to run longer into the evening than it originally needed to, and it has a much higher heat load running across it. When there's a high heat load, it raises the temperature of the whole system. A typically evaporator (the coil that gets cold) might run at 42-45 degrees, but that could rise up to 50 or even 54 degrees depending on how warm it got. When temperature increases, so does pressure, and when pressure increases, now the compressor has to work harder to compress that refrigerant vapor back into a liquid, which results in a higher amp draw, which also hurts the compressor in the long term.
We haven't even touched on humidity yet. When it's humid outside, an ACs cooling capacitance can be as much as halved, so a 3T system is now only putting out 1.5T of cooling, because the other 1.5T is just taking water out of the air. And by leaving the system set high or off entirely, the house is now sitting around 65% RH at minimum --
(which, although a separate issue, is still important, as most people don't realize 65 is the RH at which mold can begin developing depending on the dewpoint at the time, as the walls, floors, and wood in the house will begin absorbing the excess moisture in the air. This is a long term concern but it's one that a frightening number of people do not consider.)
-- anywhere where it gets remotely humid. If there are crappy dehumidifiers in the house, or if the thermostat has a dehumidification setting, now there's increased energy draw to bring that RH back down.
When we talk about maintaining temperature, we're talking about picking a temperature and keeping it there. An AC doesn't have to work harder to maintain 72 than it does 76 unless it's excessively hot and / or humid outside. When a temperature is maintained, all an AC has to do is remove the heat and humidity actively entering the building during the day, there isn't going to be excess stored heat or humidity in any of the furnishings or structure itself, and the amount entering the building doesn't change if the house is set to 72 or 76 (okay, technically the amount of humidity in the air will be higher in a building that's 76 vs 72, as warm air holds more humidity, but I'm not going to get into that level of minutia). This also means that setting an AC to just a high point during the day only means the system isn't running for a little bit longer while the house warms up, but then it will run just as it would have at 75 when it gets to 80 (again, more minutia, as a house gets to much higher temperatures, like 85 degrees and it's 95 outside, the amount of energy absorbed lessens marginally due to the decrease in temperature difference between the outdoors and indoors, but like the humidity above, it becomes insignificant for practical purposes.)
Which, yes, for any onlookers, that means you're not actually saving money at 78 vs 72 (again, unless it's excessively hot, I will get into that later). An ACs runtime is almost exclusively based on how hot it is outside. Yes, it will turn on later in the day, and turn off sooner in the evening, but assuming we're talking hot hot days, which I typically assume is what most people care about, it's not a matter of hours, and it's not going to impact the middle of the day when the heat is bearing down on the house.
This next part ties into your third paragraph as well a little, but the start up process of an AC is the hardest part on the system. If a particularly fiddly homeowner is constantly adjusting the temperature and turning the system on and off, they're putting a ton of wear and tear on their compressor (and thermal expansion valve (refrigerant regulator that actively adjusts) if the system has one, which most do nowadays). Amp draw on a compressor on start up can be 3-4x as high as its runtime amps, which on most residential system is 4-6A, so start up could be drawing 15-20A for a couple seconds. Best case is the capacitor dies early, which relatively speaking is a cheap fix, but if the AC is trying to run while the cap is dead, and the homeowner doesn't realize, because they're used to their home being warm af for their entire evening, this can damage the compressor and the condenser fan motor. Worst case is the start winding in the compressor gets burnt out and shorts (that's a new compressor by the way).
The long and short of that is: ACs are built to run constantly. They're not made to turn on and off all day. Frankly if an AC can achieve a 1 degree drop with a 10% cycle per hour, the AC is ridiculously oversized, or it is very cool outside and it would be better to just open a window. I understand that's an extreme exaggeration on your end, but cycle rates are super important for mechanical health and maintaining humidity. Cooling starts much quicker than dehumidification does, and if a system is only running for 10 - 15 minutes at a time, it's not running long enough to dehumidify. At that point we would call that short cycling. That's a great way to get mold growing around the ducts and registers by the way, as the house is humid because the moisture isn't being removed, but the cold air from the AC across the metal duct system brings the surface temperature of those parts below that of the now very high dew point and generates condensation.
Now for some niche caveats:
-- Some ACs are two stage, so they literally have a low cool and high cool mode. High end communicating systems that have variable capacity might have 20 stages of cooling. They can adjust their capacity based on runtime, how quickly the temperature increases, it depends on the system, and that gets absolutely wasted if the system is being turned off or set super high during the day. A variable capacity AC may be able to run in a lower stage and maintain temperature in the house fine, but if the house is super warm and now it's being asked to cool off 5-8 degrees, it's just going to run in high stage, and there goes the money invested into a more expensive system.
-- I also referenced a couple times that extreme outdoor temperatures can have an effect. ACs for example in the SE US are rated for 92 degrees, meaning anything above that and it's kind of hit and miss. Generally an AC can expect to get a serviced area (bunch of factors go into this, I won't get into all of them) about 20 degrees cooler than the outdoor, but if it's say 99 degrees outside, and the system is set to 72 during the day, it might only get to like 76. Well now the AC will need to cool a further 4 degrees once it's cooled off outside to reach the desired 72, so in that situation, keeping the setpoint higher would be advantageous, as one set to 76-78 would not have to run as long into the evening.
-- If a house is going to be vacant for a while, like a weekend or longer, like I said above, keeping it set higher will cut down on the morning and evening run times, but I never recommend turning it off unless the house is located in a drier climate, as that brings us back to humidity and potential mold issues. I've had some customers that had apartments over top their detached garage that nobody ever uses, and I would still recommend keeping the AC set to 78.
Forgive any typos or grammatical errors, I've been typing this up for about 2 hours and it's a little after 2AM at the moment for me. I probably forgot some stuff too. Obviously the climate a person lives in affects things a lot, this is just a general overview for most people who live in temperate climates with humidity and 85+ degree summers.
Thanks for all the info! Does this advice also apply to mini splits? As in, should we keep all of our mini splits on all the time in the summer, turning the set temperature up (but not off) during the day? We live in an area that probably gets upwards of 95 F and high 60s RH.
Absolutely, and Mini Splits are stupidly efficient anyways.
That was a very thorough response! Thanks and good luck!
In a simple box view of thermodynamics, heat ingress into the home from the outside is proportional to the temperature difference. For example, it it's 80F outside, then a home at 72F would have about double the natural heating flux from the outdoors than a home at 76F.
I presume you're rejecting or at least heavily nuancing this view? I suppose this would be less true for heavily-insulated homes, where the largest share of heating comes from interior energy use (both electrical equipment and mammal metabolism) or temperature-independent fluxes like direct insolation.
I'm an HVAC tech, not an engineer, so I only speak from having checked thousands of units' refrigerant pressures and compressor amp draws at all varieties of outdoor and indoor conditions.
In this vein, what's your thoughts on 'soft starters' which reduce the peak amperage draw? I've heard they're good for the compressor even if it doesn't draw less electricity, as well as being a sacrificial device if there's an electrical surge.
I always say if you have the know how to do it yourself, go for it. A lot of companies will use it as an accessory sell and charge a ton for it. Some manufacturers are even having them factory installed now.
And this isn't directed at you, but anyone reading this, if your AC stops working, and a company comes out and installs a hard start kit, your compressor is very likely on its last leg. Preventatively they're great, but reactively, they're band aids to limp a compressor along for a bit longer.
This was my thinking as well. I admit there might be detail's we're overlooking, but from first principles this is where I arrived as well. My naive expectation would also be to that ACs are more efficient when the inside temperature is high, because thermodynamics passively helps the heat exchange along, but again, could be details I'm missing.
Its all about that delta, for exactly what you mention. Someone with fresher knowledge of thermodynamics equations could almost certainly work out some of these thresholds in a simplified example.
I believe the AC will run very efficiently if the indoor temps are higher than outdoor temps, but there's certainly a threshold of delta and humidity where it's worth just opening the windows and doing either passive or with an intake/exhaust fan, as those will use far less energy than the compressor.
One thing I'd like clarified but never see in explanations like this is the time function of running. I accept both yours and @ACEmat 's explanations that it takes way more energy to cool the room than to maintain the temperature but let's say I only cool the room for sleeping, so it runs for ~9 hours per day is there more energy savings by the fact that the unit is turned off for the other 15 hours? At some point the time that the unit spends off would have to outweigh the need to completely cool the environment but at which point are we hitting that inflection point?
Honestly, it varies too much home-by-home to say with any level of certainty. If you're only ever cooling a single room there's almost certainly an inflection point, but the only way you'd really be able to figure that out is tracking the usage during a week with consistent weather. Do 3 days with it on continuously, do 3 days turning on/off as you describe, see which is using less energy.
A smart plug which is rated to sufficient amperage for a window unit (15 is usually enough, 20 is better to guard against spikes) will usually be able to keep track of this for you.
Yeah that's a good point and probably an exercise I'll perform next summer. Thanks!
https://tildes.net/~engineering/1j94/how_to_build_greener_affordable_ac_for_high_humidity_and_hotter_summers#comment-dsv8
Mirror: https://archive.is/ImOnf