Cost efficiency: Programmable thermostat vs constant temperature
October 27, 2014 6:44 AM   Subscribe

I've read so much conflicting information and opinions regarding this, I have no idea what to believe. Logic tells me that keeping the temperature high when we're home and awake, then dropping it when we're asleep or gone is the most efficient use of energy. But others have told me that the amount of work a furnace needs to go to bring a house (1450 sq ft for us) up a significant number of degrees daily is less efficient than simply maintaining a constant temperature. That doesn't make sense to me, but I've heard it from people whose opinion I respect. Help!

I have a gas furnace and programmable thermostat (Carrier Infinity). It has the ability to adjust the temperature up to 4 times per day. So, theoretically, I'd bump it up to 68 or so before waking up, drop it down to say 60 during the day, back up to 68 for when we get home, and then down to 64 or so when we go to bed. That seems reasonable to me.

Or I could just keep the house at 68 at all times. I'm not sure which is the better use of energy.

Side-note: We have mostly old windows in our house (we'll be updating them over time), crappy insulation and ventilation in the attic (I'll address the ventilation in the spring, no idea when I'll get to the insulation), cats at the present, and possibly children in the future.
posted by JimBJ9 to Home & Garden (22 answers total) 6 users marked this as a favorite
Do one month your way, do one month leaving it at 68 the whole time. Compare your gas bills between the two months and you'll have your answer. Of course, this assumes you either have an electric stove or you can keep your gas-stove use somewhat consistent.
posted by Grither at 6:52 AM on October 27, 2014 [2 favorites]

It is more efficient to let the temperature drop when you can. This is because the rate of heat loss depends on the temperature differential between inside and outside. You save enough energy by letting the differential shrink (letting the house cool) to make up for the burden of re-heating.

Grither's suggestion will only produce sensible data if the weather is the same both months.
posted by exogenous at 6:55 AM on October 27, 2014 [10 favorites]

Response by poster: That's also assuming that the outdoor temperature between months is constant, which it won't be. That comparison wouldn't net accurate results, as it's only getting colder.
posted by JimBJ9 at 6:57 AM on October 27, 2014 [5 favorites]

I kind of agree with Grither. However, ideally, you'd have to have two months with the same exact weather to make that comparison -- but you're not going to get that. The closest you'll come is compare make both months February, and compare your gas bill from February 2015 with that of February 2016. I'd also advise also that you compare the number of therms of gas used instead of the cost, because the price of gas fluctuates.

I'd be interested to know the answer, as this is a point of contention in my home: my wife and I have opposing opinions on this.
posted by tckma at 6:57 AM on October 27, 2014

exogenous is right - the extra energy that your furnace burns to bring the temp from, say, 62 back to 68 is the same (OK not really but somewhat) as the energy you *saved* as the temperature *fell* from 68 to 62.

There is an argument for not making the swings in temp so dramatic but they're more to do with comfort and how long it takes to warm the house back up. If you have old steam radiators that take forever to get going, you might want to just keep the temperature roughly the same, for convenience and comfort.

For what it's worth, when I use a programmable thermostat (wifi-enabled, free after rebate from the gas company), my heat basically *doesn't go on all day* while I'm away. I turn the heat to 60 or 55 when I leave, and then the furnace runs hard for 30-60 minutes before I get home to bring the temp back up. My furnace would run for a lot more than 30 minutes over the course of the day if it were keeping the house at 68 all day. This is in a very leaky old house in Massachusetts.
posted by mskyle at 7:02 AM on October 27, 2014 [5 favorites]

Depends on the rate of heat loss of your house and the degree of shift. Things that "hold onto heat" (or, less accurately, cold) like concrete slabs will also enter the calculation.

In general, though, better not to have the heater on while you're asleep or out of the house.
posted by supercres at 7:02 AM on October 27, 2014

In this Slate piece, the author answers your question and states that you save more energy by letting the temperature drop when you can.
posted by crLLC at 7:05 AM on October 27, 2014 [5 favorites]

Say a big bucket full of water represents the amount of heat in your house. Only it has a small leak, which is the loss of heat to the cold outside, so it needs to be replenished.

If the leak is constant, then it doesn't matter if you replenish the bucket constantly or if you replenish it all at once, you still have to put the same amount of water back in the bucket.

Now, heat loss doesn't work like this constant leak. The heat loss is proportional to the temperature difference. The hotter the object is, the more full the bucket is, the more rapidly the leak lets water out. You lose less water the longer that the water level is lower, and you lose less heat the longer that your house is cooler. Bringing it back up to temperature requires no more energy than you would have had to put in to keep it at a constant temperature, but you've gained efficiency by the slower loss of heat when the thermostat was set lower.
posted by kiltedtaco at 7:21 AM on October 27, 2014 [8 favorites]

Now, heat loss doesn't work like this constant leak.

Well, if the leak was the bottom of the bucket then technically it would be a perfectly good analogy, because of the pressure head effect, but your point still stands. Maintaining a higher temperature is more costly then maintaining a lower one, so allowing the heat to fall is better.
posted by Brockles at 7:29 AM on October 27, 2014 [1 favorite]

It's actually an interesting question, and there are a lot of variables. One of the main variables is the type of furnace and the type of heat delivery -- some heating systems burn out more easily when put on a set-back schedule. However, most forced-hot-air/natural-gas fired furnaces are fine on this schedule (although if I were you I would still confirm this with your furnace guy or your manual or whatever, because if you have radiant underfloor heat or something you could apparently do v. bad things to your system).

I think of it this way. As you note, your house isn't perfectly tight -- there is a transfer of heat over time. That being the case, it is more efficient to reduce the difference between the inside and outside temperature when possible. Assuming the same outdoor temp, heat should be lost more slowly from a 60 degree house than from a 68 degree house.

I have a gas furnace with forced hot air and I use almost exactly the setback schedule you suggest. It works nicely for me. I do use an electric blanket at night, and I also use Mortite around my leakier windows. If I had old windows, I'd probably add a layer of plastic or something.
posted by pie ninja at 7:31 AM on October 27, 2014

The laws of physics definitely say it's cheaper to let the temperature drop when feasible rather than maintaining it all the time, but I think it's a subtler question than that.

When you set the thermostat to, for example, 65 degrees, the only thing the thermostat tries to keep at 65 degrees is the thermostat. In reality different parts of the house and its contents will be at all sorts of different temperatures. If you set the thermostat lower for some period of time then the furnace will stop adding heat to the air, which is continually losing heat to the cold windows, walls, foundation and top floor ceiling. The indoor air temperature will drop, and that in turn will start cooling off the house's contents. When you turn the thermostat back up, the furnace dumps heat into the air, which in turn heats the house's contents. Solid objects tend to both cool and re-heat much more slowly than the air.

You can't really perceive the temperature of the things you touch. What you perceive is the rate of heat entering or leaving your body at the point of contact. If the surface of your couch is at 65F when you sit down, you only experience that sensation of coolness for a moment before the upholstery in contact with your skin warms up and heat loss from your body to the couch slows to imperceptibility. If the air is at 65F, it will continue to feel cool because the air warmed by the heat lost from your body rises away from you, and is replaced by more 65F air. If your hardwood floors are at 65F and you're in the habit of walking around barefoot or in thin socks then they'll feel quite cool if you're moving around, because even though wood is a decent insulator and its surfaces warm quickly, you're constantly placing your feet on unwarmed parts of the floor. Tile floors or a granite countertop at 65F will seem positively frigid because they have such a high specific heat and are such good thermal conductors that they will rapidly suck heat out of any part of you that touches them, and keep doing it for a long time.

Point being, if you let the house's air cool down for long enough that the house's solid contents also cool off substantially, then when you try to warm the place back up again you may find yourself turning the thermostat up higher than you otherwise would, so that the warmer air temperatures compensate for the heat you lose to the granite countertops and the hard floors on which you walk around barefoot.

Thermal Delight in Architecture is a neat little book on the subject.
posted by jon1270 at 7:47 AM on October 27, 2014 [5 favorites]

My brother has geothermal heating, and he has tested this systematically over several seasons. I want to know because I have installed geothermal right now. Brother says: constant temperature is more economical. He guesses because the heat-storage in the building is a huge factor. This would mean less in a wood-construction, particularly if this wood construction is on pillars. (My brother's house has a concrete foundation and is made of brick, both of which store a lot of heat)
In practice, he keeps the house slightly colder than their preferred day temperature, and supplements with a wood-oven. I don't know if this is an option for you. I'm thinking of putting ethanol "fireplaces" in some guest-rooms, where a real oven or fireplace isn't practical, and where I don't expect there will be much need during winter.
None of us are experts on this.
posted by mumimor at 7:48 AM on October 27, 2014 [1 favorite]

Related. I find the graph notsnot describes in the second comment particularly helpful in understanding why lowering the heat is better regardless of any variables.
posted by daisyace at 9:57 AM on October 27, 2014 [1 favorite]

If you want to do the experiment but don't want to wait for a month for the results or worry about the weather being too dissimilar, you could read your meter yourself on the days that you run your experiment (both before and after) and then see what the difference is. Switching between one way and the other every day and averaging the result over a couple of weeks should give you a pretty good idea of which one works better for your house.

My understanding is that for most situations, letting the house cool while you're away is the more efficient option since you don't have to add as much heat to the system to maintain the higher temperature while you're away, so long as you're away for a longer period of time than it takes to heat your house from the lower temperature to the higher temperature.
posted by Aleyn at 9:58 AM on October 27, 2014 [1 favorite]

Keeping your temperature constant throughout the day works for A/C but doesn't work for the heating season. Some programmable thermostats have a usage meter on them. Mine indicates that it runs the furnace less on a weekday with a daytime temperature set back compared to a weekend day with no daytime set back.
posted by dlwr300 at 10:13 AM on October 27, 2014 [1 favorite]

I, too, have had smart people tell me to leave the thermostat at 1 setting. I've lived in Maine for 30+ years, and those people are wrongity-wrong. Geothermal heat pump or other factors might influence this, but with gas or oil furnace, and air, steam or water, turn it down.

The faster you fix windows and insulation, the sooner you save and the more comfortable you will be. I had an energy audit, it was helpful.
posted by theora55 at 10:52 AM on October 27, 2014 [2 favorites]

I'd add one thing: Our HVAC installed told me that yes, it's best to have it fluctuate -- but not too much. If you want it to drop at night and heat in the morning, there's a sweet spot for the night temp. Too low and you'll lose savings to the furnace overwork; too high and you lose savings because you're heating all night. That guy said, as have others, that the differential shouldn't be more than ~5-6 degrees.
The principle makes sense to me, but other commenters have provided more science than I can.
posted by LonnieK at 11:34 AM on October 27, 2014 [1 favorite]

Response by poster: It sounds like I'll be using my programmable thermostat this year. Thanks for all the input, folks!

I'll be back later with questions on what the hell to do with my attic. :)
posted by JimBJ9 at 1:18 PM on October 27, 2014

The people who say it ends up costing you more to reheat your house are the same people who will claim it costs too much to turn fluorescent lights back on and to start your car. They're wrong, all three times (even if at some point in the past they had half a point).
posted by IAmBroom at 1:59 PM on October 27, 2014 [1 favorite]

I know you are happy with the answers you got (I'm late in joining the party) but I'll chime in anyway. Of course I support the answers that you should have the temperature drop whenever you can as it is more efficient: it's simple physics which can be easily understood using the very good approximation of Newton's law of cooling.

For those that find themselves arguing with others that disagree with them, I offer this simple thought experiment. Imagine that you have to leave your house for a year (and that the outside temperature does not get so cold that you have to worry about freezing pipes - just to simplifiy the discussion.) You are given two choices:
1. you keep the heat on for an entire year, keeping the temperature constant.
2. you turn the heating system off for the entire year, asking a friend to drop by a couple of days before your return to restart the heating system so that the temperature will be set at your preference when you come back.

Now, what's the cheaper option?
posted by aroberge at 6:35 PM on October 27, 2014

It's Fourier's Law, not Newton, but yeah, program the thermostat.
posted by bumpkin at 5:47 PM on November 1, 2014

« Older Alternative to Simplyfile for Outlook on Mac   |   Cross training for a half marathon while avoiding... Newer »
This thread is closed to new comments.