Winter heating confusion

October 24, 2011 12:21 PM

How do I use a programmable thermostat/heater efficiently in the winter? Specifically, how cold should I let it get when no one is home?

Which approach is better, from a cost perspective? I live in Chicago, with forced-gas heating and relatively good insulation in an apartment. Assume this is in the middle of winter, so something like 25 F out in the daytime and 10 F at night.

1. Heat at 68 when the occupant is home and 60 when the occupant is not home.

2. Heat at 68 when the occupant is home and 50 when the occupant is not home.

So in the first situation, the thermostat doesn't have to work "as hard" to bring it back up to 68 because the house doesn't get as cold in the daytime. However, in the second situation, when the occupant is not home the thermostat only kicks on to heat the house to 50, instead of 68, which uses less gas.

I always thought approach 1 was better (because the costs of heating the house back up outweigh the savings from keeping the house really low when no one is home), but my friend recently questioned this and my Googling has me confused.

Everyone agrees you shouldn't heat it at 68 all day long. But how low can I go? Should I really be setting my thermostat as low as possible, as long as the pipes don't freeze, when no one is home?

posted by andrewesque to home & garden (9 answers total)

I asked the dude who put my furnace in last year that same question, he said it depends on the size of the house. He didn't have hard numbers, but he said the bigger the house the warmer you should keep it. That is, if the house is big (2,000+ sq ft ish) it takes longer to heat up. I'd also guess it depends on how we'll it's insulated.
posted by Blake at 12:27 PM on October 24, 2011

#2. The extra work to keep the house at 60 rather than 50 is far, far greater than the work to heat from 50-68 vs 60-68. The furnace doesn't "work harder" when the temp differential is greater. It works longer - so your first cycle once you get home is longer, big whoop.

For a more in-depth discussion as to the *why*, think of a graph with each hour plotted out on the x axis. The y axis would have the temperature. For simplicity sake, draw a line across at, say, 15F. Now plot the two different thermostat temps you're looking at. The total work (therms) your furnace produces is roughly equivalent to the area between the thermostat curve and the 15F baseline. You can see that option #2 has a much smaller area.

There are only two considerations that need minding if you go for #2. If you have pets, make sure they have blankets to snuggle up in; our cats love to bury themselves in their fuzzy blankets when our furnace kicks off at night. Two, and this is severely minor, you may have to make the transition from 50->68 at, say, 4:30 if you plan to be home at 5, so that the furnace has time to get it up to our desired temp. That, or just plan to leave your coat on for a few minutes while the place gets up to temp.
posted by notsnot at 12:30 PM on October 24, 2011 [4 favorites]

I have a high-efficiency multi-stage/multi-speed forced air furnace. Yeah, fancy! When the furnace ppl installed it, I also asked them this too, they said not to set the difference greater than 6F - it's better to keep the house warmer at lower power than to re-heat the house at high power.

I noticed that my house holds the heat pretty well, so even when I set the temp low, rarely does the house actually cool down that much.

I also have a fancy programmable thermostat which learns how long it takes to recover the heat, and automatically starts heating earlier to arrive at the target heat at the scheduled time. No extra thinking/planning required!
posted by jpeacock at 12:47 PM on October 24, 2011 [1 favorite]

Well, you have both answers already, so I'll just add something to consider: house @ 50 = objects in house @ 50 (aka 50 degree couch, counters, clothes, kitchen chairs, floors, etc).

I find if we let our house get too cold it never really feels warm because all the stuff in it is constantly giving off cold (or sucking up heat, whichever you prefer).
posted by that's candlepin at 1:32 PM on October 24, 2011 [1 favorite]

notsnot is square on. Heat losses are proportional to temperature difference. If it's 40 outside, you'll have twice the heat flux (loss) at 60 when compared to 50. Total running time for the unit will be LESS if you let the house drop to as low a temperature as your interior can stand. (Big temperature swings can cause condensation depending on house conditions.) The math is really quite simple to work out if you need to see it proved.
posted by introp at 2:06 PM on October 24, 2011 [2 favorites]

This comes up on AskMe all the time, so it is clearly difficult for many to conceptualize. I find it easier to think in terms of the heat lost to the environment, rather than the heat added to the house. Basically, your house is a box that loses heat to the outdoors and you add heat to your house to counter that loss. Since you lose more heat when the temperature difference between your house and the outdoors is greater, you have to add more heat when the temperature is set higher. Conversely, when you set the temperature lower, you lose less heat to the outside. Simple as that. From an energy-use perspective, it simply does not matter that you have to add a bunch of heat at once to warm your place back up.
posted by ssg at 2:58 PM on October 24, 2011 [1 favorite]

Yes, it's true that the total energy expended by the furnace is less if it's set at 50 during the day instead of 60. However, as jpeacock points out, most furnace people recommend not trying to swing 18 degrees back and forth from 50 to 68. I think most recommend something between 5 and 10 degrees. It has to do with how often the furnace cycles.

I can't claim to be an expert who understands this perfectly and can guarantee its correctness, but here's how it was explained to me: When the furnace cycles on in the afternoon and runs for a while to heat the air from 50 up to 68, that only holds for a few minutes because all the (50-degree) objects in the house chill the air fast. So it cycles back on again, heats up the air quickly, and then the air cools down again quickly. It's not about the total energy spent by the furnace, I understand that the re-warming of the walls takes less energy than you would have spent keeping them warm all day. It's about maintaining the furnace and not shortening the lifetime (switch failures) or reducing performance (furnace efficiency) by too-frequent cycling.
posted by aimedwander at 3:17 PM on October 24, 2011

As a counter-argument to aimedwander's point: if your furnace's lifetime is shortened by even, say, five years, out of a 20-year lifespan, that will cost you an extra quarter of the cost of a new furnace, so what? A couple of thousand dollars, max? Over 20 years, how much more will you save on power bills by not heating the house up all day when you aren't home? I bet it works out more favourably.
posted by lollusc at 4:26 PM on October 24, 2011

At my house, heat at 65 when the occupant is home and 50 when the occupant is not home, or when occupant is in bed, with electric blanket. I hear the furnace kick on in the morning; it's doing its job effectively. Programmable thermostat means I don't have to get out of bed if I forgot to turn the heat down.
posted by theora55 at 8:38 PM on October 24, 2011