Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Electricity use:Saving by turning lights/power on/off

  1. Oct 25, 2011 #1


    User Avatar
    Science Advisor

    Hi, All:
    My knowledge of physics is very rudimentary; please be patient.

    A friend of mine's father owns a small restaurant, asked me (as a math student)
    how he could save some money. All I could think of (not knowing much about
    either general economics nor restaurant business) was the use of electricity;
    specifically, the issue of turning lights/power on or off. AFAIK, it is the process of
    turning the power on that consumes most of the power, after which the power
    consumed (therefore the cost) decreases. So, if the power were to be repeatedly turned
    on and off, this would seem to be costly. Yet, turning it on just once and leaving it on
    for , say, 12 hrs at-a-time, may not be the best idea either. So I thought taking some data
    of the number of times the power is turned on , and compute the basic statistics; mean,
    sd, etc. Basically, there may be an optimal number of times to turn power on, given average

    Still: is there some physical law describing the use of power over time after a device is turned on, and how power use decreases over time, to be able to tell if, given the data, e.g., for the bathroom light, whether it is better to leave it on all day, or to turn it off , if the bathroom is used, say, 25 times a day on average? How about for heating/cooling?

  2. jcsd
  3. Oct 25, 2011 #2
    Whenever possible turn the lights off
  4. Oct 25, 2011 #3


    User Avatar
    Science Advisor

    Would that still be true if, say, lights were turned on 100 times? Would you suggest the law/model you're using?
  5. Oct 25, 2011 #4
    From the OP I think you believe turning lights on/off costs more energy than keeping them always on. It's a false idea; there's no extra cost in turning on a lamp. I repeat my advice: turn the lights off a soon as you don't need them.
  6. Oct 25, 2011 #5


    User Avatar

    There are several 'false economies' that can arise with the otherwise honourable intent of saving electricity.

    Firstly, if you have your heating on, it doesn't really make any difference. The heat from your lighting (and any other appliance) will simply unload some of the burden of the heating system. So many kWh in = so much heating put in. The energy put in cannot 'disappear'. The issue in this case is your gas/oil versus electricity cost.

    (This is why I always use a gas-stove whistle kettle during winter months. It is 'free boiled water' because all the heat ends up in the house anyway. In fact, it is more efficient that burning gas in your boiler because on the stove you get all the exhaust gas as well - a full 100% thermal conversion.)

    Secondly, though it is an old tale that it takes more power to start up your light than to run it (well, in the case of fluorescent lights there is a small power surge, but nothing to note if you're running the lights on for more than a few seconds) the issue of turning them on and off then becomes a function of their lifetime and reliability under thermal/power cycling:

    Various reports are made when buying fluorescent (especially the 'compact' type that replace bulbs) by the manufacturers of how long the life is and how many times they can be turned on and off. I now write the date of installation on the bulb when I fit them, and estimate the on-off cycles. They achieve no where near the manufacturer's claims.

    I do not think your calculation will come to much. To save power on lighting the best thing to do is use the most efficient lighting, and as little of it as possible. In areas not regularly used you can install motion-detecting, or even ambient-light detecting, bulbs. But for a restaurant I think it could be a potentially very damaging thing in regards the impact on passing trade, if the store looks dark from the outside.

    I would think the place in a restaurant to look at energy efficiency would be the ovens (their fuel source and thermal efficiency - you want the heat in the oven, not in the kitchen!) and the same for water heating. Heating up water really takes enormous amounts of energy that make turning off bulbs look like counting paperclips.
  7. Oct 25, 2011 #6


    User Avatar
    Science Advisor

    Thanks all for your comments.
  8. Nov 1, 2011 #7


    User Avatar
    Science Advisor

    A quick followup; sorry forgot to ask previously:

    Does the same method ( turn off when possible ) help minimize energy use with other appliances, e.g., with heating ? If I were to leave my apartment for 30 minutes: should I leave the heating on, or should I turn it off and then turn it on again when I get back? Is this an issue of inertia?
  9. Nov 1, 2011 #8
    I'd say 30 minutes is a timespan short enough to leave the heating on. On the other hand, if I were to be absent for say 8 hours, I'd turn the heating off. I know the house won't be warm enough but I can wear a pullover and wait until it warms up.
    Saving energy is important to your wallet and to our world
  10. Nov 1, 2011 #9
    Any business OUGHT to turn the lights off when they're CLOSED. Drives me nuts to see a closed supermarket with every bulb (or even every other) burning away all night. It takes very little light to be theft-deterrent, if that's even really theft-deterrent.

    I wouldn't turn the heat or AC in your place off for 30 minutes. Too much thermal inertia. If you're gone all day maybe.

    I would love to see a QUANTIFIED study of exactly how much energy and money turning your heat down or A/C up really saves in a well insulated and in a poorly insulated house. I've only ever seen conventional wisdom quoted that "it saves" but never seen any evidence or even calculation...
  11. Nov 1, 2011 #10
    Mythbusters tested leaving bulbs on vs turning them off. Traditional tube fluorescents (not CFL) had the the longest break even time at 23 seconds. So if you are leaving a room for at least a minute you will use less energy by turning the lights off.

    However, I think bulb life is worth considering. They do mention testing for this in the above link, however there is other hardware to consider, eg, ballast. Still as a general rule I'd turn them off as much as seems reasonable.

    As for turning off your heater when leaving, that should also save money. If you think about it, your house loses heat at a certain rate, that rate is a function of the temperature difference between the inside and outside. In other words, the hotter the inside of your house is than outside the faster you lose heat.

    The heater will have to replace every joule that leaks to outside. The way to reduce how much the heater has to do is to reduce how many joules leak. By reducing the temperature inside the house you reduce that rate. Therefore, if you turn off the heat, the rate gradually slows down, and eventually reaches 0 (when inside = outside). When the heat is turned back on, every joule will have to be replaced, however since the rate of loss was lowered there are less joules to be replaced.

    That being said, I have no idea how significant the savings from a half hour would be. My guess is there would be very little difference. However keep in mind, a planned half hour could turn into a much longer trip.
  12. Nov 2, 2011 #11


    User Avatar

    I performed this study in my last house (3 bed 2 storey detached) and this one (4 bed 2 story detached), to determine if turning the heating off at night was worth while. These were UK industry-standard insulated houses. I determined that the saving made in turning the heating completely off at night (30% - 40% off period each day) during periods of <10C daytime and <5C night time (when you'd expect the heating to be needed most of the time) was a little more than 5%.

    As I found the cost overhead is so minimal, I leave my heating on all night, albeit adjusted a few degrees down, because I have concluded that cost of repairs from the cyclic thermal shock strains on my heating system would likely be imbalanced to the advantage of having a warm house at all hours.

    I actually came to perform this analysis, thus conclusion, because I got fed up with sitting on a cold toilet seat during the night time! The energy saved was so relatively small it was virtually within the noise of the measurement accuracy. Warm toilet seats at all hours is definitely worth the little extra!
  13. Nov 2, 2011 #12


    User Avatar
    Science Advisor
    Gold Member

    I think your house must be a lot better insulated than mine (a Victorian terrace).

    To solve the toilet seat problem just get someone to sit on it first - in true English public school tradition.
  14. Nov 2, 2011 #13


    User Avatar
    Gold Member

    I love when the old wives tale of it being more efficient to leave a lightbulb on for an hour instead of turning it off and on. If you actually took the time to determine how much power the lightbulb would have to draw in that fraction of a second it takes to turn it in, for this to be true the power draw must be near that of the capacity of a nuclear power plant.

    As for your friends father's restaurant, you really need more information to determine how best to save his place some money. Where the restaurant is, the climate, building type, what actually is the most costly to run, etc. There may even be programs for more interesting things like programs for subsidizing solar panels.
  15. Nov 2, 2011 #14


    User Avatar

    But there is more to it than just the energy it uses. How many times can you turn the light on and off? For all the manufacturer's hype over how many times these CFL's can be switched on-off, I have had some fail after (approximations here*) 100 hrs total 'on' time and, say, 400 power-ups. For a $4 18W bulb, that's 1 cent per switch on, yet it is using 0.5 cent per hour in electricity - of which 100% is being efficiently converted to heat for your house!

    *(based on me marking the date of installation on the bulb, and approximating usage profile after it has failed)

    'Old tales' work both ways. Go do some objective tests of your own and report back!
  16. Nov 2, 2011 #15
    Very true. I'm wondering if the original question had been posted in July if there would be as many responses talking about "free heating". Remember that if you are running your air conditioning (as some of us are right now), that every watt saved in lighting also saves about a third of a watt in A/C as well. While free heating may look like a deal in winter, averaged over the course of the year it is probably better to get your heat in winter from your heater instead of your appliances and lights, since it will cost you come summer to get your heat from waste.

    I think that cmb is right on track to look to the ovens as the real place to save energy, at least when it comes to looking at the amount of energy to bring them up versus power to run them and their average usage patterns. This is a good place to gather data.

    Of course, every good restauranteur knows that the best way to save energy is to move the tables into separated alcoves, provide a good wine list, and turn the lights down to a nice romantic low. ;)
  17. Nov 2, 2011 #16


    User Avatar

    Hence my caveat "if you have the heating on". In summer months, you should need the lights less, if at all, if you have good windows and if not then fit better windows... and make then quadruple glazed too, because even in summer with A/C it'll mean that works less hard and quadruple glazing is also much better at stopping annoying sounds (e.g. traffic) from getting into the building.
  18. Nov 2, 2011 #17


    User Avatar

    Are there any 'light pipe' products built/sold by anyone, that can gather solar light from the roof and use it for lighting the interior?
  19. Nov 2, 2011 #18


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    You'll shorten the lifetime of a fluorescent lamp by 1 or 2 hours each time it is turned off and back on. It has been a few years since I did this calculation, but I remember being able to fit some manufacturer's lifetime data pretty well by assuming each turnon/off cut a fixed amount off of the lifetime, and I remember coming up with an hour or two by doing a least-squares fit to the published lifetime data.

    The point is, if you're going to leave a room for just a few minutes, it's better to leave the lamp running. I calculate a "break-even" time of 9 minutes, assuming the following:

    40 W fluorescent, cost = 10.00 $US, lifetime = 10,000 hours
    $0.17 per kW-hr electricity cost
    1 hour lifetime hit each time lamp is turned on (corresponds to $0.001 cost hit on $10.00, 10 khr lamp)
  20. Nov 3, 2011 #19
    Google it and you'll find a number of such (along with some "psychedelic products" ha ha)

    Solatube is one. Our friends have one and it works pretty well, though it is not like a skylight at all, more like a fairly bright bulb with very white light as long as the sun is out.
  21. Nov 3, 2011 #20
    Yeah, I wonder even overnight or daytime, how SIGNIFICANT the savings are. From what you said, I realized that of course it will matter how cold or hot it is outside due to the temperature differential, so the answer depends a lot on that.

    But I would still like to see ANY case that had an actual measured comparison!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook