What’s the best strategy for turning off CFLs that will save energy but not reduce lifespan so much



With ordinary bulbs, I was used to turning the lights off even when I left the room for a few minutes (mostly because I know that I'll find other business and forget to return :-) ), but I've heard that for compact fluorescent lamps it's better to turn the light on/off much less frequently.

So, if we calculate not just energy, but also ecological footprint from production and disposal of the lamps, where is the threshold for "dark interval" that saves energy and ecological footprint? For how long must the light be off until I start saving energy (in comparison to having the light on all the time)?

If there are bigger differences between models/subtypes, please send a range and add which lamps are best/average/worst. If there are significant differences between estimates of certain on/off-switching patterns' impact on durability of the lamps, please give the range too. I assume that energy costs or ecological footprint of production and disposal of a compact fluorescent lamp are well known; if not, consider it while computing the range.

Anyway, I don't need it to be extra exact, just to have a good idea whether (and how) I should change my light-switching habits.


Posted 2013-01-29T21:41:21.937

Reputation: 683


Found a related question on skeptics SE: http://skeptics.stackexchange.com/q/8365/13674

– THelper – 2013-05-02T23:13:50.003



Mythbusters featured lightbulbs on Episode 69.

The conclusions they came to were

  • There is no appreciable lifespan impact from turning them off and on
  • There is no cost savings from leaving them on to "warm up"

Edit, FTFA:

Bulb Longevity

They tested one final element of this myth: frequently turning lights on and off decreases their life span, thus leading to greater costs. Grant setup a timer and relay to turn the bulbs on and off repeatedly every 2 minutes. After six weeks, only the LED bulb was still working. Based on this test, they extrapolated that it would take five years of ordinary usage to cause the bulbs to burn out.

Consumer reports weighs in:

CFLs keep burning brightly

The bulbs in our labs have been cycling on and off since early 2009, or 6,000 hours. For comparison, a typical incandescent bulb lasts only around 1,000 hours. Even after all that time, brightness and warm-up times remained virtually the same as after 3,000 hours of testing. Our results were confirmed by an outside lab.


Posted 2013-01-29T21:41:21.937

Reputation: 828

The link on CFLs and LEDs didn't help much - they said they regularly turned the light on and off, but not with which frequency. Given approximate number of on/off turns a day (or in total) they (or some similar experimenters) had and cost of the mercury disposal/recyclation, I would consider accepting the answer - from these data, I'd be able to calculate approximate cost per on/off turn myself. Well, the data doesn't have to be scientific, just the Mythbusters' way seems too oversimplificating for me. – Pavel – 2013-01-31T11:19:08.740

@Pavel You mention "fluorescent lights" in the dorms (which may not be the same as CFL's), so any longevity studies or experiments on CFL's may have little bearing on standard long tube fluorescents since, although they may operate similarly to CFL's, they are not identical. (and there are variants of fluorescent tubes like preheat-start, instant-start, quick-start, etc that can lead to differences in lifetime under repeated cycles) – Johnny – 2013-08-06T23:45:21.920

Well, I've changed my mind. Accepted. – Pavel – 2013-10-30T09:11:49.067

1Well, you affirmed that my light-turning habits are well suited for ordinary lightbulbs and probably for LEDs too, but my students' dormitory has fluorescent lamps everywhere and replacing them with LEDs is not an option. That's why I was asking about the fluorescent lamps, and I can't see answer for it in your post. – Pavel – 2013-01-29T22:10:29.257

Try clicking the link. The answer is there. – BryanH – 2013-01-29T22:32:52.353

The link proved that energy cost is marginal, which is just one (the easier and the less important, from my point of view) part of the answer. About the harder part - to determine the effect of turning on/off on longevity and compare and recalculate it to electricity consumption - your link just proved that the impact of turning on/off on the longevity of the bulb is important. Since CFLs contains mercury and other dangerous materials and the cost to recycle or destroy them ecologically is significant, this part of the question is the more important. However, +1 for the first part. – Pavel – 2013-01-29T22:57:59.863

Updated answer. Hope that helps. – BryanH – 2013-01-29T23:03:05.007

I have already read it in the link. But as the commentators there remarked, they don't say when did different bulbs break, just that most didn't survive. I'd be glad if you found answer for this subquestion, but still it wouldn't be enough to accept the answer. Mythbusters can help on a vague level, but it's not hard, scientific data. Thanks for your link, but I need something with better backup. – Pavel – 2013-01-29T23:11:27.527


If you're looking for hard, scientific data, you might check the Journal of Electrical Engineering. Have you heard of any formal studies on this topic?

– BryanH – 2013-01-29T23:25:23.823


Paul Wheaton has an interesting article and several videos on his website that expands on the Mythbusters experiment. He measures the average time it takes to do several common activities for which you'd typically need a light then measures the life span and energy usage of various types of bulbs under those conditions.

His conclusions:

  • CFLs cost more to manufacture and replace than they are worth in energy savings
  • Only turning a light on when you need it is still the best way to save money on lighting costs
  • Lighting is such a small percentage of the average electricity bill that you'd save a lot more electricity by buying a clothes line or replacing a water heater than by worrying about which type of light bulb to use.

I'm inclined to agree with the majority of his conclusions but different usage patterns will have different results. I live in Florida where the primary electrical use is cooling during the summer months. Most of my lights have CFLs in them because they give off less heat (though I usually only use ambient sunlight during the day). Paul lives in Montana and considers the heat given off by incandescent bulbs a bonus during the long winters.

Until I read his article I can honestly say I didn't realize that CFLs were subsidized. (What isn't these days?) If something requires large subsidies to hide its true cost it's hard to call it sustainable.

The IEEE has an article on bulb lifespans but doesn't give any concrete results. They do point to a worksheet published by the Lighting Research Center at Rensselaer Polytechnic Institute which you can use to do your own analysis. It includes a lamp life multiplier table for fluorescent lamps. Here is the table:

Average hours of lamp                    Lamp life multiplier
operation per start                      for fluorescent lamps
    continuous                                1.8
        12                                    1.5
         6                                    1.2
         3                                    1.0
         2                                    0.9*
         1                                    0.7*
         0.5                                  0.5*
         0.25                                 0.4*

For some compact fluorescent
lamps with "soft start" electronic
ballasts, the values may be higher
than those shown in the chart.

This table by itself does not give a clear picture but it is obvious as you cycle a fluorescent lamp more frequently it's expected life span drops drastically. For lights that you use infrequently and for only short amounts of time (a closet light for instance) the cost of replacing a prematurely failing CFL can vastly outweigh the savings gained from its use compared to an incandescent bulb, which is affected far less by cycling.

Here is an excerpt from research published by this same institute back in 1998: excerpt

Kenneth Cochran

Posted 2013-01-29T21:41:21.937

Reputation: 603

3At 0.104 US$/kWh, 13W CFLs would need to cost more than $50/bulb for $1/bulb incandescent lamps to be less expensive over long periods of time. At those prices and an average of 3 hours of use per day, the CFL is still $1/year cheaper. I did assume that advertised life expectancies were correct, but I think we all know standard 13W CFLs don't cost anywhere near $50 in the USA. – Evan Johnson – 2013-06-20T14:37:35.647

1Even in Montana, the heating benefits of incandescent bulbs are minimal at best, especially if they are mounted in ceiling fixtures. In that configuration, incandescent bulbs probably provide less heat than dedicated electric heaters because the hot air near the bulbs stays near the ceiling and the heat is lost through the top of the home. – Evan Johnson – 2013-06-20T14:48:36.973

2This article from Consumer Reports indicates that incandescent bulbs dim over time, just like fluorescent and LED lamps. – Evan Johnson – 2013-06-20T14:49:54.807

1You also didn't really answer the question. – Evan Johnson – 2013-06-20T14:50:59.687

I think I have now. – Kenneth Cochran – 2013-06-20T21:21:42.683

1From the IEEE article you linked to: "Playing with the default assumptions given in the sheet, we reduced the CFL’s lifetime by 60 percent to account for frequent switching, doubled the initial price to make up for dead bulbs, deleted the assumed labor costs for changing bulbs, and increased the CFL’s wattage to give us a bit more light. The compact fluorescent won." – Evan Johnson – 2013-06-20T22:20:15.677

1That article also mentions that dimmers and recessed fixtures can reduce CFL lifespans. The question seems to be looking for information on how to avoid prematurely killing CFLs, so information on cycling effects and proper installation (see IEEE article's comments on dimmers and recessed fixtures) would be helpful in an answer. The CFL vs incandescent "debate" is not, especially considering how questionable the Paul Wheaton article is as a source. – Evan Johnson – 2013-06-20T22:30:56.273

1@KennethCochran, "If something requires large subsidies to hide its true cost it's hard to call it sustainable." Economics are an independent issue. We can argue about which subsidies are any good (if any), but there is no net effect. The cost to a taxpayer, for example, is income to he/she who receives the subsidy. We can't "run out of money" because we heavily use subsidies. We can, however, run out of fossil fuels, or clean air/water, if we make environmentally unsustainable choices. There is a net effect there. I'd also point out that externalities are a form of subsidy, too. – Nate – 2013-06-20T22:43:31.877

I would like to give this an upvote because I think the table of multipliers is perfect for this question, but I have to downvote simply for the Paul Wheaton material. From his site: "There is also some concern about whether CFLs cause cancer." I'm sorry. Wheaton is a quack. – Nate – 2013-08-03T03:38:58.923

1960, 1978, 198290270-7), 1990 and 2005 studies supporting his claim. If you are going to criticize an answer at least do so based on it's content and not preconceived notions. Wheaton is a respected software developer and his anecdotal evidence is supported by the more rigorous studies I cited. – Kenneth Cochran – 2013-08-05T21:18:18.547