Candles vs Light Bulbs Smackdown

September 11, 2008 9:53 AM

Candles vs fluorescent light bulbs smackdown: which produces more greenhouse gases? Which one is environmentally friendlier when you figure in all the emissions?

I'm in a heated(!) debate about the folks who turned off their lights during the Earth hour but proceeded to light candles instead.

And has anyone studied the health ramifications from the combustion byproducts of candles?

posted by storybored to science & nature (19 comments total)

Just think about it - if heating up an element is an inefficient way to produce light, then just how much more inefficient must be burning wax vapour in the open air? Think how little light in produced by a candle, and how much energy needs to go into creating and transporting it (think how heavy candles are). Candles produce huge amounts of soot, relatively speaking - you can see it when you visit old churches in Europe (that's why they had to clean the Sistine Chapel several years ago).

This is to say nothing of the insufferability of people who think they're doing something for the environment by sitting in candlelight for an hour. Get over yourselves.
posted by Dasein at 10:09 AM on September 11, 2008

Also consider what goes into making a candle versus a bulb. Yes a single candle may be more friendly than a single light bulb, but how many candles would it take to equal the lifetime of a single bulb?
posted by wavering at 10:22 AM on September 11, 2008

Also consider that most candles are made from petroleum.
posted by electroboy at 10:29 AM on September 11, 2008

Also consider the risk of candles lighting a house on fire, which could release many toxins into the air depending on how old the house was. Of course there are house fires caused by shoddy wiring as well.
posted by Meagan at 10:43 AM on September 11, 2008

What Wavering said about how many candles = lifetime of single light bulb, and add in the fact that yer standard 60 watt bulb (about 960 lumens) is equal to 70 candlepower (1 candle is about 13 lumens). So a better comparison would be 1 lightbulb for 30 odd candles (I'm assuming that a candle nowadays is more badassssss than the original definition.)

You can also credit lightbulbs (I do believe) with preventing thousands and thousands of theatre and home fires. Fires were much more ubiquitous when folks had to light their way with open flame. So we're saving the environmental damage of replacement costs too.
posted by Wink Ricketts at 10:52 AM on September 11, 2008

"Just think about it - if heating up an element is an inefficient way to produce light, then just how much more inefficient must be burning wax vapour in the open air?"

Fluorescent lights don't have elements.

"Think how little light in produced by a candle, and how much energy needs to go into creating and transporting it"

Interesting that this applies only to candles but not fluorescent lights in your argument. Any idea of the embodied energy in the electronic products in the control gear of commercial fluorescent lighting, and how far you have to ship them from? What about the mercury? Or the phosphors on the tube coating?

To do all the math involved in your question would require an awful lot of information.

"yer standard 60 watt bulb (about 960 lumens) is equal to 70 candlepower (1 candle is about 13 lumens). So a better comparison would be 1 lightbulb for 30 odd candles"

I'm guessing the people lighting the candles weren't trying to evenly illuminated an open plan room. Consider, for example, an open plan office where fluorescent light is pissed all over the place just in case you need it. A candle is the ultimate task light: massively portable. You need 960 lumens to read from a fluorescent light that's three metres over head. You can read by a candle 30 cms away. Light intensity obeys the inverse square law.

Some of these answers, valuable as they are, highlight the fact that it is very difficult to take into account all factors, and definitive answers to "what's greener" questions almost always fall short. The only way to be sure what you're doing is making an impact is to consume less.
posted by nthdegx at 11:42 AM on September 11, 2008 [1 favorite]

So, here's some ordinary-looking paraffin wax candles (paraffin being by far the most common substance for candles). 3 9-hour candles weigh 5 ounces. Divide, convert to metric, that's about 5g of paraffin burned per hour (assuming, perhaps incorrectly, that all of the paraffin is burned). Paraffin consists primarily of saturated hydrocarbons, C_nH_2n+2, where n is in the 20-40 range. So roughly 12/14 of the mass of the paraffin is carbon; 12g of carbon burned generates 44g CO₂; burning one candle generates about 16g CO₂ per hour. As Wink Ricketts notes, though, you'd have to burn anywhere from 30 to 70 of these candles to generate light equivalent to a 60-watt incandescent bulb. Let's say 500-1100g CO₂ per hour to generate that much light via candle.

A bit of Googling finds that CFL bulbs equivalent to a 60-watt incandescent bulb typically use 13-15 watts. A coal-fired power plant typically generates 0.0946 kg CO₂ per MJ. Running the 15-watt CFL generates 5g CO₂ per hour--roughly equal to burning one single candle--if it's drawing electricty from a coal-powered plant; less from an oil or natural gas plant, and essentially zero for nuclear, wind, etc. Even a conventional incandescent 60W bulb only generates 20g CO₂ per hour.

So even if you're lighting just a single candle in place of a 60W-equivalent CFL, you're not reducing emissions, and people are almost certainly lighting more than one.

Of course, the above only considers CO₂ emissions involved in directly generating the light, and not the amount incurred in producing the candle or bulb, which others have already addressed. Another factor I haven't taken into account here: candles generate a lot of heat for the amount of light they produce, compared to other light sources, even incandescent bulbs. If you're doing this in the summer, you might create additional emissions as your air conditioner works that much harder, but if you do it in the winter, the effects may be somewhat (but not entirely) mitigated because the load on your furnace is reduced.
posted by DevilsAdvocate at 12:00 PM on September 11, 2008 [3 favorites]

I retract "(but not entirely)" in the last sentence in my previous comment. After further thought, it's not clear that that's correct.
posted by DevilsAdvocate at 12:09 PM on September 11, 2008

"Just think about it - if heating up an element is an inefficient way to produce light, then just how much more inefficient must be burning wax vapour in the open air?"

Fluorescent lights don't have elements.

I think that was the point, I read that as:
"Just think about it - if heating up an element [in an incandescent bulb] is an inefficient way to produce light [compared to a flourescent], then just how much more inefficient must be burning wax vapour in the open air?"

To answer the question directly, I think DevilsAdvocate has it.
posted by TungstenChef at 12:46 PM on September 11, 2008

Again, you don't need to produce like for like light output with an overhead light versus a task light.

"A coal-fired power plant typically generates 0.0946 kg CO2 per MJ. Running the 15-watt CFL generates 5g"

This is the most egregious statement in this thread. I've checked your maths, and that's fine, but what you're missing is the distribution losses in electricity which throw your figure of massively. I've seen various estimates, but this source puts the percentage of the energy generated available at the point of use at 31%, so we're looking at over 15g of CO₂ for your 15-watt CFL.

You *can't* do back-of-a-fag-packet calculations for this sort of thing.
posted by nthdegx at 2:51 PM on September 11, 2008

"so we're looking at over 15g of CO2 for your 15-watt CFL."

Sorry, I meant to say that, based on the information in DevilsAdvocate's equation, you're looking at 15g. In no way am I putting that forward as a figure. As DA says it ignores embodied energy, which is as good as saying "this is meaningless, but...".
posted by nthdegx at 2:55 PM on September 11, 2008

I've seen various estimates, but this source puts the percentage of the energy generated available at the point of use at 31%, so we're looking at over 15g of CO2 for your 15-watt CFL.

You are misunderstanding the calculations. Your figure of 31% is overwhelmingly due to the losses in the coal fired generator. DevilsAdvocate already included those generation losses in his calculation. Transmission and distribution losses were not included in DevilsAdvocate's calculation but amount to only about 7%, which does not materially change his conclusion.
posted by JackFlash at 3:38 PM on September 11, 2008

I've seen various estimates, but this source puts the percentage of the energy generated available at the point of use at 31%, so we're looking at over 15g of CO2 for your 15-watt CFL.

Thanks for pointing that out--I realized there was some inefficiency in electricity distribution, but I didn't realize it was that high.

You *can't* do back-of-a-fag-packet calculations for this sort of thing.

Sure you can. I just did. So did you. There's plenty of assumptions involved, each a possible source for inaccuracy, but at least it gives us a handle on the numbers in an order-of-magnitude kind of way.

As DA says it ignores embodied energy, which is as good as saying "this is meaningless, but...".

It's one piece of the answer. It's not the complete answer, which I think my comment made clear, but that doesn't make it "meaningless," unless the CO₂ emissions due to production/shipping is an order of magnitude or more greater than that due directly to the lighting. Which is entirely possible--I really don't have an idea at all how much that would be (although I suspect it's not for the CFL, simply on account of their very long lives). If you can make an argument that that is the case, by all means, do so.
posted by DevilsAdvocate at 3:50 PM on September 11, 2008

Re-reading my earlier comment, this:

So even if you're lighting just a single candle in place of a 60W-equivalent CFL, you're not reducing emissions

is overly precise and not justified, due to the potential inaccuracies in the assumptions. It's unclear whether lighting one candle releases more CO₂ directly than having one 15W CFL bulb burning. I feel confident saying that lighting 10 candles together will release more CO₂ than the CFL.
posted by DevilsAdvocate at 4:01 PM on September 11, 2008

Thanks for pointing that out--I realized there was some inefficiency in electricity distribution, but I didn't realize it was that high.

DevilsAdvocate, you were right the first time. The loss in transmission and distribution is only about 7%. Your figure for 0.0946 kg C02 per MJ is electricity out of the generating plant and includes turbine and generator losses. Your rough estimate only needs to be corrected by a factor of 7% which is not significant in this discussion.

nthdegx's figure of 31% is mostly due to turbine and generator losses which you already including in your original figure.
posted by JackFlash at 5:53 PM on September 11, 2008

You make a good point, Jackflash. I've seen the figure of 7% and I did misread what went into the 31% figure (which isn't the same as misunderstanding the calculation).

"DevilsAdvocate already included those generation losses in his calculation."

This isn't clarified as clearly in the linked source as I'd like, and I'm yet to come across a source that breaks down all losses from the point of generation to the point of use.

If we look at the Dpt of Energy's figures for CO2 output per unit of energy generated they state "The 1999 national average output rate,(4) 1.341 pounds of CO2 per kilowatthour generated". Doing the sums I make that 9g of CO2 emitted to run that 15W CFL for an hour. And that's taking an average of all means of electrical generation in the States - if you liked at coal I imagine that would be much worse.

And we haven't even begun to scrutinise the figures for the candle...

But the broader point I'm making, and DA has backed off a bit from this, is that making fixed conclusions on the basis of quick calculations is very dangerous. It's a fun exercise, and enlightening to a degree, but ultimately we've learned squat about the environmental benefits of candles over CFLs or vice versa. My instinct tells me that a candle using locally sourced materials (if they're available) non-intensively produced would kick a CFL's ass, but I have no way of backing that up.

I'm just concerned that storeybored's haste in marking that as the best answer means he was just looking for numbers to back up his story that he could show to his friends. I'm not saying that's the case, of course. Unless you spend a lot of money and a lot of time on a proper study, the only possible answer is "who knows?". And the whole time the green debate is framed in questions of x vs. y, people are setting themselves up for a possible backlash. I'm the first to point out the folly in people trying to be green actually having a detrimental effect through poor judgement. I'm pretty sure that the folks who turned of their CFLs during earth hour were trying to make a powerful statement, and no doubt do all sorts of other things to reduce the impact of their lives.
posted by nthdegx at 12:09 AM on September 12, 2008

My instinct tells me that a candle using locally sourced materials (if they're available) non-intensively produced would kick a CFL's ass, but I have no way of backing that up.

By locally, do you mean an oil well and petroleum refinery in your backyard? Okay, let's assume that you make your candles in your kitchen.

A CFL lasts about 15,000 hours. At 13 watts it burns up about $20 worth of electricity over its lifetime. The CFL only costs about $2 even after retail markup. So the cost of manufacturing a CFL is only a small fraction of the cost of the fuel burned during its lifetime so you can just ignore the CFL manufacturing cost as a first order estimate. You really only need to compare the carbon burned producing light in the two cases. The CFL wins handily as DevilsAdvocate showed. It's not necessary to make it complicated. A back of envelope estimate is just fine.
posted by JackFlash at 1:08 AM on September 12, 2008

Except that the Dpt of Energy's figures is 100% on top of DA's figure - it depends where you source your data. And that's just *one* aspect of the calculation.

"By locally, do you mean an oil well and petroleum refinery in your backyard? Okay, let's assume that you make your candles in your kitchen."

Who's to say the candles aren't beeswax?

"It's not necessary to make it complicated. A back of envelope estimate is just fine."

I'm not *making* it complicated. It *is* incredibly complicated. You're attempting to simplify it. I have no issue if CFLs are better for the environment than candles: on average, they probably are - in the case of locally made candles, I suspect candles can pip them. But we're nowhere near close to proving it here, though. As DA has said, he's ignoring embodied energy which basically ignores the whole fact that CFLs require incredibly industrially rigorous processes. Not all candles do.

Have you any idea the ridiculous amount of information required to calculate the embodied energy of a complicated product?
posted by nthdegx at 3:25 AM on September 12, 2008

If by "embodied energy" you mean the energy required to manufacture a product it's very simple. The CFL costs $2. That means that the energy required to make the glass tube, mine the metals, refine the fluorescent powder, produce the electronic chips, assemble it all and even including the fuel to ship it half way across the country to your local hardware store is much less than $2 since the company must make a profit. You don't need to carefully calculate each step no matter how complicated the product. Less than $2 of energy to manufacture the CFL is a small number compared to the energy used during its lifetime of use. This quick and dirty method of estimating the energy cost of manufacturing works for most products. It has to be less than the purchase price and assumes the primary energy source is carbon (coal, natural gas, or oil) which is true in the U.S to a first order guess.

So even if your candles are beeswax and somehow manufactured energy-free, they still emit more carbon than a CFL.
posted by JackFlash at 8:56 AM on September 12, 2008