Paul's a great guy, but his article, CFL Fluorescent Light Bulbs: More Hype Than Value
is just totally off base.
I tried to give it a fair shake, though it was hard to imagine how the Department of Energy, the EPA, and every frugal/green blog I'd ever read were so wrong. Well, the facts presented just didn't check out. What follows is a point-by-point rebuttal of Paul's article.
Full disclosure: I make no money off CFLs, other than saving energy
The first point is that CFLs longevity claims are exaggerated. He starts off comparing an 8,000 hour CFL
with Fiet's 25,000 hour, 100 watt incandescent
. He doesn't mention that those long-life bulbs have even worse efficiency than regular incandescents. A typical 100 watt bulb puts out around 1750 lumens. That long-life bulb? A mere 900 lumens
However, the point stands that CFL lifespan is shortened by every power cycle, more-so even than incandescent lamps. Just how big is this drawback? If Paul's numbers are correct, then each flip of the switch robs you of a whopping 0.02¢. Not $0.02, but 0.02¢. So if you sat there for 5 minutes, switching on and off every second, you would waste six whole pennies.
You might think
that rated lifespans are for continuous operation, but this is not the case. They're turned on and off at 30-minute intervals. Paul thinks this is unrealistically long, believing 2-5 minutes is more accurate. We don't actually have data on this (please someone prove me wrong), but we can work out the break-even point
, e.g. the shortest cycle time for which the a CFL costs less. The answer
works out to 3m 15s. This is an average, so leaving the light on for a half-hour makes up for 11 one-minute jots.
Side-note: the article mentions marine incandescents with replaceable filaments (which turned out to be carbon
arc lamps), without realizing that compact fluorescents with replaceable tubes
have existed for years. They cost more due to economies of scale, but they don't present the huge technical challenges that replaceable filaments do. Brightness Exaggeration
Paul cites independent testing by The Telegraph, and says that CFLs are 35% dimmer than the manufacturers claim them to be. This is total bunk.
What the article actually says is that some 11 watt CFLs are labeled as "60 watt equivalent". What he fails to point out is that the words "60 watt equivalent" isn't the brightness
. Brightness is measured in lumens, not watts. People
have gotten sloppy, equating brightness (lumens) with energy consumption (watts). It's as if everyone stopped using miles to mark distance, and instead used how many gallons of gas it took to drive there. That works well enough
, but only until someone makes a more efficient car! Measurement demands precision, and just as there is no 'standard car', there is no 'standard bulb' to define the 'correct' efficiency.
With no official meaning, it's unsurprising that less scrupulous vendors use optimistic numbers. Remember, manufacturers don't write Nutrition Facts labels for their health, they do so because the FTC mandates it.
Speaking of which, where's the FTC in all this? Answer: doing its job
claims of brightness (e.g. "800 lumens") are tightly regulated by the FTC, and required on the package. In fact, they're making a huge step forward by creating the Lighting Facts
label, analogous to the Nutrition Facts label. It puts brightness and energy information in a clear standard format, and its adoption should
straighten out this common misconception.
Side-note: I don't know about you, but I've never seen 11 watt CFLs advertised as 60 watt equivalent, only 13-14 watt bulbs. I'm sure they exist, but it's hardly an industry-wide conspiracy. The "Correction"
Paul proposes a thought experiment: Take 100 incandescent bulbs in one room, and 100 CFLs in the other. How many CFLs have to be added to produce the same brightness?
This experiment starts out badly because it assumes you didn't compare lumens at the store, but there's a deeper flaw. No-one makes lighting choices like that
. The real situation involves a fixed number of sockets, and the decision is which bulbs to fill them with. People don't generally add lights
, they buy brighter bulbs!
The objection is more than academic - it's cheaper to buy a bulb that's twice as bright rather than another bulb. Even if you accept Paul's brightness claims, the right comparison would pit a 60 watt incandescent against a 19 watt CFL (which is how I'll compare them). Starting Performance
This section opens with the following statement: "First, a fluorescent light uses about 20 times more power in the first second to get started. So, for a two minute cycle, the total power consumed is 16% higher." This statement is flat wrong. According to the DoE
, "the relatively higher 'inrush' current required lasts for half a cycle, or 1/120th of a second. The amount of electricity consumed to supply the inrush current is equal to a few seconds or less of normal light operation. Turning off fluorescent lights for more than 5 seconds will save more energy than will be consumed in turning them back on again."
Then there's the overly-pessimistic estimate of 30% for the initial brightness of a CFL. All I can say is, Mr. Wheaton buys crappy CFLs. I've never noticed anything approaching this. The more charitable, and indeed likely, explanation is that Paul was an early adopter who got burned by immature technology. This may also explain his bad experiences with bulb reliability. Cold Weather Performance
No argument there! Motion and light sensors are far more effective outdoor energy-savers. LED lights
will be good here, but there aren't many fixtures yet. Toxicity
Yes, CFLs contain mercury. Paul makes recycling them seem arduous, apparently unaware that Home Depot and IKEA both accept dead CFLs, in addition to most county recycling centers.
The "stressful story" that the article links has been debunked by none other than Snopes
, so I won't waste time rehashing it other than to say that there's no need for a hazmat team. Cleaning up a broken bulb exposes you to less mercury than eating a tuna fish sandwich
Side-note: There are people who don't eat tuna because of mercury! It's a legitimate concern, even before it gets over-blown in the media. And you know what? If someone avoids CFLs because of mercury, I can respect that. They just shouldn't fool themselves that they're saving money too. Subsidies
You might think after reading this section in Paul's paper that the federal government pours money into the laps of CFL manufacturers, and that they'd cost five times as much without them. You'd be wrong.
It's hard to know where to begin with this section actually, because there's just so little basis for it. The New York Times article mentions CFL incentives in California and Vermont (South Africa, India, and China also have programs). So unless you live in California or Vermont, or buy through your utility company, you are paying the unsubsidized price.
Objecting to tax dollars I can see, but bizarrely Paul seems to oppose utility company incentives as well, since the customers ultimately pay. However, "demand destruction" is often the cheapest way of getting "more power"
. I wonder if he has the same objection to building power plants?
Paul then goes off on the federal ban on low-efficiency bulbs, and I completely agree
. The ban is heavy-handed and breeds resentment, not purpose. We should make utilities to clean up their messes like everyone else, and let the market discover the best path. The Table
Just about everything about this table is wrong.
[li]The "unsubsidized price" is bogus.[/li]
[li]The "correction" is bogus.[/li]
[li]The price of electricity is too low. Paul may be blessed to live in a state with cheap electricity, but not all of us are so fortunate. The average price of electricity is about 12¢/kWh (as of December, 2010). This is what I used, and I encourage you to run the numbers with your own price.[/li]
[li]Paul again conflates "watts" with "lumens". The 40 watt bulb puts out 505 lumens. I couldn't find that exact long-life bulb, but this comparable one puts out only 220 lumens. Apples and oranges.[/li]
[li]We arbitrarily switch to 500 lumen bulbs. I'll stick with 850 lumens ("60 watt"), which actually helps the incandescent bulb. I used 13 watt CFLs, and for those who accept Paul's pessimistic numbers I've included 23 watt CFLs as well.[/li]
Stripping out the incorrect speculation and inaccurate "correction" yields a reality-based table:
With the 23 watt bulb we can intuitively see the earlier break-even result - the incandescent light
doesn't cost less until you get to about 3 minutes of "on" time.
We can now rewrite the original article's comically unequivocal conclusion: If you don't pay half the national average for electricity, realize there's no subsidy, have ever eaten a tuna sub, and use your lights for more than three minutes at a time, then CFLs are cheaper. Lighting Habits
Paul goes on to detail his lighting needs. And you know what? He's right that CFLs wouldn't save him much money. However, this is only because his habits make him an outlier: a single 500 lumen bulb for about 4 hours/day. It's obvious that by using less, there is less potential savings. With no lights, there would be no potential savings!
Switching to Paul's lighting habits would save more electricity than switching bulbs, but the majority of the population isn't going to make that change. That's a much bigger commitment than grabbing a different bulb at the supermarket. Maybe Paul has that kind of persuasiveness, but I know I don't! Other Stuff
Dryers, heaters, fridges, and electronics use lots of electricity. So what? It's not an either/or decision.
Some may argue that CFLs are a distraction. However, there's always been low-hanging fruit
you can make that argument about. Catalytic converters, recycling, and efficient windows were last year's distractions, but they've become this year's status quo. Some might call that "progress".
Side-note: Paul's estimated savings from filling the fridge with bottles is way off. Assuming one prevents all
the air from falling out (say 10 cu ft of 40 °F air replaced with 70 °F air), at 6 openings/day you save 0.3 kWh of heat per month, or about half that much electricity. So that $4 of savings is more like 0.8¢. The point is, air is incredibly
thin, and we're incredibly
bad at guesstimating energy savings. The only way to know is to run the numbers. "Free" Heat
the heat isn't free.
[li]If you have anything cheaper per unit heat than electric (read: everything else), the argument falls apart. Air-conditioners invert the argument, for those that have them.[/li]
[li]You don't always want heat where you want light. Of course Paul understands this concept: contrast his brilliant "lizard lamp" hand-warmers with the upward-facing 500 lumen bulb. Now imagine reversing them![/li]
The first video proves that at least one CFL flickers. No-one denies that. The CFL shown must use a magnetic ballast operating at 60 Hz. Decent modern bulbs use electronic ballasts operating at 10,000 to 20,000 Hz. This may annoy a housefly, but it's far above the range humans can perceive.
The problem with the second video is more subtle. The real question is not the time until "full brightness", but the time until "enough brightness". That's what's so dramatic, and so wrong, about this video.
Cameras are very different from our eyes -- it's why most photographers sport light meters. The big difference is in dynamic range
. Our eyes have 90 dB of dynamic range, or a billion times difference in brightness. Cameras aren't so lucky. Photographers adjust F-stops and apertures to do what the eye does automatically. Digital cameras have about 25 dB of dynamic range, and film about twice that.
What does this mean in practice? Imagine we set up a camera so the sensitivity is in the middle of human vision. When the camera displays pure black, your eyes could still see things 1000 times dimmer!
Given these differences, I strongly suspect this video is fibbing with photography. A compelling video would have a light meter in the shot, too. Conclusion
I would oppose compact fluorescent lights too if I thought they were highly toxic, half as bright as advertised, and had a shorter life than incandescents. But I know better, and so should you. It disheartens me to see "CFLs = Bad" taken as gospel in the Permies.com community
, apparently without challenge or discussion. It's time we set the record straight.
Will people continue to cite the CFL article despite its flaws? Of course they will. Hopefully they'll seek out opposing views, and maybe find this article as well.
I know I won't change everyone's mind (least of all Paul's), but I hope we'll have a lively discussion in any event. Please no personal attacks, but responses are of course appreciated.
Thanks for reading. Now go forth and save!
[size=8pt] assumptions are: 60 W incandescent vs. 23 W CFL, .025¢/switch loss, 12¢/kWh electricity price, unswitched bulb cost is 60¢/1000 hour incandescent and $3.47/8000 hour CFL.[/size]