How Many Lightbulbs Does it Take to Change the World?

[mahatmakanejeeves]

One. And You're Looking At It.

Big surprise: making them is not yet an automated process.

For years, compact fluorescent bulbs have promised dramatic energy savings--yet they remain a mere curiosity. That's about to change.

From: Issue 108 | September 2006 | Page 74 |
By: Charles Fishman | Photographs By: Christopher Griffith

Sitting humbly on shelves in stores everywhere is a product, priced at less than $3, that will change the world. Soon. It is a fairly ordinary item that nonetheless cuts to the heart of a half-dozen of the most profound, most urgent problems we face. Energy consumption. Rising gasoline costs and electric bills. Greenhouse-gas emissions. Dependence on coal and foreign oil. Global warming.

The product is the compact fluorescent lightbulb....

Making compact fluorescents is expensive and complicated, compared with incandescents, in part because of the electronic controls each bulb contains, and in part because swirls remain partly handcrafted. To make each spiral, a Chinese worker wearing gloves takes a tube of glass, holds it over an open flame, then wraps the heat-softened tube around a metal form. The job requires a deft touch so the tube doesn't become flattened while getting its spiral shape.
....

At the Wal-Mart home office, they talk about swirls with a zeal that goes beyond product promotion, as if the bulbs are a pioneering product, a new way of thinking about retailing. Says Andrew Ruben, Wal-Mart's vice president of sustainability: "We realize that we can influence big things. Energy usage. Efficiency. Dependence on foreign oil. And we realized that if we're really going to move things, it's not about our direct footprint--our stores, our offices--it's about our supply chain and our customers. So this is about selling lightbulbs, but it's far bigger. This has huge implications for the world."

[NMDemDist2]

1. 3 per household = 3.5 million car's exhaust taken out of the air annually

and Thank Heaven, I didn't have to buy my CFBs at WallyMart

[mahatmakanejeeves]

4. You can buy a six-pack

of 14-watt CFLs at Home Depot for $9.97. The price is good through September 6. This wattage CFL is the equivalent of a 60-watt incandescent bulb.

Look for the price to drop on these things for the next few months as the big box stores slug it out.

[crikkett]

2. re: the big surprise

Big surprise: making them is not yet an automated process.

I bet Wal-Mart has plans to "fix" that. Which wouldn't be a surprise at all, would it?

[LiberalEsto]

3. I use them, BUT

don't they contain mercury?

I would love to see a law requiring the manufacturer of every consumer product in the US to come up with a safe disposal or recycling plan for said product.

[NMDemDist2]

7. yes, but luckily they last 5+ years each n/t

[skids]

8. Not worth the effort.

Yes, they are a point source of mercury pollution and yes people should recycle them. They do, however, actually save on mercury pollution because over their lifetime they save enough in electricity to prevent coal plants from putting at least that much mercury in the air.

However, these products are only going to be popular for about 5 to 10 years. As we speak, LED lights are, in the lab, passing them up as far as efficiency goes. LEDs will end up being even more power efficient, as well as lasting longer and being virtually unbreakable. So the Compact Flourescent light will become obselete fairly quickly and the problem will fix itself.

People using incandescents should not wait for LED bulbs, BTW, they should replace their incandescents with CFs now and they will perhaps have to replace them once, or maybe not at all before they will be able to buy a more power-efficient LED bulb instead.

[LiberalEsto]

10. I hope they make better LED lights

We bought two strands for outdoor Christmas lights last year and the light they give off is hideous. The light seems to vibrate like a dying old-fashioned long fluorescent bulb.

[skids]

12. That's a pretty cheap design...

That means they probably just tied a load-matched string of LEDs direct to house current or a secondary coil. Very cheesy. There are numerous LED controller chips on the market which are cheap and are ready-made solutions to that problem. Eventually manufacturers will stop trying to shave pennies and use them to produce a quality product.

[Xithras]

13. LED's have heat problems, and that may be an unsolvable issue.

Basically, all light bulbs are heaters that happen to emit light. Fluorescents and LED's simply emit more light at lower temperatures. To get an LED to fire, you have to induce an electrical charge across two semiconducting materials. This charge adds energy to the semiconductors, creating a little bit of heat and a little bit of light.

The big problem with high power LED bulbs is that as light output increases, so does heat. Because LED's don't have the glass surface areas to dissipate heat like incandescents and fluorescents, they can get extremely hot as you start scaling up the light output. This is a HUGE problem since LED's begin to lose efficiency once they pass about 85 degrees. Without adequate cooling, this heat will significantly shorten their lives and will cause premature failure of the light. Those lights in the lab that you were referring to ALL have heat-sinks and fans attached to them in order to keep their efficiency up. The power consumption of the fans are not included when they calculate the efficiency of the bulb.

It's probable that the fans will be gone by the time these hit the retail channel, but the heat-sink requirements will still be there (Basic physics...you have to do SOMETHING with the heat). Because of that, many light sockets probably will not accept LED bulbs. In exposed bulb installations, seeing the large heat-sink won't be all that attractive either.

The other problem with LED's is that their lifespan is a bit misleading. Unlike fluorescents and incandescent bulbs, LED's tend to fade out as the semiconductor degrades, rather than burning out all at once. Your LED light bulb may indeed have a ten year lifespan, but it will probably lose 1/3rd to 1/2 of its light output by year 7-8. Many people will replace them long before they fade that far, giving them lifespans no longer than CFL's. While it's possible that different types of semiconducting materials might alter this problem, all of the alternatives researched so far have been FAR more expensive and required uncommon or rare elements.

LED based bulbs WILL become far more common over the next decade, but I doubt they'll ever become our primary light source. Where they will be very useful, and where they will have the most impact, will be in large industrial-type installations. They are likely to replace mercury vapor and other types of high intensity outdoor lights (including the streetlights in front of your house) simply because adding cooling fins to those types of lights will be a non-issue. Since industrial and commercial flood lighting are large consumers of electricity, they will have a large impact in those areas. Cars may also begin to see LED based headlights, since the car companies can engineer a cooling solution straight into the vehicle. But in your house, it's unlikely that you'll be replacing your CFL's with LED's unless you're also tossing your fixtures.

On the other hand, the megacorps have to be worried about sales sustainability. It's possible that they'll release a drop-in LED replacement sans cooling fins just to maintain their market share. Sure, it might run at 150 degrees and burn out in a year, but the 1 or 2 watt power draw would still make it attractive to consumers. So long as the manufacturing costs are cheap enough, there might actually be a sustainable market for that sort of product.

[skids]

14. Heat is no deal killer for LEDs.

As their efficiency goes up, the heat/light ratio becomes more manageable. If that ratio passes compact flourescents, as it has done in the lab, they need remove no more heat per unit light than a compact flourescent does. Hence they can fit in the same profile as a swirl bulb, and even smaller.

Moreover they are much more flexible in application as they need not be used a single high-lumen source but can be strung out -- CFs and incandescents are not economical to manufacture or use in that size. There are varieties of white LEDs which can even change their "warmth" based on PWM of their input power source.

[Xithras]

15. Not exactly.

Remember, the heat/light ratio is only part of the equation. As I recall, the goal of the LED industry is to achieve a production bulb that produces 100-200 lumens per watt. Since a 100watt incandescent bulb generates about 1500 lumens, an LED bulb at the upper end of their goal is still going to require about 7.5 watts to match that incandescent in total light. Remember, we're not talking about 1 watt bulbs here, these have to SCALE. 7.5 times the power means 7.5 times the heat.

The big difference between LED's and fluorescent bulbs in heat is the location of its generation. In a fluorescent bulb, the heat is generated all across the interior surface of the glass as the phosphors get excited. The heat is then radiated out of the bulb through that same surface. In an LED, the heat is generated internally, within the semiconductor itself*. Because the semiconductor is quite small, not to mention buried inside the package, this kind of passive radiation from the heat source will not work. Depending on the thermal efficiency of the LED body, the heat at the semiconductor could be far higher than the heat on the bulbs surface, which robs efficiency. You can make that surface the same size as a CFL, but since your heat is being generated at a single point inside the LED, that radiating surface isn't going to do you much good. To combat this, high output LED's need to extract the heat directly from the semiconductor, just like any other solid state device. Keep in mind also the fact that LED's require LOWER operating temperatures than CFL's in order to maintain their efficiency and lifespan. Even with 100% cooling efficiency, operating an LED light outside on a 90 degree day will shorten its life. This is why most manufacturers only warrant LED traffic lights for 10 years or so (and they are low-output LED's).

Many of the papers I've read discuss coupling LED's with thermoelectric coolers (Peltier effect) to provide an active solution to the problem. Short of mounting fans to the radiators or requiring that all installations remain open to provide adequate airflow, this seems to be the only reasonable solution to the heat problem if you want them to work in current fixtures. The problem is that an active cooling solution is going to require more power, which decreases the overall efficiency of the unit. Your 1500 lumen LED bulb will probably end up consuming about 15 watts by the time all is said and done. While that IS a huge step up from modern incandescents, it's not quite the large leap from CFL's that many people assume.



* Remember that LED's are NOT bulbs, but solid state electronic devices. Efficiency can reduce byproduct heat, but just like the chips in your computer or TV, a certain amount of heat is generated as a simple side effect of electron flow. The flow of electricity through an LED semiconductor likewise produces byproduct heat that is completely unrelated to the generation of light.

[skids]

16. Current products.

My argument was towards your point that a heat sink would have to be abnormally large. This is not true. Yes, you need a conductor to get the heat away from the junction, but the total area of the heat sink need only be as big, or smaller, as the surface area of a CF bulb.

That CF bulbs don't need to be cooled to as low a temperature as LEDs is besides the point because CF bulbs currently operate close enough to the temperature LEDs require that the difference is not drastic. So my base point stands: a heat sink, likely doubling as a reflector or scatterrer in many cases, need only be as big as a CF bulb. LED bulbs of equivalent wattage will fit fine wherever swirls do.

Further, your conjectures don't mesh with reality. Here is a photo of a current LED light product. This uses the LEDs that are not as efficient as CF bulbs.

6W, about what you were stating for a goal for 60W incandescent equivalents, and advertised to "never get hot" despite having a lower efficiency than a CF bulb:



It is not too hard to find LED bulb assemblies in both flood and lamp profile up into the 25 Watt range. Being made with lower-than-state-of-the-art LEDs they dissipate more heat than the newer ones will, and do not seem to need any huge honking heat sink to do it. Moreover they advertise respectable MTBFs and yes, I do realize they dim over time, but they still retain most of their efficiency for longer than the MTBF of a CF bulb, which is all that matters.

I don't know what research articles you're into, but they must be for specialty and industrial applications, because there simply is not a real heat problem to be had in the normal household bulb profiles. None of these require TECs.

[Davis_X_Machina]

5. I am down to two incandescents.

(The CFL's won't work in extreme cold, so the porch light and the one in the mud room are still incandescents)

Saves me about $12-15 a month near as I can figure.

[DemSigns]

11. Mine work in the cold

Although some what dimmer and slower starting. The two "100 watt" CFLs do a good enough job even during Chicagoland winters. I have never seen them not work well enough.

[Marblehead]

6. Most of my bulbs

are CF bulbs now. The next technology will be LED's for home lighting, they use even less electricity.

[hunter]

9. Wal-Mart had to explain why they make them in China, not USA...

Those clever orientals, they are the only ones with hands "deft" enough to make those glass spirals, don't you know...

:puke:

Not, "they work for cheap, and they can't complain."

But I'm sure it beats working in a coal mine, especially a Chinese coal mine, so this is probably good, all things considered... if they are keeping a close eye on the mercury... which is not a sure thing in China.

[eppur_se_muova]

17. One more time ... Banthebulb.org n/t

edited to .org from .com

[hunter]

18. What is that? How many pennies can I collect?

:shrug:

[eppur_se_muova]

19. A campaign to save money and help the environment by using energy efficien

Ban The Bulb: Campaign Aims...

This campaign aims:

1. To increase the use of energy-efficient light bulbs.

2. To encourage the taxing and phasing out of incandescent light bulbs.

3. To propose a time limit for the replacement of light fittings requiring the use
of incandescent light bulbs and for altering the shopping habits of consumers.

4. To include environmental costs in the prices consumers pay for their light
bulbs and to reward those who switch to using less polluting light bulbs.

In 2001, lighting accounted for 101 billion kWh (8.8%) of U.S. household
electricity use. Incandescent lamps, which are commonly found in households,
are highly inefficient sources of light because about 90% of the energy used
is lost as heat. For that reason, lighting has been one focus of efforts to
increase the efficiency of household electricity consumption.

[hunter]

20. There you go... thanks.

Note for the confused -- always buy the dotcom and probably the dotnet too, even if you are strictly an org.

[bhikkhu]

21. count me a skeptic

I have used compact flourescents for years to cut down on heat gain and electric bills. But the notion that they will impact on overall energy use has a serious conceptual flaw.

Take any closed system, having a stable level of resource production and a population utilizing that resource as a part of its livelihood. The population will increase as the resource increases, or the population will cause the increase of the resource to meet its needs. Where the resource can no longer be increased, population increases cease (by whatever means - reality sets in). So you have a limited resource and a limited population. If this population then finds a means to, say, double the efficiency of its use of the resource, then the population can and will quickly double. Then it is back to where we started - a limited resource, a limited population. Efficiency increases solve nothing. Their primary effect is to allow a larger population to rely upon the same resource.

[skids]

22. Last time I looked...

...power blackouts were more likely to result in childbirth than well lit rooms. :evilgrin:

But seriously, efficiency and population growth are two separate issues. Both should be addressed. It does us no good if population stabilizes but we continue to refuse to use more efficient products, just as it does us no good to become more efficient but use the overhead to expand population. It's a false dichotomy to say one or the other is needed.

[bhikkhu]

23. a false dichotomy, depending upon interpretation.

Efficiency per se is only a piece of the equation, which is balanced (as in a dichotomy) as population vs. resources. Those resources necessary for life are the most relevant to this argument, and the size of a potential population related to these resources varies both with the quantities available and the efficiency with which they are used.

The notion of efficiency not being a solution to a constraint upon resources hinges upon the tendency of any animal population to fully utilize the resources available to it, by increases of population. Increasing the efficiency with which a resource necessary for life is used and you effectively change the system so that population can be increased.

Of course, whether humanity behaves as "any animal population" is a different type of argument.